I've uncovered a ZX-10 Mitsubishi unit. It used the same SH7052 CPU but had no visible AUD port top or bottom. I've been meaning to trace out the AUD pins from the CPU to verify they are unconnected or if they go somewhere.

Given what we know about how to program the SH7052 in boot mode I'm guessing the CPU programming pins will trace out to the harness connector but I haven't checked that either.

Worst case the SH7052 can supposedly be read in an eprom programmer if you can pry it off the board without destroying it you could read the code but you would have obtain a second ECU to test the flashing on then.

BE VERY CAREFUL removing the potting around the CPU. The mitsubishi unit I uncovered had no hard epoxy around the CPU protecting the pins like the Denso units and it doesn't take much to rip those little pins right off. Found this out the hard way.

Good Luck

I did a cross reference on the Kawasaki part number 21175-0084 and it comes up as a

ZX1400A6F Ninja ZX-14 (2006)
ZX1400A7F Ninja ZX-14 (2007)

PetriK wrote:

---

Acquired recently a ZZR1400 ecu which is said to come from an US bike. History is unknown. The ecu has the following markings:

F8T98571
21175-0084
6302

Misubishi

Exterior >PBT<
Interior >PUR<

The project will begin shortly. Now I am intersted in gaining as much information about ZZR1400 as possible. Particularly ZZR1400 wiring diagram and info about other mitsu units would be of interest.

---

Hello,  I am new to your site, which I find very interesting, I was wondering if any progress has been made with the ZX14 ECU

I have started hacking into the ZX14 ECU. It's a 21175-0084 from a 2007 non-ABS model.



It's opended up and repaired, a couple surface mount components pulled off with the potting and I ripped off the 3 small elecrolytic capacitors off during the process. It has been powered up and voltages check ok. Checking the pins of the MH8206, the Vcc & Vss all match the 7052. There are 2 wires from an unamed connector going to unused pads labeled JP1 on the bottom of the board. I have attached wires to the AUD pins per the 7052 datasheet and setting up to connect to a BS2 with RR's code.






Thanks to everyone on this forum for all the info that has got me this far.

Still working on getting the AUD to work. RR has been alot of help. I'll keep everyone posted.

when soldering on top of pga chips usually I would buy a 30awg wire. Makes soldering on the pga chips so much easier. http://www.radioshack.com/product/index.jsp?productId=2062641. Just coat a little solder on the wire and line the wire up with the pin. Next apply heat for 5 seconds and be done.

I can send you the ZX14 ECU I have if you are willing to publish the .bin and physical harness to processor pin routes.

Have not had time to work on it at all...

Got your ECU in the mail today.

Shaved it down and hooked up the wires. To late tonight to give it a try. Maybe tomorrow.






By the way the trick to this is partly using the right wire.  Small earbud headphones these days have a small stranded wire with a special lacquer used as insulation. The lacquer dissolves at soldering iron temperatures.

Having three teenage children in the house there is always a busted set of iPod headphones somewhere about. I stripped the wire out then untwisted the individual strands off the nylon shock cord. You tin one end of the wire by putting a small solder ball on the end of your iron and just dipping the wire a very small way into the ball.

After grinding away the chip case I tinned all the exposed copper and then solder wicked away the excess. then it was just a matter of holding the wire end against the pin lead and pushing down with the solder iron to tack it on. Don't try to add solder when you make the joint. Just the residual from tinning the wire and the pad are enough for a tack job.

I used regular old candle wax to secure it in place. Of course I checked and double checked continuity and for shorts to neighboring pins before I sealed it.

BTW what are the power pins for the harness connectors?  17 and 34?

Tried it out and got the exact same results as ZXRACER; nothing.

So here is what else I did find.

I checked all the pins resistance to gnd and vcc powered down and they all read around 10M in either polarity to either power pin.

