My Dual TBI Camaro Project

[gregs78cam]

Ok, I have been working on this for a few months now, everything is together, the engine runs but has some issues that I am working through. A little bit about the setup: 2pc rear main, 385 i built about 8 years ago, 10.5:1, 6" rods, TFS 23* heads, jesel belt drive, roller cam with 4/7 swap. It spent most of it's life in front of a T-56 with 4.10 gears,(dynoed 340hp/370Tq, and rich) until I just got tired of always shifting. So I found a 4L80E and decided while I was at it I might as well do the EFI since it could all be done with the same PCM. Now I know I could have used a 454 TBI with 454 injectors, but they are usually either hard to find, or cost more than I was willing to spend. So one day I stumbled across a CL ad selling 6 TBIs off various vehicles. I went go to get them and it turned out to be 7 for $35. So after cleaning up two matching ones(for the linkages) my Dual TBI was born(and inexpensively too) I am using a '7427, with Rbob's 4-injector upgrade, APU1, and TPV5. I could have used Rbob's fantastic EBL, but again the cost issue.

1. It was running way too rich at a "idle", I got this issue corrected by changing the injector size from 55 in the original .bin to 120, due to using 4 injectors

2. "Idle", I use this term loosly right now (1300-2200rpm). I am pretty sure it is an IAC issue, this morning when I put the ecm in diagnostic mode(jump A/B on ALDL), the pintle on the IAC remained retracted. ??? When key off the pintle extended. I have read on another site that some oldsmobile IACs were wired opposite. IDK if that is true but tends to make sense with way the engine idled(temps went up, idle went up instead of down) So with it extended, I unplugged the IAC and will try again tonight when I get off work. EDIT: problem found and fixed in Post#9.

3. Datastream communication. I am having a difficult time getting useful data consistantly from the AutoProm to the PC. Friday, before work, I hooked up the PC, loaded TP, and got everthing connected, car started, idled high but I had good data coming out. I let it run for a couple of minutes and shut it off. At the next key on all data was wrong, or flashing stupid numbers, like 345 degrees spark advance????. Any who, my next step will be to just burn a chip with Cyl. count, injector flow, and displacement changed from the original .bin, put the chip in, and start it with the IAC disconnected. EDIT: Found that you need to do it in this order: Open bin., Change MASK ID to AA, upload to Emulator, Then Emulate.

Where I go after that depends on what happens.

[EagleMark]

Hey! Another North Idaho guy! Where you at?

What's the cam specs at .050 ? Idle is a lot of work with big cams. They usually have to idle high to get enough STEADY vacuum for the MAP sensor to give ECM reading it can deal with.

Your IAC has to be correct first of course. Also Big Block Chevy TBI came from factory wired with 2 wires switched. Should be enough diagrams in our library to figure out all you have there.

Some things I do with big cammed engines, in no particular order because you usually do them all several times is. IAC closed, set min air on throttle blades, set with bypass wire disconnected,timing, sent min air, set timing, look at vacuum reading, is it steady? probably not, but give it what it wants with no interference from IAC and timing to get an idle.

In the chip disable EGR because you probably don't have one. Disable closed loop idle. Match base timing to what you have at distributor or it will not know and you will end up with more than total advance. Look for timing advance for that motor and better if the cam manufacturer has a timing table, plug that into the chip tables. With a 427 type ECM you will plug in injectors size and cubic inches and I forget what else. But a 7747 ECM puts them all in one lump called Base Pulse Width constant and is harder to figure out to start.

If you have a wide band O2 sensor use it for idle.

Go back to first part and start all over with IAC, timing and O2 sensor all disabled and set min air, set timing. Give the motor what it wants! Get it to idle at whatever RPM even if it's 2000. Then drop min air, drop timing. Play with both of these. Since you have an AutoProm use it emulating and usually drop fuel. Then min air, then timing. There may be a set procedure but I have never seen one and I have watched experienced tuners and they did exactly what I did and learned and am telling you.

While your doing all these things motor has to be warmed up to start and then rev the motor up once in awhile to clean it out so you are getting true idle, not adjusting a loaded up engine.

HTH!

Oh yaeh I forgot...
:ttwwop:

[gregs78cam]

I am up in Spirit Lake, you are in CDA right? I am pretty sure we conversed a bit on TGO.

I did not know the BBC IAC had two wires swapped. But yes first thing is to get it to idle at a decent rpm, and not pig-rich. It idled at about 700rpm with a carb, so that is my target with efi. As of Friday, my problem was more with getting a solid data link between the PCM and laptop.

Cam is 224/230 @.050, 0.525/0.524 lift. Not real big for two reasons. I knew I wanted to go EFI eventually, and the engine was built for lots of torque down low, not HP @6500rpm.

I do have an LM-1 and it is running through the PCM in the datastream.

I am at work so i don't have any more pics but here is one from 2003 when I built the engine and another from my post on TGO.

[EagleMark]

I think I remember talking to you on TGO as well because of that adapter you made! :wtg:

What's this motor in ?

[gregs78cam]

It is in my '78 Camaro rust bucket test mule. Once I get the drivetrain sorted out, it will all go into an '81 T-top, Z-28 clone.

