MS3 na cas sync problem?
 

Been trying to get this to work for a week now... so I'm deferring, I have what seems to be the beginnings of the correct signals, but it looks incorrect. I'm running ms3/ms3x using the abeFM circuit for the crank and the simple pull up for cam (was running it through the abeFM circuit, but changed it because I wasn't getting much of anything for a long time)
(also of note, I'm not running the tiny caps on the abefm circuit, i was under the impression that they weren't strictly required)

Anyway, here's my composite log...

Change rising edge to falling edge.

You're welcome.

Nice try, its set for falling edge. Here's my msq so everyone can have a good laugh.

Did you capture the log using a JimStim or off the car?

Are you using JS10 or the MS3X cam input (pin 32)? I recommend the MS3X input.

Thats from the car... I know, it looks like the wrong pattern. I wish I had a jim stim.

wait, you're using the abeFM input circuit for a CAS input and not 99-00 crank/cam sensors???

Yes... want to eventually run the trackspeed wheel, already have the sensor... Its all Hall 5v logic stuff, abeFM circuit should work fine (even good) w/ the cas.

well, try rising edge...

it looks like it's getting the correct signals. 4-2.

I'll give that a whirl this evening. May go ahead and pick up a couple caps over lunch too.

Put it back to both inputs using the abefm circuit, added the caps (they're ghetto radioshack ceramic caps w/ the crappy temp rating sadly) And the composite log looks much better.... but I'm pretty sure I've nuked the factory coils (messing around w/ settings while trouble shooting, going high (inverted) for ms3x coils ppl) COP harness is almost done though... so hopefully tomorrow evening I can get the damn bottom bolt out of the coil bracket, it wiggles but won't come out. ugh.

Ugh, looks like sync is still my problem... got the cops in, no dice, cranks and doesn't try to do anything, thought 'well, can't hear the fuel pump, so I'll put the oe ecu back in and see if I can hear it that way'... couldn't hear it, and amazingly the thing started right up w/ big ol rx8 injectors. Its weird that it doesn't even cough or anything because it gets sync and loses it, it should squirt and spark with any amount of sync right?

switched to rising edge and sync is perfect... but i'm not getting any spark, i'm set to going high (inverted) ms3x outputs, using spark A (1+4) and spark B (2+3) (cops work perfect oe ecu) .... what could I possibly be missing? I've just double checked the harness for the 15th time... ms3x bad?

use the output test mode and check for spark on each output.

if you cant hear the fuel pump...what are you using to turn it on? what model year?

i tried that on both outputs and we never saw anything vs the oe ecu which we def. saw spark on. just tested the idle valve which is on the ms3x and it buzzed at 50hz... so itsnot the whole board, we even beeped my ms harness just in case. it is going high (inverted) for the coil signal right?

msextra.com/forums/viewtopic.php?f=131&t=38381&p=266533#p266533

yes should be inverted. you put the ribbons in correctly? You have your spark outputs connected to pin 14 and 33 on the MS3X db37?

When I get home today I'm going to quadruple check the harness and probably get some pictures to send to diyautotune. jmscortina tested my msq on his bench and got spark. I ended up buying a new tv from BB yesterday instead of messing w/ the car any... needed a break :)

You have the spark output set to MS3X, right?
If so, try "wasted COP" instead of "wasted spark" and use spark outputs C and D instead of A and B.

What coils are you using, and what does your dwell look like?

The final thing I could think of would be the ribbon cable. Continuity test the big ribbon cable if nothing else pans out. Those ribbon cables are machine assembled, not hand assembled, so I don't think there's a problem there, but ya never know...

Coils are toyota COPs, dwell is set to 2.5 ms @ 12v, 3 for cranking. I'll try the other spark outputs this evening. I was thinking there was probably a way to get it to use different spark outs. I was thinking wasted COP would be firing A+B and then C+D, it makes more sense for it to be doing A+D and B+C together though.

I really appreciate all the help.

That's not enough dwell for cranking. Turn it up. 6 ms might be more reasonable. Do this first, before moving on to re-wiring things. Insufficient dwell could be your real problem.

