commanded AFR

[wowzer]

this probably is for rob L but maybe others know these answers.

1) how can i "easily" calculate commanded AFR from the ram locations in the ecu? there is an injector pulsewidth location. can i use that and, if so, what would the formula be to get the theoretical AFR? i want to be able to compare commanded to actual from my wideband and calculate AFR error as a percentage.

2) is there a table that stores the volumetric efficiency of the engine based on rpm and map? or is the pumping efficiency table (only rpm based) the only way to adjust VE?

3) what's the easiest way to disable closed loop and also ignore all fuel modifier adaptives?

4) this spring/summer my goal is to tie in my wideband to mpscan2 to provide the ability to "autotune" the cal, or at least capture the data to do it manually. using the ostrich gives realtime access to the tables so it could be done. otherwise, you still could use mpt2 to tweak the table(s) based on the error, flash, and try it again.

[roachjuice]

Question. How in the hell do you know so much? :) lol

[zin]

If you use the S60 code, I believe it has the O2 feedback disabled, I think the adaptives too, but I'm not sure about that... IIRC, when the O2 is disabled, the adaptives go with it, but I'm not willing to bet my life on that...

Mike

PS I really like the idea of "auto-tuning", or at least getting the info logged to adjust it for yourself... I was thinking that the charge air temp could be used to log the 0-5V signal from the wide-band, while having the cal ignore that input. Actually, I think it would be great if one could use that input for wide-band control, but the I/O for that sensor location would have to be changed/re-written, so I don't know if that is practical...

[1966 dart wagon]

Question. How in the hell do you know so much? :) lol

thats what I was wondering!

[wowzer]

... I was thinking that the charge air temp could be used to log the 0-5V signal from the wide-band, while having the cal ignore that input. Actually, I think it would be great if one could use that input for wide-band control, but the I/O for that sensor location would have to be changed/re-written, so I don't know if that is practical...

that would definitely be the easiest way. if your not using the charge temp then all that routine would need to do is just read the signal every loop and store the volts to a ram location. ACTUALLY, maybe rob could just add a "flag" in one of the setup config tables that you could either have the charge temp routine or the afr routine!?!?!

ROB - what do you think?

[olk93]

I guess this has to be done in 2 steps: a) relocate the input of the temperature reading routine, so it stays in open loop. should be pretty easy to tell the routine to read from foo instead from bar. :) b) read the "now wideband input" 0-5Volts from the (connected to) temperature-input a/d location Value and convert it to a 0-1 Volt range so that the original O2-Routine can continue calculating the pwm for the injectors. In general a nice idea for stick-shifters but for those having a A-604 it is crazy since the OD wont come up in open loop, one has to confirm external requests (here: tcm to sbec) regarding the temperature to above 80C to let them stay in close loop. However pretty interesting problem to be solved here.

[zin]

I'll be happy just to be able to log the wide-band in the same file as the rest of the data, thereby keeping that critical data in-sync with everything.

Keeping the standard O2, though I believe Rob has some ideas on how to adapt a wide-band signal through the normal O2 input...

Mike

[bakes]

Would it be easer to have the mp scan/ mp tuner do the dirty work for the wide band leaning and just spimly auto flash the prom/ ostrich for now until proper code could writin for the PCM to do this by istself?????

[turbovanman²]

4) this spring/summer my goal is to tie in my wideband to mpscan2 to provide the ability to "autotune" the cal, or at least capture the data to do it manually. using the ostrich gives realtime access to the tables so it could be done. otherwise, you still could use mpt2 to tweak the table(s) based on the error, flash, and try it again.

Are you f*cking kidding me? Damn, you are awesome, :hail: :hail:

[wowzer]

I'll be happy just to be able to log the wide-band in the same file as the rest of the data, thereby keeping that critical data in-sync with everything.
...
Mike

that is the initial goal. as mentioned above the easiest way (for me) would be to have the wideband 0-5 volts run through an existing (but unused) 0-5 volt sensor. on my camaro i removed alot of the egr crap (which was a 0-5 sensor) and ran the wideband analog out into the pcm using those wires. the harder way will be to have you hook up your wideband to the pc as well as the sci cable to the pc, log them both, then merge into one file.

i already have the innovate lc-1 and just received the ngk unit in the mail to try in my charger. need to figure out the protocol stuff next. hopefully rob L will catch this and can comment on using the charge air temp port and which cals could support it.

[ShelGame]

that would definitely be the easiest way. if your not using the charge temp then all that routine would need to do is just read the signal every loop and store the volts to a ram location. ACTUALLY, maybe rob could just add a "flag" in one of the setup config tables that you could either have the charge temp routine or the afr routine!?!?!

ROB - what do you think?

Yeah, that's pretty easy...

However...

My original idea was to use the WB signal in place of the original NB signal. Then, use a routine to 'convert' the WB signal to an equivalent NB signal thru a transfer function, and then store both results. The input for the O2 sensor is a normal 0-5V A2D channel as far as I can tell. It doesn't have any fancy circuitry to 'convert' the 1V signal to a 5V signal, for example.

[wowzer]

...

My original idea was to use the WB signal in place of the original NB signal. Then, use a routine to 'convert' the WB signal to an equivalent NB signal thru a transfer function, and then store both results. The input for the O2 sensor is a normal 0-5V A2D channel as far as I can tell. It doesn't have any fancy circuitry to 'convert' the 1V signal to a 5V signal, for example.

i'm good with that. when can you have that done?:)

[shelbymonster]

hehe have been asking Rob about that for a bit of time !! i want it !!!!

[ShelGame]

i'm good with that. when can you have that done?:)

I'll have to go and look thru my files. I think I may have actually done most of it already several years ago...