[Nasty-Z]

FWIW ,

454 TBI IAC wiring :

PIN A - Coil A Low - Lt Blu/blk
PIN B - Coil A High - Lt Blu/wht
PIN C - Coil B High - Lt Grn/wht
PIN D - Coil B Low - Lt Grn/blk

This is the 454 IAC wiring at the valve , trace back to the proper pins on your ECM to be sure

HTH

TOM

[dave w]

Ok, I have been working on this for a few months now, everything is together, the engine runs but has some issues that I am working through. A little bit about the setup: 2pc rear main, 385 i built about 8 years ago, 10.5:1, 6" rods, TFS 23* heads, jesel belt drive, roller cam with 4/7 swap. It spent most of it's life in front of a T-56 with 4.10 gears,(dynoed 340hp/370Tq, and rich) until I just got tired of always shifting. So I found a 4L80E and decided while I was at it I might as well do the EFI as well since it could all be done with the same PCM. Now I know I could have used a 454 TBI with 454 injectors, but they are usually either hard to find, or cost more than I was willing to spend. So one day I stumbled across a CL ad selling 6 TBIs off various vehicles. I went go to get them and it turned out to be 7 for $35. So after cleaning up two matching ones(for the linkages) my Dual TBI was born(and inexpensively too) I am using a '7427, with Rbob's 4-injector upgrade, APU1, and TPV5. I could have used Rbob's fantastic EBL, but again the cost issue.

1. It was running way too rich at a "idle", I got this issue corrected by changing the injector size from 55 in the original .bin to 120, due to using 4 injectors

2. "Idle", I use this term loosly right now (1300-2200rpm). I am pretty sure it is an IAC issue, this morning when I put the ecm in diagnostic mode(jump A/B on ALDL), the pintle on the IAC remained retracted. ??? When key off the pintle extended. I have read on another site that some oldsmobile IACs were wired opposite. IDK if that is true but tends to make sense with way the engine idled(temps went up, idle went up instead of down) So with it extended, I unplugged the IAC and will try again tonight when I get off work.

3. Datastream communication. I am having a difficult time getting useful data consistantly from the AutoProm to the PC. Friday, before work, I hooked up the PC, loaded TP, and got everthing connected, car started, idled high but I had good data coming out. I let it run for a couple of minutes and shut it off. At the next key on all data was wrong, or flashing stupid numbers, like 345 degrees spark advance????. Any who, my next step will be to just burn a chip with Cyl. count, injector flow, and displacement changed from the original .bin, put the chip in, and start it with the IAC disconnected.

Where I go after that depends on what happens.

The '7427 has two VE tables and two SA tables. Maybe for the time being it might be a good idea to set both tables to the same value? I think I would try using one IAC, and leave the other IAC unplugged. As for the '7427 injector flow, I've had to set stock 55 lb injectos to 61 lbs, and 90 lb injectors to 85. Setting the injector flow in the PCM does not always match the injector flow rating!

Personally, I like DataMaster TTS for data logging the '7427. I use the DataMater Histogram to set the injector flow until most cells are in the ideal mixture.

dave w

[dave w]

An excellent picture of a TBI running WAYYYYYYYY to RICH.

dave w

[gregs78cam]

OK, Let's try this for the second time.

IT RUNS!!!!. The High idle was two problems. Vacuum tube on back of rear TB (MAP connection) was plugged. And in order to plug off the rear IAC passage, I gutted an IAC, threaded the inside, and used a 1/2"NF set screw with a jam nut to simulate a pintle seated against the TB. Well the set screw was not screwed in all the way, hence a vacuum leak. Idle is little high yet ~825-950 when desired is 788-850, but I think it is because the IAC won't extend past 27 counts, no matter what. It's livable though.

I am still having issues with the AutoProm. I can burn a chip, put it in and datalog though APU1 and TP, and start it as many times as I want. It's emulating that is messed up. If I disconnect battery to reset PCM, hook up AP, open TP, load .bin, and keep verifying until succeed, turn on emulate, turn on Data Aq., then start engine, I can emulate, but as soon as I turn off ignition it's done. Key on after that and nothing, won't start, no data, FUBAR.

I am going back out to clean up get it off the blocks and hopefully take it for a drive tomorrow.

[Six_Shooter]

When emulating you need to disable checksum by changing the mask id byte from whatever it is (I.E. OD, A1, 8D, 58, etc) to AA.

[gregs78cam]

Tried that, didn't work. Do I need to change it in the .bin that I load into the Autoprom, along with the .bin that I have loaded in TP? I don't think I have tried that. I just figured if it were emulating it would update the checksum just like anything else I change while emulating. I will try that tomorrow.

Another thing does any one else have problems with this website? Every time I try do something I have to login again, and start over, it took me almost an hour to get my last post to be submitted, I was going to upload some pics, but I kept having to login and start over. Anywho, I will post up tomorrow on how it goes.

Greg

[cmaje72]

Greg,
The website issue sounds like you may need to clear the "cookies" in whatever browser you are using.

Other than that I have nothing to add except that the dual TBI adapter is awsome! Cool project.

[gregs78cam]

I thought maybe the cookies too, but that didn't help, I think it is my ISP.
Thanks for the compliments, I would make it slightly different if I had do it again, but it seems to work well so far. I was kind of hoping that the extra common plenum volume would help extend the RPM range a bit. We'll see though.

