ill have to double check, one of the ones in the FAQ

ScottFW is currently testing my wiring harness and cops out on his miata. He has not been able to reproduce the problems I had been having. So this is starting to get interesting.

Is it very cold where he is testing them?

We live 15-20 miles apart, weather is pretty much same-same.

In cold weather with my COPs on my car, I have to crank for ~3 seconds, then it fires and stumbles to life. It's not a strong OEM-quality start, but it works. Because I don't drive the car much when it's below 30° my cranking PW and ASE are probably not ideal, but it's good enough to start the car.

When I swapped in shuiend's harness and COPs and tried starting the car on a cold morning (22°F in my garage) it behaved mostly the same. Crank for 3-4 seconds and the engine stumbles to life. It died after running for 3 seconds, which is when my megasquirt switches from fixed MAP mode to ASE. Again, my ASE may need tinkering at that temp. I started the car a second time and it stayed running.
I will try to repeat this test on another cold morning so I have n>1 observations.

I don't know *exactly* the difficulty Lars has experienced on his car (need video) but his COPs & harness did not behave much differently than mine on my vehicle, which is to say, adequate to start my car in the cold.

My COPs harness has 10,000 uF capacitors (two of them, which is probably one more than necessary) and has 12 AWG +12V and ground. Lars' harness has puny +12V and ground wire, probably 18 or 16 AWG, and he has no capacitor. My COPs are 90080-19015 and Lars' are a different Toyota/Denso part number that obviously work. I might expect his small AWG (16 is probably fine since that's what Toyota used to wire them, but 18 is pushing it) and lack of cap to affect operation at high rpm, but it doesn't seem to matter for simply starting and idling in cold weather.

Since the COPs & harness don't seem to be the issue, I'm trying to think of other differences that may exist between our cars, other than the obvious fact that I no haz trub0z yet. My MS has the ignition pullup resistors changed to 270 ohms (or did I use 330, I forget) while Lars' MS likely still has 1K resistors. My cranking advance is 14* and in my idle cells it's 17*, not sure what everyone else runs but it might start easier with more OEM-like timing (less advance). I've got BKR6E plugs with out-of-the-box gap (all verified to be ~0.040"). Then there are whatever differences in MS startup settings like cranking PW which I'm sure Lars has already tried tuning. Maybe his battery is marginal and not putting out enough juice in the cold, and combined with small AWG and no cap he has voltage drop issues at the COPs. I'm just throwing stuff out there. :dunno:

Is it confirmed that the factory coils start the car OK?

Lar's factory coils start the car fine, he's been running with them since it's cold and he's had this issue.

again, I started my car on a wonderful 23*F afternoon, highly untuned as it was first time trying MS-II...had absolutely no issues starting right up and idling perfectly.

Does increasing or decreasing the dwell improve the cold starting?

Have not tried that out yet. How much should I try increasing the dwell by?

When I get my cops back from ScottFW in a few weeks I will try increasing the dwell and I will also try to get a video of the car running bad.

The only difference that I can tell between ScottFW using my cops and myself when starting is I only had problems when I would leave the car outside over night as opposed to ScottFW starting it in his garage. It is a good 10 degrees colder on average right now then when I was trying, so I do not think the garage is playing to much of a roll in it.

My garage is attached to the house, but not insulated. Using my digital thermometer with remote probe (quality lab-grade shit calibrated to NIST standards) over the past four days it has consistently measured 2.0-2.5°F warmer in my garage compared to outside. That deltaT is insignificant for these tests, but it does remind me that I need to insulate that bitch and put a nice heater in there and stop living like a Neanderthal.

I can try messing with the dwell. I'm currently running 3.5/2.5 cranking/running. I'll increase the cranking dwell a bit and see if it changes anything.

I dwell settings when running with the cops are 5.5 cranking, 2.5 running and minimum discharge .5. Does that look to far off?

This has been happening to me as well, seems to be when ambient temp is below 60F. Runs great when warmed up.

It's definately in the coils. Past 3 mornings I decided to experiment, I took the hair drier and warmed the coils up...it works lol. Ran the hair drier for about a minute tops, going back and forth over the coils. Runs perfect after that.

I can't explain, but I'm thinking the resistance is high in some of the coils when it's cold. Going to try measuring them warm and cold.

shuiend try 3 mS cranking.

Does the MS not have a dwell vs battery votlage curve?

The Toy coils have a function where dwell > y when V > x causes it to terminate the pulse and fire way too early. This function may be going awry when cold.

Does the car show signs of way too much advance when cranking? (e.g. motor suddenly stops turning then re-starts, perhaps with a "chunk!" sound).

Yes the MS will trim dwell with battery volts. It is configurable. It also fixes timing and dwell when cranking. Most people lock timing to 10° fixed when cranking.

If memory serves, my settings on my 91 were something like 3 ms running and 5.5 ms cranking dwell. I never had an issue starting ever, even with a true cold start on a snowy day, with 3 really big vents in my hood.

I struggled with cold starts on the 19015 COPs until I fed the signal wire with more than 7.5mA from the 5V rail. I'm using ground switching to fire them. 10ms cranking, 2.5 running, 0.1 min.

Also, capacitors... They're there to filter out the noise. The toyota coils can be very harmful to a v2.2 MS.

I can't find my notes on how much pullup current the Toyota COPs require. However 1k seems a bit wimpy. Indeed 330 ohms may be required, and the cold temps could reduce the "gain" of the internal circuitry, necessitating a stronger pullup.

Jason,
Do you still have the dwell research you did on them? I'd be interested in all the data you have on these coils.

Sounds like their may be a solution finally after 2 winters :giggle:

I've been having some trouble starting cold too.. I wonder if it could be the coils. I'd been blaming it on my cranking PW, but thinking back IIRC it wasn't this hard to start with the stock ignition.

It also sometimes takes a couple cranks when hot, but that has to be either vaporlock, a cranking PW table in need of fine-tuning, or both.

I've already changed the resistors in my MS, I just want to make sure I understand what's going on. Is this a simple relationship of V=IR? So if V=5 and R=1K the wire gets 5mA? If R=330 then I=15mA, R=270 means I=18.5mA, etc?

Also, 10ms sounds like a lot of dwell, but that's just compared to other numbers I see strewn about the internet, not any specific knowledge on my part regarding the COPs. Probably no harm in trying to see if it gives me a more robust startup, since I don't think I'm cranking long enough to damage anything even if it is a bit on the high side. Right?

Not quite, because the input of the coils won't be at 0V. If it's at say, 1.5V, then the resistor is seeing 5-1.5 = 3V, and if R is 300 ohms then I = 10 mA.

10 mS is way too long for the Toy cops - it has a "just fire if dwell is too long" function which I mentioned in an earlier post, and that function may be invoked, and spark will be too early. Even at 8.5V the dwell should be no longer than 3.3 mS.