Blitz's Dream Dyno

[The Nutty Professor]

Mid Bike if you think this thread is out of line dump it. I thought someons BLITZ might really like the idea. I don't have the tools or know-how to do this but maybe someone else can. It's from a kart site...And here we go.


"How to Build an
Inertia Dyno

Why build a dyno?
We've all run into it before, looking for that extra "edge" in the engine department. However,when you go out looking for information often times what you find from various sources is confusing, and when you put it all together it sometimes contradicts itself.
How do you find out what's right or wrong, or what's right for your application??
It seems nobody has the answers you need, or they aren't telling.
Have you ever wondered....."what happens when you change this??"
Have you ever wondered which of those "25 Go-Fast Tips" in the articles really worked??
It's time to find out for ourselves..........

This is where TDK Motorsports was at a few years ago. The only way to find out what we needed was to find out for ourselves. We needed a dyno........
After looking at the options it seemed that an inertia dyno was the easiest to build (the theory was simple at least) and seemed the most accurate. But here again, the people that knew what we wanted weren't telling.
They were obviously attempting to keep us from building a dyno.

Some of the things we were told......
"That technology is too new, it's only been in this country a couple years."
NONSENSE!!!
"Them things is too complicated, change one little thing and it's all messed up."
Obviously not someone who understands how the dyno they sell commercially works.
"Your gonna need a flywheel as big as a house and 3 computers!"
Hey! We are talking kart engines here, not railroad locomotives.
"You can't use a computer within a mile of a Yamaha engine, it just won't work"
Really!!??
I've been doing it for the last 4 years, works just fine.
We heard it over and over!
What's the big deal?
The concept was simple, accelerate a load, measure the time to accelerate it, and do the calculations. All we needed was a load, and a way to collect the data. In the end it wasn't all that tough. The dyno is built, it works very well and is easy to use. We learn something new with each run, and have found out for ourselves what works and what doesn't, both on the track and on the dyno. The following pages outline all the parts and pieces and some of the things that caused stumbles along the way.
What is an
Inertia Dyno

An inertia dyno, (sometimes called an acceleration dyno) is a system which measures work (Torque/Horsepower) based on the acceleration of an object of a known mass. This "known mass" can be any number of items, assuming you know the weight of the object, the distance it was moved, and the time needed to move the object. For our purposes a wheeled vehicle is most easy to visualize. Any vehicle takes a certain amount of power to accelerate from one given speed to another. If you know the weight of the vehicle you can calculate how much work (torque) it takes to do the acceleration, and if you know the time it takes to do the acceleration you can figure Horsepower.
Simplified formula: 550ft/lbs of work/second = 1 Horsepower
In this case a flywheel spinning on an axle can replace the vehicle or "mass". If you determine the amount of energy required to accelerate the flywheel from one given rpm to another and the time it takes to do the acceleration you can calculate the torque/horsepower produced.
Why an
Inertia Dyno??

Inertia dynos are quickly becoming the preferred method for obtaining the most accurate "real world" results in dyno testing for racing applications. Inertia dynos more closely simulate the dynamic conditions created when accelerating an engine under load, therefore giving more accurate and repeatable results time after time. The following is a brief and general description some more traditional dynos and their advantages and disadvantages...
Traditional dynamometers are typically known as "pump" or "brake" type dynos, whether they use a hydraulic pump or a water pump they all work on the same principle. The engine being tested is run at a steady speed, load is applied via the pump until the engine can no longer maintain this speed at Wide Open Throttle (WOT), at this point the rotational force or "torque" being applied to the pump housing is measured and converted to "standard" engine output readings. This method, known as "steady state testing" is normally performed at 200-500 rpm increments across the "power band" of the engine. This type of dyno testing has been used for many years on all types of engines, however for racing applications it has a few inherent problems.
1. "Pump" type dynos must use some form of fluid to pump, be it water or oil. Everyone knows what happens to oil as it is pumped, it gets hot very quickly and the viscosity or "thickness" of the oil goes down. As this happens the power required to pump it changes dramatically, this will in turn change the power output readings on the dyno. Water has this same tendency, although to a lesser extent. Water and oil pumps also tend to mix air with the fluids, causing them to become aerated, or "foamy", changing it's "thickness" and again changing output readings.
2. "Steady State" testing is fine for equipment such as waterpumps, generators tractors etc, equipment that operates at a steady load and rpm for long periods of time. How often does your racing engine operate at a steady speed? Almost never, from the start of a race to the finish the engine operates in an almost constant state of acceleration or deceleration. The thermal dynamics of intake and exhaust flow, combustion and mechanical components are much different under these conditions than at a steady state.
3."Steady State" testing also requires that an engine be held at WOT at each test increment for a period of time while test readings are taken. The advent of computerized data acquisition systems has helped this considerably, but the engine still spends a relatively long period of time under load at high rpm and WOT.
4. Over time the internal workings of a pump type dyno wear causing it's power absorption characteristics to change. Unless these changes are carefully monitored and compensated for, the dyno can become very inconsistent and inaccurate.
An Inertia Dyno operates much differently than a "pump type" dyno. Inertia Dynos consists of one major component, a large flywheel, mounted on an axle and connected to the engine via a traditional chain and sprocket.

TOO BE CONTINUEDtwocents

[The Nutty Professor]

The test procedure is simple:
1. Start the engine
2.Warm to operating temperature
3. Accelerate from near idle through the power band to max rpm
4. Close the throttle, and apply the brake to slow the flywheel.
During the acceleration of the engine, a computerized data acquisition system is monitoring the speed of both the engine and the dyno flywheel. After shutting off the engine the data collected by the computer is analyzed and processed to produce the appropriate information. The computer "knows" the weight of the flywheel and calculates horsepower and torque values based on the amount of time it took to accelerate the flywheel from start to finish and moment to moment. (up to 30 times/second)
Notice that nothing was mentioned about: hot or cold fluids, load control valves, throttle actuators, load sensors, component wear, Etc....
These items do not exist on an Inertia Dyno, the flywheel is always the same size, nothing changes over time and temperature except the engine. The dyno's output is consistent run to run, day to day, and year to year. Output readings are very consistent and reliable. This entire process takes a relatively short amount of time, after setup, warming the engine etc, typical full throttle run times take from 8-12 sec. The engine is placed under no more stress than a run up the block or a lap around the track for each "run". HUGE amounts of testing can be performed with less "wear and tear" than a night at the races.
Mechanical Components of an
Inertia Dyno


Finally we can begin to talk about the components of the Inertia Dyno
This system can and is entirely safe if do correctly, but there are some things to watch for,
Please read the following carefully........

Flywheel Warning!!!!! Due to the size/weight of the flywheel needed, and the speed that it needs to turn, EXTREME care must be taken in its construction. Failure to do so can and will result in major damage to you, your property and the property of others, personal injury and possibly death!!
This is entirely up to you, if you do not feel comfortable at this point, turn back now. Neither TDKMotorsports nor any of our associates will be held liable for damages caused by the use of the following information.


WARNING WARNING WARNING WARNING WARNING WARNING
Please use some common sense in the construction of a dyno, your life will depend on it!!
(The following is an example of what not to do.)
Recently...
It has come to our attention that someone has built dyno using a flywheel made from railroad car wheels. This in itself is not all wrong, assuming of course they are properly balanced etc.
However, connecting them to an engine in a direct drive configuration and spinning them to 6000 rpm!!!!! is NOT a good design.
Remember the laws of physics still apply and at this RPM there is a very real chance of the flywheel becoming a bomb by exploding due to centrifugal force. Please see the warning below for more information.
Please use a bit of common sense and if you have ANY doubts as to the suitability of your design PLEASE contact us first.
WARNING WARNING WARNING WARNING WARNING WARNING

This is serious business, the flywheel you need is going to be approx. 2 ft+ in diameter, 1" thick and weighs 135lbs+, spinning at 2000 rpm or more. If it should fail in any way, it can and will cause serious damage! This can be done safely, however if you try to use some egg shaped piece of steel, strap it to an axle, and spin it 5000 rpm ... The authorities may never find you or it again! Please use some common sense, the construction of this flywheel is best farmed out to someone who can do it right. (call in all the favors from buddies who may be in, or have contacts in the machinist business) It must be perfectly round and balanced, it must have a center hub securely welded in, and keyed to the axle. Do not attempt to balance the flywheel by drilling holes in it, this only balances it for a given RPM and will also weaken it which increases your chance for disaster. You must build the frame of the dyno to include some integral scatter shield in case for some reason there should be a failure. Remember that you may need to allow for removal of the flywheel for service at a later date. The flywheel is probably the one reason, other than the data acquisition system, that people don't build their own dyno, it can be a little intimidating.
Again, this can be done safely, you just have to be careful.