Otherwise, I'm asking you to try spark outputs C and D just for output test mode, to see if you get spark. I was assuming you were running a wasted spark set up. For that matter, there's no reason why you can't wire it up sequentially.

I'll do that, its wasted spark... Not going sequential because I don't want to run more wires and there really isn't a whole lot to be gained.

Bumped the dwell up to 6ms cranking and no dice, switched connectors (swapped A+B outputs) No dice, still ran fine on oe ecu. Checked my harness (again) to verify pin 14 and pin 33 (beeped w/ my multimeter). Didn't quite have time to mess with trying other spark outputs, I'll do that tonight. Really feels like bad spark outputs on the ms3x, i haven't moved the pins at all previously so I can't see how it'd be anything I did. Really wanted to get tuning time in before the weekend because I've finally got a chance to pull the motor and getting ready for turbo, if this thing has to go back to GA I don't think theres anyway I'll have time to make it work.

(BTW, ppl run 3.5ms cranking with these toyota COPs, 2.5 running... But they also run fine on the stock ecu's so I was pretty sure 6 wasn't going to help it but wasn't going to hurt either)

BTW, MS-III is looking for the value at 13.2v, not 12v.

I use 5ms cranking and 2.1ms running.

Ah, I noticed the curve didn't match up to 12v = 100% dwell... figured it wasn't a big deal though.

I can't see how you would have 2, let alone 4, bad spark outputs.

This would be a great occasion for an oscilloscope. Or you could even use a JimStim and jump the spark outputs to IGN LEDs on the stim.

jmscortina did just that and was getting spark output with my exact msq. I don't have a jimstim or an o-scope, though I could make arrangements, what would either tell me that we don't already know? I agree its unlikely but I feel like I've ruled everything out. I mean, if it worked on the bench that still doesn't make my car go :)

My thread on msextra.com last post is where James tested it for me.

It could be used to tell if the outputs are broken or the coils you're using just need to be driven differently.

I'm using toyota 1zz-fe COPs on my 20v 4age and they work fine, but I'm driving them one coil per output.

Ken

Bumping my own post here because I didn't know if you tried this yet.

The very first MS3X ribbon cable that we hand made here didn't have continuity on a couple of the spark outputs. It looked fine, but a couple of coils failed to fire. A quick test with the DMM condemned the ribbon cable.

Now all the cables are machine produced, and should be totally fine, but can't hurt to check.



http://www.msextra.com/doc/ms3/pix/ms3-jp2.png

Guess I'll check the ribbon cable when I get home. I don't know that anyone else is in fact driving them wasted spark from the MS3x, though I think Braineack ran them that way for a little while... I guess I can also disconnect the plug from coil 4 and see if 1 starts sparking.

um wasted spark i was driving through main board. seq spark through ms3x. but Ive tested wasted spark through ms3x cause the output test mode is fun.

So you drive 2 toyota cops off ms3x spark A on the bench and they both fired no problem?

I believe I have, yes. I can't remember what I tested, but I was sure I tried wasted when figuring out how I messed up my wiring with I first went seq.

there would really be no reasons why one would spark and the other not, it's just a trigger for the coil to release its load.

I've only driven 1 at a time per output myself. If 2 together draw more than 25 mA it may not work. Also, when we originally got the first ms3x boards about a year ago, we had to change the resistors on the outputs from 1k to whatever they are now (the value escapes me) because the Toyota COPs seemed to want a pretty strong 5v to trigger. Putting 2 on one output may itself be causing the problem.

you may want to try unhooking one of the coils from the output and then try test mode to see if the remaining coil sparks.

Ken

I wouldn't think so either, but after Ken's comment I'm thinking its possible the spark outputs don't have enough nuts to trigger 2 coils. I hadn't even considered that a possibility as lame as that would be.

Turns out the ms3x cannot push enough current to drive 2 toyota coils from the same spark pin... All A B C and D are working fine, I'm thinking I'm going to steal the wires from the EGR solenoids to use as the other two triggers. No sense in running new ones when there are perfectly good wires already there :)

well that's lame.

