psst joe: capacitors in series?

C = 1 / ( 1/C1 + 1/C2 + 1/C3 ... 1/Cn )

so did you like stop on the road and bust out a soldering iron to change the caps on C1 and C2?

Unfortunately I wasn't carrying an entire sample kit with me, just a few pairs of different values in a bag. All I had on-hand at the time were two each of .0047, .00047, .033 and .001.

edit: Actually, I take that back. I also had a pair of .0082 on hand, since I was half expecting the .0047s to fail. Well, I'll stick 'em in tomorrow morning before I drive to work and have an equivalent .003uf

No. After asking myself "What would Chuck Norris do?" I focused the intense power of my mind onto the circuit and the solder melted out of fear.

:jerkit:In other words:

http://img178.imageshack.us/img178/8273/protoqx0.gif



On a more serious, I gotta say a big thanks to Abe for taking the time to reverse-engineer the Mazda OEM input circuits, sharing them, and then hounding my ass until I actually fixed the problem.



BTW, patsmx5, the reason I was being so stubborn about the plug gap is that I ran .030" for a very long time with my EMU and never had ignition issues.

that is neat. NB owners dont have that nice wide space to do things like that

And there is much happiness.
 

Drove to work this morning with an equivalent .003uf on C1 and C2 (.0047 + .0082) and it felt fine. I compared this morning's log with yesterday afternoon's, and it seems like the .003 gave me just a tiny bit more "jitter" in the RPM readout than the .0047 did. So with that piece of data in hand, I've settled on the .0047 and will be permanently installing them and closing up the MS for what might hopefully be the last time for a while.

Thanks for the suggestions everyone, and I do sincerely apologize for being a bit of a dick the past few days.

Honestly- the ECU in the NB is in such a horrible location that I cannot see why you fixed-headlight guys don't all relocate it to the pax side. I can't imagine having to crawl up under the steering column every time I wanted to touch the box- hell, one of my MAP sensors is up there, and even that little thing is a horrible pain in the ass to reach.

I've considered relocating it, just use a 4-6' DB37 and move the MS over to the passenger side. That would require effort though.

Joe, have you sent out your COPs to Ben to test yet? ;)

you know, i might JUST do that. my foot sometimes hits it while it hangs attached to the stock ECU. ill hit frys later today.

i too have been having some misfiring issues (check engine light has been coming on with P0172 (system too rich) and P0300 or P0301 (misfiring), im thinking this might help the problem. ill try out what you have there, using the 0.0047µf, il let you know how it fares on the NB. after tomorrow's exam, im going to also try to increase the amount of grounds coming from the DB harness, hopefully that might solve some problems.

that means i gotta reroute the vac line, hah, more work. oh well.

Still waiting for a PM from him with his address.

Sounds like a plan. The '94-'97 guys really got lucky with their ECU placement. I wish Mazda would have done them all like that...

It doesn't look like a lot on paper, but that darn comparator circuit winds up being a whole lot of parts. I managed to fit mine (barely) into the proto area, where it consumes about 2/3 of the space, leaving just enough room for two relay drivers and a pair of resistors related to another hack.

Euro-spec NB for the win! :)

Jim

Sorry Joe, I thought you wanted to go a different route. PM forthcoming.

Nice install- looks very clean. What is the amazingly large thing to the left of the MS?

[edit] Greece... Left-hand-drive country... Didn't know the LHD Euro NB's had a different wiring harness. Interesting.

hmm, ill look more into it for sure.

Joe. This thread here: https://www.miataturbo.net/forum/sho...hlight=ms+2004 highlights his build. That is a switch box for his parallel wired MSI.

OMG! If anyone does this, I want to know how it works. I have SERIOUS doubts that an off the shelf parallel cable will deal with 15 amps of low ohm injectors running through it! I've thought about it and rejected it on principle several times.

Also, it's not just swapping in a cap into any circuit, it's the combination of caps and a good comparator which does it. Like Pat's set up, you can't just take parts from one circuit and put them in another.


Joe,
Yes I too noticed that there's more jitter (on the scope) with big caps. Ideally you run the smallest caps which never EVER let any noise through. If they are two big the motor won't run above certain RPM.

I'm still really curious why you needed such large ones, I'd love to see your car on the scope. Pulling out the caps entirely and idling in the garage might tell a nice story. Obviously your sensor is very different from my VR ones.

I guess we just work in different industries, but I just don't consider 0.0047uf to be a huge cap. I mean, the "fundamental" frequency of the CKP pulse, at redline, is 240 Hz.

When I ran without any caps at all, it misfired just revving it in neutral.

However, the caps have been soldered in, the case is closed, and there will be no probing.
From the ECUs point of view, your sensor is same as mine. They both contain an active driver with an open-collector output.

But, mine hates me. And yours doesn't hate me. It hates you. :-)

Too bad about the probing, I like probing. You need to mount the MS somewhere easier to get to it, perhaps your forehead.

I should strongly consider sleeping longer at night.

High Load misfire especially under rapid throttle input is generally due to inadequate spark. Basically high cylinder pressure combined with a lean mixture and you get a resisitant too high for the spark to jump, thus the "0" on the tach log, as you got no spark.

Assuming this a 89-97, I strongly suggest running Nology Wires and NON resistor plugs. You will get static on the radio, but who listens to AM anyway?
Adding the Nology Coil amplifiers will also make a difference if you want to run serious boost and are having spark blow out problems.

http://www.nology.com/images/sparkdiag.gif

Doc, see the section you are posting in...it's specially related to his Engine Management.

Ok, thanks for the advice, Doc. :jerkit:


And it turns out that I fail again. I swear that it was working fine earlier this week when I had the breadboard on the floor. I re-assembled the unit, mounting those exact same two capacitors back in the proto area, closed up the case and re-mounted it, and it's doing it again... :vash:

This is really starting to get ridiculous.

I think I'm going to just say "fuck it" at this point and install a pair of .022uf caps. It was working with .033uf on the protoboard, so I know those aren't "too big", and I really just need this thing to be running properly for Sunday.


:ky:

Yeah, I envy the '94-'97 crowd in this regard...