Here are some of the considerations you need to be aware of:
Size/weight:
We've done some of the work for you here, the wheel that we are using is 24.5" in diameter 1" thick, weight is approx. 135lbs, made from plain old hot-rolled steel plate. We got lucky and came across a couple of steel disks in the scrap yard that were about the right size. We did all the rest ourselves, cut the center hole, welded in the center hubs with 1.25" keyed holes, and then spent many hours trying to make it round and balanced. We wouldn't recommend it to anyone to do, in fact we would pay someone to do the next one, it was far too much work, and we nearly messed it up a couple of times. Not having the proper tools to do this it really wasn't worth the hassle.
Checking some local prices:
Steel: 26" diameter x 1" approx. $85
Weld in center hub and machine round approx. $300
This was quoted from a local shop that is known for being pricey, if you do some checking you should be able to do better, even at these prices, doing it yourself really isn't an option for most people. Again, call in the favors.
WARNING!!!! It is possible to spin a piece of plate steel fast enough to for it to burst. The centrifugal force becomes so high that it exceeds the tensile strength of the steel and it explodes. BOOM!!! You're dead!! Not a nice thought but, this is one of the dangers you should be aware of. We have had the calculations done. We won't tell you what the actual numbers are. However spinning a 30" wheel to 2000rpm, or a 24-25" wheel spinning up to 3000rpm, should allow you all the capacity you'll ever need for 20-25 HP, and still be well within safe limits, again assuming the wheel is perfectly round and balanced! Choosing a large enough wheel to give you acceleration times in the 8-12 sec. range is about right.
1. Safe operation is the responsibility of the operator/builder. TDKMotorsports will not be held liable for any damages caused by the use of this information.

TOO BE CONTINUEDstirthepot

IF YOU GUYS WANT TO SEE THE REST OF THE ARTICLE LET ME KNOW IF NOT I'LL STOP NOW?

[redryderaus]

Keep going, mate. Interesting reading.

Cheers,

red

[Blitz$M.Inc.$]

ya i know a crazy dude that might do it

[The Nutty Professor]

Something told me this was going to go all the way and we're OFF!

More Size/weight:
If you really need to know more than the guidelines given above, here are the relevant formulas...
The formula for determining the torque is:
Torque = PM * rpm per second / 9.551
where "PM" represents the Polar Moment of Inertia of our inertia dyno's flywheel.
If you don't know the Polar moment of Inertia for the flywheel (and your flywheel has a constant thickness cross-section) we can calculate it with the formula:
PM = (W * r^2) / 32.16 / 2
where "W" represents the flywheel weight in pounds and r is its radius in feet.
(the formula for weight of a steel disk can be found in the "FAQ" page)
Once you have the torque, it is easy to calculate the horsepower with the standard formula:
Hp = Torque * rpm / 5252
Keep in mind that the rpm in the last formula must be the average rpm during the sampling period.
Say our example uses a 10 pound flywheel, 8" in diameter (thus it would have a Polar Moment of Inertia of .017 foot-pounds-second^2). If the engine was able to accelerate this flywheel from say 4,800 rpm to 5,200 rpm in 2/10 of a second (a rate of 2,000 rpm per second) that would represent a torque of 3.6 pound feet. Since our above example had an average rpm of 5,000, it produced 3.4 Hp during the test.
Finally some pictures of what we've been talking about......

*********The pictures have already been posted***********

Frame
Here is one place where a picture is worth a thousand words, or absolutely nothing. The pictures will give you a general idea, but the actual configuration is up to you.
Here are a few considerations:
1. Make it sturdy, weld all joints. 2" x 2" x .095 works great. Single cylinder engines produce an amazing amount of vibration, you don't want it to shake apart.
2. Build the frame in such a way that it will allow easy access to the motor.
3. Make everything as close to a regular kart configuration as possible, this makes it easier to take the motor off the kart and mount it on the dyno.
4. Put some rubber castors on the bottom so it can be easily parked in the corner when your not using it, also helps to absorb some of the vibrations.


Axle assembly
Use standard kart pieces here as much as possible, makes it easy to integrate everything together. Otherwise nothing fancy.1.25" kart axle, sprocket hubs, brake rotor, etc. The only exception is using cast iron pillow block bearings, just because they were easy to bolt down to the frame. You may want to get an extra sprocket hub to make it easier to swap between motor types.
The portion of the axle that goes through the flywheel and flywheel bearings is recommended to be solid.
Brakes
Kart type brakes are normally sufficient unless you intend to pull the motor against the brake in an attempt to build heat. If that is the case you may want ventilated or dual rotors. Slowing the flywheel is the only other purpose here. Listening to the flywheel spin itself down gets really irritating! We once let it do this by itself, would you believe it spun for 8 minutes and hadn't stopped yet! Attempting to stop the flywheel too quickly with a large brake can cause the entire dyno to try to tip over. Remember though, all these components contribute to the "inertia weight" which can mess with the accuracy, especially if you change something mid-stream.
One-way clutch
THIS IS VERY IMPORTANT!! After the flywheel/engine has been accelerated and the throttle is closed the flywheel will want to continue to rotate at high rpm and decelerate very slowly (see above), while the engine will attempt to slow much quicker. When this happens the engine becomes the brake for the flywheel. This "engine braking" is very hard on your engine, the vacuum caused by the throttle being closed places the connecting rod and piston in "tension", they are being "pulled" instead of "pushed", exactly the opposite of what they were designed for. Stresses on these parts can be as much as 200-500% higher than normal during this "engine braking" period. Also remember 2-cycle engines are getting no oil when the throttle is closed. A brake on the flywheel is not fast enough. A centrifugal engine clutch doesn't work very well here either, remember the flywheel is doing the driving, all the pieces in the clutch stay locked as long as they are all spinning at the same speed. (until the clutch assembly slows to below stall speed) You can try it without if you want, but it's not a pretty deal. We did alot of searching here, several options were looked at. There are available, some one-way bearings that would work great for this, just like the ones in an electric starter drive. The problem is, by the time you get a bearing that will handle the torque load and abuse of a single cylinder engines alternating power output (one of the most harsh things you can do according to the "power drive" industry, they step up the size of components by about 30%+ if they are used on a single cylinder engine) you get into some expensive pieces. We found one piece that would have worked great, it had an 11/4" keyed hub inside and a "bolt on" flange for sprockets on both sides, really cool! The price was nearly $500! Not an option as far as we were concerned! We've been told that a kart style axle clutch would work here, but haven't gotten hold of one to try. Looking at the pictures, you'll see a small gold colored deal between the sprocket hub and the brake rotor. This is our solution. It's an adapted piece of farm equipment. (Hey, we're farm boys!) It has some limitations, but it was cheap! It needed a small amount of machine work first but works great.

For those who are wondering I have no idea why the smiley keeps showing up. It does on every listing of the word s p r o c k e t?

TOO BE CONTINUED

[The Nutty Professor]

The following is from the end of the article and as I stated before...or at least I think I did I didn't write this I'm just passing it on to everyone in Blitz's name. I think this guy sells certain parts but the article is old and that may no longer be true

Update Sept-2004:
These clutches are very simple but have proven very reliable. Many dyno builders are reporting literally thousands
of runs with no failures.
I can make one of these clutches for anyone who cannot find one or cannot get the machining done themselves.
They are done on a "made to order" basis to fit your specific application however.
Price is $140 including shipping.(Must be pre-paid)

Engine Clutch
An engine clutch is used simply for ease of starting etc. Direct drive was attempted, but just didn't work out real well. The engines were hard to get started because we were trying to "start" the flywheel at the same time, the engine would start and have to accelerate the flywheel while trying to get to an idle. Putting a clutch on works much better. The only thing you have to do is set up a clutch for very low engagement in order to be able to accelerate the motor through the entire power band, before and after the torque peak. A Yamaha clutch set to stall at approx. 6000rpm, and a Briggs clutch at approx.2000rpm are used now and work quite well. Now the engine can be started, warmed up, etc. When doing a "run" the engine is brought up to the clutch rpm, the flywheel is allowed to catch up so everything is locked together, then the throttle is opened form there.
Gauge
The data acquisition system is capable of recording and displaying CHT and EGT "live" depending on the configuration(optional hardware),however if you do not have this capability you'll need something.
We use a Digatron off of a kart to monitor these in real time. You can use one from your kart or used ones can be had a swap meet very reasonable.
That should about do it for the pieces-parts explanation, take a look at the pictures for ideas on throttle controls, fuel tanks, etc.
Sorry the dyno isn't real pretty. Since we got it all working, we've been USING it, and haven't had time to tidy it up much.

Update Sept 2004
Unfortunately (or fortunately, depends on your point of view) my dyno still looks like this today.
I didn't build it real pretty originally because I wasn't really sure it was going to work...........
Now, it works so well I haven't had any reason to take it apart and "pretty" it up. NOTHING has been changed on it since it was built.

Data Acquisition for an
Inertia Dyno

We've got the dyno built now, but how does it actually work??
Now we'll work on collecting data.........