[EagleMark]

I looked into website issues and a lot of things went on at server yeasterday. Lets hope your issue goes away.

What are you doing for an air cleaner?

[gregs78cam]

Well I was lazy yesterday, and did very little on the car. But today I finished up a whole bunch of wiring. Just have to vacuum it out, put the seats in, the hood on, and take it off the blocks.

For an air cleaner, I am going to modify a standard 14" to fit over both TBs. If everything works ok then I will make another base for my triangular Edelbrock one. I may have to make the cowl induction scoop taller, or put a forward facing scoop on it, that is what I would actually like to do. I like the looks of the '67 427 Corvette scoop but nobody makes a hood so I will have to attach a scoop.

And the website seems to be better today, so far. We'll see how many times I have to write and submit this post. :doh:

I'll get some more pics tomorrow.

[EagleMark]

Did you ever measure the center to center to see if an old dual carb woud fit? They look a little to close...

[Six_Shooter]

Tried that, didn't work. Do I need to change it in the .bin that I load into the Autoprom, along with the .bin that I have loaded in TP? I don't think I have tried that. I just figured if it were emulating it would update the checksum just like anything else I change while emulating. I will try that tomorrow.

By changing this setting in TP and then uploading you have changed the mask ID in both then. I'm not sure how you would change the mask ID in the AutoProm without changing it in TP first.

[gregs78cam]

EagleMark - Standard Dual Quad measures 6-7/16 on center. My setup is 5-1/4 on center. I did that for clearance to the water neck, distributor, and to make the plenum volume as small as possible, and I knew I could then fit them in one air cleaner.

Six_Shooter - Well the last time I tried it, I loaded the bin to the AutoProm, then turned on emulation, then remembered to change the mask. That didn't work. I will try changing the mask first then load, then emulate.

[gregs78cam]

Oh yea, no driving today, gotta replace the brake master cylinder. :doh: :mad1:

[gregs78cam]

OK. got it down the road to get gas. I was able to Emulate by changing the mask id first. Funny, I never read before that you had to do that. Anyways, I got the thing to idle down, iac sits at 14 steps. Put it in gear, iac opens up to like 45 steps or so, but if you put it back into park, it only closes to like 21 or 24 steps so idle is a little high. If I restart it does it all again. I can upload a datalog if that would help. Any ideas?

I am using the following.bin, def, adx.

[dave w]

OK. got it down the road to get gas. I was able to Emulate by changing the mask id first. Funny, I never read before that you had to do that. Anyways, I got the thing to idle down, iac sits at 14 steps. Put it in gear, iac opens up to like 45 steps or so, but if you put it back into park, it only closes to like 21 or 24 steps so idle is a little high. If I restart it does it all again. I can upload a datalog if that would help. Any ideas?

I am using the following.bin, def, adx.

I've looked at the .bin file that was downloaded. I think maybe to much advance in the near idle table. The reason I say this is because I just finished tuning a 383 with 9.8:1 / Single 670 CFM TBI. I also did a '427 PCM upgrade and used the $OE. The 383 is actually running best with a stock $OD BDWL timing tables and 92 Octane fuel. No knock sensor, with this 383 so I kept timing on the safe side. I also set the initial spark advance, and main spark bias to zero.

dave w

[gregs78cam]

Thank you dave, I will give that a try.

[EagleMark]

You also asked earlier and I did not notice an answer. But yes you have to upload bin to autoprom when you change it in TunerPro. Also when you make a change to bin you need to save it, I think the command is commit! Then it is loaded to AutoProm.

[Six_Shooter]

You also asked earlier and I did not notice an answer. But yes you have to upload bin to autoprom when you change it in TunerPro. Also when you make a change to bin you need to save it, I think the command is commit! Then it is loaded to AutoProm.

You only need to use the "commit" function if you are using the hex editor to make the mask ID change. Editing the mask ID using the XDF, you just need to save the bin. You can save the single change by using the "save" button in the XDF editor box, then save the bin, as well.

[EagleMark]

Seems when I am emulating I make changes in bin, no hex editor, and still have to commit to changes for the AutoProm to recognize change. Please correct me if I am wrong, this is what I remember doing but been awhile.

[gregs78cam]

Thanks guys. It's all working now. I got the speedo pulses figured out, now both the datastream and my speeedometer match. I got the LM-1 configured and the datastream is showing within 0.2:1 of the controller, I am sure that difference is just a ground offset, close enough for me though. Only thing I can't get to work right is the PCM control of the radiator fan. I have it wired up, the code written in the .bin, but no joy. I would like to put that bit in the datastream and add it to my TP dash so I can see if the PCM is actually turning on the output. I just haven't taken the time figure it out. I still need to go though my datalog, and make some adjustments to the VE but I am getting ready for work now, so that about does it for this week. Here is the datalog from this morning. Have a look and feel free to make some suggestions, Thanks again.

Greg

[Six_Shooter]

You followed all of the additions and changes needed for your specific code in this thread? viewtopic.php?f=9&t=153 (http://www.eagle-mark.com/Forums/viewtopic.php?f=9&t=153)

Just to point out that, there is a couple hex characters that need to be changed to enable the code to work. This over writes a couple characters that are already present in the bin.