They supply 25 mA. The inline resistors there may be cutting things below what 2 coils need compared to 1.

You could jumper them but then if the two coils together are more than 25 mA you'd fry the chip.

We put 8 drivers on the board and supply the "wasted COP" output mode for this very reason though. Each output is designed for 1 coil. The same as the injector outputs. Each one is designed for 1 injector.

Ken

You wouldn't want to drive 2 coils on two ports from one wire tho would you? To me you wouldn't ever run wasted COP unless you didn't have cam sync or you only had one wire running to the coil triggers.

BTW, tested using the EGR solenoid wires (purge and vent iirc), got all 4 spark'n in order... and it started, I'm gonna sleep well tonight :)

Weird thing tho, started showing 18,900-29,800 rpm when I turned the ms on, showed sync and spark cut... figured 'what the hell' and did the spark test anyway, wasn't paying attention but it dropped back to 0 rpm at some point... When I tried to start it I didn't have the laptop hooked up (last minute, 'why not' moment. beer induced no doubt) and it clearly was reading the appropriate rpm at that point, idled right away. Any ideas what would cause that? Previously its only ever showed 0 at boot up, and it wasn't oscillating like it was an error, just a steady bizarre reading.

Wasted COP was originally intended for those who have COP but no cam sync. We never intended for anyone to have an output hooked to more than one coil.

Ken

Back to having a sync issue, now when the CLT is up to 150+ it loses sync at idle, reliably does great up till this point... smooth, (tho my idle valve isn't working, dialed the screw out a little so it'll idle around 1000) It'll be running along fine and then shows an rpm spike and stutters and increments sync loss, it seems 'fairly' random, but happens enough that it will eventually die, as it'll have errors sometimes while its recovering from the previous error.
I'm going to pull the caps back off the abe fm circuit and see if it doesn't get any better. If that fails or it gets worse I may just say screw it and pull the motor to start preparing for the turbo... If its actually my cas (which I doubt) I have a spare from a 1.6 but will prolly just setup the trackspeed wheel and tune it from there, it idles well enough now.

Tried the other cas and no dice.... I'm not going to touch the caps, after reading Joe's ordeal w/ the cas and the abefm circuit. Time to order a trackspeed wheel, sigh. Has anyone actually been brave enough to grind a tooth off, cut a tooth off the cas and run it as 12-1 on a megasquirt?

I've run abe's circuit with much success. you just need to make sure you put it to falling edge.

I'll give that a whirl tonight, pretty sure I had it set that way at first and it didn't work, but who knows. Went ahead and ordered a buncha caps from newark to mess with, it looks like its just not filtered enough, gets a big rpm spike before losing sync (to me that says its getting two pulses when it should be getting one). Do you know what values work well with a CAS? The 1nf ones that are in there work for a while, once clt is in the 145-150 range it starts doing it every 15-30 seconds, and just keeps getting more frequent the warmer it goes. Hopefully its just these radioshack ceramic caps... need to go ahead and get the trackspeed wheel coming anyway.

No great rush anymore tho, looks like we're going to put the new SC on my brothers car, prolly this weekend, so we'll have an SSM car to autocross this year. (of course its gonna need a 1.8 rear swap and clutch, weee.)

i build abes circuit to spec and it was perfect with the NB sensors...dunno about CAS. i keep forgetting you are trying to do that.

Yeah... tried 2.2nf caps, and got the same thing after it was more warmed up. I'm thinking that its just hopeless, Just don't do it. I haven't seen anyone that's actually happy with the abefm circuit on a cas...
I went ahead and soldered up the normal miata opto crank and pullup cam setup. Trackspeed wheel is coming, so I'll eventually change it back to the oe nb input circuit, but for now I'm thinking this will be enough for me to get a comfortable tune in, and drive the car a little.

So anyway, got the other circuit built (the traditional one) And now it doesn't ever lose sync, Yay!

