Burned Up Injector Drivers [SOLVED]

[Schwifty]

Hello all, so I've been having some issues with my MS2 3.0 no longer firing the injectors. It's setup for a 4 cylinder engine with batch fire. I started the car and it ran for about a second and then it shut off, no fuel. I verified that the injectors had power and the ground wires for both banks to the MS were good. In test mode I couldn't get either injector bank to fire, and when I opened up the MS could smell magic smoke toward the top right of the board. Until now I've been able to solve every problem by checking resistances, but that hasn't helped me so far this time. no traces or components are visibly burned. I compared the resistances of all the components near the part of the board with the smell with their twins, but they all read the same as each other. I have a new U4 IXDI604PI FET Driver on the way, because I don't see what else it could be. I'm currently running cheap Ebay 1100CC injectors, and I'm thinking they have a large inductive kick and that's what killed my drivers. The injectors are high impedance, but can I put a resistor on the ground wires anyway to help stop this from happening again?

[Matt Cramer]

Look under the board to be sure the 12 volt trace isn't cooked. Also see this guide: https://www.diyautotune.com/support/tec ... or-driver/

[Schwifty]

Unfortunately I don't have a stim yet, and it will be a while before I can get one. I don't see any burned traces. I suppose if replacing U4 doesn't work I'll have to wait to get a stim and follow the guide. I would like to know why it burned up in the first place, though.

[Matt Cramer]

You can use a power supply and LED test light in place of a Stim.

Unfortunately, it's a lot harder to figure out why something broke, compared to figuring out what broke.

[rickb794]

Measure the injector resistance? All of them.

Check for crossed wires to the injectors. If battery voltage gets applied to an injector output directly, it will burn the output.

I don't know anything else that would cause that.

[Schwifty]

It's been a while, but I finally got a Jimstim and started troubleshooting the board. I did measure the resistance on my injectors and quadruple check my wiring, and it all seems to be good. I've been trying to figure out what on the board is fried for hours, but I can't quite figure it out. I followed the guide step by step, and I've determined that I do not have power on pin 6 of U4. It says that it could be a failed d3 or d21 diode, a shorted C14 capacitor, or a bad trace. I de soldered D3 because it was conducting in both directions, but on the bench it was fine. D21 also conducts in both directions, but I don't want to de solder it until I know I have to. I wish the guide specified the resistance for these. The C14 capacitor reads 0.7 Mohms in one direction, and 2.5 Mohm in the other. It fluctuates like you would expect a capacitor to. D3 reads 1.6 Mohms in one direction, and 2.5 in the other. d21 reads 1.1 Mohm in one direction and 5Mohm in the other. I've been looking and probing for a bad trace for a while, but I haven't been able to find any. I feel like I'm close to finding it, however I'm just not sure how to determine which component has actually failed. Any help would be greatly appreciated.

[Schwifty]

My bad, I counted the pins on U4 wrong; there is 1v on pin 6. I decided to just hold my jumper wire on pins 21 and 22 to ground to enable the injector circuit, and after about 10 seconds there was some smoke out of resistor R15. (around the area I smelled magic smoke initially) and it's now turned black. So, obviously I need to replace R15. I also noticed that there is 10v at Q1 and Q5, and I'm assuming it should be 5v. I have a feeling when I replace R15 it still won't be fixed. I don't really want to have to buy another whole board, but that's beginning to look like my best option at this point. Would R15 being shorted cause 10v at Q1 and Q5?

[Marek]

It won't be fixed if you just replace r15. This is not the designed high current path to ground and there shouldn't have ever been a high current going through r15 at any time.

The low current control circuitry delivers either a high or low signal to pin1 of q1 and this controls whether there is conduction between pin2 and 3 of q1. That is the "switch" which turns the injector "on" and the current is supposed to flow from the 12v rail, through the injector, into the v3.0 board at >>INJ-1, in to pin2 of q1, out of pin3 of q1 and then to earth via the bottom of r37 connection.

