Well, I finally finished messing with a few things and have larger injectors installed. They are 42pph, the car is an ’87 L98 running what started as ’89 AUJM $6e code which was running well with the stock injectors (there was a slight off idle stumble for about the first 30 seconds that it was running but after that it was pretty happy).

I set both the single fire and double fire size to 42 and it won’t run.

When I crank (no foot on the throttle) it it fires instantly, right away revs up a little like it always did on its own, and then if I don’t do anything it dies. If I give it throttle it will pop a few times and then rev up, but I can’t get it to run well below about 3K rpm. After it cranks I can hit the key and it will crank and start instantly but does the same thing.

Based on that I figured that it was probably lean since usually if a car is so rich that it dies it’s hard to start after that, no gas smell either. I tried adjusting the fuel pressure to a few different positions a few steps at a time until I got to the maximum pressure of about 60 psi, and it acted the same the whole way. I tried going backwards, a few steps at a time till I had the screw all the way out at about 30psi and it was still acting the same.

On a whim, with the FP all the way down, I popped the previous prom in (exactly the same but with 22.3something in for the injector constants, whatever the stock L98 setup is), and again, it starts instantly, but it dies right after the initial rev, but if I give it some throttle I can get it to stay steady above about 2200rpm, and after a few seconds of that I can let off slowly and it will somewhat idle. By somewhat, with the engine temp it should be around 1200rpm, but I’m getting what looks like on the gauge a really rough, 3-400rpm idle which surges momentarily 1500 every few seconds (it's acting like what an aftermarket ecm acts like if you have your idle way lean).

Any ideas how to get it to run?

 

Im interested in this, too... My car did the same thing stock. Fixed it for the most part but never really found out exacly why it did what it did. Is it just the fueling? Spark?

 

Why did you go with such large injectors? If you don't any or many major mods, thats *way* over-injected.

You might try spiking your injector constants up above 42 (try 50!). Unfortunately, your BPW will probably excessively short and the injectors "lose control" so to speak.

And finally, whats the scanner saying? Rich? Lean? BPW?

 

ok not sure what all you did before installing the injectors but in some instances espically after flashing trouble codes, I would have the same problem on my Stock LB9, and Grumpy told me that some times the IAC in some instances has a 30 MPH reset Speed. So I started the car, while feathering the throttle to keep it running and took a dirve down the road, one I hit about 30 mph the IAC reset and all was well.

May not pertain to you but It is worth a shot, maybe
Just my .02

 

The engine is a dead stock L98, but I plan on bolting the turbo setup off of the engine I have sitting on the engine in the corner on an engine stand (before somebody offers me enough money to sell it off, I’ve gone down this road 3x before with this car). I’ve got enough turbo for about 800hp, I figure that the engine will be pretty unhappy somewhere in the 500-600hp range without race gas, and I figured that the 42’s are a happy middle ground for experimenting (to see if the ECM would deal with them) before I get serious and install some 77’s like we’re using on my brother’s car.

FWIW, if I can get the thing running (injector drivers capable of dealing with it) I’m splicing a MAP into the harness and running some 808 stuff with the boost…

Well, that’s the experiment that I’m running with this. Everybody and their uncle seem to have told me that “with saturated injectors you just need to set the injector constant and go, they’ll work fine.” While I’ve never seen a car that idled right with a stock ECM with >36pph. I’ve seen a few run with 36pph injector cars run, but usually they idle rich.

What’s driving me nuts is that with the stock injector constant and almost 2x the injector installed it was idling like most cars act when they are lean, sorta a low, surging idle. I’ve got a lot of experience with tuning things with aftermarket setups, but this is my first try at the GM stuff. I’m really missing the “plug the laptop in and tweak it a little and see what happens” part.

I’ll try messing with the injector constants, but I would think that I basically covered that end by changing the fuel pressure, I’ve tried rougly 30-60psi, which should make the injectors act like just about everything between 30 and 50pph injectors with the constant set to 42… wouldn’t that accomplish roughly the same thing?

For that matter, do you have a specific reason for saying 50? That should make it run leaner, where all indications are that it’s running lean now.

Scanner? I doubt that I’ve had this thing run more then about 20seconds at a time, it really doesn’t want to run right. Since there have been no changes to any hard parts in the engine besides the injectors, the only thing that I see any reason why anything should be off besides the fueling and it’s not running long enough for a meaningful O2 reading. Am I missing something, what would I see on a scanner that would be useful?

Well, it’s not running well enough to even try that. I’d hate to try to shift it into reverse at 3K rpm with the car sitting on ramps, in the garage facing my 4th gen and try make it down the driveway… and I somewhat doubt that it would stay running even there, and considering the slope of my driveway the rest of my experimenting (wrenching if necessary) will end up in the street.

