LLY Ficm

[kidturbo]

Lighting up this old tread again since it's somewhat related to my questions below. And it's not eleven years old as the one below. Which I found was fairly informative on the subject. Back when the price of a good used FICM was less than shops get to replace the $2.00 capacitors in one today...

FICM's swappable?

From a LB7 into an 05 LLY? Voltage differences perhaps?

www.duramaxdiesels.com

My main questions are, has anyone here revisited this LB7 to LLY FICM swap approach with any techie hardware like a scope or common J1939 daig tools? Or even just cracked open both type units side by side to compare the guts?

It must be pretty bad out there when people come to me looking for options.. I personally haven't messed with anything LLY since the OG Dmax boat a decade ago.. But that was my first true stand alone build, and I'm pretty certain my GM hacking skills have much improved. From what I hear, sourcing new GM replacements units might be an issue.

So before I go asking buddies to let me hack around on their perfectly good running twenty year old trucks, it would be nice to know if someone has already discovered the answers.

If it's only comm's related differences as previously mentioned, we can likely fiddle with that much easier today. Maybe a simple firmware upgrade would get ya going. If the PCB hardware between them is noticably different, then probably not worth even cracking cases open. But said I'd take a gander, so here I am..

 

[Bdsankey]

Lighting up this old tread again since it's somewhat related to my questions below. And it's not eleven years old as the one below. Which I found was fairly informative on the subject. Back when the price of a good used FICM was less than shops get to replace the $2.00 capacitors in one today...

FICM's swappable?

From a LB7 into an 05 LLY? Voltage differences perhaps?

www.duramaxdiesels.com

My main questions are, has anyone here revisited this LB7 to LLY FICM swap approach with any techie hardware like a scope or common J1939 daig tools? Or even just cracked open both type units side by side to compare the guts?

It must be pretty bad out there when people come to me looking for options.. I personally haven't messed with anything LLY since the OG Dmax boat a decade ago.. But that was my first true stand alone build, and I'm pretty certain my GM hacking skills have much improved. From what I hear, sourcing new GM replacements units might be an issue.

So before I go asking buddies to let me hack around on their perfectly good running twenty year old trucks, it would be nice to know if someone has already discovered the answers.

If it's only comm's related differences as previously mentioned, we can likely fiddle with that much easier today. Maybe a simple firmware upgrade would get ya going. If the PCB hardware between them is noticably different, then probably not worth even cracking cases open. But said I'd take a gander, so here I am..

I believe Jason (@2004LB7) was working on it with the help of another member here to do exactly this.

 

[kidturbo]

I believe Jason (@2004LB7) was working on it with the help of another member here to do exactly this.

Glad to hear. Pretty certain there is no E41 level of complexity hidden inside those old boxses. So happy to help if I can.

 

[1FastBrick]

I was going to try and get some base readings from my un-molested LLY for him.

 

[1FastBrick]

I believe Jason (@2004LB7) was working on it with the help of another member here to do exactly this.

I think this is one of the threads you were referring too which has a link to his other thread...

LB7 FICM

This will be a continuation of the thread where we dissect the failed LLY FICM. https://www.duramaxdiesels.com/forum/threads/lly-ficm.89101/ This one was graciously loaned to me by Brad (Bdsankey) (to look over and see if there are any worthwhile capacitor upgrades or other modifications that...

www.duramaxdiesels.com

 

[2004LB7]

Comparing the two side by side, there are definitely noticeable differences but also some similarities. One notable difference is the use of film capacitors in the LB7 FICM vs the electrolytic in the LLY. Possibly one of the reasons for the longer life on the LB7.

The LB7 also seems to use two separate banks of smoothing caps for each of the input power lines whereas the LLY's as one bank of caps that join the two feed together.

The boost converter on the LLY looks pretty straight forward with the inductor capacitor LC network but not on the LB7 which has no inductors. So I'm not sure how it's generating such high voltage with just the capacitors unless it's using the injectors as the inductor which would be a neat trick.