Data Collection
Probably the #1 stumbling block for most people looking to build a dyno of any kind is collecting the data. "Pump" dynos use either a gauge on the output to measure pressure or some method of measuring the "twist" applied to the pump body, seems easy, just write down or remember the gauge reading, but it doesn't usually work out quite right. With an inertia dyno the theory is to measure the time required to accelerate the load (flywheel), less time is more power. Technically you could do this with a stopwatch but the accuracy wouldn't be real good. In the case of our dyno, the time difference between 7HP and 10 HP is only 1 second. Most times I am looking for a .1-.2 gain, how do you do that accurately with a stopwatch? Even if you could, you still wouldn't be able to tell where the peaks in the powerband are located.
It quickly becomes obvious that this is a job for a computer. Well, that's all fine and good but how? Some sort of interface is needed to collect the analog data (engine rpm and flywheel rpm) and convert it into digital data that the computer will understand. Also some sort of software is needed to interpret the data and calculate the desired outputs. The "real" money around here comes from a job as an electronics technician, but even at that, this was going to be a big project. Lots of the calculations were done, some circuitry was even drawn up, but then we realized that the commonly available ports into the back of a PC weren't fast enough to collect the data at the rate we desired. Regardless of processor speed, common I/O ports aren't up to the job. This was going to involve a dedicated I/O board, which is available, but we still didn't have the software. ARRRGH!!!! At this point the dyno frame was built, complete with flywheel, etc, and we had done some running tests, everything was working, but we had no way to measure it. Something had to be done and soon. Many phone calls and internet searches were done without much success, some stuff was found, but either the price was beyond our means, or the software just didn't do what we wanted.
Finally we ran across a webpage for the Datamite Data Logger by Performance Trends. Immediately some of their products were recognized as the same ones sold by places such as Jegs Performance, Summit Racing, and Steve Smith Autosports.
Performance Trends writes some of the most well known and well respected Engine Analyzer and Drag Race Analysis/simulator programs in the industry. They had been known about for years, and now realized they had a data collection system also. Could it work for an inertia dyno? Maybe! After downloading the demo programs / product brochure and looked them over, everything looked like a possibility. The Datamite system is actually intended to be an "on board" data collection system. Sensors are connected for engine rpm, wheel rpm, etc. The data is collected in memory and then downloaded to a computer. What if you "lied" to it and made it think that the dyno was a vehicle? Sounds good, maybe it would work. We knew of at least one other dyno available that was doing the same thing. After looking at the demo for the software used on that system, we had not been impressed. The Datamite program however would supply all the info we wanted and more!
I called Performance Trends and talked to Kevin Gertgen, "can this actually work??"
He was interested in the project, and seemed to think it would work. A Datamite system was ordered.
One note: I have to stop here and thank Kevin Gertgen at Performance Trends. Kevin has been very patient, extremely helpful and great to work with on this project. He has made several "free of charge" updates to the software to make the dyno more accurate and easy to use. If you are interested, call or E-mail Kevin.


Update: May 2000
In addition to all the help Kevin has given so far,we now have the Datamite in a "Dyno Specific" version.
All the messing around with having to "lie" to the software is gone.
The program is VERY comprehensive and easy to use with even better results than before.

The Datamite system:
At this point in order to get a full understanding of the process we recommend you go to the Performance Trends website at www.performancetrends.com Look over the Dyno Datamite section pay attention to the system requirements, and then go to the "downloads" page and get the Datamite Demo Software. The DEMO software includes example files to use, and excellent "on screen" help.

The standard Datamite records up to 4 input channels at once including, engine rpm, flywheel rpm, CHT, EGT, oxygen sensors, pressure transducers, airflow meters etc.
The Dyno Datamite system even has the ability (with options) to automatically collect the weather conditions from a seperate weather station. This information is VERY important for comprehensive dyno testing. the weather numbers can be manually
entered also.
The standard Datamite system (DTM128) includes: recording module (4 channels, 128k memory), control panel with 2 pushbuttons, serial downloading cable, fully wired harness with leads for tach pickup and flywheel sensors, 4 magnets for RPM sensors, Datamite Dyno software, and Excellent user's manual.
Options needed with Datamite:
Inductive pickup for single cylinder engines (DTM-IPU), and 120v AC adapter (DTM-PS)
Approx. cost at this writing: (Feb. 2001)
Datamite system (DTM128) $559
Inductive Pickup (DTM_IPU) $ 45
Ac Adapter (DTM-PS) $ 25
Total $629
This is the base system that will give you pickups for engine and flywheel rpm. Many, many options are available
The only other item needed (besides the dyno itself) is a cheap, cast-off computer system. Bigger is better, but only marginally. Currently we use a Pentium 90 laptop, not lightening fast, but adequate. Any machine capable of running Win95/98 is perfectly adequate. A color printer is a nice addition, it's not neccesary, but with color printers now under $100 why not??


Here are some examples of many of the Datamite Dyno Software screens.......
Main Screen
This is the main screen that you see while running the Datamite Dyno software.
























Dyno Specs Screen
This is the screen that you use to input the size and material specs of all the components on the main dyno shaft.






















Graph Selection Screen
This is the screen that you use to select the type of graph that you wish to create.















Typical Graph.............
Many graphing options are available.Any recorded info can be graphed against time or rpm
Report Selection Screen
This is the screen that you use to select the type of text report that you wish to display,
as well as the parameters of the report.













Typical report.........
Again,many options are available,any recorded info can be displayed,along with averages.
All screens are printable


















Datamite Specs Screen
This screen is used to set up the type of Datamite used and the data recorded on each channel
















Live On-Screen Tach/Temp Displays
Each of the 4 datamite channels can be monitored live on-screen while recording.

















Now you should have all the parts and equipment needed for the dyno project. The following pages will explain how to make it all work together
Using your
Inertia Dyno


Making it all work together...
The Datamite system comes with excellent detailed instructions on how to hook up and run the Datamite system, follow them closely. Since these instructions are include with the Datamite we will not go into great detail about them.
Adapting the Datamite to the dyno...
Magnets and Wheel speed pickups
Only one pickup is required for the flywheel however we are presently using both channels 2 and 3 in order to assure a clean, noise-free signal. After a run is made if you find that the channel 2 data is noisy you can switch to channel 3 from within the software. Only one of the magnets is needed, epoxy it to the side of the flywheel near the outer edge (accuracy increases as you move the magnet farther from the flywheels center) and attach the pickup to the dyno frame according to the Datamite instructions to ensure good signals.
Inductive pickup (IPU)
Connect this to the Datamite harness according to the instructions. Mount the IPU box to the frame of the dyno. Use tie-raps to attach the purple pickup wire to the spark plug wire just like a Digatron pickup.

WARNING----During early testing we experienced a catastrophic failure involving the inductive pickup. We're not really sure what happened exactly but it appears that a spark may have traveled through the pickup wire and damaged the IPU box, the Datamite data recorder module, and the COM port in the computer. To prevent this from happening again, we now follow these guidelines.
1. Do not remove the spark plug wire to kill the motor, the spark will take the easiest path to ground which will be through the Datamite, install a kill switch wire instead.
2. Keep the cut end of the wire away from the spark plug, may want to put some heat-shrink over it also.
3. Unplug the serial download cable when making a run, this will not protect the Datamite, but will protect your computer from errant signals.
We've not had any problems since following these guidelines.
Update FEB. 2001
Performance Trends now has an "optical isolator" that hooks up in series with the serial port download cable, which will prevent any stray voltage spikes from getting into your computer. Well worth the investment.......

Setting up the Datamite for dyno usage:
This section was used to go into a long explanation of how to "lie" to the computer to make it think that the dyno was actually a moving vehicle, However since the "Dyno" version is now available, this procedure is no longer needed.
I will not do any explaining of how the software is actually used as the manuals supplied with the system are VERY good and walk you right through the process with very little effort.

Making a dyno "run"
1. Finally you're ready for the simplest and most exciting part of the whole deal, actually doing a dyno "run". If you've done everything correctly up to this point, all should go well.
There is a bit of technique to this, and you will need to make a few practice runs before you get the "hang" of it.
Start the engine, and warm to operating temperature.
Slowly raise the engine rpm to bring everything up to speed and let the clutch lock up.
Hold the rpm steady, push the "clear memory' button on the Datamite "control panel" this will clear all data in memory and start the recording.
Open the throttle completely and allow the engine to accelerate to the maximum rpm you wish to test
Close the throttle, push the "record" button again to stop the recording. (The record button is used to start and stop recording without clearing the memory, allowing you to record several "runs" before downloading them to the computer)
Apply the brake to stop the flywheel and shut off the motor.
Now you will select "get data" on the computer, the computer downloads the information from the Datamite recording module.
At this point you can process the data to get the actual Hp/Tq, and any other information you would like. All the data from each "run" is stored in the computer, allowing you to extract any of it at a later date if needed. All of this sounds very complicated, however after you are familiar with how to do it, the whole process only takes only a couple minutes to get the "hard numbers" from each "run".

Final Notes/List of Materials/FAQ's
1. Inertia Dyno

Some final thoughts,and some of the most frequently asked questions..........

Over 6 years old now, this project has been a very rewarding/hair raising/hair PULLING/agonizing experience, hours upon hours of calculations, phone calls, and self education have resulted in an inertia dyno that works very well, is easy to use, and has taught us a lot .
Was it all worth it?
YES!!! In every way.
Why are we sharing it?
Originally this project was intended for our own use, however after getting it up and running it worked so well that we thought that it was only right that we share it with everyone,hence the birth of this webpage, our "contribution" to the sport of kart racing.
There have now been over 75000 visitors to this website,and the number of dyno's built is well past 200 across the country and around the world.
Very little approaches the satifaction that I get when I recieve an E-mail or phone call from someone who has made their first
successful dyno runs. As of yet I have not heard from anyone who has been dis-satisfied,I only hear from those that are thoroughly enthused and excited. Many users claim very good successes in both their engine building and racing as a result of building this project. Nothing in this price range can compare.
HAVE FUN!!!.......
If you have any questions/answers/contributions/complaints please E-mail us and let us know, we are here to help.