[gregs78cam]

Yep, I changed the the entries at address 7C07 from D567 to 7120. That inserts the beginning of the fan control routine into the runnning code right?

[gregs78cam]

OK guys, somehow I go the idle issue fixed, it goes into gear, and back to park, and idles right back to normal. New problem. I was not getting any PE but I got PE to come in like it is supposed to and sometimes it runs WOT great. However it doesn't run great all the time. In the attached datalog @ 2:15 and 4:25 you can see I start to go WOT, and everything looks good, TPS, MAP, SA, BPW, injector DC, it just goes lean and drops RPM. I immediately let off and get back on and it goes like snot, WBO2 is slightly rich, but good enough for now. I don't know where to go next.

[gregs78cam]

UPDATE: I moved the external regulator closer to the TBs, the lean spike before the rich spike must have been a split second pressure drop between the throttle bodies and the regulator. I got the 1>2 shift sorted out, and I think the 2>3. The trans needs a shift kit/plate mod and the PCM needs the shift param's tweeked. I am going to take the injectors out and have them cleaned and flowed so I know for sure. AE still needs some work, but WOT is much better @ 12.5-13.2 AFR.

[EagleMark]

Mr Injector is in Dalton on Govement way.

[gregs78cam]

Thanks, I found your post over on FSC and got them done this week, picked em up Friday. Only thing I don't quite understand was the flow number he has listed on the sheet 105cc/min???? He showed me the handwritten slip that had 800 or 804cc/min. I was expecting some concrete accurate numbers. I will call him Monday and talk to him again.

[gregs78cam]

I am going to try running this thing in $0D instead of $0E. oldred95 over on FSC says he has done it, so I will give it try. $0D just has so many more trans tables defined, more .bins available and is used much more, hopefully can get it sorted out this week. I got my air cleaner baseplate modified, I'll get that mounted tomorrow.

[EagleMark]

Thanks, I found your post over on FSC and got them done this week, picked em up Friday. Only thing I don't quite understand was the flow number he has listed on the sheet 105cc/min???? He showed me the handwritten slip that had 800 or 804cc/min. I was expecting some concrete accurate numbers. I will call him Monday and talk to him again.
My recipts always come with before and after numbers. Unless he was just cleaning and forgot to right them down. He also called tech support last time I was there and they showed him the cc flow rate to PPH conversion that GM uses. Now he can use his machines cc flow rate and pressure and come out with what is expected from a GM injecotr. So far since this has happened I have only had one set done and they were very close to what GM said they would be. But you have to take into account gausges could be a little differant. His cleaning fluid could be differant than gasoline etc...

[dave w]

Thanks, I found your post over on FSC and got them done this week, picked em up Friday. Only thing I don't quite understand was the flow number he has listed on the sheet 105cc/min???? He showed me the handwritten slip that had 800 or 804cc/min. I was expecting some concrete accurate numbers. I will call him Monday and talk to him again.

I use the measured flow only as a best guess what to program the PROM with. What the PCM "thinks" is the injector flow is, and what the measured injectors flow is are apples and oranges! :doh:

dave w

[gregs78cam]

Ok, I got everything back together, It is running $0D with the 4L80E, but I still need to change one or more of the shift solenoid state entries because the manually selected gear doesn't match what is shown in the bitmasks. It is drivable though. And the radiator fan is fully controlled by the PCM also. My timing seems a little off though. It shows about 5 degrees more than what is in the timing map. I'll upload my last datalog if anyone is interested. I ordered the hood scoop this morning, I'm putting a '67 427 Corvette scoop on it.

[EagleMark]

Could the timing be where you have it set?

[gregs78cam]

Distributor and Initial advance should both be set at 0, but I can double check. What I meant was the timing on my .ads dash shows about 5 more than the SA table has @ idle. The only spark bias I don't have zeroed out is Main spark bias which is 9.8, so I don't know where the 5* difference would come from.

I think I have the trans shift solenoid states figured out, but I still have no manual 1st gear. I think it may be in one of the tables - Manual upshift/downshift MPH vs. TPS were all zeros.

Anywho I've got more stuff to try tomorrow.

[EagleMark]

Sounds like you zerod everything? I usually start with a stock chip and just adjust. There's tons of things you can mess up!

[gregs78cam]

Sounds like you zerod everything? I usually start with a stock chip and just adjust. There's tons of things you can mess up!

Well I started with what I thought would be a stock .bin off Moates site, but it had a lot of stuff zeroed. It seems every .bin I download has some stuff wrong in it. I only zeroed PE spark, and EGR stuff. I just use compare to get them right. I just looked at BJYL, BJLH, BJYL, BDWL, BJYB, and they all have all zeroes up to 93 & 100 kpa where they have between 27 and 35. BJDR has mostly zeroes but it has 127.5 from 37.5%TPS and up in the 2->1 downshift column.

[EagleMark]

Take a look here as I know most of the bins are stock and I have the gallerey set that people can download them and even upload them with a differant name and changes but origanals are protected. Most were donated by Robert!

[EagleMark]

How did you go about wiring for the VSS with this 427 ECM? Did you use the VSSB/DRAC ? I am finishing up a harness for a 427 ECM for my kids 345 IH engine. I don't need any tranny controls as it's a manual.

