JohnL

My apologies- it was late when I wrote this, and I agree that the solution is not just in adding the wheel radii. I think my fundamental point is still valid, though- you can't take moments about the tyre-to-tyre interface and multiply the resultant torque by the angular speed to get horsepower, because the system is not in equilibrium at this point- there is work being done on this point and an unknown force difference. The tractive force being exerted by the car is greater than the force being reacted by the truck tyre- if they were the same, there would be no rotation. The "unmeasurable component" [without another instrument] at this point results in a torque that is taken out by the truck brakes (and so not measured in the current load cell). Yes, the load on the truck axle should be the same as the load on the current load cell (this assumes the truck axle was truly pin jointed though).

You do need to take moments about the truck axle as I did (the system is in equilibrium there, and so resolvable), but what I didn't take into account is that there is a resultant torque that is being taken out at this point in the mounting system- caused by the S cam brakes. So I think the truck axle needs a pivot mounting, and a load cell is needed to measure what torque is being reacted at that point.

You have a really neat project Austin

John

AustinT

If I set the axle up the same way that JoBy shows in his picture, would this give me moments around the truck axle? The Pivot is outside of the axle and the axle becomes the swing arm. It would not be hard to have the axle hanging down from the pivot and the loadcell attached at the axle. The loadcell reads both tension and compression so it would not matter which side of the axle I put it on.

I have some time to rework things if I need to. I built the dyno so I can start to learn diy prom. The car now has a carburetor but I am going to fuel injection. Decided to swap the zz430, t56, differential, and suspension parts into another camaro. That car that was on the dyno was completly dissmantled yesterday. The engine, tranny, differential and front clip are out of the car. All of the parts are next to the car, leaving nothing but a shell. I thought about posting a picture stating the car had came loose during a dyno run

JoBy

I still disagree. I don't see any "unmeasurable component".

How can the force not be the same if the car does not move?

The braking tourqe of the brakes does not have to be measured. If you want you can calculate that from from the radius of the brake drum and the horsepower transfered to the axle and the diameter of the truck tires.

You know the force between the tires, it is the same force that is on the load cell. If they were not the same the car would move.

You know the radius of the wheel so you can calculate the tourqe.

You know the rpm of the wheel so you can use the torqe to calculate horsepower.

Tom 400 CFI

I disagree with you too. This phrase:Is the most incorrect. If the forces were/are the same, then you would have no acceleration either posotive or negative....not "no rotation". An impalance of forces results in accelleration either positive or negative. AustinT is using the brakes to balance the forces so there is no accelleration in either direction while a reading is being taken, so while that is happening, there is not an imbalance of forces. You said that the force exerted by the car are great than the force being reacted by the truck... If this were true, then the truck and car wheels would be accellerating. The brakes ensure that they don't.

mikesz

The power the car generates at the rear wheels is dissipated by various friction components of the truck axle ( bearings, brakes, etc) and the friction of the truck and car tires. The sum of these forces is equal and opposite to that measured by the load cell.

This means that for the load cell to accurately measure engine power the resistance of the truck axle generated by these friction components must be equal and opposite to the power generated by the engine. The variable here to control is the trucks brakes. Due to the current inability to accurately control the truck brakes and the absence of a computer control, you can forget about running tests which involve a sweep of engine RPM.

For now my suggestion would be test at a certain single RPM, say 5000. I would ramp up throttle and brake until you can STEDILLY maintain 5000 RPM at WOT. Any fluttering around this number would be like trying to weigh yourself on the bathroom scale while jumping up and down. Once the RPM is stabilized with the brakes at WOT have your measurement system take continuous data as fast as it can. When complete find a large set of data points with a continuous RPM of at least a few seconds or a few thousand points and average them together. This is your number.

If you want to do a RPM sweep you need to be able to steadily and slowly increase RPM over the entire test range while holding the throttle at WOT and only manipulating the truck axle brakes, anything else will give a false reading. The steadier the better, though any fluctuations would average out in the data if they weren’t to bad. The slower the better, if RPM rises to fast the wheels accelerate faster and the rotational mass of those big truck tires comes into play. I would put this data into excel and create what’s called a trend line with a 2nd possibly 3rd degree polynomial. It will figure the equation of a 2nd degree polynomial of the form ax^2 +bx +c that best matches your data and will yield a much prettier graph to look at.

You may want to look at what’s called PID control to operate the brakes. PID control needs an input, setpoint, and output. The input could be an RPM pickup on one of your plug wires. The output would manipulate the truck brakes. The setpoint would be the desired engine RPM. In this case it changes. You would need to simulate an increasing RPM signal. A PID controller would then measure the difference or error between the generated RPM and the actual RPM via the plug wire pickup and automatically manipulate the truck axle brakes to the error or difference between the setpoint RPM and actual RPM is minimal.


mike

wanarace

Haha, I think I mentioned this a month ago.

