LLY Ficm

[2004LB7]

AMIS-30421

5v stepper motor controller. Sounds right, fast response. Used on both lly and lb7 models. It's likely a keeper unless something better exists now.



Sent from my SM-S901U using Tapatalk

There are really good dedicated boost converter ICs now with integrated FETs that I'd recommend for new designs. Likely get less ripple and no need for a microcontroller to run things. Still have an enable pin if we had to shut it down for whatever reason.

But I suppose given the voltage conversion doesn't appear to be a failure point, why mess with success?

Sent from my FlashScan V2 using Tapatalk

Either of you manage to find the datasheet for this chip?

Google seems to indicate that it is a 30421 BOSCH QFP-64. I have one board with, what looks to be a dead pin that should be turning on bank 2 power but isn't. No signal coming out of the pin. I get signal on bank 1 from the same chip. Bank 1 is turned on from pin 37 and bank 2 is pin 38

The only datasheet that I can find is one that has the wrong number of pins so I can't rely on that

If these chips don't require programming then I might try replacing it. Found one on eBay

 

[kidturbo]

That other flash chip had a bosch # option also. Would check under that manufacturer site. Some stuff Google bots don't catch.

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

That other flash chip had a bosch # option also. Would check under that manufacturer site. Some stuff Google bots don't catch.

Sent from my SM-S901U using Tapatalk

No luck

 

[2004LB7]

more info. current draw test to see what these things really take. I'm thinking that the fuse might be too large for the best protection. the use of the smallest fuse that meets the requirements of the FICM's demand will provide better protection from burning up the board if something shorts out. thus making repairs easier.

I used a 1 ohm resistor in place of the fuse. convenient because one volt = one amp.

peak current was about 8 amps but its not continuous. it was in 96Hz pulses. almost looks as if the pilot and main pulse are the driving factor in the current waveform shape. but what's weird is it didn't seem to change much when I revved it or power braked it. the pulse got slightly wider but didn't increase any. I'm going to try out a 10 amp fuse. if that doesn't hold then i'll bump it up to a 15. currently (haha get it, current...) it runs on a 10 amp fuse.

this waveform should directly represent the actual current draw. note the negative current. is some of the current form the capacitors flowing backwards?

 

[TheBac]

Dang. Kole was just asking me at lunch if we knew what the current draw was.

 

[kidturbo]

I just found that like our flash chips, there is a slight difference in part #s between IC 30402 models.

Here is what's on lb7 along with his little ST brother 3043548 [middle back side of board] which I traced out, and sets between the 2 canbus pins on 30402, and our canbus pins in the case.

So even though I, or Googles best AI, found no specufic data sheet on this little 3043548, just based off part # likeness, and position in the circuitry, we can probably assume it's job is as a canbus transceiver for our big Bosch chip.

So we pretty well know that Bosch 30402 is our main workhorse on this job. He's also only chip with direct heat sink attachment, and lights up as hottest thing on the board in IR. While totally idle...

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

more info. current draw test to see what these things really take. I'm thinking that the fuse might be too large for the best protection. the use of the smallest fuse that meets the requirements of the FICM's demand will provide better protection from burning up the board if something shorts out. thus making repairs easier.

I used a 1 ohm resistor in place of the fuse. convenient because one volt = one amp.

peak current was about 8 amps but its not continuous. it was in 96Hz pulses. almost looks as if the pilot and main pulse are the driving factor in the current waveform shape. but what's weird is it didn't seem to change much when I revved it or power braked it. the pulse got slightly wider but didn't increase any. I'm going to try out a 10 amp fuse. if that doesn't hold then i'll bump it up to a 15. currently (haha get it, current...) it runs on a 10 amp fuse.

this waveform should directly represent the actual current draw. note the negative current. is some of the current form the capacitors flowing backwards?
View attachment 117614

Remember in the video from guy mapping out the dead injector circuit, whe he talked about voltage spike, or flyback after the injector fires? He had a pickup coil on the injector wires and caught it. Would suggest check his work and compare what you are seeing on scope. Or put a pickup on injector and log that signal together with your power supply scope capture.

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

Remember in the video from guy mapping out the dead injector circuit, whe he talked about voltage spike, or flyback after the injector fires? He had a pickup coil on the injector wires and caught it. Would suggest check his work and compare what you are seeing on scope. Or put a pickup on injector and log that signal together with your power supply scope capture.