[Six_Shooter]

You should be able to use a 2000 PPM VSS connected to F12 and "VSS not from DRAC" set.

The Painless VSS should work for that.

[gregs78cam]

EagleMark, yes I used the DRAC/VSSB for the signals to the PCM and AutoMeter Speedometer.

Well after playing with $0D for a day, I am going to back to $0E. I just have had no luck getting the shift solenoids right. While re-doing a new .bin for $0E, I found the error in the cooling fan code. I put a $90 in a spot that should have been $96. That should fix it. Now I will have to figure out how to get the 1>2 shift tightened up. That was the reason I wanted to try $0D, it actually has a .xdf parameter "desired shift time vs. shift" that is not in $0E, along with quite a few others. That's about where I'm at.

[gregs78cam]

Update: After setting up a new .bin based on a stock BJDR.bin with cooling fan code, flags set for wideband input, desired AFR, and all the other cr@p that needed turned off or on, It runs pretty well. Shifts into all gears, even TCC works. The actual fan on and off temps don't exactly correlate to the .bin values, so that is one issue to sort out. I also changed a couple trans line pressure variables, I will try those out next week. I think I am going to get the Camaro registered and insured before I go into work tomorrow.

[EagleMark]

Some day when you have time would you do a write up on changes needed for the wideband ?

[gregs78cam]

As far as the wideband goes I just used the Advaneced_$0E_TP5_v250.xdf and .ads. these are already set up with the parameters written to insert the WBO2 into the datastream. You just have to follow the read_me directions at the top. You can also datalog commanded AFR (in place of Battery V) from the PCM by changing a couple of bits in the .ads I just inserted a scalar to make it easy to change. I can upload my current files later when I get home.

Also here are some very helpful tuning notes that I gathered from several post by HaulnA$$ on the FSC site. These really got me going in the right direction.

[gregs78cam]

So BJDR 0E-1 is what I have loaded now, and 1.1 has the transmission line pressure changes to try this week, with a gauge on the trans obviously. I found that the acceptable pressure ranges are 37-171 for drive gears and since I didn't put a shift kit in it or drill any feed holes then I figure what pressure changes I do make shouldn't be as harsh. It was set to 90psi max allowed line pressure, so I am going to work up little by little to see if I can get this thing shifting a bit firmer. Running it in $0D it had set a transmission error which defaulted to full line pressure and the shifts were nothing short of BRUTAL, it shifted so hard I didn't even dare try it at WOT. So I know it can get better, I am just going to take my time and sneak up on it. I increased the Load based Pressure offset vs. TPS, PE line pressure vs. Input speed, and max allowed line pressure to 125. I will run it on -1.bin to get a current baseline and if it does max out at 90 psi, then I will try the -1.1.bin version and post back with the results.

[EagleMark]

Hey did you go to Car dAlene? Or going tonight?

Thanks for documenting this tranny stuff I'm sure it will help others! :thumbsup:

[gregs78cam]

No, I didn't make it to Car d'alene. I usually don't get to because of the hours I work. But the up side is I have lots of time during the week to work on my stuff. Those line pressure changes seem to have made a big difference. I also modified the kickdown stuff so it drives much better. I made some VE changes thursday night and have been datalogging my drive back and forth to work. It's kind of neat that the engine likes the fuel curve a certain way whether it's EFI or carbed. It's just much quicker and easier with EFI. It doesn't need a lot fuel for cruising, but the instant you start to open the throttle it wants LOTS of fuel and it wants it NOW. I still don't know how WOT really shifts because all it does is spin the tires in 1st (basically bald 295/50-15 Cooper Cobras)

My next step. After driving it a few days ago and idling up my half mile driveway as usual I got out and the TBI adapter was so cold it had moisture condensing on it when it was about 70 outside and the engine fully up to temp. I then let it idle while looking over the DL and making a few changes and the idle started to creep up with the IAC completely closed. Got out and the adapter had warmed up from the fan blowing on it. So after reading a bit on FSC I decided I am going to put a MAT sensor in the RPM Air Gap manifold. From looking thorugh it the PCM already has the code for it and I still have a couple of extra wires in the harness so it shouldn't be that hard.

I am also trying to fix a small issue with the 60mph cruise. It is kind of wierd. At 55 and 65 the engine sounds happy, runs great. But right at 60 it just feels rough. I added more timing and that seems to have helped a little but I am almost maxed out in those cells and I don't think it will completely get rid of it if I do go to 41.2 degrees SA.

And the last issue is that while going through town @35 or 45mph, it requires so little throttle like 2.4 -3.1% that I thought I was bouncing off the DFCO or Decel Enleanment threashold. I was looking through the DL from my drive to work Friday and could see the lean spike (20:1 AFR) as TPS (2.4%) and MAP (28.1) remained completely steady. I changed the TPS for DFCO down from 3.9% to 0.4% and no luck. I have the "time before DFCO" set to 1 second so I will try up-ing that and see what it does. I don't know if anyone is still watching this thread, but any advice would be welcomed. I can go out and get the laptop if anyone wants to see the log.

[gregs78cam]

Also one question for you guys with a test bench. I know that the injectors alternate firing with DRPs, but does each one always fire on the same DRP, from one engine start to the next? Or when going from Asynch back to Synch fire?

