As logging engine data has become so easy for me actually, I thought I could give this old street dyno idea a go again. My question is: is there a rather defined time span between the single entries in the .raw file? For instance 2000 rpm now, 2200 half a second later, 2400 again half a second later etc.?

Thx

No I don't think there is currently any useful timing. Improve the realtime nature of the data is something I want to do in the future.
Something like a set of data per engine cycle.

http://img97.imageshack.us/img97/8295/wotrun.jpg

Shown is a WOT run, taken this morning. The bigger the distance between the dots, the better it pulls, or so ;) Of course you know the principle, it's all but new.
Because the whole thing is more or less wild guessing anyway it would be perfectly enough if ECUControl could store a timestamp every half or quarter of a second.

FWIW I did some empirical testing of this a year ago or two (it was before the higher baud rate & BT) and as far as I could tell, I got a constant data rate from the ECU regardless of RPM. I don't quite remember now but I think it was something like 3 Hz - that is, three complete sets of data (each set consisting of three different kind of data strings IIRC) per second. I remember you (Cliff) said it wasn't really a constant rate - but in reality it was constant "enough" for me.

If I search a little (or maybe a lot) I can probably find some notes. It may even be mentioned on this forum. But it may be easier to just re-test it. Very easy, blocking read from serial port, print with hires timestamp. Ride it for five minutes like you stole it  :)

thx for the info. In the meantime I re-read the GSF Dyno help files. That's all one needs. It'll give you the curves you want (literally ;) )with no more effort but recording a wave file. Could be taken from the alternator, the exhaust or - my idea ;) - from the tach pulses.

Link

You can have it in German and French. And English.

Today I could give the GSF Dyno a try. For me the easiest solution actually was borrowing an old ZEN V with Line In from my son and attach it with two wires to one of the spark plug cables (4cm parallel to the cabel and fixed with tape). Aim was to check if and how the whole thing works. I recorded 6 WOT runs in 3rd gear, varying only the spark advance by +- 10 % as I'm running the MyEcu closed loop actually. That means that the map was as lean again as it was before I started gambling with the BT module.
Here are the results:

http://img192.imageshack.us/img192/3361/runstandardplmin10prozs.png

It's not the actual max. values that count, nor is it really the form of the curves. The point is that GSFDyno seems to deliver constant values and so is an excellent supplement to MyECU and vice versa. Do 4 runs, change the map (quick and easy on the road), do another 4, change again and so on 'till you think it's enough. Then, at home or on the laptop in your buddy's car, you can see at once what change changed what.

Hubert

What a cool idea for software. I need to try that out!

Please do :) I'd like to find someone I could talk to about my findings (or not-findings). You need a recorder with line-in. My HTC Magician unfortunatey lacks one.

The picture shows today's experiment:

http://img16.imageshack.us/img16/5298/vergleichkoen.png

The red curve is 20% more fuel in open loop (dark blue is my standard closed loop) plus 15% spark advance (up to 40deg - OEM still is 44!)

It seems that fueling is not that important once the map is smooth and around/below L=0,9. Also it looks as if these little Mivv Ghetto-Blasters are not the first choice power wise.

This day's curve (middle from 6 runs because of the windy whether):

http://img710.imageshack.us/img710/79/lameronivsmivv.png

Open, non street legal Mistrals instead of the officially street legal Mivvs, map based on the red curve, not  yet adjusted.
Next exhaust will be my modded Quat-D Ex-Box again. I remember the Quat being better around the dip. I hope GSF-Dyno can approof this :)

For those maybe interested in how it is done:

http://img405.imageshack.us/img405/5036/install1s.th.png How the wires are connected

http://img225.imageshack.us/img225/5552/install2.th.png The other side

http://img88.imageshack.us/img88/1010/zenq.th.png The recorder (get one now, they vanish)

http://img402.imageshack.us/img402/5581/impuls.th.png Typical spark

Hubert

From reply #8:
http://img710.imageshack.us/img710/79/lameronivsmivv.png

Green one sticks out being better at 4500. What was the difference with that one? Just the wind?

no, the lower 4 curves are the ones from the day before (sorry for mentioning the 6 runs). The green one sticking out is the result of these mentioned 6. The difference is the Mistral exhaust.

The original looks like that:
http://img41.imageshack.us/img41/1479/lameroni.th.png

Any one tried the iPhone app Dynolicious? Have tried in in my pick up but not yet on the Sport 1100i.
It uses phones accelerometers and you need to give it the mass being moved (large in my case!) and transmission drag estimate. May try later this week.

I have not, not the app and not an I-Phone.
But I have reinstalled the Quat-D over the weekend. Shouldn't have it removed before. Here is how the Mistral map works with the ExBox:
http://img200.imageshack.us/img200/5715/quatd1.th.png
The nice thing is that exactly around the dip in the torque curve the map now is really lean. You can see the vertical blue bars. As the engine normally crosses the map from the lower left corner to the upper right one, I assume that the first lean column (positive compensation) leads to a rich one (negative comp.) and then to the next  really lean one. Looks a bit as if the compensation was oscillating.
For the next run I'll change only the lean areas a bit, tonight I then will see whether this will have a positive effect on the rich zones also.
Hubert

Here's the GSF-Dyno comparison (Blue Heather is the Exbox):
http://img225.imageshack.us/img225/6628/mivvmistralquatd.th.png
and this is how the map actually looks:
http://img31.imageshack.us/img31/8018/mapquatd.th.png

Torque looks nice and flat, alas it could be a bit higher. Driveability is very good with this setup.