ExtremeX ECU Stays Powered at IG-Off - Backfeeding Help

[TC951]

Hello,

Hoping someone can point out what I'm doing wrong or give me some suggestions on what to check…

I'm using a G4X ExtremeX with a custom wiring harness. 

Here is how my power supply is wired:

 

My problem:

Powered up with IG-On, engine off, ECU reads 12V and operates normally.  At IG-Off, ECU battery voltage drops to ~8V and it remains powered up.  My Sub-Relay remains energized as well. 

If I remove the Sub-Relay the ECU battery voltage drops to zero. 

If I remove the fuses or disconnect the connectors for my idle control valve and my electronic boost control MAC valve, voltage drops to zero. 

If I remove only the MAC valve connector, voltage drops to 6V.

If I remove the battery positive from pin 30 of the sub-relay the backfeeding stops.

If I place a diode inline with the power supply side of the ECU supply wire the backfeeding stops. 

 

So it appears that the idle control valve and the boost control valve are where the backfeeding is taking place.  However I have them wired per the wiring diagrams in the help sections as far as I can tell.

 

Here is what the help sections shows as the correct way to wire a solenoid:

Based on this diagram, my idle control valve and/or boost control solenoid should both be wired same as the "Relay or Solenoid" in the diagram above. 

So it seems like either:

1. The ECU is somehow still supplying a small enough amount of voltage to partially energize the sub-relay so that the 6-8V can flow through pins 30-87 of the sub-relay?? 
2. The EBC/ICV are somehow backfeeding due to some wiring error.

 

Does anybody have any suggestions on how to check #2?  I've already confirmed both solenoids are wired how I have them shown.  Power coming from sub-relay pin 87 and ground to the respective aux output pins on the ECU.  Is there something else I can check??

Alternatively, is it possible that somehow #1 is occurring from some hardware issue or wiring issue with the ECU itself??

 

I've attached a log file and my harness pinout document.

 

Any help or suggestions is MUCH appreciated!

PC Datalog - 2021-06-29 5;43;05 pm.llgx

Link G4X ExtremeX Wiring Diagram_Dist.xlsx

[Adamw]

Have you got a schematic that shows how the main relay works internally?  Any different if you disconnect pin 87b and 85b from the main relay?

[TC951]

So I've actually tried removing the main relay completely and the backfeeding continues.  That's how I came to the conclusion that it has to be coming from the ECU side or from the ICV/EBC somehow. 

 

The Main relay is an OEM Porsche DME relay which is used as a main relay for powering the OEM ECU and Injectors as well as powering the fuel pump by a ground output from the ECU.  I'm re-using it to do the same with my G4X.  Here is a schematic:

[Adamw]

I think what is probably happening is the sub relay is slower to open than the main relay, so you end up with the main ecu power disappearing, but if the sub relay is still closed the 12v will flow through the solenoids, into the aux, through the internal flywheel diode and exit the ECU 14V pin which is connected to the sub relay pin 86 and keeps it energised.

 

The quickest fix I see would be moving the sub relay pin 86 to the ign switch.

[TC951]

Ah OK, that makes sense I suppose!  I'll give that a try.  Thank you!

 

Just out of curiosity, would using a diode like I tested originally be a bad idea for a fix?  In the end it's a band-aid rather than a true fix, but it does eliminate the backfeeding from having a path back through the ECU and is a bit easier to permanently wire up compared to moving sub-relay 86 to ignition switch.  

 

This is my first time building a harness like this so I'm not super familiar with what is considered acceptable conventions vs. a ghetto fix...

 

[essb00]

@TC951 Diodes have a considerable voltage drop. The voltage reaching ECU must as much as possible be the same as the voltage reaching injectors, coils. etc. - for correction of injector dead time, ignition coil dwell, etc.

[TC951]

2 hours ago, essb00 said:

@TC951 Diodes have a considerable voltage drop. The voltage reaching ECU must as much as possible be the same as the voltage reaching injectors, coils. etc. - for correction of injector dead time, ignition coil dwell, etc.

Ah!  OK, very good point.  Thank you

[TC951]

FYI - Running the same switched 12V supply that the main relay uses for 86 to also supply the sub relay's 86 worked.  So thank you Adam!!

 

I still don't understand why the sub-relay isn't able to open fast enough.  It's a mini Song Chuan 35A 14VDC relay that I see pretty widely used... P/N 301-1A-C-R1 U03.  --> https://prowireusa.com/p-3892-35-amp-spst-12-volt-relay-iso-280-mini-resistor.html

 

It is a very inexpensive relay I guess, so is it just a cheap "slow" relay???  Or is it somehow possible I've wired power or grounds in such a way as to cause the EBC/ICV to want to back feed so quickly at IG-Off??  Just seems odd. 

 

 

 

 

[essb00]

The EBC/ICV/etc. being driven to low would back feed to the ECU's internal flywheel diode when the ignition switch is (or in you case main relay) is opened. See the orange trace in the diagram.
You can actually even just supply everything in the sub-relay via the main relay so everything surely turns off with the ignition switch and it is simplified with only 1 relay (ECU itself does not take too much current anyway).

[Adamw]

9 hours ago, TC951 said:

Just out of curiosity, would using a diode like I tested originally be a bad idea for a fix?

Electricity needs to flow in both directions through that main wire.  Everytime an inductive device such as a solenoid or relay is switched off it sends a large voltage spike back into the ECU - these can be quite large ~100V or more.  So when correctly wired these spikes are passed back to the battery through the ECU power supply wire.  If you had a diode in this wire the voltage would have nowhere to go which makes the valves/solenoids/injectors very slow to colapse as well as high voltage potentially damaging electronics.

 

2 hours ago, TC951 said:

I still don't understand why the sub-relay isn't able to open fast enough.

It will likely be to do with the inductance and mechanical design.  When you cut voltage to a relay it doesnt just intsantly switch off.  You can look up inductance on wiki or similar for a more accurate description, but basically inductance makes a device resist any changes in current flow - so when you remove the voltage source an inductive device will try to keep the current flowing the same by generating its own voltage using the energy stored in its magnetic field.  The relay will stay closed until there is not enough stored magnetic field left to hold the contacts closed.

 

We did discuss that diagram from the aux output wiring page a while ago at Link internally and chaged it a little to reduce the chance of this happening but for some reason it looks like the updated one never actually made it into the help file.  It may be my fault because I remember saying I was going to test it before comitting.  I will follow that up.