RomRaider

Most hacks are inserting a hook into the middle of an existing routine. I don't see why you would need to rewrite an entire routine to achieve any of the common hacks out there.

No doubt, I was wondering if you're using just the IGN and 1 ground or a few others as well.

That's probably true for "real-time" hacks. But hacks to incorporate speed/density (Freon) or substituting LC-1 in place of stock AFR sensor (cboles) probably do require the rewrite of an entire routine.

You are only dealing with an alternate method of determining base airflow, leaving everything else unmodified - airflow comp, load calculation, etc. So, not really. Of course, there are a number of ways you can implement any hack, but I'm of the opinion that you should implement them in a way that requires the minimum number of modifications to the existing code to minimize the introduction of bugs. It is much easier to debug a set of standalone routines that you developed than major changes to the existing factory code.

Did I really post this in general discussion?

My bad. I think I meant to move another post.

Haha, gotcha.

Would that be "Fear of the unknown" ?

Bump.

Before I can go any further with my speed density tests, I need to get my spare ecu up and running.

I've narrowed the issue down to either wds issues (p407 on shbootmode diagram), or issues power up the ecu (Proper pins, order of connecting circuit/power/ecuflash).

To check the wds, I was hooking up a headphone with a large resistor, and comparing it to a test tone. Just to be sure, I hooked it up to a scope, and it's coming in around 150hz, not 125hz, as a square wave between 0V and 2V (crossing the 1.5v TTL threshold), this is based on the scope's calibration port.

I found a 555 timer equation and plugged in the schematic, giving 145.8hz, so I'm pretty sure the wds is good enough.

I used the same setup to tap into the send and receive lines of the circuit, and I get an intermittent click whenever ECUflash is sending the boot command, but zero response, which matches the logs.

So, either i'm not powering on the ecu properly, or the wds is off too far. I'm leaning towards the former, specifically the order of doing things. I've tried powering the 5v first, then the 12v, waiting a second, and not waiting, to no avail. I'm really at my wits end here, paying tactrix and buying new ecu's is not an option... any advice would be greatly appreciated.

Forgot to mention, I'm also running a ground wire between the ecu ground the the circuit ground as mentioned on openecu, and I've tried using a battery @ 12.5V and an adapter at 12V.

So, anyone who's successfully shbootmode'd an ecu, which and how man ecu pins did you use for power, and in what order did you apply the 12V, 5V, and WDS, and how fast.

Thanks.

Here are some pictures of the SHB adapter from the scope.

Pictures!

Did some more testing today.

When my adapter isn't hooked up to the ECU, everything reads:
VCC: 5v
GND: 0v
TX/RX: 5v
WDS: 0v - 5v Square Wave @150Hz.

All this looks fine to me, I'm not sure what the TX/RX lines should read though.

Anyway, when soldered up to an ECU, things change.

Vcc => P405
GND => P413 => ECU pins D1,D2,D3,D7,A5 (connecting adapter and ecm grounds as Colby mentioned on OpenECU)
TX => P409
RX => P411
WDS => P407

This is while running:
12V POS => ECU pins B5,B19
12V NEG => ECU pins D1,D2,D3,D7,A5

I get the following readings with or without the 12V source applied
Vcc: 5v
GND: 0v
TX: 2.5v
RX: 2v
WDS: 0v - 2.5v Square Wave @ 150Hz

So, it looks like the ECU may be broken, or the pad numbers aren't corresponding to the proper CPU pins. I found a spare 04 STi ECU to test, will see how that goes, and I suppose I could scrap one of my ecu's to trace the pads if my cable is indeed working.

Can anyone who's sucessfully booted an ECU post the above readings when connected?

Thanks!

You do have a problem with powering the processor.
Voltages measured prove that there is 2 V power instead of 5 V required for the chip.

Probably the ECU voltage stabilizer (12V->5V) chip is dead.

Is it possible that your 12V wiring is incomplete?

I'm pretty sure my 12V is good. I talked to Colby @ Tactrix, he said my power adapter is sufficient (12.0v, 1.5a), and the pins and ground connection are correct. I'm using every POS and GND connection I can find in the Subaru manual, and it matches what Colby is using. I also tested with a car battery running 12.5v or so.

I'll have the 04 STi ecu shortly for testing, but in the meantime I'll check over the 7058 docs to find a way to test the voltage stabilizer.

Tested the 04 STi ecu, no luck.

The ecu was in working shape, has a working rom on it and all. I'm using a ground strap and a grounded soldering iron, so It's safe to say there's nothing wrong with it.

With or without 12V power, the WDS, TX, and RX lines all show a big voltage drop, down to ~2V, just like my 06 WRX ecus.

Not sure what to make of this, perhaps I have something setup wrong with the adapter's voltage setup on the DLP-USB chip.

I sucessfully SHbooted 2 bricked ECU. Euro AL340(sti 06) and AK710(fxt 06).
ECU powered with 13.1v by two pins - ign(b134-7) and gnd(b134-5) in my case. Also provided strong ground between ECU, adapter and 13.1v power source. 5v is VCC on schematic. Uploading rom time in shbootmode is about 15min. I made USB-serial adapter on FTDI ft232bm chip and ne555n timer.
Here ecuflash logs.