Hondata

I'll try to explain this without being too long winded, but no promises.

I am out of ideas on this issue, so any help would be much appreciated:

I am running a built NA B16 with TWM ITB's on a P28 ecu with a s300 installed.
Also the ITB setup is using 270 cc injectors and the MAP has been removed.

My tuner installed the Hondata and loaded it with a stock ITR calibration modified for the 270cc injectors and also set up to use the TPS based look up tables.

The car absolutely would not run with this calibration. It would fire, but as soon as it did it would die.

If I change the calibration to run off of the MAP again, the car will fire up but will run rich and load up real bad.

At this point if I lie to the ecu and lean it out by telling it that I am running 440cc injectors, the car runs decent.

However, if I tell it to run off the TPS again at this point, the car dies instantly.

Now this is the part that really gets me:

If I load a stock JDM P30 calibration into the s300 I get the same result (car barely runs & loads up real bad)

If I take the ecu with the Hondata out of the car and plug my old stock JDM P30 ecu back up, the car starts up and runs pretty decent. The thing is, the P28 with Hondata and the stock P30 calibration should be identical to this, but it wont run.

This convinced me and my tuner that there was a problem with the P28/Hondata ecu so I sent it back to him. He bench tested the unit and found no problem, and also plugged it up to an integra he had on the dyno and ran it with no problem. However when I got it back and tried it again I get the same problems.

The only thing that makes some sense it that the integra it ran on was still referencing a MAP sensor, while mine is not.

I did plug the P28 up to a car and run it to make sure it was good before I had the Hondata installed.

Also, the TPS appears to be working ok because I can see the input scale up and down as I move the accelerator when I have the car running.

Any thoughts? :( :?

Ok so I was long winded.

Please post the calibration, and a datalog if possible - the problem is almost certainly in the TPS settings in the calibration. Also it is critical that the TPS reads 0% to 100% at full sweep.

Attached are 2 calibrations and 2 datalog files. The TPS270 files are the ITR calibration modified for the TPS and 270 cc injectors. the MAP440 files are where I trimmed the fuel and set it back to look for a MAP signal (though there is no MAP sensor) to get the car to start and run somewhat.

The TPS270 datalog may not show much because the car will not run like this but I repeatedly cranked it and let it die several times over 30 seconds or so.

Also, my TPS signal will not drop below 12% and there is no adjustment in the sensor setup (slotted holes, etc). Can the signal be biased in the software somehow? If not, I'll set the intake up on the CNC and make it adjustable :)

Thanks for the help.

Noticed another interesting thing when I went back and looked at the datalogs. In the map440 datalog, looks like the injectors are running full scale all the time.

The TPS sensor needs to read 0% with the throttle closed. It ideally should also read 100% fully open, but this is less critical than the closed reading. Most ITB TPS don't have the same range as the Honda TPS and you end up adapting a Honda TPS.

The MAP sensor is not reading at all (it is pegged to 173 kPa) - check wiring and the MAP sensor itself.

Also the tps270 calibration doesn't show any settings for TPS based map lookup. It usually is better to get the car running from the MAP sensor first, and if that is unsatisfactory then use the TPS.

Sorry, I posted the wrong calibration. See attached for the right one. I did have the "use modified TPS signal" option set under the MAP parameter settings. Sorry for the confusion.

I currently don't have a MAP sensor wired up at all on the car. The only place I could draw vacuum from would be one of the intake runners and I figured it would pulse so bad that it would be useless.

Concerning the TPS on the ITB's, I should be able to check the ohm range on my TPS and see if it matches the Honda, right? If so, what is the correct ohm range on a Honda TPS?

And concerning the TPS signal not going to zero, can the signal be biased somehow in the software?

You need the MAP sensor connected and working even if you are using alpha-N - the ECU will still us it for some settings. Generally taking vacuum from all four runners into a small vacuum canister provides a fairly stable MAP signal.

If you want to match the TPS with the Honda TPS you need to look at the resistance from closed to fully open vs the resistance across the whole TPS. Throttle closed should be 0.10 and fully open 0.90 to give 0.5 and 4.5 volts. There is no set number, it is a ratio, so it might be easier to do in the car measuring the voltage with a voltmeter or the ECU.

I've got something to work with now anyway. I wasn't aware that the MAP still had to be wired up. I'll hook it back up and verify my TPS signal with a multimeter and see what happens.

I'll let you know how it goes.

I appreciate the help and the quick responses.