Sent from my SM-S901U using Tapatalk

My cheap scope doesn't seem to like the injector or control signals. Something about it seems off. Only shows a few volts at most when I can get it to show. Probably something having to do with it being a negative side switch. Might try some other pins to see if it was just a fluke (and not the DVM type ) . If I can get it to trigger properly then I can measure the current at the same time and see if they match

 

[2004LB7]

So we pretty well know that Bosch 30402 is our main workhorse on this job. He's also only chip with direct heat sink attachment, and lights up as hottest thing on the board in IR. While totally idle...

Sent from my SM-S901U using Tapatalk

I noticed that with the board open and exposed. While running barely gets warm. You can touch all the chips and none feel hot. So this idea of needing fuel to cool it is busted. Not sure what the original intent was from the engineers. Maybe the plate on the backside with fuel running through it was just a heat shield to protect it from radiating heat off the engine???

 

[kidturbo]

Don't measure the pins, wrap some bare wire around both lines at injector and measure the pulse signal that way..

At least that works on my bench scope..

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

Don't measure the pins, wrap some bare wire around both lines at injector and measure the pulse signal that way..

At least that works on my bench scope..

Sent from my SM-S901U using Tapatalk

Good idea. I thought about doing that earlier on but figured I couldn't measure how much voltage it actually was unless I had a carefully wrapped and calibrated. But just to see the pulses and see if they line up with the current draw would work

 

[DAVe3283]

The fuel to cool it was always nonsense. I've got 60,000 miles on my FICM running dry in the stock location, and 350k miles (10,000 hours) on it total. Sometimes I think designers and engineers get bit by bureaucracy or just make an honest mistake, and we end up with fuel cooled injector drivers for no actual reason.

The LB7 was a rushed design, I believe it was the quickest inception to production engine in GM history at that point. I bet some engineers got in an argument about the lifespan of the FICM and the testing would take months, and a manager asked if liquid cooling it would help. Probably got a "yes" and a "won't do anything", and under a time crunch those distill down to "liquid cool it and move on".

Sent from my FlashScan V2 using Tapatalk

 

[kidturbo]

Couldn't agree more.. I was told by a GM tech over a decade ago that had to be fuel cooled or it would burn the board. That's what he was told in class...

But only that one chip position has direct contact to the backplane. And not directly to the chip, but pcb under the chip. like they were worried about it burning the board more than cooling the chip...

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

I just sent off a reply email to the folks at Bosch Repair again. Asking for any schematics or diagrams available for both versions. I explained that since they offer no options to replace or repair, we are looking to fix what's currently in existence. Hopefully it reaches a sympathetic ear.

Also while searching for more details on the 30402, i stumbled across a site which lists it in some schematics. And guess what, the schematics are for EDC15 and 16 version boards, used in TDI's.. That said, they want 100Euro for the group of files. But I did look thru what ya get, and all our boards cousins seem to be in that list.. So, while I search by the doc titles for these schematics in other locations, it's looking up on finding the ones we need. At worse, we start with closes match available to our layout, and guestimate the rest..

ECU Circuit Diagram for Bosch (ECU Schematic) - AUTODTC.NET

ECU Circuit Diagram database - download 400 documents : ECU wiring diagram, schematics for cars and info for Bosch EDC16, EDC17, EDC15, ME71, ME75, MED7.

autodtc.net

Ah yeah, they are all written in German..

 

[kidturbo]

Ok I've gone way deep down the rabbit hole, and found most of what we need exists out there for free... Well if ya sign up for a trial svc at https://www.scribd.com and start digging around by chips numbers, or ECDxxx, you will discover good documents uploaded by someone.

Which is likely better option than I just received from Bosch. Either way, plan to take up German as a second language today... LOL. But with a little help from Google Translator, Wir haben diese Scheiße...

Ok first, here is the Bosch reply in English. Sending reply back with, "happily join, once someone can tell me how to translates linked site to English.." But for now, here is where they pointed me.

thank you for your enquiry.

It is not possible to send this information as it is part of the chargeable diagnostic software (ESI[tronic]).

If you do not have a paid subscription, please contact a Bosch Car Service near you with your request:
https://www.boschcarservice.com/de/de/werkstattfinder/?searchLocation=

Mit freundlichen Grüßen / Best regards

---

So while waiting on above reply, I joined a German auto tuning page listing that complete schematics dump, available at only $30. Just have to figure out the subscription plan, and rules to DL. But all 450 documents look to be there. Will grab, right after I ask a buddy about this RAR.

