Quote: Originally posted by JPrevost You can also pass in closed loop without the emissions junk if you modify the PID routine . I can "average" anywhere from a 12 to 16 afr. Agreed, that is why I threw in the "if everything is working as designed" statement. I was just trying to define the two ends and middle of the "emissions" scale. A "tight" PID aproaches open loop. The tighter the PID servo loop, the closer to open loop you get.

JP,

Hows the NOx running that lean? Im sure it wouldnt matter as much with my low compression motor. Even better is delete the cat, but possibly use the linear EGR to control the NOx if its a problem. Man would be great to get rid of the cat.

I like the idea of tweaking the PID routine to target a lower AFR. Id probably disable or limit the prop. term and O2 error params as well as the O2 window so I dont get punted out of closed loop. Also set the BLMs not to be cleared and scale them to all be in the areas that the car will be inspected at. Drive the car around for half an hour or so beforehand so the BLMs have time to settle out and it should blow a steady lean AFR.

The last time I tested, I blew a 988ppm (1184 standard) NOx at 15MPH and 848ppm (1082 standard) at 25MPH with my S-10 4.3 daily driver. It has a stock bottom end (9.3:1 comp) and a .508/.520 lift, 215/220 dur., @ .050 cam and mild ported stock heads with 2.00/1.56 valves.

EDIT: Oops!, I didn't see that you were asking JP. Oh well.

The way they handle power enrich in these is really cool. Its no longer just go full rich and waste gas when the TPS is above a certain point. Instead it may not necessarily richen the AFRs, but just exit closed loop if the throttle is somewhere in between. It also doesnt enter PE right away, but rather brings the AFRs smoothly in with time and/or ammount of throttle. The PE spark is also brought in smoothly over time as teh time in PE increases. There is also a routine to estimate if the cat is overheating. It limits how lean the AFRs can go if this state exists.

The computer reverts back to the old way of quickly going full rich if you stab the throttle down to the floor or youve been flogging the car/truck.

Anyone know what pin is for the MAT? How about the A/C clutch control? Or better yet, pinouts for a CPI blazer would also work.

Finally got off my lazy *** and bought some connectors from rosesandwrenches and made up a short harness so I can run/test the PCM on the bench before it goes in the car. But, still missing a few things...

Yesterday I decided to swap over from the 7730 $8D combo I've been running to the 7427 PCM $31 mask to start getting my feet wet with the new unit. I was reading that a hardware change has to be made to the netres to allow for currect PFI mode enable (idle and low loads have a very annoying surge which I assume is to the hardware change not being made). Also I can't get HaulnA$$'s fan enable to function under the $31 mask (I think I made the change at the currect address) as well put the fan relay control on E3 (which is the pin for the Tuning Valve on a 95 CPI blazer). I will attach my current Bin if anyone want to help me find my mistake on the fan control and any other obvious problems - including PFI enable. Also could someone verify that the Injector Constant is set currectly for 19Lb/Hr units.



Current bin Disclaimer ---NOT TO BE USED ON A VEHICLE OTHER THAN THE ONE OWNER BY ME---

Prom/NETRES in the PCM is a BMHM
http://www.moates.net/files/2)%20Cus...es/305BMHM.BIN

First of all, congrats on upgrading to the PCM. Second, without the mode mod to the NETRES ,you are still firing in TBI mode and I see several problems with the .bin but I will start by saying that the injector constant looks good as well as PFI enable although I am not sure about the synchronous fuel at idle bit bieng only for TBI. If I were you, I would also disable the MAT sensor as none of the tables are calibrated. Get it running well, then enable the MAT and tune for it. Next is EGR. Again, I would disable EGR and zero ALL the EGR tables but if you want to switch the .bin to a non linear EGR valve, then you must also turn on the back pressure EGR bit as well as toggling the Linear EGR bit to EVRV and tuning the assosciated tables. They operate on different sections of the code and cal. Next, to get it to run right you will need to tune the injector cals for high impedance injectors. Lastly, the fan control stuff is wrong. I am not a fan of modifying code with a hex editor (calibration stuff is no problem) as many times, as in this case, you will need to delete or add a byte to make it work, and that is next to impossible to do without disassembling, making changes, and re-assembling. If this setup were as easy as flipping a few bits and wiring the port setup, I would have had it done months ago. I'm not trying to flame, just pointing outsome of the obstacles that need to be overcome to make it all work. HTH

Hauln Can you explain what has to be done to the netres. Also thanks for noticing the bits that were misset. What exactly are you refering to when you state the injector calirations? The fan is not important right now as it is running off of a homemade controller. I need to spend a bunch of time on the disassembly and reassembly as my attempt for this did not work (disassembled file was over 2GB! when I came back and closed the dos prompt) I will learn it though.