FAQ's
We have received a multitude of questions about our inertia dyno. Most of these questions involve the use of different types of flywheels, brakes, gearing, clutches, etc. Keep in mind that there are many different/better methods for doing something, however these are the methods we used. Many of the parts we used were selected based on what we had on hand or were readily available, most of which were kart parts. Assuming that most of us have some extras lying around, why not use them? Here are some of the questions we have received and some answers to them:
Just how accurate is this dyno?
"Accuracy", for many reasons, some of them mentioned in this website, is a very relative term as far as dyno's go. Just how accurate is anyone's dyno? Take 2 motors, test them on 2 different dynos, one comes out at 8hp, the other comes out to 9hp, but they run the same lap times. Which dyno is "accurate"? Which one is giving you the correct numbers? Horsepower is simply a calculated number. Dyno 10 engines on the same dyno at 8HP, put them all on the same kart one at a time, and they all run the exact same lap times. Is the dyno accurate? No, the dyno is however "repeatable". "Repeatability" is what you want in a dyno, test the same engine time after time and get the same results. Take one of the 10 engines from above, change a pipe or a cam so you get 5% more power, run it against all the others again. It runs 5% faster, you redyno it and it still says 5% better, now you have an "accurate" and more importantly, a "repeatable" dyno. If you follow the advice in these pages you will have a system that is every bit as accurate as you will ever need. A couple hints: Always start your tests at the same engine temps and settings. Pay close attention to weather conditions. Get yourself a good barometer, thermometer, and humidity gauge, and enter the conditions into the program each time you make a run.
The Datamite only records at 60hz (60 readings per second). Shouldn't it be more than that??
No,not really,for a couple reasons.......
1. The faster you collect data,the faster your hardware(computer) has to be,and the more likely you are to have problems.
The hardware needed to collect data at 500-1000 samples per second would quickly send the price into the $15-20,000
range,and would actually be more prone to electrical noise and breakdowns.
The Datamite system is very immune to electrical noise,and very rugged,especially when you compare the price.
2. Other systems may actually collect more data,but very little of it ever actually makes it to the screen as a graph.
Many of the software programs do not actually look at all the data that is collected,and what is used has been through
so much averaging and filtering that you never see most of it anyway.
Remember,in a dyno situation you rarely look at individual data points. What you are looking for is the shape of the curve,which is the average of all the data taken anyway. Having more data rarely does alot of good here.
Plain and simple,the Datamite is hard to beat.
What size flywheel should I use?
This one place where bigger is not neccesarily better. Problem #1 is getting a wheel made. As you go up in diameter,the availability of equipment to machine the wheel gets smaller and harder to find.There are alot of lathes around that will handle a 24" wheel,but very few that will handle a 30" wheel.
The wheel on our dyno is 24.5" in diameter which is more than big enough to handle most any kart engine with the possible exception of some of the "open" motors. If you plan on running these motors go with a bigger wheel,probably in the area of 26-28 inches.
What gear ratio should I use between the engine and axle??
The best advice here is to use the gearing that gets you closest to the amount of "run" time you want. Your looking for total acceleration times in the area of 8-12 seconds.
A 4 to 1 gear ratio works very well for most Briggs motors with a 24" flywheel,and a 6.5 to 1 works well for most 2-strokes using the same wheel. If you go to a larger wheel you will put more load on the motor,which means you have to g to a higher gear ratio. (5-6 to 1) this will slow the wheel down and shorten the run times.
Can an automotive/diesel truck/tractor or other type of flywheel be used?
Yes, you could use most anything for a flywheel, however there are a few items that need to be taken into consideration...
The flywheel must be heavy enough and spun fast enough to produce acceleration times long enough to collect reasonable data, yet not so long as to induce unneeded stress on the engine. Acceleration times in the 8-12 second range seem to work very well. Changing the gearing to spin the wheel faster or slower will raise or lower the load on the engine and influence the acceleration time. Diameter is actually much more important than the weight of the wheel. With a larger diameter the weight is farther from the center of the circle, making it a larger "lever", which will be harder to turn. Take a look again at the basic formula for inertia of a flat disc:
"I" = Inertia
I=1/2MR^2 Where: I=inertia M=weight of the wheel in pounds divided by 32.2(the constant for gravity) Weight= volume of wheel in cubic inches (radius squared x pi x thickness) x .2833(weight of steel/cubic inch) R^2= radius squared
In this formula, the diameter is the multiplier and holds the biggest deciding factor. If you pick a couple of numbers and run through this formula, you will see that changes in diameter add up quickly. Example: Our wheel measures 24.5" in diameter x 1" thick, approx. weight = 133lbs Running through the formula you'll find that "I" equals approx.310
Now, try a wheel 18" in diameter... With a little backwards math, you'll find that in order to get the same weight in lbs. the wheel will have to be approx. 1-7/8" thick. If you can handle this then run the numbers through the formula again and you will find that the "I" of this wheel is only 168! What this means is that you'll have to spin the wheel nearly twice as fast to achieve the same loading on the engine.
Try the other direction... 30" wheel 1/2" thick Approx. weight only 100lbs Approx. "I"= 349. Now you can actually spin the wheel slower with the same results.
Once you start thinking about this, something like a flywheel from a car quickly becomes less attractive for a couple reasons...
1. Weight/diameter-most car flywheels are only 12-15" in diameter and approx. 1" thick. Most weigh less than 50lbs. You'll have to spin these really fast to get much of a load on the engine, this requires a major gearing change, which means you'll have to go out and buy several sprockets/chains etc instead of using a larger wheel which allows you to use standard kart parts.
2. Be careful with automotive flywheels (and some others for that matter), some of them are not "neutral" balanced. Many manufacturers use flywheel/balancer combinations that are counter-weighted to help balance the internal assemblies. A Chevy 400 is one example, most Fords also are "externally" balanced. Ford had 2 systems, one used 50oz.of weight, the other used 28.5. Spinning up a flywheel with 2-3 lbs. of counterweight could be really exciting!
3. Most automotive/truck/tractor flywheels are made of cast iron, assuming you'll be rummaging through someone's junk pile to find a flywheel, be careful, there is usually a reason something is in the junk pile. You don't know for sure what is wrong with it. Having seen the damage caused by an exploding cast iron flywheel, I can tell you that, you don't want it to happen to you!
How do I calculate the weight of the flywheel??
The formula looks/works like this:
Pi x R^2 x thickness x weight of steel/cubic inch
Or
Pi(3.14) x wheel radius inches squared = area of round wheel in square inches
x thickness of the wheel in inches = volume in cubic inches
x .2833(weight of steel(lbs) per cubic inch) = weight in pounds
What kind of hub do you use to attach the flywheel to the axle?
The flywheel hubs are also a piece of farm equipment. They are designed to have a sprocket welded to them to create a gear for use on farm equipment. Just weld them to the flywheel. Here is a picture of them as purchased.

















These hubs are manufactured by G&G Manufacturing as part of their "Weldasprocket"
line. Website here: http//www.ggmfg.com/products/subcategories.asp?cat=weldahubs
You are looking for their "XX" series hubs which are the largest outside diameter.
Most bearing supply houses and farm-fleet type stores carry G&G products
1. Can an automotive caliper be used directly on the flywheel for the brake?
Again, yes you can. Theoretically. But consider these points:
1. A flywheel weighing 130lbs spinning at 1500-2000 rpm is a lot of stored rotating energy. Our flywheel will spin for approx. 10 minutes if left to slow down on it's own. Considering the fact that the dyno frame/engine combination probably weighs less than 200lbs, you don't want to stop this kind of stored energy too quickly. It is entirely possible that if you clamped the caliper down on the wheel too quickly the entire frame could rotate around the wheel and land on your toes!
2. Some automotive calipers are not the easiest things to mount to a frame. Most have some goofy mounting system, plus most need to be made to "float" so that both brake pads will clamp evenly. Considering the ease with which a standard kart caliper mounts up, they are much easier to deal with.
3. Cost of a complete Paul Martin brake system including caliper, master cylinder, brake rotor, rotor hub and lines retails for $150-175, brand new. You could easily find all these components at a swap meet for far less than this, and still have matching components. By the time you round up a car caliper and a suitable master cylinder, then figure out how to mount and plumb it, the kart components become much more attractive.
What is the one-way clutch that is being used?
As we said, the one-way clutch is actually an adapted farm equipment piece, commonly called an over-running PTO clutch. They have been used for years on older farm tractors without "live" power take off systems. The clutch was used with "high inertia" implements such as brush mowers or balers that continued to spin after the power was removed, which without the over-running clutch would continue to drive the tractor forward even with the engine clutch disengaged, quite an exciting and dangerous ride if you are not prepared for it! These clutches should be available at most farm-fleet or bearing supply houses for approx.$60-80. Basically they look like a male-female coupler with a large ratchet inside, most have a six spline male end approx. 1-1/8" outside diameter and a corresponding same size female end. What you will need to do is to machine the splines out of the female end, large enough to slip-fit over the axle being used, and machine the male end to fit the inside of the kart axle. The only real tricky part here is that the splines inside the female end are VERY hard and a bit difficult to machine, however most any reasonably equipped machine shop should be able to make quick work of it. The clutch is designed to be "roll" pinned to a shaft which is what we did and it seems to work fine. Here are a couple pictures to show what it looks like after it has been modified for dyno use.










