bobworkman

Revisiting a thread from the past.
https://www.thirdgen.org/forums/diy-...49-memcal.html
In post #14 jwscab provides some drawings of the resistor networks inside the Memcal units, and in post 21 jp86ss
combines them using color to discriminate the upper and lower levels, but there are no labels on the resistors.
In the article on Injector control at
http://www.fortunecity.com/rivendell...ECM/inj_op.htm
table 17 refers to 30 resistors. Can anybody associate the resistor # from the 'ref des' column with these drawings?
In post 34 jwscab has board layout drawings and there are Rn labels on it. The question is do they correspond to the
table 17 values. Also some are hard to read.
thanks
Bob

utilityguy2

iTrader: (1)

see post 26 28 and 30. What is your purpose to the questions. All netres resistor packs are different depending on the engine. They are simply resistors and changing the value is pretty simple but many are interconnected and perform different functions. I'm sure RBob will jump in here sometime.

bobworkman

Read the article in the second post!
This article talks about how the different parameters are set by the resistor network.
It uses r1,r1....r30 as labels for the different resistors in the network not specific values.

Yes each application has different values for these resistors and these values help determine such things as Injection mode, limp mode pulse width etc.

Knowing the labels and using the info provided a better understanding of the Memcal can be obtained.

utilityguy2

iTrader: (1)

I have read the article many, many times. You are resurrecting a 5 year old posting and you tell me to read the article? I personally have researched the netres a lot and I simply wanted to know what you want to do with the info and how are labels are going to help you. There are electrical engineers on here that have all the answers but if you have no question that you are willing to share, how can they? As to your point, I don't see where it said the fortune city article had any connection to the netres on a tpi memcal. Not many of the resistors I have measured come near to those values.

bobworkman

I don't want this be to construed as a flame war. But if I have not made my self clear then let me clarify:

I know this is an old thread, I ask because I expect that others have proceeded beyond the posting scope and can shorten my research.

my question was "Can anybody associate the resistor # from the 'ref des' column (in the injector control article) with these drawings? (posts 14 & 21)

You have provided one possible answer "I don't see where it said the fortune city article had any connection to the netres on a tpi memcal" [that is to say there is no relation.]

However the injector control article does state that mode 5 requires a grd reference to select 8cyl TPI and the common mod for TBI to PFI is to Ground the pin 12, leads me to believe that there is in fact a correlation.

My 2nd question can probably only be answered by jwscab. In post 34 he presents a "test board" to allow him to play with the resistor networks. I asked if the resistor lables on his test board where consistent with the injector control article (on the assumption that there was a correlation).

As a result of this conversation, I do have a 3rd question, if there is no correlation between the article & the drawings, what data did jwscab uses as the base for his testing and experimentation and can this info be shared (even though it is 5 years old).

As to why! I love to research and understand as many aspects of any project I get involved in. I find developing a good base knowledge both personally rewarding and usefull, even though some of may not be used.

utilityguy2

iTrader: (1)

Now those are some good questions! As I said, I also recently researched this and got nowhere but Rbob did give me some very good information and I had hoped he would have chimed in on this by now. Rbob is the guy behind www.DynamicEFI.com and has a very good understanding of the netres and knock filter. I would like to see this go further as a thread because it is interesting and will help lots of people with the memcal shortage problem. I do have charts and such of ohmed values that I can put on here but that is probably not what you want as it seems you want to duplicate the netres and my level of understanding of that is nowhere on the level of Rbob or an electrical engineer. I don't think there is an easy way to duplicate it. As for the injector control article, I don't think there is a correlation as it appears it is for a different type of ECM and is TBI. It seems to be for possibly the 1228746 ECM or the like which have a socket netres. But I'm not sure, so I don't know if labels would apply.

Lets hope others have better and more updated information.

JP86SS

iTrader: (1)

Other have asked for everyone to share such info so we all can figure out the puzzle. My 350 doesn't like stock resistor values and my 406 certainly will not either. That's why I started the thread to begin with.
Nobody has posted any info concerning which values change the limp PW etc. if they found something.
Until people start sharing what they know, we are all inthe dark (and will remain there)
So please post what you have.
Rant off.