If your running the 16197427 ECM ($0D mask) here is the wire diagrams off GMECM http://www.diy-efi.org/gmecm/ecm_info/16197427/

Injector calibrations refers to biases, offsets and PW modifiers. As for the mode mod, I don't have that info handy as I am on the road working and using my laptop. The info for the mod is on my computer at home. As I recall, one of the pins on the NETRES needs to be grounded to change to PFI mode. P.M. or e-mail me this weekend when I should be home or maybe someone else can chime in. HTH

No MAT on those, unfortunatly, so its not documented.

Try B9 for the MAT.

Oddly enough, none of the wire diagrams have MAT's in them. I uploaded some wire diagrams for you guys on Moates site:
16168625 PCM - 93 C1500 Pickup 4L60E - Wire Diagram
16197427 PCM - 95 C1500 Pickup 4L60E Wire Diagrams

One intresting thing I noticed is that the 16197427 PCM, $0D uses two knock sensors.

Only for the 4.3L V-6. The 4 cyl. PFI applications use no knock sensor and the V-8 applications only use one KS.

Nice guess, thats the one.

Yep, i was feelin' lucky

Do all V6 apps use the dual KS', or is it just something thats on the high performance CPI option?

All of 'em, including my '95 4.3 TBI (soon to be MPFI) S-10 daily driver. My '94 C1500 TBI V-8 only has one. HTH

I'm just taking a stab in the dark with this one but I have a feeling the dual knock sensors had something to do with the balance shafts in the 93+ 4.3L v6's. I can't think of any other reason to have it unless they had code that would do individual cylinder retard like the LS1's.

BTSOM. That sounds like a reasonable theory but there is definetly no code for individual cylinder retard as a distributor is present. The knock sensors on the V-6's are in the heads just above the exhaust ports aimed right at the balanc shaft. The V-8 applications have the KS in the usual spot, in the block near the starter. HTH

Update for those of you who are using the PCMs.

Ive finally gotten to the point of making a massive dent in the side of this thing. All those thousands of uncommented lines of code in the original $0D hack where hiding a dirty little secret...

A very large, full PID idle routine with multiple derivative terms, sophisticated subroutines to construct the desired RPM, integral, proportional, and derivative terms, full learning capacity, and logic to detect saturation and integrator windup due to IAC faults. It even has a proportional control routine for the battery volts to speed the engine up and increase alternator output if the battery is draining. All of that and theres still lots of idle stuff left to be commented.

Needless to say I wasnt expecting that. I had thought it would have had something like the half-assed stock idle routine in the first TBI ECMs with some extra add-ons. Instead they ditched all the previous stuff and started over.

Gone is the method of just looking up the IAC counts, kicking down, and adding/removing steps depending on what the engine speed is. Now its all done in percentage of IAC flow to give linearity and only at the end are the actual needed IAC counts retrieved from the airflow. Not only that, but theres always some form of closed loop action and it enters closed loop idle much sooner.

It looks like a massive pain in the *** to tune but once comple it should practically think for itself.

Anyway, Ill see if I cant get something posted soon so we'll have a better idea of how it works.

Bringing this post back up. I know it has been a long time coming, but I finally finished the TPI setup using the $0D mask. Besides the fact that I have been busier than a one legged grape stomper, what took so long was researching all the code differences between TBI and MPFI then testing the changes on the PCM bench. I wanted it to be a painless swap and it was. It runs great. It needs just a minor amount of tuning, mostly AE and startup. It is my sons '94 C1500 305 with 160k miles and a 16196395 PCM. The truck is stock except for exhaust and all the TBI tuning I did on it. I made sure to eliminate as many variables in the swap as possible. I had the injectors (19lb. SVO's) cleaned and flowed. I used all new connectors and made up my own conversion harness, more on that later. I got a new serpentine accessory drive setup from SDPC to clear the throttle body and I ran steel braided fuel lines with AN fittings and adapters. The IAC was replaced and a new Walbro 190 was installed and I got a new throttle cable from Jim's Performance. The mechanical part was all plug and play so I will spare the details.

