Tweet
D15 i think was the 3rd gear switch.. i'll have to check later though.
-
GM thought ahead...
for us 4 speed auto guys, it's a 4th gear switch, indicating top gear.
for the 3 speed guys, it's a 3rd gear switch, indicating top gear.Comment
-
If I want to start shopping for a MAF sensor can you recommend a donor? I was working on parts for a 96 3.4TDC anyway and was wondering if that would be a good donor? Any idea what pins you will utilize on a 7730? I have EGR wired to the engine but did an EGR delete, Guess I could always repin those. At one time you had mentioned wide band. What would be involved number of wires ect if I need to add wires I would like to add for future mods as well.
MikeComment
-
ideally, the MAF you want is one that is capable of measuring just slightly more air than the motor will be able to pull, ensures that you will have the highest accuracy possible.
that being said, i've made a little spreadsheet of the OBD2 MAFs i've found so far... IIRC, some of them are even the same sensor, just that GM made the tables a bit different...
slight update for test vehicle: used steering rack cost $80(ouch), but when the weather isn't so incredibly horrible, it should be in and back to progress.Comment
-
SO, news:
it'
s raining/snowing in michigan for the next few days, so rack swap may have to wait...
however, in my downtime, i've been gathering information...
i decided to hack through at least some of a $DF BIN to confirm my suspicions, then referenced TGO as a teriary source... turns out the 9396 PCM doesn't just have double the RAM of a 7727/7730.... those have 512Bytes. the RAM module found in a 9396 is ACTUALLY 2,048Bytes, for a total of 2,560 bytes of RAM. the external RAM is mapped from $1800 to $1FFF, which is what confused me so much while looking at my A1 hack for F31 related items, since there were numbers going all over the place in that range.... in $DF, the eRam module is used almost exclusively for stack purposes.
if someone were to make themselves a little battery backup and fit it inside the PCM(and succesfully solder it to the RAM chip), then it WOULD be possible to store partial tunes on the 2K of RAM, and essentially emulate using just an ALDL cable(similar concept as the NVRAM module moates sells)... but that would require considerable code rewrites, and i don't know much about the reliability of such an operation...
and for the 1227730 guys: as i was researching this, i came across an interesting link.... it seems the 1227808(underhood ECM, looks like a 7727, used on Quad4s in W's), and the 1227807(in-cabin ECM, looks like a 7730, used in anything else Quad4) may have been the precursor to the 16149396(since they came out just a LITTLE earlier).... it ALSO has eRAM module as found in a 9396, so you should be able to literally drop that in place of a 7730 and basically have an in-cabin 9396... another benefit is that is seems to have the upgraded injector drivers that would have been found in a 1227749, meaning you SHOULD be able to run up to 6 peak/hold injectors(like the Sy/Tys) with this ECM, should you want to. or up to 18 regular(saturated/high-Z, however you refer to them) injectors, i think.
i seem to remember them missing an I/O chip though? i'll definitely have to look again, they seem to be a great option for the in-cabin ECM cars, or for replacing/upgrading from a 9396, should it be necessary, but the Quad4 Ws aren't exactly common....
end transmission.Comment
-
update time: GP now has new PS rack and pump, seems to be working alright, so test vehicle is now up and running again.
looked into the 1228707 stuff a little more, apparently, it's missing an I/O chip compared to the 1228708? i'm not sure what to think of that.... i'd like to get ahold of one to verify, but for now, guaranteed to work with ALL revisions, past and future, of nAst1 is currently just the 16149396.
that's all for now, maybe i can get the damn single fire code tested soon, huh?
Comment
-
ok, finally went out and got the single-fire code tested!
interesting note from the logs i generated: when in single-fire mode, the final BPW value shown in the datastream DOESN'T get updated! 8mS at idle seemed wrong to me. i'll need to fix that, otherwise, the only values you'll see are when above the max single-fire BPW limit(cranking or lots of throttle), then when back in single-fire it will quit updating.
