think i'll take a look at this when I take the squirt out next

 

Worth doing for the sake of 40p mate, my car's been flawless since I altered it to Joes circuit.

 

Someone needs to cut all the BS out of this thread and provide a shopping list and a HOWTO for my MSPNP9495. I can solder alright, just don't know much about ICs.

 

Just did this mod last night, and everything runs the same as before. I'd call it a success.

Thanks for the great writeup Joe!

 

I will have a trackday on sunday. next week I will do the mod. thanks a lot.

(I'M SICK OF THIS @$%#$!! POP .!!!)

 

/me needs photo howto with MSPNP... anyone?

 

Sorry man, I wish I could help. Although I am proud that the mod is super clean on my board and only took up a 3x5 hole square in the proto section.

 

I have no idea how to do this on a pnp since their boards dont have a proto area. I'd send DIY an email if I were you and reference this thread.

 

FYI, with this setup, if you accidentally load in a config with SPARK OUTPUT INVERTED set to NO, it'll get your ignitor reaaally hot, and the car will run like crap. But no apparent damage to the car.

I haven't experienced this myself.... But it's actually comforting to know that it won't necessarily blow your coil pack.

Even so... anybody who uses this setup should put a BIG DISCLAIMER on any MSQ's that they share publicly!

 

Ummm it'll kill a 1.6 igniter pretty quickly, in my case it actually only initially killed the tacho part, but as it wasn't my car I didn't want to 'fix' it by having the MS drive the tacho instead and then the Igniter dying later down the road.

So in short whatever setup you're using having the output wrong will damage something!

 

Were you/he running the circuit described in this thread, or the original circuit?

 

Well, I decided to swap over my spark output to this circuit when I moved my MS from my 93 to my 96. In the process I blew one of my igniter/coil assemblies. I'm not blaming the new circuit. Joe was very clear to change Spark Out Inverted to YES and I did just that on the base map I started up with (a modified PNP map). I was having some problems with my idle and while sitting on the side of the road with a dying laptop battery, I decided to load up my old map from my 93 and see how it did. In the process I forgot the change the Spark Out Inverted setting on the old map and apprantly it blew one of my ignitor/coil assemblies. Limped it home and checked and I was only running on cylinders 1/4. Disconnected my MS outputs and returned spark and fuel control to my stock ECU. Still only running on 1/4. Checked continuity and voltage to the 2/3 coil and they look fine so it's pretty much gotta be the ignitor/coil. I may go ahead and upgrade to COPs or PlanetMiata has a used coilpack for $75. We'll see.

It's funny, one other time I stupidly reloaded the firmware in my 93 (with the old circuit) without disconnecting my ignitor and nothing blew. I guess my luck finally ran out.

All that said, do make sure you have your Spark Output Inverted set correctly in Megatune and be careful when using various old maps or maps from other folks.

 

Got excellent feedback from DIY. Now need to find time to sort this further...

 

Hmm. I wonder what the "harder to use" mod I mentioned was?

 

This:

Well, more or less - it wasn't driven off the CPU output pin itself, but through a more complex circuit.

 

Oh yeah. I nixed that theory after putting some brain power into it. You'll see it's crossed out from my original post above. The problem there is that a PNP needs to see some actual negative base current in order to operate, not merely the absence of positive base voltage. With the CPU "asleep", the output pins are all high-impedance (neither pulled up nor grounded), so while the circuit would work in inverted form during normal operation, it would not have satisfied the original design goal of keeping the outputs off when the CPU is not active. I might have been able to make it work with a pulldown resistor on the base, but that's just more parts and more complexity and it'd still sort of a balancing act between providing enough pulldown to operate the transistor vs. sinking so much current that the CPU can't turn the thing back off when it needs to. I like the double-NPN design better as it seems more robust to me.


Sadly, I see that your prediction of someone frying their ignition by loading an old .MSQ has come true...

 

Joe,
Forgive my "I haven't picked up a textbook in a decade" thought here, but couldn't we just jump the resistor/diode, and swap in a PNP, taking the signal off the other leg?

You might have to dig around the layout a bit, but that should give you the inversion you were looking for, without adding parts?

 

I borked his ignitor straight after updating his MS to the new design I forgot to update the software!

 

I did a little schematic of the mod on a 3.57 (i.e MSPNP) board.



Could someone please confirm that it's O.K.

Greets

 

It would be nice to have a simpler way of doing this for new builds!