On the PCM/code side is where I spent the most time (the last year or so) figuring stuff out. What I ended up doing was modifying the P&H injector drivers to drive saturated injectors. I also modified the Memcal mode select for an 8 cyl. PFI setup. All the pertinant calibration data was changed to run the TPI like injector biases and offsets, select bits, etc. In the startup .bin, I deleted EGR and CCP until I get it tuned then I will add it back as well as install a MAT sensor and tune for it. As for the wiring changes, they were minimal. I had to replace the TBI injector connectors for port connectors. I ran 4 injectors off of each driver. I also had to extend the IAC and TPS wires about 10 inches and swap the TPS connector for the old style flat type vs. the round type. Thats it. It runs great and idles smoother that any other TPI I have seen. Once I get it tuned, I will make some 0 to 60 runs and post the differences. Here are some before and after pics. Enjoy!

That is awesome! I have been wanting to do that with my '94 c1500 for a while, But just dont have the experience to dive into the code side of it. What exactly did you have to do to the injector drivers? Have you driven it enough to notice a mpg increase? I bet it is a treat to drive!
----------
Is that TBI you where running on the 305, a BBC unit, it looks like it in the pictures?

Without getting too technical, I basically just turned the injector drivers from "Peak & Hold" to just "Hold" drivers. It has only been a week since we finished the swap and it is on my 18 year old sons truck, who never fills the tank completely so the gas milage hasn't really been tested, but I will eventually. I must say that I had done a lot of tunung on the TBI setup and it ran really well before the swao and got 17MPG city and 21MPG highway so I don't expect much increase if at all. The main difference is the throttle response. It is much crisper and more responsive on part throttle. WOT at this point is only very slightly improved but it still needs to be tuned.

No, it is the stock 305 TBI with the ultimate TBI mods. HTH

Looks like it started out as a SBC TBI. It maybe bored though. Looks to be just a standard TBI that has been treated to ultimate TBI mods. I am guessing that it is untouched though, the idle speed screw still has the factory cover over it. With that style, it will not come off in one piece!

This is a 454 TBI



This is a stock SBC TBI.



Another

Huh…

Huh….

Oh Crap… the gears are turning…

Now you got me thinking… my ’92 K1500 (it’s even the same color) has to come apart very soon, as in it has a blown head gasket or cracked head and I was planning on just swapping everything heads and up…

Huh…

The original plan was some vortec heads, adapt some intake, keep the TBI and convert to a ‘747 because the lack of info (and my relative stupidity with ecm stuff in general) for the ‘299 stuff is driving me crazy.

Oh man… what’s making this worse is that I have a whole new top end for my ’98 Formula 350, including new heads, cam, converted victor jr…, so I could potentially have pretty much all the hardware to do this except a spare throttle body but for the cost savings it would be very easy to get a nice aftermarket one for the formula and use the stocker… what I didn’t really have was a harness and was wondering if I could adapt the factory truck harness to work with one of the 730’s or 749’s that I have sitting around…

I really don’t need another project, tell me that this isn’t worth really considering… I have to say, as much as I basically hate the TPI setup, it does look really cool… That would give the truck a totally trick look under the hood. My gut is telling me that the TBI is still happier from idle to about 2500, which is 99% of the time that this truck is running.
----------
WHY? A normal peak and hold driver should run saturated injectors just fine… did you experience a problem or did you not even try it the stock injector driver?

First of all, this won't work on the '299. It will only work on the '93-'95 PCM's that control the electronic tranny's meaning the 16168625,16196395, and the 16197427 and maybe one or two others so whether it is worth it or not for you seems a lot less likely.

