Tuning progress so far, help much appreciated!!

[yorkshirespud]

I thought I'd convey my attempts so far to tune my rebuilt big block with Edelbrock multi-point TBI based intake manifold. The ONLY result I'm aiming for so far is to get a reliable idle with a reasonalble AFR (Air Fuel Ratio), and be able to set the distributor base timing. So far I've had variable results ...

The new engine has bigger than stock injectors (to cope and supply with higher top end horsepower), so I have assumed I need to alter the ECU tune/map to at the bare minimum stop the idle from running way too rich with the bigger injectors. As such I have also installed a modified ECU with accompanying Moates Ostrich (V1) real-time emulator connected to my laptop running TunerPro. Also connected to the laptop and TunerPro is a USB ALDL diagnostic cable, and another USB cable to the WideBandCommander with wide band oxygen sensor installed, monitored by the supplied WideBand software.

http://www.ntalbot.demon.co.uk/DSC00362.JPG

As I said ealier, all me and my able assistant Dr Frank are trying to do at this stage is get a reasonable idle so we can finalise the distrubtor timing. I am using (an internet sourced version) of the AZFT standard bin as a starting point. I would rather use the Edelbrock chip that came with the Edelbrock multi point intake manifold coversion (which itself was a replacement for the stock 1987 ACLP chip) as a starting point, but I have never been able to get a 'good' reading of it. I obtained a duplicate chip from Edelbrock, and removed the chip holder from one of the two identical Edelbrock chips and put it in my Pocket Programmer (USB edition) chip reader/writer, but I've been able to read and re-verify the chip successflly. Also, what I have read seems to have some odd values when viewed using TunerPro. Therefore I decided to start with AZFT for tuning the new engine.

For the initial engine 20 minute break-in fireup, I didn't want to have extra computers involved, and didn't need idle as I was after 2000 rpm for 20 minutes to break in the cam. So I used the existing unmodified ECU (with Edelbrock chip) for the break in. As expected this ran way too rich because of the bigger injectors. Once break in was complete, I switched to the modified 'real-time emulation' ECU, with mixed results ...

The connectivity of the Ostritch input and ALDL output isn't as reliable as I would like. Every time I attempt to tune the truck in a particular session (e.g. an afternoon), the same pattern occurs. After turning the ignition to position 2 (run, but nor crank yet) I upload the AZFT bin to the Ostrich, verify, and emulate. This always works ok. Interestingly, if try to connect ALDL output before this, ALDL doesn't give any output until the bin is uploaded. Then I test the ALDL before cranking by pressing the gas pedal, and checking the ALDL dash of TunerPro registers the TPS (Throttle Position Sensor) change. This works ok too.

The ONLY variable I am altering in TunerPro is 'BPW Constant' in an attempt to get the ECU closer to reality with the bigger injectors. The truck will usually fire up and run, although the fuelling and ignition seems to be all over the place. In the first minute or so of running the ALDL dash gives all the info expected, TPS, MAP, RPM, 02 (narrowband) voltage, O2 cross counts. Also, if I alter the BPW constant and update immediately in TunerPro while emulating and keeping the throttle constant, the engine note certainly does change, indicating the emulating is working. Without fail though, after some time (not sure how long, a 2 or 5 minutes maybe) the ALDL stops working, and emulation changes seem to have no effect, even if I kill the engine, swutch back to off and to 2 again, restart TunerPro etc. While writing this post, it has occured to me that the only reason it all works again if I leave it all and come back the next day is because we disconnect the battery leads whenever we're finished at the end of the day. Maybe that is resetting something in the modified ECU.

Anyway, with the above setup, we can usually get the engine running, and idling eventually after it has warmed up. I must note that the running characterstic of the engine seem wildly different when cold and warm. There's no chance of it idling when cold, and then starting when hot seems difficult. Actually, the enginr running seems all over the place, sometimes it seems to be running and sounding great, other times it feels like it's missing cylinders, and the timing is all over the place. Using a timing light, we've had everthing from 60 degrees out, down to almost the 4 degrees it shoudle be at. Admittiedly, we keep having to move the distributor because in one position it will start but not idle, and another will idle but not start. The WideBandCommander I have installed is also showing massive swings in the fuel mixture all over the place.

I have also installed a MSD 6A ignition, but I disconnected it pretty quickly and changed back to stock to simplyfy setting the timing with a timing light. I'm assuming the modified ECU with Ostrich is working correctly, as it the engine does fire up with it connected, but really it's another variable that I'm hoping is 100% ok but of course I don't really know.

After a couple of now failed attempts trying to set the timing at idle, I'm open to suggestions and helpful hints. My current plan is to strip as much back to basics and remove as many possible red herrings as possible. I'm going to:

Change Oil and Filter again (probably a lot of petrol in it now)
Change the plugs again
Fit the new plug leads I have (been using a pre-rebuild set so far)
Readjust the roller rockers again
Put the dash back in again
Put the old ECU back in again AND swap the new injectors out and put the old smaller back ones in again

This, HOPEFULLY, will let me get the truck idling enough so I can set the timing and clamp it down to remove that as a variable. Then I'll put the bigger injectors back in and the modified ECU.

