EDIT: The thread title says 7730 but actually this has evolved into a thread about replacing the 1227730 ECM with a 1227165 with NVRAM.
To all those who are running a VIN "Y" (LD8) Northstar on a Ryan Hess 7730 ECM with a manual trans.
How many of you would like it to run better? Or have you sorted it out already?
Not sure of the protocol of resurrecting the old thread but the problems raised on it were not solved. So continuing on the archived thread: https://www.fiero.nl/forum/A...130314-2-111250.html Problem: The 7730 ECM modified for Northstars by Ryan does not run well. Problems I have experienced are: running very rich, poor idle, hunting IAC, lack of DFCO. I spent a fair bit of cash and time on a dyno using Tunerpro and got the richness down, Hp and torque up but the other issues still seriously plague drivability. From the prior thread Ryan posted about modifying the XDF file with data taken from the BFUZ A1 Disassembly File. Looking at this file we see just how much is missing from the XDF we are using. This file is huge compared to the Northstar XDF file This must be what the problem is? I have added the lines of code Ryan mentioned and several more lines too from the A1 file to the XDF but am unsure of the conversion formulas required for some of the variables so I have not run the file yet. it will need more work before I try it. I am looking for cooperation from others with the same problems who will help resolve this for all our benefit.
Any participants?
edit: Upon re-reading this, I got the impression some might feel I am knocking Ryan for how his ECM conversion runs the N*. I am not. I think he accomplished lots, but I also think there's more to do from where he left off to make it run correctly. edit: added VIN Y(LD8) engine type. ------------------ Daviero - 88 N* GT
[This message has been edited by Daviero (edited 01-12-2014).]
Just to confirm - this is for a N* engine only setup with a manual trans.
quote
Originally posted by Will:
The Shelby computer is another option. The 7730 is pretty basic.
Will - I am not familiar with that. Swapping the 7730 for another is a thought I was going to ask about. What is the "Shelby computer"?
quote
Originally posted by jdv:
Have you looked at $12p ? They use the 77165 so your 7730 will not work.http://pcmhacking.net/forums/viewtopic.php?f=27&t=356
Will the $12p and the 77165 run the N* though? Are there other codes that can run in the 7730 that will run the N*? I think that with the N* we are rather limited - the same old discussion about how to run engine only on one. Is there an OBDII option I don't know about?
In a way I hate to toss the fuel table tuning I have done so far, but I would for other gains. The dyno tuning can be redone.
Will - I am not familiar with that. Swapping the 7730 for another is a thought I was going to ask about. What is the "Shelby computer"?
Is there an OBDII option I don't know about?
In a way I hate to toss the fuel table tuning I have done so far, but I would for other gains. The dyno tuning can be redone.
The Shelby computer is the GM 09354896 ECM from '99-'00 LS1 Camarobirds and trucks with the program from a '98 Shelby Series 1 sportscar. The .bin is available from AllData as a "1998 Oldsmobile Shelby". Being a 1998, it's obviously OBDII. It can be tuned like an LS1 by anyone who has HPTuners, which is almost every dyno shop you can find. It is ALSO designed for a manual tranmission, and has a clutch switch input to prevent stalling when clutching from an overrun. It uses MAF based fueling.
PM Darth Fiero on this forum. He can set one up for you.
Originally posted by Will: Start reading here: https://www.fiero.nl/forum/F.../000121-19.html#p727 I go over some of the ins and outs of installing the Shelby computer on my car in place of the OBDI Caddy computer.
Thanks Will. Interesting and have to read again. Changing the ECM as you have done would seem to eliminate the problems with the 7730 setup. Can you tune using EFI Live or similar program? I have done quite a bit with Tunerpro V5 and a Moates emulator and would like to still be able to do that. I guess I will be contacting Darth.....
I have added 58 new lines from the A1 disassembly to the XDF file using Tunerpro but I do not know if I have the "Hex to real world" conversion expressions correct. I will post them here - does anyone else have any experience with them? Say for example for a percent, conversion= X/255*100 - Ryan told us that in the last thread, so that is correct. For rpm functions I have used conversion = 25*X but Ryan has conversion= 12.5*X for max block learn rpm and 25*X for fuel cutoff rpm. What gives? For MAP functions I have used =0.39625*X For temperature I used conversion = .75*X-40 For time functions: I have used conversion=1*X For count functions I have used conversion=1*X For voltage I used conversion = X/255*13.6
Can anyone comment on these? Are they correct or not?
