IT Guy, gearhead, dipping my toe back in the GM EFI world

[89S10_Project]

Hello all, I was somewhat active on the old GM EFI and DIYEFI newsgroups years ago. Yes, I'm older than a few of you and have now proven it....

I built an ALDL interface cable back when laptops actually had serial interfaces. It worked very well for my purposes ( a then-stock 4.3L 1989 S10 Blazer) and was mostly built for bragging rights. While I intended to do some ECM tweaking, it quickly became apparent I had no business messing with that in the truck I was driving every day to work.

That was something like 20 years ago. Now the same truck has been modified slightly (ok, the frame is mostly stock, the body is stock, and it still has the same transfer case.... everything else, however, has been messed with). SOA, Dana 44 axles front and rear, springover lift, hydroboost, a replacement 700R4 (soon to be upgraded).... and a 96 GM Performance Gen II 5.7L V8. I'd intended to use the original ECM, but a good friend on another forum donated a raft of parts to the project- including a pulled ECU that plugged right into the same harness as my 1227747 V6 ECM and already had the MEMCAL and proper chip for an auto 5.7 V8.

I've got the truck running, just not well. I am having driveability concerns that appear to come and go- starts up fine, and after the 30 second idle-up it starts to exhibit a stumble in the idle... once things have warmed up the stumble is REALLY apparent.

I had this issue upon initial build (truck has been down for a year due to the need to go Springover up front and ensuing domino effect with shocks, steerage, etc).... I figured it to be the IAC. Pulled one off my old 4.3 as I knew it was recently replaced- and things appeared to be fine. Now it's a year later and the issue is back.

I thought I'd found the issue when discussing matters with same friend, I mentioned the VSS needed to be extended as I'd noticed the wire hanging off the tailshaft of the tcase. DUH, he said- that's required input for the IAC, that'll fix the issue. I investigated and found that I'd never dropped the harness side of that connection down from the ziptie I'd used to secure things on initial build- ie, it had never been hooked up and the truck ran fine once upon a time. Extended the wires (solder, heatshrink, and quality tape and loom, so there's no shorts causing this- I solder and build my electrical to MilSpec), reattached and the problem persists.

At this point instead of approaching the issue with the Parts Shotgun (a few hundred bucks of new parts at the shop, to no avail), I want data...

At present I am not looking to burn chips. This isn't a race car, and I don't know what I'm doing there. I just need to get the data pulling, and prefer to do it without a special cable.

To that end I came across info last year on using an Arduino MAX232 USB to (TTL?) to interface with the ALDL wires A and E (GND and 160-baud data). I have the parts needed, soldered the TXD and RXD wires together, tested by attaching the MAX232 GND to the ALDL Gnd and the TXT/RXD combined to pin E (160 baud data).... and nothing worked (using Tunerpro, not convinced I'm doing THAT right either. I'm an IT guy and I do all manner of complicated things- but Tunerpro is even moreso!). I suspect it may be the lack of a 1KOhm resistor and a diode for the RXD/TXD wires respectively. I do recall having to use those to get the old serial interface to WinALDL working. Problem- there's nowhere in town to grab these bits now, as Radio Shack is no longer in the area. I foresee a weekend spent raping old electronic bits laying about the house of resistors and diodes....

If anyone can advise me on a way to use the above equipment to speak to TunerPro to datalog as I testdrive the truck, I'd be forever grateful. As it is, I have not yet paid for Tunerpro, though I do certainly intend to do so once it's done something useful for me. (ok, ok, also, no sooner than next Friday- just started a new job and Friday is our first full payday- I'm broke as a joke at present!)

I realize there are prefab cables that I can buy. However, while that may well be the solution for most, I can't go that route- I'm VERY much a hands-on guy (I've built this truck myself, with some wonderful advice and tips from friends). I also am able to solder, etch PC boards, etc- pretty much everything but code. I'd rather not get into that- I'm doing this for relaxation and accomplishment, if I have to learn to code as well it becomes too much like a day at work (and I have plenty along that line I have to do for my career, so no thanks!). When I'm at the point I'm receiving data realtime, I'd like to work the solution underdash so that I can simply plug a USB cable under the dash and into a laptop on a stand in the passenger seat and do what I need to do. When that's done, unplug the USB and carry the USB cable and the laptop in the house- the rest of it stays in the truck, hardwired in.

Future expansion probably holds a replacement ECM that can control fans, run a WBO2, etc- but that's a far part down the road... I need to get the old beast running and getting me to the new job, ASAP.

Also, HI, TOM! :)

[89S10_Project]

I should add:

-No CEL is activated. No codes pulled with the paperclip method other than 12 (over and over 6 times before I quit)

-Once the truck gets around running temp, the idle gets to the point where I have to apply throttle to smooth it out, and the truck will often cut off if I don't.