[EagleMark]

I don't know?

Just pulled the harness and ECM from my bench as it was done for a freind. Next time I get one on there we can look into that.

If your ever down to CDA stop by and we will try it out?

[EagleMark]

Wish I had some earth shattering news for you but sounds like your heading the right direction. You sure it's not hitting lean cruise?

Have you thought of a hot air intake? When I installed one on my otherwise stock 1990 TBI Blazer , that someone put an open element on, it made the world of difference in anything below totally warmed up on an 80 degree day. GM has so many available from factory that need no hoses or lines. Closed and draws heated air off headers when cold. Open to all cool air when warm.

No, I didn't make it to Car d'alene. I usually don't get to because of the hours I work. But the up side is I have lots of time during the week to work on my stuff. Those line pressure changes seem to have made a big difference. I also modified the kickdown stuff so it drives much better. I made some VE changes thursday night and have been datalogging my drive back and forth to work. It's kind of neat that the engine likes the fuel curve a certain way whether it's EFI or carbed. It's just much quicker and easier with EFI. It doesn't need a lot fuel for cruising, but the instant you start to open the throttle it wants LOTS of fuel and it wants it NOW. I still don't know how WOT really shifts because all it does is spin the tires in 1st (basically bald 295/50-15 Cooper Cobras)

My next step. After driving it a few days ago and idling up my half mile driveway as usual I got out and the TBI adapter was so cold it had moisture condensing on it when it was about 70 outside and the engine fully up to temp. I then let it idle while looking over the DL and making a few changes and the idle started to creep up with the IAC completely closed. Got out and the adapter had warmed up from the fan blowing on it. So after reading a bit on FSC I decided I am going to put a MAT sensor in the RPM Air Gap manifold. From looking thorugh it the PCM already has the code for it and I still have a couple of extra wires in the harness so it shouldn't be that hard.

I am also trying to fix a small issue with the 60mph cruise. It is kind of wierd. At 55 and 65 the engine sounds happy, runs great. But right at 60 it just feels rough. I added more timing and that seems to have helped a little but I am almost maxed out in those cells and I don't think it will completely get rid of it if I do go to 41.2 degrees SA.

And the last issue is that while going through town @35 or 45mph, it requires so little throttle like 2.4 -3.1% that I thought I was bouncing off the DFCO or Decel Enleanment threashold. I was looking through the DL from my drive to work Friday and could see the lean spike (20:1 AFR) as TPS (2.4%) and MAP (28.1) remained completely steady. I changed the TPS for DFCO down from 3.9% to 0.4% and no luck. I have the "time before DFCO" set to 1 second so I will try up-ing that and see what it does. I don't know if anyone is still watching this thread, but any advice would be welcomed. I can go out and get the laptop if anyone wants to see the log.

[gregs78cam]

Well from looking at this morning's log, it seems that while cruising @33mph, 3.9%TPS, 34.7Kpa, I have the Synchronous AE flag ON. Then just as the lean spike is starting the Asynch pulse flag goes ON, Synch AE goes off for one sample while the Asynch remains on for another 5 samples.

But now my question is why is Synch AE on all the time? Is it the lower vacuum that my cam makes.

[EagleMark]

Saw you on TGO, what was it?

[gregs78cam]

Well after looking at the datalog for a couple of hours, it looks as if the synch AE flag in on whether AE is in effect or not. That would be the default AE mode. Then if more AE is called for it switches to Asynch AE. Who knows maybe the "Synchronous AE" flag should really just be labeled "Synchronous Mode"?

[gregs78cam]

AS far as the lean spike goes, I still have no idea. I was wondering if I was at such a low TPS that maybe PCM thought I was trying to idle, and because RPM was too high for an idle maybe it was cutting fuel. IDK. Interesting thing though, when I put the TBI engine in my '74 truck, I had the power to the ECM running off the resistor wire from the original points ignition, It showed 12+V with engine off, but then dropped to 5V when started. This naturally set off the LOW BATTERY code and opened the IAC all the way. Engine would Idle very high, and to keep RPM down the ECM would cut fuel completely until RPM dropped low enough. Now that is the '7747. I will have to look at the log again and see if the "Idle Flag" is set when the lean spike occurs. I'll check it when I get back.

[gregs78cam]

UPDATE:

Lean spike only occurs in closed loop. That is not such a bad thing though. I have the VE tables set pretty well (rich enough@WOT) so that I don't have to worry. It drives so much better in open loop. One thing I don't quite understand is, correct me if I am wrong, but when in closed loop as you get into WOT doesn't the PCM basically go open loop and ignore the O2 sensor and meter fuel strictly off the VE table and associated adders? I ask because WOT is actually richer in OPEN loop than it is in Closed loop. I always figured they would be practically the same.

Also, trans tuning is coming along slowly. It takes a lot of time to get the shift points(up and down) and TCC to happen when what you want is far from stock. One more ?. How do I set this thing up to keep the TCC locked while decelerating?

[EagleMark]

Not sure the exact procedure that your ECM follows but yes open loop like when you enter PE is by fuel tables. Some guys on TGO say yes others say there are still some O2 feedback? Some of the best tuners I have worked with in person tune totally open loop. When done switch on closed loop and all data is spot on and your back to 14.7 to 1. Enter PE and goes to that fuel ratio. There are always little checks in stock bins to check PE has not been entered to long to overheat cat. Sounds like you got er close?