As far as your original question, I don't believe his board relates to anything other than his own numbering scheme.
There is mention of a document that could open this up but has never surfaced. It will take a test bench, a bunch of resistors/jumpers,and someone to use a scope on the ECM outputs to figure it out.
I'm figuring someone has done this and is not sharing.

utilityguy2

iTrader: (1)

Part of the problem lies with the fact that files cannot be attached to the forum, at least I don't see how. I have numerous documents and spreadsheets but no way to post them other than as a image.

bobworkman

Here I may be able to help, and it might get me some new data. I have a Rapidshare account where I could host the files. A pointer on the board would point to the files and they could be donwloaded from there. You would need to email the files to me as an attachment, I would upload them to my account and provide a pointer to them, downloads (size limited) are free to the downloader at a slower speed than account holders. After the upload you could post to the board and use the provided link.

If this interests you let me know. I don't want to post my email in a message, so let me know and I will PM it to you. Maybe my email address is available in members area I don't know.

DENN_SHAH

iTrader: (1)

the files can be uploaded to Moates file manager and i believe EagleMark also has an area for uploads over at http://www.gearhead-efi.com/

RBob

iTrader: (1)

Zip them then attach to a post.

RBob.

utilityguy2

iTrader: (1)

I knew Rbob was there somewhere. I'll get right on it and hope there are others that join in and maybe we can share some info.

utilityguy2

iTrader: (1)

Ok, here are the files that I have collected. I also included some files for the knock filter. I don't represent any of these as fact just as a collection. Use with your own discretion.

netres pinouts info.zip

FILTER BOARD RESISTANCE.zip

bobworkman

Okay, my turn to share my recent acquisition. I queried the gmecm mail list and brentmirchandani returned the mythical memcal.pdf.

This file contains 3 items: 1 page with 2 pictures of ?? and 2 schematics.
The schematics are for a netres and a Memcal resident ESC/Knock board.

With some of the other data already available here this could be a gold mine.

utilityguy2

iTrader: (1)

Looking at the file I guess I didn't know what I was looking at. I saw though that it had a web address on it. It seems these 2 chips are for a GM family Lotus Carlton in the UK. Here is some info I gleaned from the site.

" the Carlton started as a lowly General Motors family sedan--namely, the Vauxhall Carlton. Despite the humble GM origins, anybody assuming the Carlton was a lightly-massaged civilian like the "Handling by Lotus" Isuzus would be tragically mistaken. the Carlton was and is a true world-class sports sedan. Smooth, muscular, and understated, the Lotus looked the part of a deadly serious super sedan. Under the skin, though, things got even more lethal.
Besides the inevitable handling tweaks, Lotus married the Chevrolet Corvette ZR-1's six-speed manual transmission with a twin-turbocharged 24-valve DOHC inline six pumping out 377 horsepower and 419 pound-feed of torque. That's stunning output today, and 18 years ago that was more than all but the top supercars could produce. The result was a comfortable sedan that still ranks as one of the fastest Loti ever made.
The Carlton could accelerate from 0-60 in 5 seconds flat and top out at 176 mph."

I don't think the ECM of that 1990 car was the same as it appears that it's maybe just a fuel netres and the complete memcal? I don't see knock or IO. Do you? They don't appear to have same number of pins do they?

Takes more than me to figure that out. But, good subject hope we can keep going.

cliffharris

The PC board in the picture appears to be a discrete version of the MEMCAL. EPROM in the upper right corner, Z1 - Z4 are the resistor networks, and my guess is that the 26500 IC plus Z5 & Z6 is the knock sensor. I just remembered that GM frequently left off the first couple of numbers on their ICs because there wasn't room for the full number. The real number is probably 1026500. There appears to be a date on the EPROM of February 26, 1991, so this is pretty old info.

I drew up a schematic based on the memcal.pdf file. I drew this pretty fast so there might be errors. Let me know if you find any.

The numbers in the circles are the pin numbers of the MEMCAL carrier. They correspond with Ludis Langens' 1227165 ECM schematic (what's in my car). The numbers outside the circles are the pin numbers of the resistor networks. I also put in the power and ground connections. Edit: I have put in resistors Z4B, Z4C and Z4D for completeness. Also replaced the attached file with a link to the schematic on my web site so I can maintain it easier:

MEMCAL-ECM.pdf

bobworkman

Dave Allen posted this on GMECM.
Additional link to Memcal.pdf and some other related files.
Always good to have a 2nd source.