-Driving the truck, it's VERY sluggish at low RPM, wants to fall on its face.... I apply a bit more than "moderate" throttle and it's off to the races. I want to say the fuel pressure should be good in this case. Don't have a way to test FP at present, that's one of the next upgrades- bore out the TBI for JIC fittings and run pushlok hose butt to nutt with firesleeve and addel clamps in all the proper places. Fuel pressure gauges will be permanently mounted at the TBI.

-I've checked for vacuum leaks by spraying ether all along the vacuum hoses and tbi unit as the truck ran, and listening for differences in RPM. No indications of leaks.

[Kitch]

My 1994 G20 van had a problem with a faulty aftermarket ignition coil, as the engine got up to temperature and the coil warmed up the van started to run rougher and occasional stall at idle. The problem slowly got worse and worse until I tried a spare ignition coil and then the fault was gone.

[sturgillbd]

From what I gather from your posts, you are using a '96 up vortec (L31) engine. If so, the heads etc are going to require more Acceleration enrichment (both tps and map based). The VE tables are going to be somewhat different that a standard tbi engine. The vortec heads also will want more spark advance than a normal tbi engine. There are several threads on here that cover some of the differences. I am on my smartphone at the moment and it is inconvenient to search and paste some links. I can send some links when I get home from work this evening. You can use google search on the site by typing site: gearhead-efi.com and then what you are searching for. Look for L31 spark tables and also for threads on AE
Hope this helps,
Brian

[89S10_Project]

From what I gather from your posts, you are using a '96 up vortec (L31) engine. If so, the heads etc are going to require more Acceleration enrichment (both tps and map based). The VE tables are going to be somewhat different that a standard tbi engine. The vortec heads also will want more spark advance than a normal tbi engine. There are several threads on here that cover some of the differences. I am on my smartphone at the moment and it is inconvenient to search and paste some links. I can send some links when I get home from work this evening. You can use google search on the site by typing site: gearhead-efi.com and then what you are searching for. Look for L31 spark tables and also for threads on AE
Hope this helps,
Brian

Brian-

I may have misspoke. When I ran the numbers this showed as a 96 Gen II Smallblock, but it has iron heads so it's not the high-zoot AL-headed smallblock GM offered back then (I want to say the ZZ3 is the AL-head version, can't recall.

This came out of a 1989 C1500 that was being used as a junkyard's yard truck. It was immaculate inside the engine once I pulled it- I got a steal of a deal for $500. Best guess is the guy that used to drive the truck bought a crate replacement motor and then put his intake and/or heads on it... and then the trans gave out. I believe I have the head numbers and block numbers in my build thread over on S10Extremists if needed.

Not to argue the point as I have no data at present to back it up, but I don't believe this is my issue. I think it's going to come down to a bad sensor, or a vacuum leak I haven't been able to find (though really I have NO idea how it would have escaped my ether testing).

[1project2many]

Howdy! And welcome to the forum. I'm a former diy-efi member also.

Sounds like a very frustrating experience.
Do you have a link to your thread on s10extremists?

Feel free to answer any questions below which are not already answered in the other thread.
Are you using the 5.7L throttle body?
How many bolts hold the intake on?
Can you tell if the engine is too rich (smells of fuel, black smoke) or too lean?
What is the three or four letter code on the memcal?
How did you set the timing? What spec did you use (6 deg. or 0 deg.)?

I might be able to help with Tunerpro but I don't use it much so hopefully someone with more experience can chime in.
Thanks.

Shannen

[89S10_Project]

My thread on the S10 site is a cluster- lots of pondering and pontification... last check it's 40 pages long. But if you're a masochist, or just extremely bored, will provide later....

-I'm using the 5.7 TBI unit and injectors

-Can't recall, but believe the intake has 8 bolts. Will verify later.

-no indication of rich/lean. I have installed a brand new O2 sensor in the crossover pipe.

-No clue re: the code on the memcal.

-I set timing after disconnecting the ESC wire in the passenger footwell, set to 0 degrees. Do need to verify, and make sure I reconnected the ESC wire.

Regards my interface, I found that I need a 1K resistor and a diode on the TXD and RXD wires to make my Arduino MAX232 USB interface work with 160-bit ALDL. I am going to run to the next town over tonight and get these (thans for going out of business, Radio Shack!)

[89S10_Project]

Oh, one other datapoint, and this may be a headsmacker...."OH, Well thars yer problem!"....

The exhaust crossover pipe is tapping against a frame member. I'm thinking it just might be possible the knock sensor is hearing this and adjusting for "knock" as the knock sensor is nothing more than an acoustic transducer (microphone).