Are there any plans/drawings/parts lists available?
1. For a couple years I promised to have drwaings and plans available for this project, however the more
I thought about it the more I had reservations. The mechanical portions of this project are relatively
simple for most anyone that has any basic mechanical knowledge.
Quite frankly, if a person cannot figure out the majority of it on their own I have to question their
ability to do a safe job of building, and also have to wonder just how much use they will get out of it.
That said, if you have a question about something that you just cannot figure out please feel free
to contact me.
Can something other than the Performance Trends Dtamite software be used to collect the data?
Yes, you can use whatever data aquisition system you can come up with.
Remember that your final results are highly dependant on your abilities to manipulate the data.
Also, if have any intentions of marketing your engines or your dyno services, you need to be able to present
your customers with clear and concise data tha they can understand, and in a format that if nothing else looks
impressive.
A quote on the virtues of the Datamite system from George Clausen of Clausen Racing Engines, Bettendorf Iowa:
"I have used/owned many of the data aquisition systems on the market, comparing most of them to the
Performance Trends Datamite system is like comparing Pong to PlayStation"
Says it all....................
Is the inertia dyno something you can drive/ride on?
NO
Has anyone built one of these dynos from the information provided on this site?
Yes! Go here to see what others have done with the information on this site. (Several pictures please be patient)
So what if I don't want to build a dyno,can I get my engine dyno'd???





Why Dyno?
1. By
Craig S. Walker
"Those dynamos just ain't repeatable. You can only be accurate to a couple
horsepower with 'em." Or maybe you've heard, " I like to use the seat of he pants dyno. I know when I 've made a significant change." It is hard to argue with comments like the above. Not that they are correct, but that they usually come from folks who just can not be argued with. Raising your voice won't change their minds. What will change their minds is when you pull past them on every straight stretch of road or power past them coming out of the tightest hairpin at you favorite track.
Some people, even the brightest engineers mind you, do not believe in dyno testing engines. That's a fact. At this point we'll leave behind the nay sayers and move on to the positive reasons for dyno testing. Don't worry about proving the dyno point to someone who doesn't get it. Just move on, knowing in your own mind you and tens of thousands of others have seen the light. Dynamometers are tools. They are good for collecting data. Not every type of dyno is perfect for every application, but as with any tool, when the right one is used the job becomes easier. Use of a dynamometer does not guarantee an engine at its optimum state of tune, nor does having or using one guarantee that you will be able to tune out any anomalies that arrive in your power curve. Some engines just by their nature (due to cam timing, port size and pipe length) may have a severe flat spot in the mid range that no amount of carburetor jetting or ignition timing can eliminate. Using a dyno however, can reveal important information about what an engine is doing and display the engine's out put in a meaningful format. It is the interpretation of that data that is important. In many cases you don't even have to very experienced with a dynamometer chart to make good tuning choices.
When I first experienced dyno tuning one of my street bikes on a chassis dyno the operator had no good input for me based on the chart my bike generated. Instead he just told me that a bike with similar modifications made 10 more horsepower peak. Out of the kindness of his heart he offered to fix the problem with his cylinder head port work for several hundred dollars. When queried about flow specs or power numbers he and his partner (They were brothers in a small speed shop) offered me no specific figures but insisted I would "feel the difference." I turned them down and went home where I installed larger main jets in the carbs and retarded the ignition timing. I returned a week later to post respectable numbers more in line with the "mystery engine" the operator spoke of at my earlier session. Simply amazing when you consider I didn't need to use his magical port work! Fortunately I wasn't fool enough to fall prey to his porting scam. I later booked dyno time with my nitrous oxide system primed for testing. I had made some modifications to an off the shelf kit and was eager to see the results. Almost on que the operator told me my nozzles were not in the right location and that I could hardly be getting any reasonable performance from the system. Again out of the kindness of his heart he offered to fix my bike with a special set up they use (For a nominal fee of course). I realized how little experience they had with nitrous when the operator need an explanation of the operation of the system and later stopped in the middle of a run when he needed more coaching on the use of the system.
Once I trained the “Nitrous Expert” the test session began. The bike jumped forward on the dyno from the nearly 70 horsepower hit and let out a bellowing sound that attracted everyone in the shop. "That's not a standard nitrous set-up. How did you make that work?" the brothers asked. I declined to comment and instead mumbled something about maybe coming back later for their superior methods and modifications.
I switched shops and decided to bring my machine to another location for future tuning. This shop specialized in a completely different type of motorcycle and therefore the staff was very up front and honest about "not being able to offer any tuning advice," but instead said I was more than welcome to tune and tinker all by myself. Even with my minimal chart interpretation skills I could still be successful by making a change and looking to see if the change helped or hurt. I fumbled along using my old 1950's methods of checking plug color and exhaust pipe residue to get a reasonable idea of which way to go with jetting. I checked for aluminum specs on the plug insulators for signs of ignition timing too advanced. Regardless of whether I was right or wrong in my guesstimated processes, the dyno told the story. Higher power was good. Lower power was bad.
Playing with cam timing showed why the seat of the pants was not all it's cracked up to be. Moving lobe centers around based on the local track hero's suggestion yielded a bike that felt like a monster. It would gladly loft the front wheel in huge wheelie fashion whether I wanted it to or not. All I had to do was attempt to accelerate. Surly this was the hot set-up. The dyno said NOT. I was down 3.5 HP at peak. The cause for the unwanted unicycle action was the huge valley in the torque curve between 4500 and 5500 rpm. The other side of the valley met with a virtually straight spike up to where the torque should have been all along. Result…..big wheelies, slow bike. Adding some overlap gave a smoother torque curve with more peak power, even if it was less exciting to ride. It was at this point that I truly banished the thought of "Seat Dyno" testing ever again.
A friend and owner of a Pro-Twins AMA race bike/team, had spent considerable time and money tuning on a Superflo 901 water brake dyno. I was further convinced of the viability of testing when his bike adopted silencers. The AMA began putting noise regulations (read limits) on the bikes. At first it was easy to pass the sound tests since they only required a static test with the engine revved to 50% of redline. Since the professional machines do not run standard instruments, a non-factory tachometer was employed to keep watch of engine rpm. This non-factory tach had a simple piece of red tape attached to the face at an appropriate red-line rpm. Simple, yet it also allowed for convenient red-line changes, say if a different cam were installed, if the bike were to run an endurance race rather than a sprint race or if an artificially lower redline were required to pass the sound test. Not that I've ever seen this done of course, but if we had to…….
When the bikes owner decided it just might be easier to fabricate some cans for the exhaust, he figured if they didn't cut power too much he would just run them all the time. During a dyno test session a set of crudely cobbled together canisters were mounted right up on the back-side of the reverse cone megaphones. Next run was up 4 hp peak with no losses elsewhere. Would we have known that without the dyno? Certainly not! We never expected it. We were prepared for a loss, the dyno knew better.
Why on earth would anyone forsake these tools? Some say they are not accurate. Accuracy has little to do with it. I could care less if the read outs were in torque, horsepower or pumpkins-per-fortnight. All one absolutely requires is repeatability. Repeatability, or lack thereof, lies mostly in the user. Engines themselves need to reach a stabilizing temperature before their output becomes repeatable run to run. Also chassis dynos put their own variability into the equation when we start to deal with tire side wall flex and heating, tire tread slippage on drive roller(s), drive line variability (o-ringed and grease filled chains operate in an efficiency zone based on temperature as do belt drives), etc. The repeatability of a test can be verified by performing another run identical to the one previous. Often a machine may need to be run through several dyno "pulls" before all these variables stabilize. These will appear as runs of increasing value. Runs should be performed back to back until at least two to three identical pulls appear. It is at this point that the tuning changes can commence and they nay sayers theories about repeatability are put to rest.
Given the availability of an unbiased tool why would one choose to do without? The physician doesn't guess our weight like some clown at the town fair, instead we step on a scale. When we drive our cars we believe the speedometer and the tachometer (although most shouldn't). We don't let the utility company estimate our electricity and water usage. We even believe our oven temperature is regulated correctly based on a dot on the plastic dial. You wouldn't wave your hand in that same oven and say, "that seems good for turkey, I'll just watch the sun in the sky to estimate when it will be done." Why guess at engine performance? Many trust quite faithfully the gauges of everyday life yet there are many motor sports enthusiasts who can't seem to believe that a mechanism could be made to reliably measure the twisting force of an engine.
It is hard for me to imagine doing any sort of tuning now with out access to some sort of dynamometer. Data acquisition is so low cost now, that anyone with a basic PC can pop on an external Data logger capable of 250 samples per second for less than $100.00. Acquisition cards internal to PCs can run from $200.00 up, but allow much faster sample rates. I recently acquired a card for less than $50 at a computer show. The card is considered out dated because it "only" runs at 4000 samples per second. Top of the line dynamometers from 10 years ago didn't have that capability and most today still don't. This type of power in a personal computer can allow the average racer to gain a technological advantage over his competition. Home-made dynos are a more viable option than they once were with the advent of low cost rapid accelerating data acquisition technology.
This site already assumes you don't need convincing of the viability of testing. It will focus on the types of dynamometers from the old and simple to the new and complex to the new and simple yet again. The information is provided to give an understanding of how dynos work and how one can use or perhaps construct a dyno of their own. Dynos have been around since the construction of the first engine. I would venture to guess they will be around as long as engines have a place in our society. Understanding what makes them tick can make them an asset to any true motorsports enthusiast.

Dyno Specs Screen
This is the screen that you use to input the size and material specs of all the components on the main dyno shaft.

Graph Selection Screen
This is the screen that you use to select the type of graph that you wish to create.

[The Nutty Professor]

Typical Graph.............
Many graphing options are available.Any recorded info can be graphed against time or rpm

Report Selection Screen
This is the screen that you use to select the type of text report that you wish to display,
as well as the parameters of the report.