AE fueling: according to my assband O2 sensor(which trumps narrowband and wideband combined), whenever i made quick throttle transitions, they seemed just as strong and consistent compared to double-fire... i also made sure to try and stay IN single-fire during some of these transitions, and purposely go into double-fire as well. gotta say, i'm happy it works so well here.
however
it WILL expose any weakness in your tunes. in double-fire, i occasionally got a rolling idle whever SLOWLY decelerating from about 10MPH to stopped(like no brake at all kind of decel), it only tended to bump the idle around by maybe 100 RPM in either direction and when fully stopped, it went away entirely. with single-fire on, i actually managed to get it to stall twice. seems i need to play around with my stall-saver AE stuff, huh?
but, overall, a relatively good report, just have to fix the update BPW in single-fire issue.Comment
-
1228707
I have located a 1228707 and should have it in my possesion within a week or so. Is it something you want to inspect?
MikeComment
-
i imagine if you could pull it apart and take a few high-res pics, i would be able to spot the location the I/O chips would have been in, assuming they are indeed missing. i also have the Delco part numbers for some of the ICs used inside of P4 based ECMs.
it SHOULD look nearly identical internally to a 1227730.Comment
-
1228707
I will take the hi-res photos and e-mail them to you as soon as I get it. I am excited to see if it could replace the 7730 have the additional space to run "maybe WB, maybe MAF" and be able to emulate. Not changing connectors is a major plus for me.
MikeComment
-
The BPW value in my single-fire TGP datalogs was updating... wasn't it? Or was I imagining things?Originally posted by robertisaar View PostComment
-
well, WB for use in datastream alone, it will work fine, MAF should be able to be done as well with a 7730 or similar.... it's the later revisions that won't be able to run on it due to needing more free RAM to do so. emulation can be done on any ECM that the ostrich supports.Originally posted by mswenson289 View Post
maybe... keep in mind 8F is a little bit different for it's BPW calcs than A1 due to running a multiplier for the 2BAR...Originally posted by Penglii View PostComment
-
went over my logs a little more thoroughly, and for 3 samples at one time, i was in single-fire for WOT! :O
O2 sensor was showing a good amount of fuel as well, so i MAY be able to extend the allowable single-fire range even further.... when it did kick back into double-fire, the BPW updated to 10.6mS. considering i had the minimum allowed double-fire pulsewidth set to 9.99mS, i'd say it works well.
i'm also 99% certain i have the single-fire BPW update issue fixed as well... for some reason, when in single-fire, the ECM updates the HUD value for it(GM instrumentation module would be a better description, very neat stuff), but not the ALDL value.... but since it's pretty much 100% certain none of us will ever see a GM HUD, i just changed the update location for when it updates the HUD (if it were connected) and changed it to the ALDL BPW location.
should have a new log session done tomorrow or the next day to confirm, then it will be posted.Comment
-
ahh, progress...
BPW update in datastream is fixed, and i added a few new values to the stream, including but not limited to stuff for single-fire... most important there is that i added an injector duty cycle while in single-fire value and histogram, so now you should be able to see exactly how close you are to running into the limit for single-fire operation. if it gets near 100% and you're still in single-fire(check the bit that displays it), you need to back down the max allowable pulse-width for single fire, otherwise you can hit a wall when the ECM switches from a fire cycle to a non-fire cycle and have a lean condition result. now, 100% DC in single-fire is equivalent in reality to 50% for double-fire when it comes to injector on to off ratio, but when in single-fire, the ECM will only command the injectors on for a maximum of 1 full revolution before shutting them off for 1 full revolution. weird how it works, but it's useful.
you can use the histogram to quickly run through a BIN to see what the peak DC was. when the ECM switches over to double-fire, the BPW shown in the stream will cut in half, and since it's not possible yet to use only a single value to represent multiple injector firing states, the DC of single-fire will cut in half as well, but then the double-fire value will be correct. i actually ran a little above the 100% mark a few times today, and considering i'm running a stock engine and injectors, i'm going to have to play with the max value a bit before releasing BINs...
so, at least you know what is in the works, even though you can't play with it for the next couple of days...
Comment
-
just had another thought.... (they're rare, i know) :
anyone else have a reason to add IAC steps when their brakes are applied?Comment
Comment