[Six_Shooter]

At WOT, the O2 feedback is not applied, but if the BLM is below 128 before entering WOT, that value is used for WOT fueling.

[EagleMark]

At WOT, the O2 feedback is not applied, but if the BLM is below 128 before entering WOT, that value is used for WOT fueling.
Is that for all ECMs?

I don't know what the exact procedure is for WOT but have read some conversations on TGO were some of the most experianced EFI guys argue that O2 is used and some that say it is not used.

From my humble opinion I beleive you are correct but the BLM value just before WOT is used to calculate the AFR commanded before entering PE. Right? You have to enter PE at WOT.

NOTE to Greg: I left you a message on your phone. Also got your PM a day late? Now I have responded to your PM but they are stuck in my outbox not sent and I can't figure out why....

[Six_Shooter]

At WOT, the O2 feedback is not applied, but if the BLM is below 128 before entering WOT, that value is used for WOT fueling.
Is that for all ECMs?

I don't know what the exact procedure is for WOT but have read some conversations on TGO were some of the most experianced EFI guys argue that O2 is used and some that say it is not used.

From my humble opinion I beleive you are correct but the BLM value just before WOT is used to calculate the AFR commanded before entering PE. Right? You have to enter PE at WOT.

NOTE to Greg: I left you a message on your phone. Also got your PM a day late? Now I have responded to your PM but they are stuck in my outbox not sent and I can't figure out why....


As far as I know, all GM ECMs function this way. Like I said if the value is less than 128 before entering WOT, that value is used, but if it is above 128, then the value of 128 is used. I haven't seen any indication to the contrary from people that I would consider as "reliable sources."

[EagleMark]

Actually the way you explain it makes them both right. It uses it at first, then not.

When Dimented24-7 and Rbob disgree or should I say have a differance of opinion it is a good thread to read...

[gregs78cam]

So as promised here are some close up shots of the installation. When I take it apart this winter to add an MAT sensor I will take some more pics of it.

[gregs78cam]

[attachment=2:142vu560]DTBI 008.jpg[/attachment:142vu560]
Another throttle cable mount pic.

[attachment=1:142vu560]DTBI 011.jpg[/attachment:142vu560]
An overall view of the back of the setup.

[attachment=0:142vu560]DTBI 013.jpg[/attachment:142vu560]
You can see how I pieced together the air cleaner base from two and some scrap 20ga with a spot welder.

[Six_Shooter]

I'm told that an O2 sensor plug will thread into the TB, then just cut the pintle to length that it blocks the port when the pug is tightened.

[gregs78cam]

In the last pic above I spot-welded a 1/4-20 nut between the two flanges of the air cleaner bases to screw a stud into. I also put riv-nut inserts into the front bosses on the top of each throttle body to accept the 1/4-20 stainless bolts you see in the pic.


I'm told that an O2 sensor plug will thread into the TB, then just cut the pintle to length that it blocks the port when the pug is tightened.

Well, an O2 sensor is 18mm x 1.5, and an IAC is 20mm x 1.5 so I don't think that work???

Anyways one last pic of the beast this engine is in.

[attachment=0:3cewvd25]DTBI 004.jpg[/attachment:3cewvd25]

[gregs78cam]

Well, I thought I should update this thread, so here goes.

The car is running AWESOME!!!. I have been working 70+ hours for the last two weeks so not much tuning lately, just an odd datalog to work or back, and a few changes in the evenings. But really it is running so well now that I really don't even feel the need to go any further. I did put a 2" air filter on it and raised the back of the factory scoop so it would fit, and much like I expected it ran a little richer across the board, and shows a slight vacuum (compared to KOEOff). So I am leaving a little power untapped but I don't mind. I have been thinking about taking it back down to the dyno soon if funds permit. It is just so fun and comfortable to drive now.

The other day my turn signal lever broke, so I decided to replace it with one from a '90 van with the Cruise Control switch in it. I made an adapter stub to mount it and will be running the wires down the column this week. Luckily when I was making the harness I included the cruise control plugs and wires and got the servo control of the van that the 4L80E came from. So i am one step closer to having cruise control on this thing.

That is about all I have to add. Below are the .xdf, .adx, and .bin that I am working with now. I did add Trans Fluid Temp to the engine datastream, which is really nice to have now.

[EagleMark]

and I got to see the car in person!
:confused1:


If you dyno you can use my WideBand O2 Lc-1-AutoProm-Motes O-Meter...

[gregs78cam]

Cool. Thanks for the offer, but I already have my LM-1, and AutoProm hooked up. I think before I head to the dyno, I am going to work on a .bin that will lock the Trans in 3rd gear, and lockup the TC, you know so there is no slip and no kickdown when I hammer the pedal. That should make it a fair comparison to when I had the T-56 in it.

[EagleMark]

What kind of numbers did you get then?

[gregs78cam]

What kind of numbers did you get then?

Did you mean the last time I had it on the dyno? It was 340hp @5500, and 370lb/ft @3500 to the tires. If I get at least that much with the 4L80E then that would mean that the engine is actually making more than it was last time (hopefully due to the EFI) because I am SURE the auto is soaking up more than the T-56 was.