Keep up the good work
Steve

AustinT

I agree computer control would be nice but I do not need it. I can run sweep tests right now with the setup I have. The rpms of the engine under wot can be changed in 100-200 rpms and held.

The innovate software works great to average out the run. It will average , I think 4 seconds worth of data so the graph is smooth.
My last dyno runs were so smooth I did not even have to average out the data for a good clean run.


Would a cruise control work for this. I thought about getting the audio vox cruise control to control a linear acutator using the dyno wheel speed

Thanks for your input

BJM

Can we see a plot from one of these nice smooth runs? Are you at WOT now?

mclaren18psi

i think that your having an idiot problem in your results i have read through all your guys post on this whole subject and looked at all the photos and it seam the idiot is going to be in some thing you eyes cant see i think you might be lightly flexing you load cell post i know that the angle of the cell is important so any flex will change the geometry or some item in the line of the cell in what manner is the cell atached to the post and can you have a laser line on it like 2 ways you could check it would be in drawling a dead center line on you post and dot it with the laser and use a video to watch any flex on the post or if a laser line not avalible you could go crud and run a taught string a cross on the back side of the post to the point under preload it touches the post but just that so flex could he visible i see that you all have a large amout of brain power working on the problem and i may not have a degree of sorts but i have ran a mustang dyno for a round 6 months with my buddy at his shop and worked with the crew to install and i think your problem is a simple over looked irregularity as it usaly is you calclations all look good so i would look to a hard wear issue i want to aplaud you efforts thus far and say its a good design from the ideal that kicks throught the head on this one also and i want one in my back yard lol


o i had on thought i would check the calibration of the cell agiant a lighter weight and a heavyer one whats the small rear bucket on you backhoe weigh just to see if theres inacurisys ok its time to sleep 2 engines installed to day and up for 35 hours makes me the runner up for wost speller so let me knowwhat you think on all of this and maybe a video of a run that would be cool andy

comp

newbe just found this and need to read all info..
Looks vary interesting

BJM

Is this project working well now? I haven't seen an update in a while.

AustinT

Sorry I have not checked this thread for a few weeks. Its been a couple of months since I did the last run. I swapped the motor, tranny rearend, and all of the drivetrain and suspension parts into a 89 camaro that had less miles. I also went back to school and have not found the time to try another run since the drivetrain swap.
As for the horsepower numbers they are correct. The car is only making 250 hp. I spent some time with a engineering Prof at the university and he verified the numbers. I have not gave up on the dyno. My next step is to go back to fuel injection and start burning some chips. I might try a couple of more runs before I go with fuel injection just for fun, but the dyno will not be real usefull to me until the swap is made.

comp

cool looking for more info

onebinky

iTrader: (1)

Keep us updated on the changes, if you make any. I'd like to do something like this once I get the land for it

AM91Camaro_RS

i just recently found this and was reading through it. that is awesome! i want to build one!

i haven't been able to do much research on load cells yet. and, my question may have been answered somewhere and i didn't notice it but, what type of load cell do you need for something like this? i looked at them on ebay and there seems to be some different kinds. also, some are dirt cheap and some cost a couple hundred dollars, what the difference, just quality?

7Point4

This may have already been mentioned. If so, sorry. But on www.dragnews.com the following is posted...

"10/29/04 - What the?
These images have recently circulated around the internet of some serious gearhead ingenuity regarding the use of a discarded semi-trailer axle and some rather high dollar pieces of data recording stretch gauges and software. All of this compiles into a home-made chassis dyno as shown by the hot rod Camaro set and ready to make a "pull". See more images here. From the use of multiple air line connected to the tractor/trailer brakes, the unit actually appears to apply dyno-type rolling resistance. From the looks of the soil, Drag News detectives venture to say the anonymous dyno is located in the Southeast, but who knows. If anyone has any specific data on the interesting dyno concoction, please let us know."

RBob

iTrader: (1)

As seen on TGO

RBob.

92Z-666

iTrader: (2)

..as ORIGINALLY seen on DIY_EFI circa 1997... {edit: the concept not the text that is} and possibly earlier on the Hotrod list.. and possibly somewhere in Porsche's bloodline and copernicus before that.. then there were the egyptians....

no, really.. "Truck parts dyno" thread on _EFI is where I saw it first... around '97. Purely from memory but thats still very solid

On a side note, I recently saw that alot of "information" sites have perminantly plucked ad re-archived alot of the early stuff from the Hotrod list including stuff from some famous guys that went PRO. thought that was interesting.

I found it while looking for Motronic 1.1 stuff. an original s/w package offered by one of those pros has disappeared from the net. All links are dead.. I don't wanna hijack the thread so if anyone has an old copy of Jim C's initial free package give me a PM! (Historical interest.. I have some decent info already)