In that case, I'll weld in a flex point upstream, and bolt a standoff to the pipe and crossmember. I have to fabricate a better crossover pipe at some point, but this would fix the issue should this pan out.

It's all coming down to- I need to see what the datastream says, or I'm just spitting in the wind.

[Nasty-Z]

Welcome Phil , Glad to see you over here .

I see you are already in good hands with Brian and Shannen .

Let me know if I can be of anymore help.

TOM

[89S10_Project]

Hey there Tom- you need more cars. :)

... and why've you been trying to talk me into swapping out the 7427 ECU? Seems like it's pretty popular in your stable.. :)

[1project2many]

The exhaust crossover pipe is tapping against a frame member. I'm thinking it just might be possible the knock sensor is hearing this and adjusting for "knock" as the knock sensor is nothing more than an acoustic transducer (microphone).

^^^^^^ Definitely a possibility. I had to diagnose a power issue in a plumber's truck once. The pipes banging around in the back of the truck were fooling the KS. That took a bit.

-No clue re: the code on the memcal.

-I set timing after disconnecting the ESC wire in the passenger footwell, set to 0 degrees. Do need to verify, and make sure I reconnected the ESC wire.

I would suggest getting the code. Some of the early cals were supposed to be set at 6 degrees. Also, chips from the early boxy R/V trucks definitely behave differently. When possible I opt for later cals every time.

Also, if you have an 8 bolt intake then you have "Vortec" or L31 heads and Brian's tuning advice is right on target. If you have a 12 bolt intake then a stock calibration should reasonably well.

[Nasty-Z]

Hey there Tom- you need more cars. :)

... and why've you been trying to talk me into swapping out the 7427 ECU? Seems like it's pretty popular in your stable.. :)

Yeah , like a hole in the head .....

Swapping to the '7427 is like a date with the prom queen , once you try it , everything else is second best 😊

TOM

[89S10_Project]

OK, this is why I document things... my memory sucks.

Don't know where I got the idea I had an 8 bolt intake. It has 6 bolts on each side, so I'm thinking NOT a vortec.

Knock sensor? Well maybe not. I just remembered I was never able to locate the wire to go to it- so while it is bolted onto the motor, it's not plugged up. So I'm thinking that's likely not the issue unless it's one of them there newfangled WIFI sensors (and communicating without an AP via the new PFM protocol.... :) )

so I am back to square one... to which I need help here-

I have an Arduino module MAX232 unit. It is supposed to be usable with Tunerpro by way of connecting ground to ground on the ALDL, and TXD and RXD get tied together via a 1K resistor on the RXD and a diode on the TXD. I am using a 1N4001 diode on the TXD with the stripe toward the PCB. Not convinced that is right.

At any rate, can't get the unit to test with Tunerpro in the first place.

I have the truck running now, having had the battery unplugged for a few days in the faint hope the ECM needed to be cleared. Nope, it's still running like crap.

ESC wire is verified plugged back together (brown wire in the pass footwell).

I have no CEL active, and pulling codes with the "A to B paperclip trick" shows a repeating 12, which means no codes in memory.

I'm completely stymied here.

I am rapidly running out of patience. I'm pretty much dead broke till payday, would really be helpful to get this thing squared away in the next day or two since the wife is going out of town and I need to rip the suspension apart on her jeep to replace the shot control arms.

[89S10_Project]

More info as I was just farting around with it-

coolant temp, oil pressure, volts all check as fine on the dash (yes I realize how GM dashes are notorious for being wrong).. I've verified they're damn close with a DVM for the battery/alt, mech oil pressure gauge for the oil, and the CTS (gauge sensor) is brand new. Also, I am running an electric fan setup using a BMW dual zone switch and aftermarket fan controller (SPAL, one each for the high and low circuits) running a dual fan, dual speed setup from a Dodge Intrepid. My cooling system is just NOT the problem, pretty much willing to bet any of you a bottle of fine Scotch on this one (thanks to Tom's extremely helpful input!).

TPS, IAC are questionable. Need to look into backprobing them since I can't get data out of the ECU.

O2 is brand new. It's positioned just aft of the joint where the banks come together in the y-pipe. This emulates the factory 5.7s of the era as far as I can tell.

I do not detect an odor of rich exhaust.

Back to the drawing board...

[89S10_Project]

OK, tested the other unit (FT232RL, I think?) with the TXD and RXD wires bridged into one which connected to pin E (160 baud ALDL data) and then GND to pin A (ALDL GND). Am still not able to poll data, but it does pass the test process in Tunerpro.

Believe I am doing something wrong in tunerpro, need to climb a bit farther up the knowledge tree on this....