Even if this works, I still am a bit in the dark about getting any further with the engine tuning as it seems a bit all over the place. Hopefully with settled timing, tuning the igntion and fuelling will be easier, although the 5 minutes only of emulation control and ALDL output is a worry. Does anyone know if the AZFT is a good or reasonable place to start? Also, I think it might have open-loop idle (no use of o2 sensor at idle, run from tables only), but I don't know for sure, anyone?

Any and all comments, suggestions and help welcome!!!

http://www.ntalbot.demon.co.uk/DSC00365.JPG

[Red94Chev]

Need a few more specifics. Which ECM? How was it modified? Which Mask ID? Do you know the hp output of your engine?

The Edelbrock injectors that came with the unit should be 19#/hr (but actually flow 20.5#/hr) and will support hp levels up to 300 hp. The Edelbrock MPFI unit can be modded to work with an aftermarket AFPR to increase fuel pressure and therefore increase the supported hp supplied by the injectors. Did you switch them out for the 29#/hr injectors?

The first step would be to put in the correct value for the BPW constant. Most people think with larger injectors, you need to put a higher value in. That's not the case. A lower value is what you want. I'll assume you have a 7.4L (454 cid) engine, since you mentioned big block. For the 19#/hr injectors, the BPW constant should 146. For 29#/hr injectors, the BPW value should be set to 93. If you are using a larger or smaller engine, the BPW numbers will be different.

BPW constant = 1461.5 * (Cyl Vol /Flow Rate), where Cyl Vol = displacement of 1 cylinder in Liters, and Flow Rate = Grams/second for each bank of injectors.

If you want to confirm my calculations, there is a website with a BPW calculator: http://www.fortunecity.com/rivendell/oddworld/24/gm_inj.html

Just so you know, even though you have a MPFI setup with the Edelbrock unit, your ECM is controlling your engine using the TBI injector firing strategy and therefore your injectors are firing twice as often as batch injection strategy used in a TPI and other MPFI engines. This means that it may be difficult to achieve a reasonable AFR at idle since the computer is unable to adjust the injector pulsewidth small enough to achieve its target AFR. The result is the engine will run rich at idle and operate normally at part throttle in closed loop operation. Using larger injectors may make this problem even worse at idle, but will support higher hp levels at WOT and high RPM.

In some cases, the ECM can be modded to the batch injection firing strategy. The only one I know about is with the 16197427 PCM. Fast305 and HaulnA$$ have done the conversion: http://www.fullsizechevy.com/forums/general-discussion/performance/tbi-tuning-87-95-obd-i-ecm-pcm/289473-tpi-tbi-pcm.html. Ken at HPTuners did this too with the conversion process detailed in his post. Lucky for you, he has also supplied a downloadable BIN for the 7.4L engine running in PFI mode. The link can be found here: http://www.hptuners.com/forum/showthread.php?t=9228

[yorkshirespud]

Hi Red! Thanks for the reply, I'm all ears for any advice.

Need a few more specifics. Which ECM? How was it modified? Which Mask ID? Do you know the hp output of your engine?

The ECU is a bog standard DIY EFI 1227747 ECU. In TunerPro I'm using some version of 42.xdf, which I believe is the match for that ECU. The 'modifications' I've made are the standard Moates suggestions to allow the ECU to be interfaced to an Ostrich emulator, i.e. desoldered the PROM mount and soldered in a 24 pin socket to allow the Ostrich ribbon cable to be connected.

http://www.fullsizechevy.com/forums/general-discussion/performance/tbi-tuning-87-95-obd-i-ecm-pcm/297188-finally-getting-somewhere-truck-engine-tuning.html


The Edelbrock injectors that came with the unit should be 19#/hr (but actually flow 20.5#/hr). Did you switch them out for the 29#/hr injectors?

For 29#/hr injectors, the BPW value should be set to 93. If you are using a larger or smaller engine, the BPW numbers will be different.


Yes I've removed the existing Edelbrock 8 x '19'lb/hr injectors and installed 29lb/hr ones. The engine is indeed a 7.4/454, overbored by .030, making it really a 7.5/460.

I'd realised that I needed to trim the BPW constant, and I seemed to get the best results about 85-95 set in TunerPro (when the emulation seemed to be working), so it's good news to hear you've come up with the same numbers.


Just so you know, even though you have a MPFI setup with the Edelbrock unit, your ECM is controlling your engine using the TBI injector firing strategy and therefore your injectors are firing twice as often as bank injection strategy used in a TPI and other MPFI engines. This means that it may be difficult to achieve a reasonable AFR at idle since the computer is unable to adjust the injector pulsewidth small enough to achieve its target AFR. The result is the engine will run rich at idle and operate normally at part throttle in closed loop operation. Using larger injectors may make this problem even worse at idle, but will support higher hp levels at WOT and high RPM.


I hear what you're saying about idle. I'll worry less about getting a perfect idle and AFR, and make do with a rich idle. There have been moments when the engine has been idling really well, unfortunately it doesn't seem consistent at the moment (which I hope/think is related to the short lived ALDL/Ostrich connectivity). Your comments seem to lend weight to my idea about putting the pre-rebuild ECU and smaller injectors back into the engine JUST to get the base timing on the distrubutor set correctly (and change it back again). I've got to get the timing set to remove another variable before I can get any further.