 

Actually, all this reminded me of a couple of questions:
- “single fire size @ 40 PSI” (same name for the double fire)… does that imply that the factory injectors are rated at 40 PSI, or is that just what someone typed into the ECU file and stuck?
- Is single fire once per revolution or once per power stroke? I’m assuming double fire is 2x that…

 

While I haven't played with the 165, really large injectors can take more work.

I wound up having to close down the IAC. Both max, and the park positions. Might be due to needing to keep the intake velocity up when cold and the fuel ain't vaporizing. With the shorter pulse widths the fuel is going to be more dribble then spray.

Typically there's a number of items that need to be adjusted, on the 31T code there is a base crank PW value, then all your crank PW vs ref pulse, and PW vs temp. THEN you can get into the choke type stuff.

 

I’m not sure that I’m following. The $6e has a “target speed deadband” for P/N and for drive. The value is RPM (drive set to 12.5 and P/N to 50), and I assumed that was the acceptable error before the IAC starts trying to compensate, but I could be wrong.

There is also a min and max “keep alive learning temp” which I assumed that this has no part in this since them min was set to 185* in the original bin.

Finally there is a “steps added for fan anticipate” which makes no sence to me since the only effect that I could imagine the fan having is lowering the battery voltage and I’d expect that that should be compensated for the batter voltage compensation.

Did you mean one of these or something else?

Do these effect anything besides the initial starting conditions? Like I mentioned, I hit the key and it cranks and starts beautifully, it just doesn’t want to run past that.

 

This doesn't help at all but the biggest I've worked with are the SVO 42's on $8E code at 43psi and it idles great. So, I know that code is capable of providing a very nice idle. I haven't worked with those size injectors on MAF so good luck. If you are going to run boost then why not just convert over to the $58 (749 or 730) BEFORE working with your injector swap? This way you can practice your tuning on a known working combination with the new ECM first. THEN do your injector swap. Everything might be for naught since you have to swap ECMs anyhow (to run boost effectively). However, with THAT much turbo it sounds like you might need the 3-bar $58 code?

ALSO - do you know these injectors are 100% good? Did you get the new or used? How old are they?

Tim

 

Actually it does help a little if you ran it in a 730, since the injector driver is really similar to the 165. I’m assuming that this is with the stock injector driver?

Because I have no real intension of converting to $58 (at least not initially). I’ve bounced the 808 in a 165 stuff back and forth with JoBy and it seems to work quite well if you just scale the maf tables for whatever MAP that your using. I plan to rescale the 2 bar stuff that I’ve got that’s worked out to work with a 1.75bar map and that should be a fairly nice combination for initial tinkering as well as providing the best possible resolution. I doubt that the stock L98 will live happily for very long with more then 11psi boost anyway.

Used, no idea of the age but they tested fine on my flowbench, all within 2% of each other. I’m fairly certain that they’re still happy in the car since it runs smoothly if I give it enough loud pedal to keep the R’s up.

 

Yes, stock injector driver.

Tim

 

Starting is different from the IAC stuff when warm or after start.

As the temp approaches fan turn on, the IAC cracks open a few counts, so that the engine doesn't stall or trigger the stall saver.

There's a huge shot of fuel during crank, and with the IAC open, it'd have to be a real mess not to even burp to life,
But, with the big injectors, the burp is really large, and with the *choke* AFR, it might be just instantly drowning the engine in fuel. So much so that it doesn't even get a chance to look rich.

 

Ok then, I'm running 60 PPH injectors in my 749. And have also run 40s, and 55's with the PnH driver swap, and even ran the 55s with a normal saturated driver 148 for a while.

 

If you still haven't gotten it running at idle, and you think the IAC is fine, turn your injector constant down (try 36 to start) and idle in open loop (with an open loop offset of 13% or more). That big of an injector is causing the idle pulsewidth to get extremely small. Saturated injectors are very non-linear (read: could fire no fuel at all, or dribble it, or fire too much but not likely) under 2 ms and with those injectors, even in single fire mode are well below 2 ms if the ECU is commanding a 14.7:1 A/F. It's going to idle stupid rich, but it's not the ECU's fault, it's those injectors that are twice the size of stock. You could also turn your fuel pressure down to help. That's why peak and hold injectors are better for wider dynamic range (or dual staged injectors, or an FMU/rising rate regulator on smaller injectors). I'm pretty sure the 165 doesn't have any injector linearization calibration to it as some other GM ECMs have, it only has the quasi-async / single fire mode.
You could also try raising the idle area on the MAF table just to get the PWs up to where the injectors fire, and keep it out of closed loop.