The pinout is identical for the two including the naming of the communication pins. My guess is the actual message being sent to and from the FICM and ECM is different as opposed to the communication method. I am wondering if detaching the comm wires from the ECM will get it to work because apparently the ECM will output a trigger signal wether the FICM is connected or not. Maybe it's just a wakeup message to get the FICM to turn on. You would have to tap in to those wires and decode the messages and the difference between the two FICM's.

Personally I think the best route to take is to build one ourselves. May not even need to communicate with any other module. Just triggers the injectors based on the ECM trigger signal. That is assuming that the trigger varies based on PW and timing. If the comm signal is what varies the PW then that can make things complicated. I doubt the timing is through the comms because there would be no need for eight individual trigger wires from the ECM to the FICM if so. Hopefully the trigger contains both the timing and PW and the communication wires are just for codes (and maybe turning on and putting it to sleep)

 

[darkness]

Source: https://youtu.be/4k-G1dMgFUo

Source: https://youtu.be/Grz_rmClH40

I watched this the other day. He does a lot of scoping and different tests. Was pretty interesting. Might shed some light? I don’t know.

 

[kidturbo]

Personally I think the best route to take is to build one ourselves. May not even need to communicate with any other module. Just triggers the injectors based on the ECM trigger signal. That is assuming that the trigger varies based on PW and timing. If the comm signal is what varies the PW then that can make things complicated. I doubt the timing is through the comms because there would be no need for eight individual trigger wires from the ECM to the FICM if so. Hopefully the trigger contains both the timing and PW and the communication wires are just for codes (and maybe turning on and putting it to sleep)

This is basically same approached I was presented with also. If word on street is correct, probably the correct path to a lasting solution. So a replacement that covers both models would be the ticket.

They can't be much more than a simple signal Amplifier, because the J1939 protocol bus speed is way-to-slow to move anything more usable than some basic engine parameters like RPM load, APPS, temps, and boost. I'm certain that's all there is, cause I learned to sniff canbus on my old LLY... Bosch added that crank signal wire in there, obviously to keep things timed correctly.

The Ford 6.OL FICM was my first though for a replacement strategy, since it ran similar voltage, and is still more boneyard common than finding a good running 6.0L engine.. lol. But a little digging, Ford actually powered the coils to close the valves on every stoke also.. Which explains alot, if ya know much about 6-Leakers... They do however share a common point of failure with the Bosch units. Burnt out Caps..

So we have 8 signal/control wires that we need to scope out the old fashion way first. Your assumptions that dwell is controlled by those lines, is also where I feel we should start comparing. It's totally plausible there could be differences in the voltage or pwm signal. But I doubt Bosch reinvented the wheel for only a 2 year deal, while they were also prepping the E35A PCM model that followed. It's way more likely some simple Wake The F Up command is our only major difference between those early GM designs. I did just see where LLY used pin [52 GY/BK - 2219 - FICM Relay Control] to light the fuses. Vaguely recall having to wire that one up in my stand alone build.

All I have to offer right off the top. See no harm in sniffing some bus, and comparing all the J1939 packets between a LB7 and LLY build. GM could have tossed something proprietary in on that line, but seriously doubt it. So that leaves us 8 control wires to compare, and shouldn't take more than sniffing one on each to catch the diff.

Let me know if anyone wants to meet up at a local shop around here one evening or over a weekend to hash it out. My favorite local GM guru moved on to electric lifts, but he still has a couple running trucks, and plenty of boxes of LLY parts we can hack around on.

 

[2004LB7]

This is basically same approached I was presented with also. If word on street is correct, probably the correct path to a lasting solution. So a replacement that covers both models would be the ticket.

They can't be much more than a simple signal Amplifier, because the J1939 protocol bus speed is way-to-slow to move anything more usable than some basic engine parameters like RPM load, APPS, temps, and boost. I'm certain that's all there is, cause I learned to sniff canbus on my old LLY... Bosch added that crank signal wire in there, obviously to keep things timed correctly.