[gregs78cam]

I got the NBO2 installed and hooked up a few weeks ago, enabled closed loop and it ran pretty good. There were no lean spikes like before when running the simulated NB off the LM-1. It cruised nicely and I got some pretty good data out of it. However, I did a few WOT runs and noticed it going much leaner than I want. After watching the replay, I noticed that the BLM was pegged at 120, Desired AFR was right on 12.8, and it was still running in Sync mode, 29% DC, with lots of PW still unused. I have read about the WOTBLM Locker, and I am wondering if it is possible to implement something similar in $0E, if it doesn't have the ability already. Or the other option is to just fine tune the VE and SA in closed loop, then switch back to Open loop when done. Here is the log, and my ADX.

[EagleMark]

I don't even know how the WOT BLM locker works? It's a program you run on your $EE bin. Maybe if you download the program you can see what it does to figure out how to apply to your mask? Never mind it's all done in visual basic and C+ so can't see what any of the files are doing?

Would be nice to know what it does in hex. Should be a performance option on any mask!

[EagleMark]

I just ran it on a bin then used the differnce tool in TunerPro and only 5 bytes are differant, Items were not defined and I don't have a complete dissesembly of an $EE bin.

But thinking of this now I could probably add a hack to TunerPro to do this because of what you taught me! :happy:

So for a guy who knows hex with dissembly, should be able to figure it out? If you go into PE or in EE case WOT and BLM are at 128 well you know the outcome. But like what's happening to you is going into PE at 120?

[gregs78cam]

Mark can you post up the two .bins? Or even just what the 5 bits changed from to. I was going to try exactly what you did, but couldn't figure out how to open/run the program at 2AM this morning.

[EagleMark]

Like the new avatar!

Here's the bins, it also updates checksum when done. It's an old program that does .LT1 and .bin so it may have been part of LT1 edit or that other free one that went away years ago, I have it in a seperate folder when I installed it, may need to do it again with a rightup or something because I don't see a read me file with it. Does not look like I have compatibility set up on it?

[96lt4c4]

Wow, this is a cool project. Makes me want to build one. Sorry to dig up an old thread.

How is the car running, how much HP are you making?

You may get a little more air flow if you install the 1/4 pod spacer. That helps get the injectors up out of the bores a little. Also machine the top of the TB flat helps too.

[gregs78cam]

Thanks. It runs pretty well. I have been wanting to change the fuel pressure a bit and start on a new tune for it and see if I can get a couple of very small drivability issues eliminated, but I just don't have the time and I can't justify $100 in gas just to make it run a little better. It made 340hp on a chassis dyno with carb and T56. I don't drive it much, and have actually contemplated selling it to finance another project.

As far as air flow, right now the limiting factor is the small air filter I have on it so I can keep the scoop on it. The two stock TBI's together are capable of way more flow than this engine needs. The idea behind running two was more for the fuel flow that this thing needs, without having to run BBC injectors at a crazy pressure.

[96lt4c4]

Thanks. It runs pretty well. I have been wanting to change the fuel pressure a bit and start on a new tune for it and see if I can get a couple of very small drivability issues eliminated, but I just don't have the time and I can't justify $100 in gas just to make it run a little better. It made 340hp on a chassis dyno with carb and T56. I don't drive it much, and have actually contemplated selling it to finance another project.

As far as air flow, right now the limiting factor is the small air filter I have on it so I can keep the scoop on it. The two stock TBI's together are capable of way more flow than this engine needs. The idea behind running two was more for the fuel flow that this thing needs, without having to run BBC injectors at a crazy pressure.

I had a problem running a carb manifold, I think you need the heat from the stock type TBI manifold to keep the TBI warm. I was running a Performer RPM air gap and could not get it to run right. I think the TBI was freezing up. I ended up boring out the stock manifold and TBI to 2 inch and running the BBC injectors on my 383. Much better drivability. This is just my guess as to what was going on, it could have been my lack of tuning skills too...:rockon:

[96lt4c4]

Did you look into trying this with a dual 4 barrel carb intake?

[gregs78cam]

Once I got the injector flow correct for the injectors it did still take a lot of AE to get it right.

I thought about a dual quad intake but it just wasn't in the budget.

[gregs78cam]

Just a little update. I was noticing last year that at a warm idle the Injector PW was as low as it would go, and the 5.7L injectors were still putting out too much fuel, which you could see in the WBO2. After Mark figured out the Injector Duty Cycle was off, I fixed that in my ADX, and did a log. I found that I was only reaching around 60% DC. So that means I have the ability to add another almost 50% more fuel through pulse width. The idea was to remove the 5.7L injectors and put in a full set of four 4.3L injectors, which is a 25% decrease in flow at the same pressure and pulse width. Yesterday before work I swapped them. I drove it to work today, and it ran a LOT better. It idles warm right around stoich in OL. I don't know why yet but it dropped out of CL in the parking lot, otherwise I would have checked the BLM/INT at idle to make sure it was good. Decel Unleanment, and DFCO work better now. I think it still needs a little more AE from the smaller injectors, but so far results are good. I need to eventually get WOT checked, and probably re-mapped, but that will come later. If I end up needing more fuel than I have PW for at WOT then a VRFPR will be put in, and re-tuned.