If you are thinking about changing the ECM maybe give the Cavalier ECM 12210553 a thought. I know people that have used them on Northstars with no issures.
Did a search on the Tunerpro list if masks and none is listed for the 12210553. What does it run? It would have to be editable. This is a new one......
For rpm functions I have used conversion = 25*X but Ryan has conversion= 12.5*X for max block learn rpm and 25*X for fuel cutoff rpm. What gives?
The multiplier changes depending on the RPM range. I don't know the exact ranges. This gives the reported data higher resolution at lower RPM. Also, the max reportable RPM via 25*X is 6375. The Northstar can rev higher than that. IIRC, Ryan changed the high RPM multiplier to 37.5.
Though there aren't too many responses to this topic yet, just know that I (as well as others, I'm certain) are keeping a close eye on how this develops. I am not at the stage where I can contribute anything useful to the discussion except cheering on the sidelines. Go guys, go!
[This message has been edited by Bloozberry (edited 10-10-2013).]
The multiplier changes depending on the RPM range. I don't know the exact ranges. This gives the reported data higher resolution at lower RPM. Also, the max reportable RPM via 25*X is 6375. The Northstar can rev higher than that. IIRC, Ryan changed the high RPM multiplier to 37.5.
Resolution correction was a suspicion I had for the differences. So the A1 diassembly file is really only half the required info - intuition is not good enough. Is there a compilation of conversions required already sorted? This would be golden. To do it myself I could use another XDF for something else as a reference and look at every variable one by one.......but very time consuming. For this option - any opinion as to what XDF to use for reference? Either the "A1_tpv5.xdf" (1991-1993 LH0) or the "Super_8dm2.ecu" (but this is 90-91 LB9/L98 F-Body and $8D not $A1). Both are 7730. Or another one?
I feel I am really groping at straws. Maybe LS1 is the best option.
Will - I read your LS1 Shelby has no DFCO. How come? The original file would've had it?
I have also looked at the LS1 pcm options, they seem quite valid. If DarthFiero can not help with those options I can offer some help.
I will say that if you are looking for a MT option... Megasquirt is very valid and has tons of recently upgraded features that makes it not anything like it was years ago.
Below is a list of the parameters in the 99 Shelby Series 1 4.0L Aurora V8 tune used in the 99-00 LS1 9354896 OBD2 PCM as well as their stock settings:
DFCO Enable Delay Time 0.8 Sec. DFCO Enable Delay Time (CAT Overtemp) 0.8 Sec. DFCO Enable Coolant Temp. - Clutch Based 139.9 Deg. C DFCO Disable MAP - Clutch Based 50 RPM DFCO Disable %TPS - Clutch Based 10 %TPS DFCO Disable MPH - Clutch Based 30 MPH DFCO Enable Coolant Temp. 80 Deg. C DFCO Disable Negative Delta RPM -75 RPM DFCO Re-enable Delay - Clutch Based 1 Sec. DFCO Re-entry Delay 1 Sec. DFCO Enable Vehicle Speed Hysteresis 2 MPH DFCO Enable RPM Hysteresis 2.9 RPM DFCO Enable Max MAP Hysteresis 1 Kpa DFCO Disable TPS Hysteresis 1 %TPS
Not having one of these cars/applications here to get my hands on has prevented me from tinkering with this issue further. There are a number of parameters I can see in the stock programming that could prevent DFCO from working, but some experimentation would be required to fix the issue.
-ryan
------------------ OVERKILL IS UNDERRATED Custom GM OBD1 & OBD2 Tuning | Engine Conversions & more | www.gmtuners.com
[This message has been edited by Darth Fiero (edited 10-10-2013).]
Does anyone else find it odd that DFCO isn't working? Is it possible the ECM isn't reading mph or is running so rich it just seems to not be working? It seems odd to me because I don't understand why anyone would mess with it or at least I never had to.
I have posted many times about the hunting idle (IAC) issue. It's a throttle plate adjustment. It has to be opened up a bit, you can't run the stock position.
Stalling issue is the same one.
Cheers,
.