Good Luck

 

yes, and I asked exactly that. It cranks and fires, but doesn’t run after that. You suggested “closing down” both the min and max. I listed the IAC settings I have in the 165 trying to figure out what would change what you’re getting at. But I’m still in the same place as I was before. I don’t understand what you mean by “closing down” min and max, and I still don’t know if the reason that it dies after firing is because that initial fuel/IAC is insufficient or I should be looking someplace else.

yea, I said that already also. I understand that, what I don’t understand is why it would stall. What effect does the fan have on the engine running besides maybe the drop in available voltage which should be compensated by the voltage compensation in the code?

If this was the case then wouldn’t it be a bear to restart? Every time I’ve ever seen an engine running so rich that it dies the engine ends up flooded and then it’s either difficult to start or impossible without waiting. This is why I’ve specified more then once that after it dies I can just hit the key and it fires right away.

 

Nope, I got so disgusted with it that I just let it sit since last night.

How would I know?

Ok, that should make the engine try to deliver more fuel then it thinks it is… how is this different then me turning the fuel pressure up to 60psi? Or trying it at with the injector constant set at 22 and the fuel pressure at about 35psi?

I can try it, it’s no big deal, but effectively you _should_ be accomplishing the same thing and if it does work I’d like to understand why so in the future I don’t have to bug you guys with “dumb questions.” (and maybe all the people that keep PM’ing me to explain what it going on in plain English could, heaven forbid, just look in the freaking archives and learn what they need to know to get things running.)

I understand what you’re getting at, and what you’re telling me (and I’ll try it after dinner), but it does not match what I’ve seen on an injector flow bench and aftermarket ECM’s. A perfect example of this is with a haltech we went from 36pph saturated injectors to 77pph saturated injectors, having done some reading we figured that same way and cut the fuel maps in half (figuring that would somewhat compensate for the lack of linearity and at least get it running well enough to start tuning). Well, it was rich everywhere. Finally when we cut the fuel to 45% (almost exactly the change in injectors), the whole fuel curve was just about perfect, the car ran beautifully and the only additional tuning that it needed was at the top end where before the injectors were going static so obviously the previous maps weren’t right there. I’ve seen the same deal on the flow bench. I know for a fact that these injectors are still fairly linear at the pulse widths that we’re talking about.

This is also the reason why I’ve posted my previous concerns about the OEM injector drivers. If I know empirically that the hardware that it’s connected to is working right, then that only leaves the wiring, the code or the actual ECM.

huh, why would closed loop make a difference here? I haven’t even managed to keep it running long enough to see what it will do in closed loop.

You know, we’ve been down this path before. Assuming that your 749 still has the OEM injector driver in it (does it?), then you have 2x the injector driver and 2/3 the injectors, basically, you’re driving 3 injectors with the same driver (or at least similar, the circuit is slightly different) that the rest of us (730 and 165) are trying to drive 8 with.

Essentially, as far as I can tell, the injector driver circuit is very similar to a simple mosfet audio amp. What happens when you wire too many speakers in series to the output on an amp? First the response gets sluggish/muddy as you exceed what the power supply can deliver, and if you keep going you let the blue smoke out of it.

Now based on what Tim said, apparently this is not the case, especially if he managed to get it to work with the FMS 42’s which are so slow that ford dumped them for 2 36’s per cylinder.

 

It's the Pulse WIDTH that is what I'm thinking is causing the problem. When it's below 2 milliseconds (no matter what fuel pressure) on saturated injectors it gets very non linear very quickly. The injector constant will get the idle pulsewidth up to a linear range (although really rich, likely) but it'll run. On my stock TPI setup with my company's ECU controlling the fuel in double fire mode (batch - 50% per injection) any pulsewidth below 2 milliseconds at idle (closed loop on) and the car almost dies. I can manually force the pulsewidth to whatever I want and it idles fine at 2.1, 2.05 and at 1.9X it just leans out and dies. And that's with a wideband watching the A/F be fine up til that point, and with an oscilloscope watching the injector signal. If I wanted to, I could idle rich, but instead I'm having my coworker code some "quasi-async" (single fire - 100% per pulse - we already have it for TBI single injector 4 cyl setups).
Changing the injector constant would just be the "easy" test to see if that's the problem. If it is, then I think there isn't much choice but to mess with the MAF tables at idle, and keep closed loop off at idle (since it'll be trying to keep 14.7 on a steep slope of injected fuel per pulse duration, where the A/F would swing rapidly with a little change, and the car would likely oscillate and die.
Another test could be to put some load on it, such as put it in gear (while keeping your foot in it) and see if the extra load brings the PW up enough to keep it running and if the IAC is working you can take the foot off and watch it idle.

Unless there is something else mechanically wrong with your setup, like IAC, or something blocked, or not all injectors firing yet (sitting on a shelf can cause them to stick), or something not re-installed correctly, I think this would be the most likely reason for the troubles.

