Megasquirt users thread to discuss all things Squirty

[MWP]

I would be a bit worried about using a glue like that. It might soften up when it gets hot.
Would hate to have that magnet bouncing around in there, you would be up for a new dizzy.

[toyo-al]

You are totally correct in saying that, although I did say that it was only for testing it works, ie spinning by hand and a drill slowly to check wiring, etc..

And I just got back from testing it then. We got a nice consistent reading using the drill (limited by a minimum of about 900rpm using the drill), though spinning it by hand we saw readings as low as 50rpm which were as consistent as can be while spinning it by hand haha

Overall, pretty happy and I'll be using a proper heat proof resin tomorrow before it goes on the car!

I'll report back with details

- Al

[jabbatron]

..........

[MWP]

You are totally correct in saying that, although I did say that it was only for testing it works, ie spinning by hand and a drill slowly to check wiring, etc..

Sigh, thats what i get for not reading full posts, sorry
Its good to hear it works well!
Thats quite a nice, easy to DIY fab job youve done. Im sure others will copy it.

[toyo-al]

It's all good man! It would be pretty bad if it did come loose from not properly securing it.. On that note, how hot do you guys think it will get in there?

We've run into more issues with the coils and spark not being fast enough. Having another stab today if my mate is free.

[toyo-al]

Triple post sorry...

[oldcorollas]

well, if you consider the block is cooled by liquid at 100C and the oil splashign around is like 100-120C...
probably not more than that

i can't remember water ever sizzling off a dizzy, so prolly less than 90?

how's your dwell/spark time? is signal inverted?

[toyo-al]

TRIPLE POST! I'll make use this space though..

I have used araldite which is rated up to about 80 deg Celsius before it starts to get a little soft. It's just till I find something stronger and while it's tested. It won't be run for a long time (if it ever even runs haha) anyway.

Haven't checked dwell. Apparently it was all set before I got my hot little hands on the engine. I know I shouldn't rely on it so it will be checked at some point.

[thechuckster]

you'll get a small amount of heat from the sparks going from rotor to posts - but most of it will be the engine heat. I have encountered dizzis that wrere way too hot to touch.

I would use a 2-part epoxy like devcon or JB weld. 24hr araldite really doesn't like heat (not knowing any better, I once tried to fix a coffee jog/percolator with some araldite - was an epic fail as the filter cup literally fell apart on the first brew).

[oldcorollas]

the code reduces both dwell and spark at higher revs or similar(?) so depending on your settings, it might be running out of one or the other

unless coils are fubar? or need better ground/wire for the coil current?

[toyo-al]

Watch this space... We have lift-off. Never grinned so big in my life!!

Edit: Yeeehaaa!!!

So happy! That was after a little bit of tuning to get it idling. Main thing was getting the timing spot on which took a bit of fiddling around.

The megasquirt has been great so far!

[XR Pilot]

Figured I may as well post some more details on Alex's setup, I did most of the ecu/wiring related stuff. We're using a pair of coils off a U12 CA20E-powered Nissan Pintara along with the original bracket and Mitsubishi J121 ignitors they come with, both ignitors are fired together. I locked up the advance plate on the factory 3TGTE distributor and we've fitted a Jaycar hall effect sensor on a bracket Alex made as I had trouble getting the VR sensor to read below 3000rpm using a LM1815 for conditioning (yes, using properly shielded microphone cable with the shield earthed at the ecu). Rather than mess around with it I decided to go the hall effect route, works beautifully. I had some issues with what seemed like spark randomly being dropped at cranking, turned out when I had enabled overboost protection in the software its actually absolute pressure, not gauge so the ecu was going into overboost mode even before 0" of vacuum. Woops. Fixed that, spent a bit of time stuffing around getting the rotor in the dizzy phased in the right spot and it started and ran fairly well. I still need to wire up the fuel pump relay, tidy up the remainder of the wiring sort out the thermofan wiring and its ready to go to the exhaust shop, then we'll get the wideband on it and do some proper tuning. We're also going to swap the throttle body for an RB30 item off a Skyline so we have a proper modern style TPS, as I want to use that for acceleration enrichment rather than MAPdot based which it's using in the video above.

I also need to chase a potential vacuum leak as I can't get it to idle below 1500rpm but it could also be the throttle sticking open slightly. Idles nice and smooth and starts easy as is, we just need to check the cold start and we're in business

EDIT: Dwell is currently set at 4ms on the Pintara coils, using the 'are the coils getting warm' method as I don't have access to a scope. They seem to put out a reasonable spark, as Alex found out

[oldcorollas]

good stuff fellas!

[toyo-al]

Haha I won't upload the video of me getting electrocuted... But thanks for bringing that up

It sounds beautiful! Can't wait to tune it!!

[oldcorollas]

hmm, so i put dizzy and toyota ignitor (Bosch 9 222 067 021, from AE82) onto the scope tonight...

found that the VRs red to pin 3, and white to pin 7 gave correct triggering on the zero crossing slope.. (dizzy is US 4KE starlet...)
but previously, i had it running in car (as direct coil control) the other way around... i guess that would have retarded the ignition slightly with rpm???

also, had issues with the scope showing both traces with both negative probes hooked.
(had pos and neg of channel A to the VR pins, ie 3 and 7, then had +12V to pin 15 (neg of both probe and power supply to case), and a 6.8K resistor to pin 16 (which usually grounds the coil) and pos probe of channel B on that pin....)
that way, the ignitor would fail, and would only get the VR trace....

so i pulled off the neg probe from the reluctor, and the ignitor would function, and would get both traces on screen..... weird.... 890ohm resistopr would also phail... will have to see whats up later...

anyway, traces.
the way i had it running before, triggering on upward slope


and the way it should have been running before (triggering on the zero crossing)


is a dumb ignitor, so it only grounds to fire coil, so input to MS optocoupler will be +ve and ground for signal.... so the opto will be on, and then off for signal...
i wonder if that is inverted or not


edit: this was with cordless drill spinning the dizzy... badly cos the 3 jaw chuck doesn't hold onto the oil pump drive tang, so shits flapping and flailing around