[This message has been edited by aaron88 (edited 10-10-2013).]
In the old days they had to drill out the throttle plates on 4 barrel carbs to get the things to idle right.I know ob2 is more advanced but for ob1 there some options. explore them.$12 is one so is Sajupv4 with you can tune through the aldl on the 7730. the n* hack was a start but even with ob1 there is more out there.
I am still on the fence. I'd like to give the 7730 a shot - $A1 or otherwise but with no success, then the LS1 is to be done. Not interested in megasquirt.
I am looking at the original $A1 definition file to confirm the conversion from Hex to real world data.
The definition file written for the N* 7730 had no defo or enleanment - that's why it goes fat rich when you take your foot out of it. It is a basic file that runs the engine. That's it. At least that's what I have found. Its these parameters that are required in the definition (XDF) file. But I'm a bit in the dark, that's why I am asking for some guidance.
Why the suggestion for a $12 definition? It is not an XDF file Tunerpro can use. Why not $83?
Thanks for your post Ryan - I will use this as compare data too. You may sell me an LS1 setup yet.
I don't have a dog in this, but after tuning V8's with OBD1 and OBD2, I wouldn't ever consider going back to OBD1.
DFCO (on OBD2 applications) is quite sensitive to engine parameters and flow. On my LS4/F40 swap, I experimented using the DFCO parameters for the LS2/Manual (to get factory manual transmission DFCO parameters) and while it still worked, it wouldn't completely exit DFCO and would hold the timing change until it activated again. When I changed it back to the LS4 DFCO parameters then it worked as it should and turns completely on and completely off. I suspect Will's issue is the DFCO parameters are for a 4.0L, not the 4.6 and it isn't meeting a needed threshold.
I don't have a dog in this, but after tuning V8's with OBD1 and OBD2, I wouldn't ever consider going back to OBD1.
Point taken - dog or no dog lol. I agree it would provide for the best tune.
The missing DFCO isn't the main problem as seems to have become the interpretation though. That's just an example of the missing code in the definition file as the N* definition file is small compared to other definition files. Can't tune it if it isn't there. My motivation is twofold -the 7730 is already installed and I already have experience with Tunerpro via my Moates emulator. If I can make the $A1 code work reasonably well with some computer work, that would be great and I save purchasing hardware and software. Top performance is not essential, but better than now is.
For now I will carry on with the $A1 comparison and post more later. We'll see where that goes. Winter is long up here.....
Does anyone else find it odd that DFCO isn't working? Is it possible the ECM isn't reading mph or is running so rich it just seems to not be working? It seems odd to me because I don't understand why anyone would mess with it or at least I never had to.
I chalk it up to the Shelby's bare-bones emissions state and the fact that people who buy high-end cars want them to "pop" on overrun. I read that in Ferrari forums all the time.
I don't have a dog in this, but after tuning V8's with OBD1 and OBD2, I wouldn't ever consider going back to OBD1.
DFCO (on OBD2 applications) is quite sensitive to engine parameters and flow. On my LS4/F40 swap, I experimented using the DFCO parameters for the LS2/Manual (to get factory manual transmission DFCO parameters) and while it still worked, it wouldn't completely exit DFCO and would hold the timing change until it activated again. When I changed it back to the LS4 DFCO parameters then it worked as it should and turns completely on and completely off. I suspect Will's issue is the DFCO parameters are for a 4.0L, not the 4.6 and it isn't meeting a needed threshold.
This disables DFCO:
quote
Originally posted by Darth Fiero: DFCO Enable Coolant Temp. - Clutch Based 139.9 Deg. C
That parameter is how the car shipped from Shelby. Shelby disabled the DFCO in all cases.
Originally posted by Will: That parameter is how the car shipped from Shelby. Shelby disabled the DFCO in all cases.
Yeah, I didn't review all the settings from Darth's post. However, in the event you do enable it, don't be surprized if it doesn't work exactly right w/o some changing of the parameters.
Even the idle follower setting isn't quite right... The Shelby has an 18# flywheel, while I have a 14# flywheel. That in combination with the engine being larger and having higher compression means that it still stalls every so often when I clutch on overrun.
Why the suggestion for a $12 definition? It is not an XDF file Tunerpro can use. Why not $83?
