Mystery on Load Bearing vs Inertia Dyno

Hi guys,

Hoping someone can give me some insight into a really strange phenomenon I've seen on one car only in my career thus far and struggling to give a concrete reason as to why what's happening has happened.

# of dynos involved - 4

Engine is a Nissan SR16VE in N/A form. I've tuned this car on two different load bearing dynos, and seen the same thing. One was a dynapack hub Dyno and the other a Mainline rolling road load bearing dyno. Both load bearing dynos a WOT ramp rate of 400-500rpm/sec was used.

Optimal ignition timing was found using the both dynos, this was double and triple checked on both dynos, there was no further increase in torque under WOT conditions.

The same car with the same tune was then taken to two separate dynojet inertia dynos, and on both of the inertia dynos, the engine took an additional 4 degrees of timing across the board and showed significant increase in torque versus the map that the load bearing dynos said were optimized.

AFR running in closed loop and properly tuned VE table in all instances.

I've tuned several other cars on both load bearing dynos and then tested on the same inertia dynos and not seen this behavior. Can anyone shed some insight into what may be transpiring with this specific car?

What was RPM rate you saw on the inertia dyno? What if you used that rate on the load-bearing dyno?

Does this car have a particularly heavy or lightweight drivetrain (ie, low-mass small clutch / flywheel, lightened crank, etc or large / heavy brakes and or transmission).

Hi Dave,

I'll have to time/calculate the ramp rate on the inertia, but from the sound of it, I don't reckon it was much off from the 4-500 on the load dyno.

As for engine and trans, no none of those.

Any chance the fuel (octane / brand) wasn't the same between different dyno types?

Other variables may be atmospheric - air temperature, humidity, and elevation of the dyno' shops will affect the oxygen content and timing possible, as will the enine's coolant and oil temperatures.

Oh, and the gear used may also affect the specific loading on the crankshaft, but I would expect that to be relatively minor?

Dave everything the same, it was done on the same day. As for varying conditions, our island is a dot on the world map, it's all at sea level so that's not the issue.

I've done this exact same test with other cars and this is the only car that's done this on two separate occasions

Certainly a mystery.

By the way- can anyone explain why there is a difference in dyno readings of the same cars? I have a collection of videos where the same cars make dyno pulls on Mustang dyno and Dynojet and Dynojet is always showing higher numbers... What makes me even more frustrated is that this situation is ok for many people, including high level professionals. How come we are ok that it's so different?

What ECU are you using? could you post logs of the pulls?

Georg1970, different dyno companies calculate power differently, different strap techniques, different ramp rates ETC. The main thing to remember is that a dyno is nothing more than a tool for tuning, the important difference is that from start to finish of the session

Shota, I'm okay with my dyno reading lower than the others (I have a Mustang AWD). Like Chris said, there are some difference on how the power is calculated. But for me, a chassis dyno is not a tool to give "accurate" horsepower/torque. It's a comparison tool to see if the change you made to the tune helped or not. Repeatability between runs is what I'm after, and I'm happy as my dyno is great with this.

Mustang dynos are easy to "adjust" numbers if you want them to read higher, some people even have channels that gives dynojet numbers. It's great if you are developing aftermarket parts as it looks good, but for tuning it's totally irrelevant in my opinion.

Guys, I totally understand the purpose of dynos- I also think of it just like nothing but a tool to monitor changes and not interesting in numbers. However, what I was trying to point out is that for some reason it's ok to have different numbers by the different dynos for the same car despite there is only one real HP (I know there are a few but they are not that different) in physics so one of the numbers must be not relevant but people still accept it (NASCAR official dyno is Dynojet). I once red the post from one famous tuning company - 400 hp evo on our Mustang dyno runs 11.5 just like 450hp evo on Dynojet runs also 11.5... That's the part I don't get - why there is no standard related to physics which should represent the real power.

It sounds like the concern is MBT appears to differ depending on the dyno, which is really a reminder that doing WOT pulls in a single gear with a single amount of load does not result in accounting for the full range of operating conditions a car will see on the road, track, or even different dynos. On an NA car if your load axis is Throttle position, perhaps baro compensated, you don't have a way to account for this. If the primary load axis is MAP, then you usually have to use extra resolution in the axis values to tune for the small but significant variance in engine load as conditions change and optimize ignition accordingly.

Also, this depends on bit on your definition of MBT. If you're talking about optimal timing for a given knock limited fuel, then this is especially important.

The ECU is Maxx ECU street. I've developed a bad habit of not saving logs since I just "pause" the real time log on the max and do my interpretation then just hit play again. As such I don't have the logs.

I'm sure there was nothing different in terms of sensor readings to account for what I've seen. There must be a reason though.

The difference is the amount of load, and time at load. If you're reviewing throttle position it will read 100% whether the car is in 2nd or 4th gear, going up or down hill, loaded with cargo or not, but what the engine is working against is changing quite a lot. Even if you run the car on the load bearing dyno with different load settings, rather than switching dynos, you can test the effect.

Gather some logs. Maxx has great built in logging so.. comparing dynos is worthless. Would be great to have all factors and variables for all 4 dynos