Thanks for the information guys!
All files are in this zip file and should stay up more or less permanently.
Later,
David

http://68.209.87.173/Temp/NetRes_CALPAK.zip

utilityguy2

iTrader: (1)

This is really great stuff! Now the question is, What are we going to do with it? I'm going to quote whoever wrote this in 2005:

"Update 01/19/2005 - I have kind of given up on trying to "reverse engineer" the 14-pin resistor pack on the 165 V8 MemCal, at least for the time being. After I finished the 16-pin, I was all excited about decoding the 14-pin, but ... the headaches ... The following documentation is left here as reference as it still may provide some other useful information and maybe someone will take this over. The problem with this type of hacking, is that there are many different variants of the resistor pack and even if we were to decode this one, there are others to consider to make this worth while. On the same token, from what it seems, it wouldn't be feasible to create a home-made resistor circuit to replace the chips as it would take up a lot of room.
Perhaps, one day, GM will stop carrying replacement memcals and from there, we would probably need to think of something. "

He was certainly prophetic but that doesn't change anything. We now know how this works but does it change anything? Those resistor packs are still unavailable to us. I don't think the limp home stuff is that important. Do new cars have it? I convert v6 memcals and from all my testing, they will still run a V8, albeit a little rich, but enough to get it off the road. Is that good enough? I think this is a great discussion and with Eaglemark's permission will transfer the data to the Gearhead-EFI website and forum so it is more widely distributed and may get more input. What do you think guys? Where do we go from here?

utilityguy2

iTrader: (1)

Great work! Thanks

utilityguy2

iTrader: (1)

My other question is, Do we actually think this is for a V8 netres? Or from the lotus 32 valve 6 cylinder? We need to be sure as every engine has a different netres. The 1227165 was used in lots of cars and trucks and in foreign GM vehicles it seems. Just a thought.

bobworkman

We in fact know that this was NOT a V8 netres as the fueling mode select
aka cylinder count is set up for 6 cyl probably the Lotus Carlton as that is what is printed on the diagram. See earlier post for discussion of this car.

Even if it was a V8 netres, there are several different variants. However the same physical resitor network sets the parameters ( see link to injector control article in 1st post). So having a layout makes 'reverse engineering' a specific setup more likely.

So What? Well other than the cylinder select no real value unless you are an Info Junkie.

utilityguy2

iTrader: (1)

The 86-88 resistor packs were the same for all 5.0 and 5.7 V8s. In 89 the i/o pack was the same but netres was different (possibly because of 9th injector delete). And for the SD 90-91 the 5.0 and 5.7 V8s were the same. So really only 3 different variants for those memcals. Below is the resistor pack cross reference in case anyone needs it:

1986-1988 5.0-5.7 TPI ---------16055375 16055376
1989 5.0 TPI --------------------16072665 16072667
1989 5.7 TPI --------------------16072665 16133420
1990-1992 F body 5.7 TPI -----16072665 16133420
1990-1991 vette 5.7 TPI -------16072665 16133420

utilityguy2

iTrader: (1)

So, because I am not an electrical engineer, please correct me if I am wrong. Because the 16 pin netres pin 56/13 sees 22.5k ohms to ground the ECM knows that it is a 6cyl. If it is infinity then V8?

cliffharris

I have updated my original post #16. I added the resistors I left off the original schematic.

Keep in mind that the MEMCAL has to work with the ECM it's plugged in to. That restricts the kinds of changes that can be made to the arrangement of the resistors. The obvious one is the cylinder select on pin 56. All the other connections go to the "limp mode" IC (U11 on the 1227165 schematic, Delco 16054995). If we knew what was inside U11 we'd know how to change the resistor values to get the results we want.

Theoretically your car will never go into limp mode anyway... ;-)

In the original post #1 is a link to a thread where jwscab created a PC board with surface mount resistors that replaced the resistor networks. He sold out all he had made and didn't buy more. I happen to know that somebody else is currently working on replicating those PC boards.

utilityguy2

iTrader: (1)

"I happen to know that somebody else is currently working on replicating those PC boards."

