Accelerometer datalogging and spark correction for max hp?

[gixxer92]

I did a search and didn't find much on it.
Is anyone doing anything of the sort with an accelerometer to adjust spark?
Is it possible to get use the data to adjust spark advance vs. knock to best hp, without having to go to a dyno (for those that are not close to a dyno)?

If anyone is using a serial data acceleromter and has some insight to better ones, please let me know.

[joecar]

Dyno is better because it lets you find the MBT point, and reveals any tendency/preference to knocking that the engine may have.

Without dyno, you can use 1/4 mile ET/TS and apply rules of thumb...

I don't see how accelerometer derived HP (g-Force SS logger) will help with spark, because engine performance is influenced by too many variables (which can be kept fixed by dyno usage)... with accelerometer, you cannot find spark/fuel that gives best TQ (without knock) since a log duration is only seconds, and you can't maintain constant load on engine. Anyone who can do it will already be very experienced in 1/4 mile tuning sessions.

With accelerometer, how will you ensure that spark is safe for engine (LSx do not need as much advance as SBC/BBC).

With 1/4 mile ET/TS you can adjust fuel from peak TQ to higher RPM to achieve best HP, but it takes multiple runs.

[gixxer92]

I guess I should elaborate a bit further here. I was asking the question with the assumption that all of the initial tuning had been completed (VE & MAF datalogging and adjustment).
Then logging the Serial data from the accelerometer along with other variables like spark, knock, etc. using that to adjust after runs and watching the accelerometer for peak G basically through the RPM range, adjusting the spark and doing the run again basically like a dyno run. I understand what your saying Joe but I think there are variables to both it's not 100% consistent on a dyno either, but there are fewer variables. I just think it may be possible to use the accelerometer to do the function.
I'd looking for peak hp through the entire HP/Torque curve. Essentially it would be the same as the car was sitting on the dyno, instead of using a eddy current dyno to map the values, the use of an accelerometer to essentially do the same thing?