Typical report.........
Again,many options are available,any recorded info can be displayed,along with averages.
All screens are printable

Datamite Specs Screen
This screen is used to set up the type of Datamite used and the data recorded on each channel

Live On-Screen Tach/Temp Displays
Each of the 4 datamite channels can be monitored live on-screen while recording.

Now you should have all the parts and equipment needed for the dyno project. The following pages will explain how to make it all work together
Using your
Inertia Dyno


Making it all work together...
The Datamite system comes with excellent detailed instructions on how to hook up and run the Datamite system, follow them closely. Since these instructions are include with the Datamite we will not go into great detail about them.
Adapting the Datamite to the dyno...
Magnets and Wheel speed pickups
Only one pickup is required for the flywheel however we are presently using both channels 2 and 3 in order to assure a clean, noise-free signal. After a run is made if you find that the channel 2 data is noisy you can switch to channel 3 from within the software. Only one of the magnets is needed, epoxy it to the side of the flywheel near the outer edge (accuracy increases as you move the magnet farther from the flywheels center) and attach the pickup to the dyno frame according to the Datamite instructions to ensure good signals.
Inductive pickup (IPU)
Connect this to the Datamite harness according to the instructions. Mount the IPU box to the frame of the dyno. Use tie-raps to attach the purple pickup wire to the spark plug wire just like a Digatron pickup.

WARNING----During early testing we experienced a catastrophic failure involving the inductive pickup. We're not really sure what happened exactly but it appears that a spark may have traveled through the pickup wire and damaged the IPU box, the Datamite data recorder module, and the COM port in the computer. To prevent this from happening again, we now follow these guidelines.
1. Do not remove the spark plug wire to kill the motor, the spark will take the easiest path to ground which will be through the Datamite, install a kill switch wire instead.
2. Keep the cut end of the wire away from the spark plug, may want to put some heat-shrink over it also.
3. Unplug the serial download cable when making a run, this will not protect the Datamite, but will protect your computer from errant signals.
We've not had any problems since following these guidelines.
Update FEB. 2001
Performance Trends now has an "optical isolator" that hooks up in series with the serial port download cable, which will prevent any stray voltage spikes from getting into your computer. Well worth the investment.......

Setting up the Datamite for dyno usage:
This section was used to go into a long explanation of how to "lie" to the computer to make it think that the dyno was actually a moving vehicle, However since the "Dyno" version is now available, this procedure is no longer needed.
I will not do any explaining of how the software is actually used as the manuals supplied with the system are VERY good and walk you right through the process with very little effort.

Making a dyno "run"
1. Finally you're ready for the simplest and most exciting part of the whole deal, actually doing a dyno "run". If you've done everything correctly up to this point, all should go well.
There is a bit of technique to this, and you will need to make a few practice runs before you get the "hang" of it.
Start the engine, and warm to operating temperature.
Slowly raise the engine rpm to bring everything up to speed and let the clutch lock up.
Hold the rpm steady, push the "clear memory' button on the Datamite "control panel" this will clear all data in memory and start the recording.
Open the throttle completely and allow the engine to accelerate to the maximum rpm you wish to test
Close the throttle, push the "record" button again to stop the recording. (The record button is used to start and stop recording without clearing the memory, allowing you to record several "runs" before downloading them to the computer)
Apply the brake to stop the flywheel and shut off the motor.
Now you will select "get data" on the computer, the computer downloads the information from the Datamite recording module.
At this point you can process the data to get the actual Hp/Tq, and any other information you would like. All the data from each "run" is stored in the computer, allowing you to extract any of it at a later date if needed. All of this sounds very complicated, however after you are familiar with how to do it, the whole process only takes only a couple minutes to get the "hard numbers" from each "run".

Final Notes/List of Materials/FAQ's
1. Inertia Dyno

Some final thoughts,and some of the most frequently asked questions..........

Over 6 years old now, this project has been a very rewarding/hair raising/hair PULLING/agonizing experience, hours upon hours of calculations, phone calls, and self education have resulted in an inertia dyno that works very well, is easy to use, and has taught us a lot .
Was it all worth it?
YES!!! In every way.
Why are we sharing it?
Originally this project was intended for our own use, however after getting it up and running it worked so well that we thought that it was only right that we share it with everyone,hence the birth of this webpage, our "contribution" to the sport of kart racing.
There have now been over 75000 visitors to this website,and the number of dyno's built is well past 200 across the country and around the world.
Very little approaches the satifaction that I get when I recieve an E-mail or phone call from someone who has made their first
successful dyno runs. As of yet I have not heard from anyone who has been dis-satisfied,I only hear from those that are thoroughly enthused and excited. Many users claim very good successes in both their engine building and racing as a result of building this project. Nothing in this price range can compare.
HAVE FUN!!!.......
If you have any questions/answers/contributions/complaints please E-mail us and let us know, we are here to help.


FAQ's
We have received a multitude of questions about our inertia dyno. Most of these questions involve the use of different types of flywheels, brakes, gearing, clutches, etc. Keep in mind that there are many different/better methods for doing something, however these are the methods we used. Many of the parts we used were selected based on what we had on hand or were readily available, most of which were kart parts. Assuming that most of us have some extras lying around, why not use them? Here are some of the questions we have received and some answers to them:
Just how accurate is this dyno?
"Accuracy", for many reasons, some of them mentioned in this website, is a very relative term as far as dyno's go. Just how accurate is anyone's dyno? Take 2 motors, test them on 2 different dynos, one comes out at 8hp, the other comes out to 9hp, but they run the same lap times. Which dyno is "accurate"? Which one is giving you the correct numbers? Horsepower is simply a calculated number. Dyno 10 engines on the same dyno at 8HP, put them all on the same kart one at a time, and they all run the exact same lap times. Is the dyno accurate? No, the dyno is however "repeatable". "Repeatability" is what you want in a dyno, test the same engine time after time and get the same results. Take one of the 10 engines from above, change a pipe or a cam so you get 5% more power, run it against all the others again. It runs 5% faster, you redyno it and it still says 5% better, now you have an "accurate" and more importantly, a "repeatable" dyno. If you follow the advice in these pages you will have a system that is every bit as accurate as you will ever need. A couple hints: Always start your tests at the same engine temps and settings. Pay close attention to weather conditions. Get yourself a good barometer, thermometer, and humidity gauge, and enter the conditions into the program each time you make a run.
The Datamite only records at 60hz (60 readings per second). Shouldn't it be more than that??
No,not really,for a couple reasons.......
1. The faster you collect data,the faster your hardware(computer) has to be,and the more likely you are to have problems.
The hardware needed to collect data at 500-1000 samples per second would quickly send the price into the $15-20,000
range,and would actually be more prone to electrical noise and breakdowns.
The Datamite system is very immune to electrical noise,and very rugged,especially when you compare the price.
2. Other systems may actually collect more data,but very little of it ever actually makes it to the screen as a graph.
Many of the software programs do not actually look at all the data that is collected,and what is used has been through
so much averaging and filtering that you never see most of it anyway.
Remember,in a dyno situation you rarely look at individual data points. What you are looking for is the shape of the curve,which is the average of all the data taken anyway. Having more data rarely does alot of good here.
Plain and simple,the Datamite is hard to beat.
What size flywheel should I use?
This one place where bigger is not neccesarily better. Problem #1 is getting a wheel made. As you go up in diameter,the availability of equipment to machine the wheel gets smaller and harder to find.There are alot of lathes around that will handle a 24" wheel,but very few that will handle a 30" wheel.
The wheel on our dyno is 24.5" in diameter which is more than big enough to handle most any kart engine with the possible exception of some of the "open" motors. If you plan on running these motors go with a bigger wheel,probably in the area of 26-28 inches.
What gear ratio should I use between the engine and axle??
The best advice here is to use the gearing that gets you closest to the amount of "run" time you want. Your looking for total acceleration times in the area of 8-12 seconds.
A 4 to 1 gear ratio works very well for most Briggs motors with a 24" flywheel,and a 6.5 to 1 works well for most 2-strokes using the same wheel. If you go to a larger wheel you will put more load on the motor,which means you have to g to a higher gear ratio. (5-6 to 1) this will slow the wheel down and shorten the run times.
Can an automotive/diesel truck/tractor or other type of flywheel be used?
Yes, you could use most anything for a flywheel, however there are a few items that need to be taken into consideration...
The flywheel must be heavy enough and spun fast enough to produce acceleration times long enough to collect reasonable data, yet not so long as to induce unneeded stress on the engine. Acceleration times in the 8-12 second range seem to work very well. Changing the gearing to spin the wheel faster or slower will raise or lower the load on the engine and influence the acceleration time. Diameter is actually much more important than the weight of the wheel. With a larger diameter the weight is farther from the center of the circle, making it a larger "lever", which will be harder to turn. Take a look again at the basic formula for inertia of a flat disc:
"I" = Inertia
I=1/2MR^2 Where: I=inertia M=weight of the wheel in pounds divided by 32.2(the constant for gravity) Weight= volume of wheel in cubic inches (radius squared x pi x thickness) x .2833(weight of steel/cubic inch) R^2= radius squared
In this formula, the diameter is the multiplier and holds the biggest deciding factor. If you pick a couple of numbers and run through this formula, you will see that changes in diameter add up quickly. Example: Our wheel measures 24.5" in diameter x 1" thick, approx. weight = 133lbs Running through the formula you'll find that "I" equals approx.310
Now, try a wheel 18" in diameter... With a little backwards math, you'll find that in order to get the same weight in lbs. the wheel will have to be approx. 1-7/8" thick. If you can handle this then run the numbers through the formula again and you will find that the "I" of this wheel is only 168! What this means is that you'll have to spin the wheel nearly twice as fast to achieve the same loading on the engine.
Try the other direction... 30" wheel 1/2" thick Approx. weight only 100lbs Approx. "I"= 349. Now you can actually spin the wheel slower with the same results.
Once you start thinking about this, something like a flywheel from a car quickly becomes less attractive for a couple reasons...
1. Weight/diameter-most car flywheels are only 12-15" in diameter and approx. 1" thick. Most weigh less than 50lbs. You'll have to spin these really fast to get much of a load on the engine, this requires a major gearing change, which means you'll have to go out and buy several sprockets/chains etc instead of using a larger wheel which allows you to use standard kart parts.
2. Be careful with automotive flywheels (and some others for that matter), some of them are not "neutral" balanced. Many manufacturers use flywheel/balancer combinations that are counter-weighted to help balance the internal assemblies. A Chevy 400 is one example, most Fords also are "externally" balanced. Ford had 2 systems, one used 50oz.of weight, the other used 28.5. Spinning up a flywheel with 2-3 lbs. of counterweight could be really exciting!
3. Most automotive/truck/tractor flywheels are made of cast iron, assuming you'll be rummaging through someone's junk pile to find a flywheel, be careful, there is usually a reason something is in the junk pile. You don't know for sure what is wrong with it. Having seen the damage caused by an exploding cast iron flywheel, I can tell you that, you don't want it to happen to you!
How do I calculate the weight of the flywheel??
The formula looks/works like this:
Pi x R^2 x thickness x weight of steel/cubic inch
Or
Pi(3.14) x wheel radius inches squared = area of round wheel in square inches
x thickness of the wheel in inches = volume in cubic inches
x .2833(weight of steel(lbs) per cubic inch) = weight in pounds
What kind of hub do you use to attach the flywheel to the axle?
The flywheel hubs are also a piece of farm equipment. They are designed to have a sprocket welded to them to create a gear for use on farm equipment. Just weld them to the flywheel. Here is a picture of them as purchased.