It is the stock injedctor driver, just modified. A stock P&H driver is designed to drive one TBI injector or two low impedance port injectors with a peak current of around 4 amps. This is not a problem if you want to drive one, two, or possibly even three saturated injectors but four poses a problem. My injectors measured right at 14 Ohms each which means four in parallel measure 3.5 Ohms. At 13.8 Volts they would draw 3.94 Amps. This is right at the Peak to Hold threshold of the driver. Very bad. If the driver ever detects the peak current, it will switch to hold mode and lower the current to the injectors which, depending on the fuel pressure, is low enough to actually let some saturated injectors close before the injector pulse is over. In order to eliminate this possible problem, I modified the driver to never detect peak current. HTH

Yea, I realize that, but I’m looking at it as:
- I was planning on swapping PCM’s, just going in a different direction, it will cost me the same either way
- I need to pull everything off the engine down to the heads anyway because of the headgasket leak or cracked head
- I’ll have all the TPI hardware sitting around anyway, and using it would be cheaper then the originally planned swap
- The TPI looks cool
- It’s just a cool swap, ecm and hardware wise

The negatives:
- I don’t know that there is any real gain to be had with the TPI setup for the application (again, used as an actual truck, If I could have all the low end in the world below 2500 I would sacrifice over 3K in a second)
- Really, this thing is a truck used as a truck. The number one concern for it is that it be reliable and cheap to keep on the road, and past that as much low end for towing/hauling and as high an MPG as possible… I don’t know that this change would serve that
- “wasting” extra time tinkering with is is not high on my list of things to tinker with, I don’t have time to get done with my “real” f-body projects…
- I hate working on the TPI setup

So basically you left it as is but disabled the hold part, right? I haven’t looked at the circuit in those PCM’s but I would assume that each injector driver is actually 2 drivers one doing peak and the second doing hold, or is there more to this then that?

The circuit consists of an injector controller IC with a transistor output. The IC monitors injector current and once it reaches the "Peak" level, which with low impedance injectors happens very quickly, the injector current is lowered to the "Hold" level until the pulse is complete. This opens the injector fully very quickly yet keeps heat in the coil to a minimum. HTH

Its would probably be worth doing. The PCM is perfect for this as it uses true speed density calcs and has lots of goodies in it, not to mention that it also supports the E-trans. Only downside is that its somewhat complicated compared to the earlier ECMs, but it also offers alot more functionality.

BTW, that TPI does look cool. Nice work on the swap, HaulnA$$

On a side note, Ive gotten most of the code (sans transmission) done for the $0D. Im thinking of writing a patch to allow me to read in the freqency of the LS1 MAF that I have. The code is such that it would be a fairly straight forward conversion. Just need to figure out how to best divide the frequency a bit to keep the number of interrupts reasonable.

Man, you’re almost close enough for me to drive over and smack upside the head for putting ideas like that in my head…


What’s worse is that I’m also really wondering about how this would work with my formula project (port injected with a 4L80e tranny), the big sticking point that I’m seeing is that it would need to play nice with a 2 bar (and probably eventually a 3 bar) map sensor.

Man, I wish that there was a book or something that I could sit down and study and figure out some of the real ECM hacking (well if you’re a computer guy it should be cracking, not hacking, but oh well…) so I could dive into this kind of project with some of you that have a real clue what is going on (for the life of me I can’t figure out how you’re figuring out what the lines of hex do in the first place), or at least if one of you were close enough that I could sponge info off of you till I was really up to speed (hell, even get a fraction as good as I am with things mechanical…)

I have a partially completed hac that I can post later on. I posted it here before, but Ill post the most up to date one since its always changing. Most of the engine stuff is complete, but I havnt done anything with the transmission side of things as I wont be using a CC trans, or any form of automatic for that matter.

Here is the latest

Thanks, but this is just a means to an end culminating in a boosted 4xx cubic inch port injected small block truck with DIS and a 4L80E all running on the stock PCM. Beauty is, the basis for the code is inherent to the mask already except for boost which has been done already. The future looks bright but I might need a little help.

It's awesome to see that you have made more progress on this setup. If you don't mind can you explain what needs to be changed on the actual PCB driver circuit to not allow the driver to enter the hold mode as well as the actual changes that need to be done to the netres set.

Unfortunately, I do mind. I'm not trying to be a d!ck but I support my family by working in the electronics trade. If I make public specific instructions on how to modify hardware that someone else designed, I put my livelihood at risk just as the software engineer who told you specifically how do modify someone else's code does. I am more than willing to share information, like my LM-1 interface, definition files, etc., stuff that is my own intellectual property. However, all is not lost. If you *search* hard enough (which should not be too hard), you will find all the information you require to "do it yourself". On the other hand, should your PCM suddenly end up "broken", I might possibly be persuaded to "fix" it for you, as a favor of course. HTH

I finally got a cheap-*** function generator to begin poking at the inputs. I finally confirmed that the transmission output speed is tied to actual honest to god pulse counters. They appear to run fine all the way up to around 40 kHz, or more, on a standard TTL signal.

This should mean that I can plug the frequency output from an LS1 MAF right into the TOS input, and read it directly without any complicated code. Mmmmm, airflow readings that are accurate to a few 10'ths of a percent. It also should be a simple plug and chug mod if its as easy as it appears to be. These PCMs definatly look like the next have to have item for the DIY'er.

I finally got my son out from behind the wheel long enough to hook up the O-Scope and record some pictures of the waveforms. Pardon the quality of the images, for some reason the floppy drive in the scope would'nt format my disk so I used the digital. The first pic is of the synchronous injector timing vs. DRP. The top waveform is DRP's from the IC signal on pin F11, the bottom is Inj. 1 from pin A16. As you can see, the injectors are firing every 4th DRP. The second waveform is both Inj. 1 and Inj. 2 together.

I just noticed this. With the PCM, from what I can tell, you have to use pin F13 for the VSS if no E-trans is present. Dont know if this was caught already or not. The others (pin F12/pin F??) are for the actual 40 pulse/rev output from the trans output shaft speed pickup pulses while pin F13 is for the output from the VSS buffer box. That output is used for the MPH term throughout the engine code. The only place pin F12 would be used otherwise is in the shift light code. Dont know why they did it that way rather then calculate it from the TOS revs but I cant say I can complain. It gives me another free input.

From what I gather, there are two seperate MPH terms used in the code and they are derived from the two seperate VSS inputs. The one at F13 is a 2000 PPM signal as derived/calibrated from the DRAC module and is used for the engine code. The second at F12 is a 40 PPR signal and is a 5V representation of the actual VSS. This signal is used for the shift light and tranny code and is calibrated by the term at L5D18. If you log transmission and engine data simultaneously, you can see the two seperate ALDL MPH outputs. These can actually vary from each other depending on how far off the calibrations are from tire size and or gear changes. If you do a gear swap, you must calibrate the DRAC and calibrate L5D18 to match, BTDT. I thing the reason GM did this was because all the TBI engine code up to this time had the MPH term derived from a 2000 PPM signal so it was easier to keep it that way. By using the VSS directly for transmission code, calculations are not dependant on DRAC calibration. As for free inputs, don't forget about pins F2 and F3 which are the Trans input speed VSS Hi and Lo which are only used on the 4L80E apps and $0E and $31 code. HTH

The only problem with the F13 input is that its sampled in the main fuel loop at 80 Hz if I recall. This means that after around 75 MPH, theres aliasing. IOW (for the rest of us non-EEs), itll start reading random speeds, and could even read near zero at certain conditions. Im not quite sure how that will effect a car when flying down the highway. That was sort of what caught me as odd, is that the input can be easily overrun. I guess they figured that GM owners are all law-abiding citizens that dont drive over 55.

What sort of signal does that accept? Ive seen the pulse accumulator and counter reg. in the code as an alternate to the 5V signal thats supplied in the other apps.

From the diagrams I have, it accepts the signal directly from the sensor which appears to be a reluctor type. I don't see any reason why a 5V signal would'nt work but I have not played with it although a voltage divider may be needed. HTH

thats cool. My TKO has a 12-tooth reluctor with a pickup. I would assume I could wire the pickups output to the high/low pins and use it to read in the vehicle speed?

Should work flawlessly, with the proper code, of course. That input was originally designed for the 4L80E code to directly measure tranny input shaft speed for slippage calcs vs. the output/VSS speed instead of relying on the RPM input which is obviously pre-converter. HTH

Good to see this thread still alive!
I have now been running my 383 TBI with a '8625 for ome time and its fantastic!
I intend to switch to PFI mode soon, and once its working properly on the 383, I will swap it for the 406 I am building at the moment..

I also recall seeing another VSS TOS input in the code as well. Have you seen any additional pin or pins for this listed?

Nope, all the stuff I have shows F12 and F13 for all tranny's and add F2 and F3 for the 4L80E. HTH

Are pins F2 AND F3 both for the pickup on the input shaft? Or is one for the pickup on the input shaft and one for the output shaft?