But with the file name list of said documents attached below, I've been searching at Scribd.. And guess what I found for ya Jason?? Probably everything ya need to know about Mr. IC 30402, and how to test it on page 8.. So here is the brief translation about it, along with link to one EDC15-pdf that covers the testing procedures for board and chips. These should keep your meter busy for a couple hours at least. While I'm gonna renew my subscription to Babbel, and go make some new friends.

Bosch 3042 Voltage Regulator. The voltage regulator that this ECM has is from the manufacturer BOSH with number 30402 When searching for information, the datasheet about this integrated device is not easily available. but to have an idea we are going to perform voltage measurements on each pin, to establish a table of values and being able to decipher the operation of each of these.

 

[kidturbo]

Starting to think I should have taken that Blue pill.




All those schematic files are also available here, some free. But don't go paid, cause site is heavily overloaded and dialup download speeds. 15 minutes, 2% on 60mb file. $30, and red pill life not looking so good..

Repair Every Car ECU And Immobiliser 🚗 | ImmoOFF

Thousands of Files And Software For Repairing Every Car ECU & Unlock All Car Immobilisers. Mechanic, Bodybuilder, Locksmith, Car Enthusiast.

immooff.net

 

[2004LB7]

Good stuff there Ken. Don't worry about those Rar. Files. They are just an alternative to Zip files. Another method of compressing files into a folder like package. I've been using them for at least the last 20 years. Just download 7zip, WinRAR, WinZip, or a handful of others that can decompress them. Or better yet, just use an online Rar extractor and let it do the heavy lifting

I was thinking , worse case if you can't get bosch to give you the schematics then we take a bad board and slowly peal back the layers. Unsolder all the components. Take some high resolution photos, lay some sand paper on a piece of glass and slowly sand away the top layer. Take more photos. Rinse and repeat until we can all the layers photographed. Then we can follow the traces from each layer. It would suck. Take forever and have some risks but it would work as a last resort.

But I think before we even get that far we are probably better off making our own FUCM board. Even if it doesn't have the communication or crank signal lines.

 

[1FastBrick]

What about an X-ray of the board though??? He can do a CT scan but he said you get shadows from the Solid objects like the caps which it can not see through.

 

[kidturbo]

Good stuff there Ken. Don't worry about those Rar. Files. They are just an alternative to Zip files. Another method of compressing files into a folder like package. I've been using them for at least the last 20 years. Just download 7zip, WinRAR, WinZip, or a handful of others that can decompress them. Or better yet, just use an online Rar extractor and let it do the heavy lifting

I was thinking , worse case if you can't get bosch to give you the schematics then we take a bad board and slowly peal back the layers. Unsolder all the components. Take some high resolution photos, lay some sand paper on a piece of glass and slowly sand away the top layer. Take more photos. Rinse and repeat until we can all the layers photographed. Then we can follow the traces from each layer. It would suck. Take forever and have some risks but it would work as a last resort.

But I think before we even get that far we are probably better off making our own FUCM board. Even if it doesn't have the communication or crank signal lines.

No issue opening a rar, just didn't want to spend $30 if I knew someone who had it... lol

In that zip file I posted above is name of all the board images used to print the PCBs, and the full schematics for each. Directly from Bosch.. Many of the names, match up with the actual board picture o shared couple days back.

So thinking we start with those one we thought look closest to our ficm, and should give us the overall routing, along with all the chips between our bigger ICs. It's a big dump of files, 50/50 these are in there but i can only search by unit designated name, not the bosch 280 #.

Sent from my SM-S901U using Tapatalk

 

[2004LB7]

What about an X-ray of the board though??? He can do a CT scan but he said you get shadows from the Solid objects like the caps which it can not see through.

I talked with a company that does high resolution CT scans of internal parts. They said they've scanned many PCBs before. They can work around components still being on the board but said they would get better results with them removed. Especially any high density ones like chokes/inductors and some ICs. So best bet is to remove as many as possible. Probably something best done with a failed one where the board looks fine but one or more chips have failed. Since we would have to remove the chip anyways.

I sent a few photos of the board so they can get me some quotes. We will see what they say