AUDRST and AUDMD are both supposed to be inputs only (and therefore should never be tristated) With the chip powered shorting either to vcc or gnd thru and Ampmeter shows no current flow (or less than the 5uA resolution of my meter)

AUDMD according to the manual is pulled high internally there is no sign of that. In any of my readings.

Also if you have ever scoped out digital inputs and output lines with a scope they are seldom actually 0.0000V when low. When touching digital lines they usually have a little bit of leakage voltage (0.05 - 0.1V) There is none of that on any of these lines all these lines are below low, they are 0V

According to the DC characteristics specs all the AUD pins should have 500uA of internal pull up except for the AUDRST which has 700uA of pull down current. Also the specs on tristate leakage current does not list any of the AUD pins which indicates they are not tristate ports.

The physical characteristics of these pins do not match the specs or functions they should exhibit as AUD pins. In fact I wondered if they were even connected at all. Then I noticed that they all go high to 5V (internal pull up?) during the main CPU power up reset cycle.

Now there is a register in the 7052 at 0xFFFFF708 called SYSCR. It can be used to disable the AUD port pins. Its default state out of reset is AUD enabled but if you write to it and set the bit then...

Setting the AUDSRST bit to 1 places, the AUD module in the power-on reset state.
It is kind of cryptic but it doesn't sound good. 

I checked the several examples of the Denso 7052 code and it never refrences the SYSCR meaning it leaves it in the default AUD enabled state. Could it be since they didn't hook it to anything those bastards at Mitsubishi disabled the AUD via software? Maybe what I'm seeing is the pins working just out of reset and then the software turning them all off.

So where do we go from here?

*I think first thing to do is patch a Denso ecu with an AUD disable instruction and see if the AUD pins then behave on that ecu as they do on the ZX-14. It would really suck if that is what is going on but at least we would know.

*Another thing I think we need to do after exhausting all other ideas is a post mortem on this board by cutting the chip off and checking to make sure that none of those 8 pins we think are AUD are actually being used as some other kind of port pin. This will not prove that it is an AUD port but it would prove that it wasn't which might be of value before we cut another one of these open.

*Find a manual with this actual CPU number on it. I traced back the origin of the 8206 is a 7052 cross reference and it leads to a comment by Colby over at openecu.org. Does anyone have any more concrete proof than that?

The pins do match up and the CPU my very well be an SH2 core but these MCUs consist of a core and a varying selection of I/O. It could be a very close cousin of the 7052.

*Along these lines would be to identify the function of every single pin on this MCU possible, not just power and ground and then compare it to all the SH2 variants.

Any one else have any suggestions?

-- Edited by RidgeRacer at 00:49, 2008-11-26

Back up plan - chip out and read it as an EPROM ?

Could it still be protected ?

Stay on it Petrik... I think these boxes are generic in the hack. I mean, once you find the rewiring, you are in be it a busa, 1000, 14, they are all the same basic parts, different jobbers.

Good luck is sticktoitivness.

How about this. I have now a ZG1400 Concours engine and the computer witht he wire harness. Only, I don't have the gauges and the big thing, the Remote key and it's attendant computer. Does anyone think the key codes are in the computer and can be blown out so I I can just use a regular key?Oh, it's actually going in an '81 Suzuki GS750E frame. After I make some swingarms for the shafty drive.
I also thought it would be really neat as it is having variable intake cam timing too. Otherwise, I may be forced to convert to solid cam timing and graft some Busa F.I. on it.
Thoughts?
Tanx
Greg

I can flash all the ZX-10, ZX-14, Concours ecus but have not released it to the public domain.

We will be offering a commercial flash cable and software in the near future.

Here is a dyno Rickey Gadson posted of a back to back run of a stock ECU and one we flashed for his 2011 ZX-10  (24hp gain at 13,100 rpm)

-- Edited by RidgeRacer on Thursday 2nd of June 2011 12:43:02 PM

Put me down as second

Any chance of boost referenced fueling ;)

RR any updates on this cable and software ??

Thanks :)

put me down for one too :) Excellent news RR !!!