That's an interesting development. I hope you tell us more if you find anything out.

cliffharris

He didn't swear me to secrecy or anything, but I just don't want to reveal who it is to maintain his privacy. This is a back burner project for him so I have no idea when it will be revealed. I don't want people to expect something soon that may not happen for a while.

Let's just say that it's somebody you know... ;-)

utilityguy2

iTrader: (1)

Figured as much. Thanks

bobworkman

utilityguy2
actually a Voltage dividing resistor network

between `53 (or 62) [ground], 56 [cal56], and 58 a reference voltage [VREF] determines the memcal mode.
Vout = Va x (R2/(R1+R2))
if the voltage at cal 56 = the reference voltage then single injector TBI mode is selected,
if the voltage at cal56 is 2/3 of VREF then Alternating TBI or 4 cyl PFI mode is selected.
if voltage at cal56 is 1/3 of VREF then 6 cyl PFI mode is selected.
if the voltage at cal56 is 0 or GND then 8 cyl PFI mode is selected.

so if R2 between 52 & 56 is 15K and R1 between 56 & 58 is 7K the voltage at pin 56 is 2/3 of VREF and 6cyl PFI is selected.

by grounding pin 56 (chip pin13) to pin 62, a GND then the voltage at pin56 is ground (zero not infinity) and 8cyl PFI is selected.

Of course reversing the Resitors will set 4 cyl PFI mode or alternating TBI.

utilityguy2

iTrader: (1)

Thank you very much for the very good and detailed explanation. I understand it much better now. Sure wish I had taken up electronics years ago. Thanks.

cliffharris

I went over my schematic (Post #16) and discovered I REALLY screwed it up. Hopefully it's now accurate.

utilityguy2

iTrader: (1)

You're fantastic. I can't believe how much work you've put into this. Thanks

cliffharris

I have the gift (or maybe it's a curse) of extreme focus. Last night I was working on an article I am writing for my web site. When I got done I realized it was 3:00 AM. One time I was working on something and it started getting light outside...

utilityguy2

iTrader: (1)

I know the feeling. I think you helped out here tremendously.

bobworkman

Just a small label issue with the latest revision of CliffHarris' great schematic.

You have labeled Cal56, pin 56 of the Memcal ( pin 13 of the 16 pin chip) as cylinder select. Cal56 is more accurately labeled Fueling Mode Select,
or mode/cyl select as it is usded to select between 5 different modes.
Two TBI modes and three (cyl dependant) Port Injection modes.

I have verified on 2 physical Memcals, both for TBI, one for a 6 cyl and the other for an 8 cyl. They are both set to the same Mode, Mode 1 alternating TBI which is set by a voltage output at Cal56 of 2/3 Vnet.

utilityguy2

iTrader: (1)

Are you saying that on the TBI that 8 and 6 cylinder mode/cylinder select is the same? no difference? Whereas on TPI there is a difference?

bobworkman

You Win the 64,000 prize.

In TBI cyl count does not matter as the injectors fire into a common Plenum so they pump fuel every Distributor Reference Pulse.

In TPI the fuel is pumped in closer to the valve and a more complex Injector Pulse program is used which is cylinder count dependant.

If you refer to the Injector Control article I referenced in the 1st post it gives a good description of this process. (

On the Binder Planet board http://www.binderplanet.com/forums/f...splay.php?f=75

They talk about TBI to MFI conversions and even if you start with an 8 cyl Memcal, you must jumper 16 pin chip pin #13 (memcal pin #56) to ground to get the injectors to fire correctly. There are changes required in the 'Bin file' as well. Skip the jumper and you will be in 4 cyl port injection mode.

As I said I actually measured 2 TBI Memcals to verify there was no difference.

utilityguy2

iTrader: (1)

Well that does make perfect sense about TBI. But still confused (always) about a statement you make: "Skip the jumper and you will be in 4 cyl port injection mode".

What about 6 cylinder? They have no jumper and it works. I know that the 8 cyl TPI fires on intake and exhaust or bank firing. is that not more similar to 4 cyl than 6 cylinder would be. Like I said "confused"

utilityguy2

iTrader: (1)

By the way, thanks for the IH site reference.

bobworkman

Okay, I will be talking about pins on the 16 pin chip, I may reference memcal pin # by including them in ().
So we are talking about the resistor network between pins 12 - Vnet or reference voltage, pin 13 (cal56) - mode/cyl select, and pin 3 -Ground.