Guys I've lost track of the photo's so some may be I've posted them twice. I also realize you have to match the photo's with the paragraphs. I tried to post just the article but it was to large

TOO BE CONTINUED

[The Nutty Professor]

:cool::cool:These hubs are manufactured by G&G Manufacturing as part of their "Weldasprocket"
line. Website here: http//www.ggmfg.com/products/subcategories.asp?cat=weldahubs
You are looking for their "XX" series hubs which are the largest outside diameter.
Most bearing supply houses and farm-fleet type stores carry G&G products
1. Can an automotive caliper be used directly on the flywheel for the brake?
Again, yes you can. Theoretically. But consider these points:
1. A flywheel weighing 130lbs spinning at 1500-2000 rpm is a lot of stored rotating energy. Our flywheel will spin for approx. 10 minutes if left to slow down on it's own. Considering the fact that the dyno frame/engine combination probably weighs less than 200lbs, you don't want to stop this kind of stored energy too quickly. It is entirely possible that if you clamped the caliper down on the wheel too quickly the entire frame could rotate around the wheel and land on your toes!
2. Some automotive calipers are not the easiest things to mount to a frame. Most have some goofy mounting system, plus most need to be made to "float" so that both brake pads will clamp evenly. Considering the ease with which a standard kart caliper mounts up, they are much easier to deal with.
3. Cost of a complete Paul Martin brake system including caliper, master cylinder, brake rotor, rotor hub and lines retails for $150-175, brand new. You could easily find all these components at a swap meet for far less than this, and still have matching components. By the time you round up a car caliper and a suitable master cylinder, then figure out how to mount and plumb it, the kart components become much more attractive.
What is the one-way clutch that is being used?
As we said, the one-way clutch is actually an adapted farm equipment piece, commonly called an over-running PTO clutch. They have been used for years on older farm tractors without "live" power take off systems. The clutch was used with "high inertia" implements such as brush mowers or balers that continued to spin after the power was removed, which without the over-running clutch would continue to drive the tractor forward even with the engine clutch disengaged, quite an exciting and dangerous ride if you are not prepared for it! These clutches should be available at most farm-fleet or bearing supply houses for approx.$60-80. Basically they look like a male-female coupler with a large ratchet inside, most have a six spline male end approx. 1-1/8" outside diameter and a corresponding same size female end. What you will need to do is to machine the splines out of the female end, large enough to slip-fit over the axle being used, and machine the male end to fit the inside of the kart axle. The only real tricky part here is that the splines inside the female end are VERY hard and a bit difficult to machine, however most any reasonably equipped machine shop should be able to make quick work of it. The clutch is designed to be "roll" pinned to a shaft which is what we did and it seems to work fine. Here are a couple pictures to show what it looks like after it has been modified for dyno use.

Are there any plans/drawings/parts lists available?
1. For a couple years I promised to have drwaings and plans available for this project, however the more
I thought about it the more I had reservations. The mechanical portions of this project are relatively
simple for most anyone that has any basic mechanical knowledge.
Quite frankly, if a person cannot figure out the majority of it on their own I have to question their
ability to do a safe job of building, and also have to wonder just how much use they will get out of it.
That said, if you have a question about something that you just cannot figure out please feel free
to contact me.
Can something other than the Performance Trends Dtamite software be used to collect the data?
Yes, you can use whatever data aquisition system you can come up with.
Remember that your final results are highly dependant on your abilities to manipulate the data.
Also, if have any intentions of marketing your engines or your dyno services, you need to be able to present
your customers with clear and concise data tha they can understand, and in a format that if nothing else looks
impressive.
A quote on the virtues of the Datamite system from George Clausen of Clausen Racing Engines, Bettendorf Iowa:
"I have used/owned many of the data aquisition systems on the market, comparing most of them to the
Performance Trends Datamite system is like comparing Pong to PlayStation"
Says it all....................
Is the inertia dyno something you can drive/ride on?
NO
Has anyone built one of these dynos from the information provided on this site?
Yes! Go here to see what others have done with the information on this site. (Several pictures please be patient)
So what if I don't want to build a dyno,can I get my engine dyno'd???