But, it doesn't want to start when warm. fucker idled like a damn champ and I got a little ve analyze live tuning done by dragging the clutch with the ebrake on (lol, I have a new clutch ready to go in and I didn't really do it that much).... I was all set to take it for a drive and it doesn't want to start, tries to start almost immediately, but then falls on its face when it gets off cranking pulse. Looked/acted really lean. so I bumped up the fuel in the idle cells. bumped it up some more... up to about 30% and it seems like its close to wanting to run now. but I'm at 95% ve. fail.

I pulled the coils off and they're still firing like champs. I'm thinking something wonky w/ the injectors? I f'd w/ cranking pulse, ASE, WUE.... none of it made a significant difference (adjusted within reasonable limits) It was running great then nothing... gotta be something hardware related, imo. Pulsewidths were where it had been when it was idling earlier, its just like its lean as crap now.

So I have some other injectors, prolly going to try those tonight... or I might clean some stock ones and try that, idk.
It just doesn't make sense, and I feel crazy blaming the injectors... but I can't point at anything else. As it was getting up to temp, I killed and restarted it a few times and it'd light right off... The other thing that makes me thing I need to try different injectors anyway is that I'm basically using the diyautotune 94-95 sequential maps, and had to set my req_fuel to 14+ to get it run right. I'm using rx8 yellow injectors (that I have cleaned in my ultrasonic cleaner while pulsing them with a basic stamp driving a tip120.... cleaning actually helped smooth it out a lot)

Looking to build an MS3 + MS3X for a '95 1.8 soon.
Thinking about the spark outputs, would using ABCD from MS3X and splicing them in pairs to 1g and 1h on the 64 pin work? That would be using the OEM coils.

No, take 2 of the MS3X's outputs and connect them to the 2 coils - each coil fires 2 plugs, but it's wasted spark so you only need 2 signals.

If/When you convert to COPS you can use 2 more of the MS3X spark outputs and run full sequential.

Okay, so A=1&4 to pin 1g and B= 2&3 to pin 1h. Or is it A=1&3 and B=2&4
Tunerstudio set spark to "Wasted Spark".

Trying to get the build straight in my head first.:nuts:

1&4, 2&3 just like it is stock when you look under the bonnet ;)

Has anyone actually done that before though? I never tried it. If it was me I might build the traditional LED spark outs and not use the ms3x for spark until you can go sequential. Post here if it works though :)

No, since only like 4-5 guys are even on MS3X and all went seq.

but that's how you'd do it, use spark A & B and wire to each pair. Use the LED circuits for something else... why go out of your way to wire up circuits that are already built on the ms3x board?

This, there's 8 available circuits on the MS3x you can use for driving ignition, might aswell put some of them to use.

Thanks guys, that's one question answered.
As I wont need Joe's spark circuit, should I just follow the standard GUIDE Step28 a-g ? or do I leave this step out?

that's not even the correct IGN output mod anyways.

I wouldn't follow that writeup personally per say...just populate this:

http://westfieldmx5.devocht.com/star...ents_empty.gif

So basically follow Franks' build leave out step 65 completely?
What's confusing me is following Franks guide and not needing Joe's IGN circuit, do I then install as per step 65 on the Megamanual or leave that out as well, due to using MS3x sparks?

Back on to CAS, same as above in step 50 ????? Franks CAM/Crank circuit or leave out?

No one has written a guide for MS3 exclusively. but most of the MSII write-ups can be used, since there isnt much difference between the two, but firmware and features.

I didn't say follow Frank's guide, but I did say populate only those that I pictured from his guide. You'll need to then add the tach input circuit for a 4g6e CAS setup and you can pretty much be done with the mainboard, unless you wanna use the three LED circuits for something.

The MS3X will handle: idle, fuel, spark, and any other inputs/outputs you might add extra.

Many thanks Brain, you're putting me on the right track.
Bit of a :noob: in electronics but smart enough for "monkey see, monkey do" assembly.:laugh:
Might change R3 and R6 to 1% as Frank says for TunerStudio.
I might just be the first with MS3 + 3X in the UK.:D

UK maybe, EU no.

Never tested my voltage reading on a DMM vs. TS.


Maybe one of these days I need ot make a MS3x NA & NB writeup

:bigtu: