I am often asked for power numbers on specific combinations of products. A lot of people refer to forum posts and frankly, I don’t have much confidence in what is seen there. Some may have done formal testing, but it seems that a lot of the numbers are wild guesses pulled out of the air like fishing tales. Other considerations are drivetrain loss which will be unique to the vehicle if performing a complex swap, environmental factors such as altitude, and if tested on a loaded or unloaded dyno. If it’s a loaded dyno, what’s the load?
For those that are new to BenzForce or have not heard of the dyno car, it is simply a bare-bones test vehicle to do product demos and testing on. If you are interested, the car has its own page with its build details listed: https://www.benzforce.com/pages/dyno-car. An engine dyno would be ideal for this type of testing, but I have not found one within a decent radius and I can only imagine the cost of testing would be considerably higher. Having an engine dyno would eliminate a lot of variables and give the best results but it simply isn’t in the cards; thus the dyno car!
For our journey into data collection, we were very fortunate to have found Torqqe Performance. They will be our dyno partner through the testing of the om617 and hopefully beyond.
Shortly after posting the results on YouTube or Facebook, I received a message saying the performance should have been higher. For what? How many [heavy] BMWs are out there to compare dyno results for a stock om617, AX15 manual, with a 3.23 gear ratio at my altitude? If so, was the load setting on their dyno the same as the one we used? This is where controlled testing is important. Anything on the car that could affect performance is to remain static throughout all the tests related to a specific engine. Regardless of all the variables, only one input, such as injection pump, turbo, etc will be changed at a time. This will give an accurate percentage change for that upgrade.
Power is only one aspect though. Other factors are very important such as intake manifold temperature, boost, etc. Two cars with the same boost level may perform differently if one is running an intercooler. The intercooler will lower temperatures and increase air density. This is where the CAN data comes into play. The dyno car setup will vary from yours, but you will at least see what intake temperatures, boost, etc. the power levels were achieved at.
How the Data was Collated
The primary data was exported from the AEM system at 1000HZ. That means the data snapshot used 1000 measurements per second. Keep in mind that I can have different sensors sampled at different rates. It makes no sense to have water temp sampled at a rate like that. Regardless of the sensor, this was the frequency used for export. After export (and import into Excel), I removed duplicate lines. Because of the massive quantity of data, I did not think all intervals were worthwhile. In fact, I think there is still an excessive amount of data. With the de-duplicated data, I matched the AEM RPM to the horsepower and torque readings from the Dynojet output.
Most dyno results I have seen are horsepower and torque in relation to RPM but that is not how a data logger sees it – its time-based. The AEM can calculate estimated horsepower and torque but the dyno should be more representative. The Dynojet can also show data relative to time but that is not normally how users consume dyno data results.
The Road Ahead
- Improved testing: One feature that both the Dynojet and AEM have in common is OBD. I need to investigate the option to integrate my system into the dyno at runtime. The Dynojet software will allow me to then plot all the channels I want totally in sync. Unfortunately, I am not convinced that the data analysis and export capabilities of the Dynojet software are on par with the AEMData software but it would give an exact match on RPM and make syncing easier.
- Next test: I have a pump in the queue at Bosch getting the 7.5mm upgrade so a fuel upgrade will be the first upgrade to test. In addition to timing, I am not sure whether I want to adjust it or leave it at the bench settings. I will need to be sure to document whichever I choose to do.
The Results
Most people will look at the horsepower and torque below and discard everything else. The results are corrected power ratings for consistency across future runs.
If you want more numbers, you can download the data at http://content.benzforce.com/om617/om617-baseline-results.xlsx.
@ Max HP 78.34:
- Boost: 8.8 PSI
- Manifold Air Temp: 226 F
- EGT: 1225 F
@ Max Torque 124.26:
- Boost: 8.5 PSI
- Manifold Air Temp: 182 F
- EGT: 1001 F
@ Max Boost 9 PSI:
- RPM: 3180
- Manifold Air Temp: 196F
- EGT: 1106 F
@ Max EGT 1272 F:
- RPM: 4404
- Boost: 8.7 PSI
- Manifold Air Temp: 236 F
@ Max Manifold Air Temp 237 F:
- RPM: 4404
- Boost: 8.5 PSI
- EGT: 1272 F