Good Luck.

 

Well, I did burn 2 more chips before heading off to the gym. Both had the injector constant set to 36, one of them had the single fire flag set (I double checked the fuel pressure also and found that it was set to 45). Both ran (I use the term loosely) exactly like the other combinations.

I did figure out why the chip with the constant set to 22 ran differently, when I changed the fuel pressure I forgot to reconnect the vacuum line to the fuel pressure regulator, but that doesn’t really clarify what is going on, since the vacuum leak should have made it run leaner but the higher fuel pressure should have made it run richer. Overall it should have seen the same mixture but less(?) load (I’m thinking the maf isn’t seeing all the airflow…)

R&G, what injectors are you running that you’re seeing sub 2ms pulswidths? I’d have to double check but I’m fairly certain that my brother’s 302 is just barely under that with 77pph injectors in a 302.

On only a somewhat related note, is there a problem with running single fire (I’m assuming that it is actually once per revolution)?

 

I'm not sure, but I'm thinking you may be a little confused about one thing. Larger injectors don't affect how much injector driver current is needed. The larger flow injectors just have a larger orifice at the restriction point, but the coils that open and close them are generally similar. If you measure the impedance (measure the resistance basically) then anything above 12 ohms will be considered a saturated style injector and your ECU will handle it just the same as a 19 pound per hour injector. I'm not sure what those 77pph injectors were, but as I said, they get non-linear below 2 ms, which doesn't mean they have the flow has to go down, it just isn't predictable, and the flow isn't proportional to duty cycle. The injectors I was speaking about were my stock injectors and they are exceptionally sensitive, but I've calibrated some cars that seemed to run just fine below 2 ms, but they still had troubles when it got a little lower (I'd say 1.5 is the limit on saturated injectors probably 1.8). Oh, and the scantool reading of 2 ms is not 2 ms of pulsewidth. It's more like 2.5 ms.

Single fire and double fire are the terms used in the TunerPro hack (I usually use quasi-async vs batch). Double fire is the normal mode, where there is one injection on each injector per revolution which equals 2 injections (at 50% of the total fuel required each) per intake stroke per cyl. Single-fire fires the injector once every other revolution (100% of the fuel - once per intake stroke). The only problem with single fire is that on one cylinder there is going to be an injection on an open intake valve which doesn't allow time for the injection to sit and vaporize on the valve (many people have the misconception about sequential injection and when it's supposed to inject, and the correct time is to inject on the closed intake valve) and the emissions may suffer a little, and that valve may run hotter, and that plug may read differently (little leaner from less vaporization), but it probably won't be noticable.

 

I understand that, but what I was getting at was that the more current that is available to the injectors the faster they will snap open. The OEM injector driver (for that matter, all the connections on the oem ecms) are current limited, the injectors are not allowed to draw whatever their impedance would allow them to. After taking a look at the board I doubt that you would hurt the OEM driver plugging in P&H (low Z) injectors, they would just be very sluggish, if they opened at all.

They have the same coil and moving assembly as FMS/SVO 36pph injectors with different metering plates. Their coils measure the same resistance (I seem to remember 12-14ohms, but it’s been a while), the only real difference is that the metering plates give them more of a GM style spray pattern then the original ford/bosch pattern (they spray in a narrower pattern rather then fords wide fan spray)

I’ve run the 77’s down to 1.2ms on a flow bench and I just called my brother and he says that he has them idling at 1.4ms. No doubt that some injectors don’t play nice under 2ms, but I know that the 77’s and the 42’s that I’m trying now are OK. I suspect that it’s more that “some injectors get sloppy down there and some injector drivers get sloppier). I had a knock sensor strapped to the injector on the flow bench and then hooked the driver signal to one trace on an o’scope and the knock sensor on the second so I could compare the signal to what the injector is doing. Of course, I wasn’t using a mosfet driver like the OEM ecms use, just a pair of darlington transistors (only one was necessary, the second was wired to provide the “hold” for peak and hold injectors, so you could set the time and limit the hold current separately to measure operating thresholds)

Thank you for the explanation.

Now on a positive note, before I went to sleep last night I finally had a cold idle, I don’t know what it will do when it’s warm or if it’s driveable, but it did idle smoothly (it was 4am and I actually was messing with it with the garage door closed to prevent unnecessary hassles with the neighbors. I probably let it idle for 45s or so (I was standing outside of the car, so I couldn’t work the throttle), shut it down and left it alone.

On a negative note, I don’t know what I changed that fixed it. It is idling with the original chip that I burned at the beginning of all this, same setup as before the injector swap but with the injector constant set to 42. I did go through and double check all the vacuum lines and sensor connections that I had off when R&Ring the manifold to swap injectors and replacing the water pump, but nothing appeared to be loose before I messed with it.