Why Dyno?
1. By
Craig S. Walker
"Those dynamos just ain't repeatable. You can only be accurate to a couple
horsepower with 'em." Or maybe you've heard, " I like to use the seat of he pants dyno. I know when I 've made a significant change." It is hard to argue with comments like the above. Not that they are correct, but that they usually come from folks who just can not be argued with. Raising your voice won't change their minds. What will change their minds is when you pull past them on every straight stretch of road or power past them coming out of the tightest hairpin at you favorite track.
Some people, even the brightest engineers mind you, do not believe in dyno testing engines. That's a fact. At this point we'll leave behind the nay sayers and move on to the positive reasons for dyno testing. Don't worry about proving the dyno point to someone who doesn't get it. Just move on, knowing in your own mind you and tens of thousands of others have seen the light. Dynamometers are tools. They are good for collecting data. Not every type of dyno is perfect for every application, but as with any tool, when the right one is used the job becomes easier. Use of a dynamometer does not guarantee an engine at its optimum state of tune, nor does having or using one guarantee that you will be able to tune out any anomalies that arrive in your power curve. Some engines just by their nature (due to cam timing, port size and pipe length) may have a severe flat spot in the mid range that no amount of carburetor jetting or ignition timing can eliminate. Using a dyno however, can reveal important information about what an engine is doing and display the engine's out put in a meaningful format. It is the interpretation of that data that is important. In many cases you don't even have to very experienced with a dynamometer chart to make good tuning choices.
When I first experienced dyno tuning one of my street bikes on a chassis dyno the operator had no good input for me based on the chart my bike generated. Instead he just told me that a bike with similar modifications made 10 more horsepower peak. Out of the kindness of his heart he offered to fix the problem with his cylinder head port work for several hundred dollars. When queried about flow specs or power numbers he and his partner (They were brothers in a small speed shop) offered me no specific figures but insisted I would "feel the difference." I turned them down and went home where I installed larger main jets in the carbs and retarded the ignition timing. I returned a week later to post respectable numbers more in line with the "mystery engine" the operator spoke of at my earlier session. Simply amazing when you consider I didn't need to use his magical port work! Fortunately I wasn't fool enough to fall prey to his porting scam. I later booked dyno time with my nitrous oxide system primed for testing. I had made some modifications to an off the shelf kit and was eager to see the results. Almost on que the operator told me my nozzles were not in the right location and that I could hardly be getting any reasonable performance from the system. Again out of the kindness of his heart he offered to fix my bike with a special set up they use (For a nominal fee of course). I realized how little experience they had with nitrous when the operator need an explanation of the operation of the system and later stopped in the middle of a run when he needed more coaching on the use of the system.
Once I trained the “Nitrous Expert” the test session began. The bike jumped forward on the dyno from the nearly 70 horsepower hit and let out a bellowing sound that attracted everyone in the shop. "That's not a standard nitrous set-up. How did you make that work?" the brothers asked. I declined to comment and instead mumbled something about maybe coming back later for their superior methods and modifications.
I switched shops and decided to bring my machine to another location for future tuning. This shop specialized in a completely different type of motorcycle and therefore the staff was very up front and honest about "not being able to offer any tuning advice," but instead said I was more than welcome to tune and tinker all by myself. Even with my minimal chart interpretation skills I could still be successful by making a change and looking to see if the change helped or hurt. I fumbled along using my old 1950's methods of checking plug color and exhaust pipe residue to get a reasonable idea of which way to go with jetting. I checked for aluminum specs on the plug insulators for signs of ignition timing too advanced. Regardless of whether I was right or wrong in my guesstimated processes, the dyno told the story. Higher power was good. Lower power was bad.
Playing with cam timing showed why the seat of the pants was not all it's cracked up to be. Moving lobe centers around based on the local track hero's suggestion yielded a bike that felt like a monster. It would gladly loft the front wheel in huge wheelie fashion whether I wanted it to or not. All I had to do was attempt to accelerate. Surly this was the hot set-up. The dyno said NOT. I was down 3.5 HP at peak. The cause for the unwanted unicycle action was the huge valley in the torque curve between 4500 and 5500 rpm. The other side of the valley met with a virtually straight spike up to where the torque should have been all along. Result…..big wheelies, slow bike. Adding some overlap gave a smoother torque curve with more peak power, even if it was less exciting to ride. It was at this point that I truly banished the thought of "Seat Dyno" testing ever again.
A friend and owner of a Pro-Twins AMA race bike/team, had spent considerable time and money tuning on a Superflo 901 water brake dyno. I was further convinced of the viability of testing when his bike adopted silencers. The AMA began putting noise regulations (read limits) on the bikes. At first it was easy to pass the sound tests since they only required a static test with the engine revved to 50% of redline. Since the professional machines do not run standard instruments, a non-factory tachometer was employed to keep watch of engine rpm. This non-factory tach had a simple piece of red tape attached to the face at an appropriate red-line rpm. Simple, yet it also allowed for convenient red-line changes, say if a different cam were installed, if the bike were to run an endurance race rather than a sprint race or if an artificially lower redline were required to pass the sound test. Not that I've ever seen this done of course, but if we had to…….
When the bikes owner decided it just might be easier to fabricate some cans for the exhaust, he figured if they didn't cut power too much he would just run them all the time. During a dyno test session a set of crudely cobbled together canisters were mounted right up on the back-side of the reverse cone megaphones. Next run was up 4 hp peak with no losses elsewhere. Would we have known that without the dyno? Certainly not! We never expected it. We were prepared for a loss, the dyno knew better.
Why on earth would anyone forsake these tools? Some say they are not accurate. Accuracy has little to do with it. I could care less if the read outs were in torque, horsepower or pumpkins-per-fortnight. All one absolutely requires is repeatability. Repeatability, or lack thereof, lies mostly in the user. Engines themselves need to reach a stabilizing temperature before their output becomes repeatable run to run. Also chassis dynos put their own variability into the equation when we start to deal with tire side wall flex and heating, tire tread slippage on drive roller(s), drive line variability (o-ringed and grease filled chains operate in an efficiency zone based on temperature as do belt drives), etc. The repeatability of a test can be verified by performing another run identical to the one previous. Often a machine may need to be run through several dyno "pulls" before all these variables stabilize. These will appear as runs of increasing value. Runs should be performed back to back until at least two to three identical pulls appear. It is at this point that the tuning changes can commence and they nay sayers theories about repeatability are put to rest.
Given the availability of an unbiased tool why would one choose to do without? The physician doesn't guess our weight like some clown at the town fair, instead we step on a scale. When we drive our cars we believe the speedometer and the tachometer (although most shouldn't). We don't let the utility company estimate our electricity and water usage. We even believe our oven temperature is regulated correctly based on a dot on the plastic dial. You wouldn't wave your hand in that same oven and say, "that seems good for turkey, I'll just watch the sun in the sky to estimate when it will be done." Why guess at engine performance? Many trust quite faithfully the gauges of everyday life yet there are many motor sports enthusiasts who can't seem to believe that a mechanism could be made to reliably measure the twisting force of an engine.
It is hard for me to imagine doing any sort of tuning now with out access to some sort of dynamometer. Data acquisition is so low cost now, that anyone with a basic PC can pop on an external Data logger capable of 250 samples per second for less than $100.00. Acquisition cards internal to PCs can run from $200.00 up, but allow much faster sample rates. I recently acquired a card for less than $50 at a computer show. The card is considered out dated because it "only" runs at 4000 samples per second. Top of the line dynamometers from 10 years ago didn't have that capability and most today still don't. This type of power in a personal computer can allow the average racer to gain a technological advantage over his competition. Home-made dynos are a more viable option than they once were with the advent of low cost rapid accelerating data acquisition technology.
This site already assumes you don't need convincing of the viability of testing. It will focus on the types of dynamometers from the old and simple to the new and complex to the new and simple yet again. The information is provided to give an understanding of how dynos work and how one can use or perhaps construct a dyno of their own. Dynos have been around since the construction of the first engine. I would venture to guess they will be around as long as engines have a place in our society. Understanding what makes them tick can make them an asset to any true motorsports enthusiast.

And that marks the end of the article. I knew when I found this I was not the person to build it. Maybe someone out there is. The Nutty Professor knows his limitationscheers.

[swheels]

Man that thing is cool.But if i had the skills to build that.I'd be blowing motors left and right.I'd be more interested in squeazing all i can get out of it.

[Blitz$M.Inc.$]

ya thats a big project but he gives some good insight to it
how can you diss a dyno?
my bud has a integra he turboed himself and he tunes it with his laptop in the car

[The Nutty Professor]

So you guys like stuff like that? I see what else is floating out there.

[Blitz$M.Inc.$]

ok glen but let my brain recover from that book lol
i didnt know if i was gonna finish it but i did
thankfully some of it is a double post and i skimmed over it :P

[The Nutty Professor]

ok glen but let my brain recover from that book lol
i didnt know if i was gonna finish it but i did
thankfully some of it is a double post and i skimmed over it :P


Told you I was losing track because of the size. If I ever do something that big again I think I'll plan it out better so I don't screw it updunno. Glad you guys liked it and don't worry I'll wait before I do it again.

[The Nutty Professor]

Here's a dyno video from the kart guys.

http://www.youtube.com/profile?user=JimAbbett

chief_rocka

[Blitz$M.Inc.$]

that one was open, cool
id like to see the thing spin too
BWWWAAAAAAAAHHHHHHHHHHH

[rene13]

Hey guys I found this I figured this might help.

It's a dyno kit for testing small motors. Here is the link http://207.234.198.78/small_dynamometer_kit_to__test.htm

[redryderaus]

Nice find, rene.

Unfortunately a 350W dyno won't be much good on motors that start at 6100W and go up from there. Plus those prices are highway robbery.

However, a DC generator is another way of putting a load on our engines so we can measure output power. :)

Cheers,

red

[The Nutty Professor]

Nice find, rene.

Unfortunately a 350W dyno won't be much good on motors that start at 6100W and go up from there. Plus those prices are highway robbery.

However, a DC generator is another way of putting a load on our engines so we can measure output power. :)

Cheers,

red

Red I know this is real simple but...there's always a but. Could you use a DC generator as a primitive dyno. This is what I'm getting at. Use a DC gen attached to a test rig. A s p r o c k e t on the gen arm matching the rear s p r o c k e t you are currently running. The output of the gen would be measured at each throttle opening or gradual opening. If you could record those reading isn't there a formula to calculate the horsepower generated? In otherwords converting watts to HP? That would be a very simple dyno. How accurate it would be I have no idea? Of course the gen would have to be able to generate power above what your motor could produce and the gauge would have to read that high also. The one drawback might be the extra safety needed because of the electrical power that results from the test run.twocents

[The Nutty Professor]

I'm trying to stick to one subject (Tires) but I can't help myself:

http://www.ls1tech.com/forums/showthread.php?t=150901

http://www.eng-tips.com/viewthread.cfm?qid=84432

When it's time I'm building a dyno. With what I've read it's not hard to build a DC gen dyno and they are as accurate as the builder wants them to be.All I want are ballpark figures so that should do. I want to know if the change I just made was good or bad and by approximately how much?

[redryderaus]

Red I know this is real simple but...there's always a but. Could you use a DC generator as a primitive dyno. This is what I'm getting at. Use a DC gen attached to a test rig. A s p r o c k e t on the gen arm matching the rear s p r o c k e t you are currently running. The output of the gen would be measured at each throttle opening or gradual opening. If you could record those reading isn't there a formula to calculate the horsepower generated? In otherwords converting watts to HP? That would be a very simple dyno. How accurate it would be I have no idea? Of course the gen would have to be able to generate power above what your motor could produce and the gauge would have to read that high also. The one drawback might be the extra safety needed because of the electrical power that results from the test run.twocents

Simple answer is YES!

You need several things:

1/ A DC generator capable of generating more power than your engine

2/ A load for the generator. Something like a large heating element in a barrel of water.

3/ A way to calibrate your generator and load. Any DC generator has an efficiency rating. Typically for a permanent magnet generator this would be in the region of 60-70%. This would mean if your engine was producing 10kW of power, the generator would put 6-7kW into the load.

Measuring is then simply a matter of the voltage across the load x the current flowing through it. Divide that by the efficiency and you have the output power from your engine.

You are right when you say ballpark figures. What we really want from a dyno is the ability to measure change.

Cheers,

red

[The Nutty Professor]

Yeah everyone is caught up in 1/2 horsepower or a quarter when measureable change in the right direction is basically what you need to know. Yes precision is a plus but fast and loose is a beginning.

[#84]

cool nice write up. rep +1