Hello, Taslack submitted this question here on my behalf and introduced me to this website.
I currently believe that SC370606DW is a Quad Channel gate driver. The chip was produced by Motorola/Onsemi from about 1995 to at least 2011 for Chrysler to use in their engine computers. i do not know how many different engine computers they were used in
My primary goal was to determine if a suitable replacement could be sourced for a replacement board for one of Chrysler's early ECUs using this chip. further detail below.
With the legal requirement to transition to ODBII Chrysler overhauled electrical systems on many of their vehicles in 1996. Namely for Dodge Trucks, Jeeps and the Dodge Viper they created the JTEC module (Jeep-Truck-Engine-Computer)
Through each generation of JTEC unit that was produced there were two variants commissioned by Chrysler, but we are focusing on the First Generation (1996-1998).
One variant was Designed in-house by Chryslers own internal engineering group and produced under the code-name "Huntsville" (Unit SNs start with "TEH")
The second variant was designed and Produced by Motorola (Unit SNs star with "TEM")
The two units share many primary components/elements on their boards (most of the SMD ICs and transistors made by Motorola) but each is designed and constructed very differently. (photos provided)
The Motorola unit appears to be very smartly designed for use inside a vehicle engine bay where it would be subjected to many extended thermal cycles over the vehicles lifetime,
It consists of a Clam-shell aluminum casing that houses the entirety of a single double-sided circuit made like a ribbon cable and adhered to the outer shell of the casing via some sort of clear semi-flexible adhesive, then coated in a thin layer of silicone for waterproofing. Later generations of motorola style JTEC ECUs would continue to use this design.
By contrast, the Gen1 Huntsville unit appears to have been designed someone hastily with very poor or no considerations made for CTE in some areas.
It consists of a standard ceramic Main board encased floating in potting compound inside a stamped Tin shell, with a secondary daugtherboard consisting of a single sided circuit which was printed directly on an Aluminium panel (with a thin layer of ceramic insulation) that also serves as the lower portion of the front cover of the unit. this circuit is protected in a vat of a sticky-jello like silicon gel surrounded by a foam wall that spaces to the potting compound encasing the main board. Gen2+ Huntsville JTECs used a single ceramic board partially mounted(heat transfer from D2PAKs along the top edge of the board) in a molded aluminum casing encased in potting compound and a single aluminum front plate. the unit was similar in dimension to the Motorola JTECs
The Gen1 Huntsville units have a habit of developing a fault where the ignition coil control circuit will start to cut out and fail while the vehicle is under operation, remain inoperable for 5-10 minutes (cant start the vehicle) and then miraculously start working just fine for the remainder of one's journey. My vehicle had one of these units, and After finding reports that buying remanufactured ECUs to replace a faulty one with exhibiting the same problem weeks or months later i decided to try to repair the unit myself since i could not afford a "remanufactured" unit at the time.
following a guide i disassembled the unit and re-soldered all of the wiring harness and daughterboard pin connection joints on the back of the main board (as this was believed to be the culprit according to some) and that did seem to fix the problem, at least for a time. over the next several years i continued to try various "fixes" based on the idea that there was some problem with the wiring harness connector pins or the daughterboard connector pins only to be met with limited success. each time the problem would go away only to come back again within a few weeks or months.
Since it wasnt working i kept thinking over it in my head and realised that the problem was obviously related somehow to thermals, as it would always occur after some amount of time operating the vehicle, length of which would vary depending on the outside temperature and how hard i drove it.
Another thing I did was buy a unit from a different year model (a gen2 JTEC) that would run my engine in an emergency if needed (Chrysler changed the signaling protocol for the dash cluster on the Gen2 units so my dash was useless if this unit was in the vehicle) and that unit NEVER experienced the problem.
I had tried multiple times to find a motorola variant programmed for my vehicle, but Chrysler rarely used the Motorola unit in lower spec vehicles, so eventually i did try to buy a "remanufactured" unit.
the unit i received began exhibiting the same problem the day I first installed it, and i ripped it out after my vehicle refused to start when i tried to leave for work 3 days later. I put in for an RMA/return with the vendor i bought it from and sent it back, only for them to tell me that the unit tested fine on their end and i would not get a full refund, and that the problem must be with my vehicle. well if thats the case why doesn't the problem show up when i plug the gen2 unit in? they don't test the units under any sort of thermal load that will reveal this problem, so of course they do not see it and im out $100 for their ignorance.
How many vehicles have been towed to a junkyard after an auto shop TRIED to replace a faulty ECU with another faulty ECU thinking that it was good due to improper testing and assuming the problem must be in the vehicle's wiring harness instead?
Now, i did finally manage to find a Motorola unit that was programmed for my vehicle, and that is when i began the journey of trying to identify an actual fix for these faulty Huntsville units.
Since the problem was obviously thermal related, the secondary daugtherboard immediately stood out to me as a huge potential issue due to its design.
Copper traces Fused directly onto an aluminum substrate, exposed directly to the extended thermal range and cycling of an Internal Combustion Engine, seems like a combination destined for failure
I already assumed that the daughter board was connected to the ignition system in some way as i had once tested the unit with it detached and the vehicle would not start.
I asked for some help reverse-engineering the board on reddit and have been working with someone to make a replacement PCB to use the original components, through that process i have learned a lot.
The components on the daugtherboard are responsible for: Ignition Coil, Alternator Field strength, and an Underdrive solenoid in vehicles with automatic transmissions(which doesnt apply to me)
Pin 1 of SC370606DW is connected to the gate terminal of an IGBT, MGB20N40CL which is wired directly to the ignition coil.
Pin 2 is connected to the source power side of the IGBT, and pin 3 is connected to the drain side next to a current sense resistor.
Pins 5, 7, 8, 10, 15, 17, 18 and 19 are all connected to ground. pin 16 appears to be VCC and 15 GND while the rest may be inputs (based on my reading other datasheets about how to connect unused inputs)
Pins 10 is also attached to a test point, likely for reading GND, but pins 4, 6 and 9 also have test points connected to them despite being left floating otherwise.
Based on the lack of components and the low voltage seen on the incoming pin of the board, i believe that pin20 is the input signal used to drive the output at pin1
The gen1 Huntsville unit is only capable of driving 1 ignition coil and 8 injectors, but the motorola unit, which was used in the SRT10 Ram and Dodge Viper, is capable of driving 5 Ignition Coils and 10 Injectors.
The Motorola unit uses 2 SC370606DW to drive the 5 Ignition coils, and the pads for the second chip have the same ground pin connections as the one in the Huntsville unit has, and pin1 on that chip is connected to the 5th Ignition coil IGBT gate terminal
The Motorola unit only has 3 sets of Current sense resistor Pads.
Based on all of this, i believe i have a potential partial pin-out for SC370606DW, see the attached.
i do not know yet what functions pins 11, 12, 13, or 14 serve but if known they may enable the use of replacing this chip with a single channel gate driver on a replacement PCB
Home
Help
Search
Login
Register