What you are looking for is a low impedance path to earth which ought not to be there. This is how a high current is allowed to flow. Transistors/FETs typically have a very high impedance across them so no current flows until they are turned "on", so you should expect a very high ohms reading across pin2&3 of q1 and across pins 1&2 of q14 if those components are good.

You need to be more specific about voltages on q1 or q5 when you say "10v" or "5v". If q1 is good and it is not commanded to turn the injector on, then pin2&3 are not conducting to each other. This means anything on the pin2 side of the circuit of q1 is simply tied to the 12v rail and showing ~12v, whilst anything on the pin3 side of q1 is tied to ground and showing ~0v. Pin1 of q1 will sit at wherever pin7 of u4 tells it to. When q1 is commanded "on", the injector load now dictates where the voltages sit and everything on the 12v rail side of the injector loom connection is at ~12v, whilst everything on the other injector loom/pin through to the MS and to the chassis will now be at ~0v - pin2&3 of q1 are conducting and so completing a circuit.

In essence, you want to turn the power off and diagnose where are all of the low impedance connections between all different points (including to chassis ground and to 0v and 12v rails that you are and aren't expecting) and where are all of the high impedance point to point connections.

Your problems appear to have started when you swapped the injectors out. I would take them out of the circuit and replace them with LEDs plus resistors - i.e. a higher impedance and therefore lower current 12v load. If q1 can't work happily switching 20mA on and off, it's unlikely to work switching 3amps on and off. I would also disconnect the loom and diagnose or verify the MS unit and loom/injectors separately before reconnecting them.

Appendix A of the hardware manuals gives the schematic for the v3.0 board.

Smoke escapes because there is a lower impedance path between 12v and 0v than designed for and too much current flows for the trace thickness or component's ability to conduct away the heat that this generates. The circuitry is quite robust. People with v12 cars run MS2 batch fired with six low impedance injectors per channel.

kind regards
Marek

[Schwifty]

Okay, I will be more exact with my measurements. I'm pretty new to board circuit diagnostics, so I'm mostly lost; I know what the basic components do and that's about it. I'm using a AC adapter on my stim, and the reading on the 12V rail from the DB37 is 10.93v. I'm not sure what the pin designations are on the transistors, so Reading the pins from left to right with the "Bowling & Grippo" text on the bottom, Q5 has 9.75v on pins 1 and 3, and 9.43v on pin 2. While Q1 has 0v on pin 1, 9.42v on pin 2, and 9.76v on pin 3. So I'm no longer reading 5v from them. Resistance across pins 2&3 on Q1 is 2.323 Mohms. Pins 2&3 on Q5 read 2.350 Mohms. I only read 9.50 ohms across Q14, so maybe that's the cause?

I'm assuming R15 and R20 are independently responsible for INJ1 and INJ2, so I wonder why just one would fail and not the other? Neither of the injector outputs are functional. Would it be wise to just go ahead and replace every component in the injector driver circuit instead of trying to chase which exact components have failed? I don't believe it would cost much. I'm going to reinstall my original injectors on the car before I plug my MS back in. I'm pretty sure the new ones are the cause, even though they worked for quite a while. I'm redoing the whole harness pretty soon too.

[Schwifty]

I replaced literally every part listed in the troubleshooting guide, but it only fixed driver A. Pin 4 on U4 didn't have power, there was a bad trace somewhere in the circuit. So, I repaired the circuit with some wire as seen in the image, which did get power to pin 4, but now driver B is stuck on.

There is 6.67v on pin 4, which seems a little high. Q15 and Q13 also get very hot when it is left on. Q5 also has 5v power on all 3 pins. this is getting stupid. Am I stupid? No, the board is stupid. I don't even know anymore.

[DGA]

I see SO MANY soldering problems on that one pic, it's hard to know where to start......

[Schwifty]

It was working fine like that for a long time, nothing is shorted or anything.

[Schwifty]

Well since one driver works, I think I'm just going to connect all my injectors to it and run simultaneous staging for now. I'll just buy another board when I can.