Call the resistor between pins 12 & 13 R1 and the resistor between pins 3 & 13 R2 the the % of the voltage applied at pin 12 which is output on pin 13 is expressed as V% = R2/(R1 + R2).

if V% is 100% (right fig) then single injector TBI mode is chosen.
if V% is 66% (right of center) either alternating 2 injector TBI mode is chosen or 4 Cyl Port injection is chosen. ( nore later)
if V% is 33% (left fig) then 6 cyl Port injection is chosen.
if V% is 0% (left of center), there is no connection to Vnet so no voltage is available at 13, so 8 Cyl Port injection is selected.

There is a flag that can be set in the BIN to select TBI/PFI, I assume that is what discriminated between Alternating TBI (mode 1) and 4 Cyl PFI. Setting this flag to PFI without the jumper would put you in 4 Cyl mode.

The 'Hack' to change a 6 cyl memcal involves jumpering pin 13 to pin10, another ground this forces pin 13 to ground. Purists will cut pin 13's leg on the chip and jumper the memcal header where it connected Cal56 but just jumpering without cutting the pins leg is reputed to work.

cliffharris

I just heard from Craig Moates, and he told me it's OK to reveal that he's the guy.

utilityguy2

iTrader: (1)

Lucky guess for me

utilityguy2

iTrader: (1)

Very very nice and clear writeup Thanks.

Just an couple observations:

"There is a flag that can be set in the BIN to select TBI/PFI, I assume that is what discriminated between Alternating TBI (mode 1) and 4 Cyl PFI. Setting this flag to PFI without the jumper would put you in 4 Cyl mode."

The flag must only be in 4 cylinder bin as have never seen it in 6 or 8

"Purists will cut pin 13's leg on the chip and jumper the memcal header where it connected Cal56 but just jumpering without cutting the pins leg is reputed to work."

Call me a non-purist, it does work having sold over 100 with great results.

bobworkman

"There is a flag that can be set in the BIN to select TBI/PFI, I assume that is what discriminated between Alternating TBI (mode 1) and 4 Cyl PFI. Setting this flag to PFI without the jumper would put you in 4 Cyl mode."

The flag must only be in 4 cylinder bin as have never seen it in 6 or 8
See post #2

http://www.hptuners.com/forum/showth...light=tbi+mpfi


"Purists will cut pin 13's leg on the chip and jumper the memcal header where it connected Cal56 but just jumpering without cutting the pins leg is reputed to work."

Call me a non-purist, it does work having sold over 100 with great results.

Yes, but Ohm's law say 1/R3 = 1/R1 + 1/R2 so if R1 =15 and R2 = 0 then R3 = 1/15 + 1/0, but 1/0 if mathematically undefined so the result is invalid. In real life it works but to a purist it is invalid.

cliffharris

Zero ohms in parallel with anything is going to come out zero ohms (that's a short).

The other version of that equation is R3 = (R1 * R2)/(R1 + R2) = 0/15 = 0

utilityguy2

iTrader: (1)

"There is a flag that can be set in the BIN to select TBI/PFI, I assume that is what discriminated between Alternating TBI (mode 1) and 4 Cyl PFI. Setting this flag to PFI without the jumper would put you in 4 Cyl mode."

The flag must only be in 4 cylinder bin as have never seen it in 6 or 8
See post #2

http://www.hptuners.com/forum/showth...light=tbi+mpfi


There is no such sync or async scalars in any of the TPI V8 bin files. Must be bin specific or?

the blur

My 1989 5.0 TPI has 16072665 16133420
which you have listed for the 5.7 motor.

utilityguy2

iTrader: (1)

What is the BCC on the Prom?

the blur

ANYX

utilityguy2

iTrader: (1)

ANYX is for a 1989 5.0 5 speed with the G92 optional 3:45 gear ratio. It is a pretty rare memcal and not found in any of the major databases. Not sure why it would have the netres the same as a 5.7. Don't think it matters much either. Here is some more info on your memcal:

www.thirdgen.org/techboard/diy-prom/652812-anyx-bcc-hard-find.html

the blur

and I think it only applies to the N10 dual cat cars.