The Ford 6.OL FICM was my first though for a replacement strategy, since it ran similar voltage, and is still more boneyard common than finding a good running 6.0L engine.. lol. But a little digging, Ford actually powered the coils to close the valves on every stoke also.. Which explains alot, if ya know much about 6-Leakers... They do however share a common point of failure with the Bosch units. Burnt out Caps..

So we have 8 signal/control wires that we need to scope out the old fashion way first. Your assumptions that dwell is controlled by those lines, is also where I feel we should start comparing. It's totally plausible there could be differences in the voltage or pwm signal. But I doubt Bosch reinvented the wheel for only a 2 year deal, while they were also prepping the E35A PCM model that followed. It's way more likely some simple Wake The F Up command is our only major difference between those early GM designs. I did just see where LLY used pin [52 GY/BK - 2219 - FICM Relay Control] to light the fuses. Vaguely recall having to wire that one up in my stand alone build.

All I have to offer right off the top. See no harm in sniffing some bus, and comparing all the J1939 packets between a LB7 and LLY build. GM could have tossed something proprietary in on that line, but seriously doubt it. So that leaves us 8 control wires to compare, and shouldn't take more than sniffing one on each to catch the diff.

Let me know if anyone wants to meet up at a local shop around here one evening or over a weekend to hash it out. My favorite local GM guru moved on to electric lifts, but he still has a couple running trucks, and plenty of boxes of LLY parts we can hack around on.

I would put money on the 8 control signals from the ECM to the FICM are identical for the LB7 and LLY. The video above seems to show a 5 volt signal. To bad he didn't rev the engine to see if that pulse got wider or changed in any way. It did appear to match the on duration of the injector which makes me believe it's doing the PW and timing. If that's true, how relevant would the crank/signal wire be for timing the injectors? Interestingly the video also shows a 20 or so ms difference between the ECM signal and the actual injector. I wonder if that is a fixed value or is that what the crank signal is for.

 

[juddski88]

This is basically same approached I was presented with also. If word on street is correct, probably the correct path to a lasting solution. So a replacement that covers both models would be the ticket.

They can't be much more than a simple signal Amplifier, because the J1939 protocol bus speed is way-to-slow to move anything more usable than some basic engine parameters like RPM load, APPS, temps, and boost. I'm certain that's all there is, cause I learned to sniff canbus on my old LLY... Bosch added that crank signal wire in there, obviously to keep things timed correctly.

The Ford 6.OL FICM was my first though for a replacement strategy, since it ran similar voltage, and is still more boneyard common than finding a good running 6.0L engine.. lol. But a little digging, Ford actually powered the coils to close the valves on every stoke also.. Which explains alot, if ya know much about 6-Leakers... They do however share a common point of failure with the Bosch units. Burnt out Caps..

So we have 8 signal/control wires that we need to scope out the old fashion way first. Your assumptions that dwell is controlled by those lines, is also where I feel we should start comparing. It's totally plausible there could be differences in the voltage or pwm signal. But I doubt Bosch reinvented the wheel for only a 2 year deal, while they were also prepping the E35A PCM model that followed. It's way more likely some simple Wake The F Up command is our only major difference between those early GM designs. I did just see where LLY used pin [52 GY/BK - 2219 - FICM Relay Control] to light the fuses. Vaguely recall having to wire that one up in my stand alone build.

All I have to offer right off the top. See no harm in sniffing some bus, and comparing all the J1939 packets between a LB7 and LLY build. GM could have tossed something proprietary in on that line, but seriously doubt it. So that leaves us 8 control wires to compare, and shouldn't take more than sniffing one on each to catch the diff.

Let me know if anyone wants to meet up at a local shop around here one evening or over a weekend to hash it out. My favorite local GM guru moved on to electric lifts, but he still has a couple running trucks, and plenty of boxes of LLY parts we can hack around on.

Where is "here" these days?

 

[kidturbo]

Where is "here" these days?

Southeastern Ohio today. But subject to change on short notice..

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

I would put money on the 8 control signals from the ECM to the FICM are identical for the LB7 and LLY. The video above seems to show a 5 volt signal. To bad he didn't rev the engine to see if that pulse got wider or changed in any way. It did appear to match the on duration of the injector which makes me believe it's doing the PW and timing. If that's true, how relevant would the crank/signal wire be for timing the injectors? Interestingly the video also shows a 20 or so ms difference between the ECM signal and the actual injector. I wonder if that is a fixed value or is that what the crank signal is for.

I reviewed your work in the other thread, but never got around to watching those videos yet. Doing it now.

But off the top, yes the dwell time should be incoded in that ecm signal somehow. Guessing the pin 94 would be rpm ref from ecm that we used to tie in analog tach adapters.

Sent from my SM-S901U using Tapatalk

 

[darkness]

So off topic but about those videos. The 2 firings of an injector so close together, is that the pilot then the main injection?

 

[kidturbo]

Source: https://youtu.be/4k-G1dMgFUo

Source: https://youtu.be/Grz_rmClH40

I watched this the other day. He does a lot of scoping and different tests. Was pretty interesting. Might shed some light? I don’t know.

Someone buy that guy a beer...

And maybe send him a brief descriptions of pilot injection. But then we can just send him an LB7 truck to test on, and never even need to open a hood to answer these most difficult questions... lol

He really did a nice job of sorting that for someone who'd never scoped out a diesel injector circuit. Comparison of the control circuit voltage and patterns between LB7 and LLY units is what's need to see first. If those are identical, then only a CANbus message difference would stop it from firing.

That 22ms delay between Control and Commnd output times is likely a constant, and due to processing times in FICM. And I wish he would have directly compared the 5V Control circuit dwell times in ms, to the injector values, but seen enough to say was exactly as I perceived.

5v triggered 48v, and pulls about 20amps. Got it..

 

[kidturbo]

So off topic but about those videos. The 2 firings of an injector so close together, is that the pilot then the main injection?

Correct. I have some prettier scope captures from my LMM bench setup I posted on here a while back. In the video he showed 2 pulses in, 2 pulses out.. Dwell time of each, along with time between Pilot and Start of Main Injection Pulse, is all controlled in the ECM tunes.

 

[2004LB7]

So if we manage to make our own module, we would either have to make sure to purposely put a dwell time in. Or as an alternative you would have to adjust the tune timing to compensate for it. It would be easy to do custom dwell with an Arduino or other programmable chip but I would prefer to do it with discreet components for repairability and reliability. I think we should make something that has no programming on it and it just receives the ECM signal amd fires the injector. If we can do it with through hole components instead of surface mounted even better

Assuming the crank timing circuit is not actually needed, and assuming the ECM signals are reliable and consistent with timing themselves then the circuit is pretty straightforward and not that difficult. If you have to time the injectors off the crank signal then we probably have no choice but to go to programmable chips

 

[2004LB7]

Someone buy that guy a beer...

And maybe send him a brief descriptions of pilot injection. But then we can just send him an LB7 truck to test on, and never even need to open a hood to answer these most difficult questions... lol

He really did a nice job of sorting that for someone who'd never scoped out a diesel injector circuit. Comparison of the control circuit voltage and patterns between LB7 and LLY units is what's need to see first. If those are identical, then only a CANbus message difference would stop it from firing.

That 22ms delay between Control and Commnd output times is likely a constant, and due to processing times in FICM. And I wish he would have directly compared the 5V Control circuit dwell times in ms, to the injector values, but seen enough to say was exactly as I perceived.

5v triggered 48v, and pulls about 20amps. Got it..

At first I was thinking this guy is an idiot. Making all the wrong assumptions and applying gasser knowledge to diesels, etc. But in the end he did do a good job for someone who has no prior experience or knowledge of how the system was supposed to work. I also think his electrical diagram is either wrong or he is interpreting it wrong. He went around and around on this injector power wire no having power. It was a little difficult to see for sure but it almost looked like it was showing the power being supplied by a single wire per bank then in the harness it split to each injector. But we know that all this is inside the FICM and there are individual pins and wires for each injector. He could have potentially brought that dead injector circuit back up by splicing one of the other injector power wires into the dead one. Internally they are all connected being supplied by the same boost circuit. He also didn't check the resistance or continuity of the harness between the FICM and the injector. But I guess probing the FICM connector and seeing no voltage was enough for him. What if it was a bad pin?

Oh, and the way he says "FICM" bugged me. Instead of saying each letter. Imagine if we read all the different modules that way. ECM, TCM, BCM, PCM, etc would sound ridicules. phonetic transcription of each of these is to say each letter as they are and not as a single word. Sorry for that last rant

 

[kidturbo]

So if we manage to make our own module, we would either have to make sure to purposely put a dwell time in. Or as an alternative you would have to adjust the tune timing to compensate for it. It would be easy to do custom dwell with an Arduino or other programmable chip but I would prefer to do it with discreet components for repairability and reliability. I think we should make something that has no programming on it and it just receives the ECM signal amd fires the injector. If we can do it with through hole components instead of surface mounted even better

Assuming the crank timing circuit is not actually needed, and assuming the ECM signals are reliable and consistent with timing themselves then the circuit is pretty straightforward and not that difficult. If you have to time the injectors off the crank signal then we probably have no choice but to go to programmable chips

At risk of jinxing myself, believe I may already know why the FICM's won't interchange. Hopefully someone can scope this out quickly, or maybe verify from past experience. I'm sure I posted on this somewhere back in the 2009 era also..

Back when I purchased the OG PPE boat, the nice new Autometer Tach wasn't hooked up. It also had a couple functional J1939 MFD's, so I guess nobody got around to running that wire. I proceed to try and connect it straight to the ECM. No luck. Ended up getting a PPE or DD Tach Adapter. To which I recall there being some difference between the ECU output signals? Square wave vs sign wave Maybe? Or possibly it used a different signal wire on ECU than PPE suggest. But I do recall there being some differece between an LB7 setup, and a LLY when it came to wiring up an aftermarket tach. Which would certainly come into play if that same signal is required by the FICM, to say control the injector Lag time between our Control and Comand lines pulse per say... If I'm correct, all else is equal on the 5v Control pulses, CANbus data, then this is a super easy hack..

SGI-100BT: Universal Speedometer and Tachometer Interface

Dakota Digital manufactures digital instrumentation and accessories for the automotive, motorcycle and car audio enthusiast.

www.dakotadigital.com

On starting from scratch on new board. Your on the right course.. Lets build an stupid Driver Amp for starters and run it on bench with a scope, some injectors, and compare. I believe the Bosch FICM is likely just simple amp, with signal in equals signal out. The timing and pulse width is in that 5v signal from ECU. Amp it up, push it out, repeat. All the other chips on that board, likely diagnostic and canbus related. Way to tell, cut the canbus wires with one running..

 

[kidturbo]

Found it. Mind like an elephant. Google that is..
From May 2010 on another channel.
--
FYI: Tach signal is ECM black plug "middle" pin 63 Lt Blue wire. Complements of the guys over at PPE.
CORRECTION: Correct pick up for the tach is C3 "gray" ECM plug, Pin 37, white wire.
--

Ok, so we see that the LLY has "2" [Engine Speed Signal] wires listed in the pinout.
C2-63
C3-37

The LB7 pinout lists a single [Engine Speed Signal] wire.
C2-76

Now I somewhat recall testing both those wires on the LLY with the aftermarket Tach, and the white wire on pin 37 was the only on which worked. Not certain if that DD adapter was actually required or not.. So question is, why does the LLY have 2 pins? And what's the difference? See where I'm going?

While comparing the control wire signals, I would also suggest we compare the signal pattern on Pin 94 - Lt Blue of the FICM, between the LLY and LB7 model. That will knock out the hardwire portion.

 

[TheBac]

Nothing to add other than Im astounded to both of your knowledge.

Oh, Jason....I call it a "FICkeM"