I believe what jdv is referring to is the OSE$12P code from pcmhacking.net. It does have an XDF file that is used with tunerpro. With tunerprort, an NVRAM module, and a slightly modified 1227165 ecu, real-time tuning is possible. I can send you some files if interested. I have used this combo on many swap in the last few years, including Fieros.
I believe what jdv is referring to is the OSE$12P code from pcmhacking.net. It does have an XDF file that is used with tunerpro. With tunerprort, an NVRAM module, and a slightly modified 1227165 ecu, real-time tuning is possible. I can send you some files if interested. I have used this combo on many swap in the last few years, including Fieros.
ahh the bump. Locating a 1227165 ECM to give it a go.Turns out they are not as plentiful as I would have thought - they are older. I don't want to pay the Vette price off Ebay. More to come as I proceed.....
So I've started to gather for the change to the 1227165:
1 pcm # 1227165 @ $40.00 from a salvage yard, list vehicles that use the '165 is here 1 NVRAM board c/w VS1245 chip from VL400 on pcmhacking.net (Australia) for AU$170.00 delivered Pinout for the '165 pcm downloaded from pcmhacking.net Bin and DXF files downloaded from pcmhacking.net for use with TunerproRT
Spark and other initial settings from the 7730 code will need to be set up in the new OSE12P code. Other details yet unknown but sure to arise....
The ostrich and your 7730 memcal will work to tune. I have used this setup on my 3500 swap and it works well. The nvram board will do the same thing with the auto tune but you can tune on the car with out burning chips with the ostrich.You can use your wideband o2 as well.
[This message has been edited by jdv (edited 10-25-2013).]
Understand that I know little about ECM/PCM tuning, so if I say something stupid, please forgive me. Back when I first read about Ryan running the N* on them 7730ECM, I came to 2 conclusions. First, that the coilpack/module acts as a translator between the dual CKPs and the ECM converting the dual signals to give both a 4X and 24X signal to the Caddy PCM. Ryan's setup only uses th 4X as the 7730 runs on batch fire only. The N* also provides a 1X cam signal as backup to the CKPs. The second conclusion is the module act as a distributor, both setting the firing order and adjusting timing with input from the ECM. The system operates much like the L98,LT1 and Vortec v8 found in '95 up trucks and vans.The reason I bring this up, is the Chevy guys are retrofitting the earlier engines with the 411 PCM used from 2000-20004 on the LS,Vortec v6 and more importantly the 2001-2002 full size vans with the Vortec v8. The van program requires a 4X crank and 1X cam signal, both of which can be provided by the N* module. It fire the injecters sequentialy, so you would have to change to the N* firing order. These are flashable PCMs that can be programed with a laptop and somekind of software like EFILive or LS1Edit. The 411 can also control any GM electronic trans and I know that EFILive offers 2bar and 3bar capability. It would seem to me that this system is what the N* swapper has been looking for. A highly tunable PCM that has a great deal of support from the LS world, can be bought at most salvage yards for about $50 and can be used with MAF or speed density alone. This is where my ignorance shows, are the Vortec tables close enough to get the N* up and running so it can be tuned properly or would you have to start from scratch? Ryan was able to come close with the 7730 and I know some of the Vortec guys are using it too. I know I must be missing something. I've decided to plant this idea here in hopes that someone can fill in the blanks for me. I'm from the old carburetor days when all you needed was fuel and spark to get up and running, I'm still learning this new fangled stuff.
.....It would seem to me that this system is what the N* swapper has been looking for. A highly tunable PCM that has a great deal of support from the LS world, can be bought at most salvage yards for about $50 and can be used with MAF or speed density alone.
Interesting idea. For now I am going to stick with the 7165 and am partway through already. I have tuned the 7730 with an emulator - just did not get the desired results due to the limitations of the modified $A1 code.
The 7165 should be better - I want to get a stock timing map to look at though.....anybody have one?
.....It would seem to me that this system is what the N* swapper has been looking for. A highly tunable PCM that has a great deal of support from the LS world, can be bought at most salvage yards for about $50 and can be used with MAF or speed density alone.
While the Shelby setup is tunable and loggable via HP Tuners, I'm having trouble logging it via my OBDLink MX because it doesn't use the same PID definitions as the rest of GM's stable.
