I also have this issue, My setup is an engine dyno with 1:1 gearing so there is even less variables in play. I have recalibrated the loadcell to its full range and it made no diffrance.

@hondarulez,

So you are doing the retardation run to compensate for speed related losses? What do you use for power correction? At coast down, the power losses are much less than under full load, so you need to compensate for both to get an accurate result. The right number to use will vary from car to car. Your numbers still look on the low side though compared with your expectation. 

When testing the MOI, there is a possibility that the engine is stronger on the way down than up in case you have a turbo car, but you may have thought of that already. To remove that possibility, make a test at full power static RPM and see if you get the same number. Other than that, a common mistake is to get the trigger wheel setup wrong (RPM will still be right but HP will be wrong), but I don't expect you have such a basic setup problem. 

@kristovey,

Your setup has much fewer variables. What are your results? You will still need to add compensation for MOI, unless you test at fixed RPM.

You can do a steady state test over a couple of secs at max power. In this test does not need the MOI. The Torque you read when the RPM has steadied at the max power RPM will be exactly the torque the load cell sees. So if it reads correctly in the calibration window, it will read correctly at steady state. Let me know how it goes.

I have the same problem.

On low power cars, upto 250bhp, the numbers are slightly low.

When I tune high power cars (F90 M5 etc) the numbers are sometimes 150-200bhp below expected.

@brynllwynog2, what dyno do you have, what settings do you use on hp correction? Are you following the tips above? 

Yes, I'm following the tips above.  I have a Bosch FLA206.

The coastdown looks to be as expected, just the wheel power seems to be low.

I recently tuned an N55 335i with hybrid turbo, methanol, fuel pump upgrade etc.  It ran 416bhp on our dyno, and 612 on another dyno.

@hondarulez what brakes are these exactly?

I have done an up and down run to then work out the MOI and I get 25% less power than expected.    The engines I build have been measured at 45-50hp for years on various other chassis dynos. I'm only seeing low 30s with the correct MOI.     Load cell calibration is correct and the torque arm length is correct.  I have even got as far as using a laser level to make sure the tourqe arm and load cell are perfectly perpendicular.    My setup is inherently noisy as the rotating mass is so low (~0.14)     do you think the filtering could be knocking off the peaks of raw loadcell data and lowering the average tourqe?

@kristovey, since you have an engine dyno and the readings are so far off we should be able to figure out what is going on.

1. Open the Run window, and with the engine off add a calibration weight and verify that the torque gauge reads correctly. The weight does not need to be on top of the load cell, you just need to know the torque that the load applies to the brake in Nm or LbFt. 
2. Remove the weight. Turn off the MOI correction. Do at run and keep the engine at the max power for a few secs and check the torque gauge. If it is noisy, look at the results afterwards in Results vs Time instead. 

It is exactly the same electronics, software and setup involved so if one is correct then the other should be too in these two tests! Let me know what you find. 

For the noise question, try to re-run a raw log file with the noise filtering turned off and you will see the how the filtered and unfiltered run look. High filtering will smooth out an abruptly changing curve, which can include the actual peaks and troughs. But it is easy to check with the raw log file. 

Posted by: admin

@kristovey, since you have an engine dyno and the readings are so far off we should be able to figure out what is going on.

1. Open the Run window, and with the engine off add a calibration weight and verify that the torque gauge reads correctly. The weight does not need to be on top of the load cell, you just need to know the torque that the load applies to the brake in Nm or LbFt. 
2. Remove the weight. Turn off the MOI correction. Do at run and keep the engine at the max power for a few secs and check the torque gauge. If it is noisy, look at the results afterwards in Results vs Time instead. 

It is exactly the same electronics, software and setup involved so if one is correct then the other should be too in these two tests! Let me know what you find. 

For the noise question, try to re-run a raw log file with the noise filtering turned off and you will see the how the filtered and unfiltered run look. High filtering will smooth out an abruptly changing curve, which can include the actual peaks and troughs. But it is easy to check with the raw log file. 

Thanks you for the suggestions, I will go away and try those. 

@Walter AD72-00, 2K Nm/each

@admin, I will collecting some numbers from different cars in the next few days.

at this moment as we see the problem comes above 100Hp and the diffrerence is increases as the power range increases.

yestersday's test as follows:

Civic 1.4L 90 Hp, full stock except a slightly tune.

we doing the up/down/run tests in gear3 and gear4, the final Tq and HP was same in the diffrerent gears except the MOI values ofcos.

the result was 87 HP/117 Tq without coasting and 95 HP and 123 Tq with coasted data in both gear.

also we do static RPM test at 3,4,5 and 6K RPM. the static Tq values always between the coasted and not coasted Tq values.

as example the 4K static Tq was 111 Nm at the run test with no coast data was 109Nm, and with coasted data was 118Nm @ 4K RPM.

so this result was exactly as we expects.

But measuring stronger cars 150-350 Hp we got around 15-25+% less than we expects. of cos the up/down (MOI) test set correctly.

do you can post dyno sheet of MOI calculed? 

after to check information about your brakes, each one has 1.9, mine has 2.9.

Moi of my hyb dyno is 4.9 each roller (9.8 total) , yours 2.89 each (total 5.78). i dont know if maybe it has to do, i know where are other factors that could make a bit more difference, just difference of inertial per brake is about 34-35%, but MOI is +50% difference than your dyno and mine. i dont know if you can post about how is made your dyno, mine is too simple. just 2 bearings and heavy steel adapter plates.  so maybe MOI is not calculated correctly?

@Walter mine HUBs are simple too probably same as yours :), for the 90 Hp test result  the MOI values was 4 for gear3 and 3.28 for gear4 on each retarder. very little changes on the MOI values causes high HP/Tq differences in the results.

@hondarulez, that MOI change is larger than I would expect. The difference is only how fast the engine + flywheel spins, does that really make that big of a difference in the two gears... Could you do a linearity check on your load cell readings? Try two different weights.

What power %-age correction do you use? 

Posted by: HondaRulez

@Walter mine HUBs are simple too probably same as yours :), for the 90 Hp test result  the MOI values was 4 for gear3 and 3.28 for gear4 on each retarder. very little changes on the MOI values causes high HP/Tq differences in the results.

75339556_1021419068198303_3034709284446273536_n.jpg

What another car do you can test? maybe an ep3? it's just for compare with my dyno, maybe that could help, what drivetrain power numbers do you get with 90hp car? (civic ej9 maybe?)