Even after upgrading to MS2 I am still getting a loss of sync or random dropouts, not as often as with MS1 but still enough to annoy me. While looking around I have found what seems to be the end all solution to the problem, which is making a second optoisolator circuit. I have found a diagram posted by Brain floating which I posted along with a picture of it on a board. Please verify that it in fact looks correct.

 

is the above yours?


do you have a composite log of your current sync loss?

 

The above is not mine, just one I found and scribbled out what I believe does not pertain to the circuit and changed pin 2 from spr1 to IAC1A. I don't have a current log with it happening as my old tuning laptop's motherboard fried. I can probably get one next time I have the car out but the way the weather has been lately that may be a week or two.

 

I did purchase 15ft of shielded wire which I am going to use to rewire both CAS signals directly to the Megasquirt box to see if it makes a difference.

 

I'd like to see a composite log of your inputs from the CAS.

Circuit above seems correct, Been a while, just follow the diagram, did you need it again?

 

A second opto is a second band-aid.

If you want a proven reliable input circuit, build this:

 

even better and still incredibly easy.

 

I'm not quite good with circuits, and that appears simple, almost looks like it would be easier to do on a separate wafer. Does anyone have a picture of this done?

 

You could easily build it in the proto area, assuming it's not already filled with a bunch of other stuff. But yeah, a separate board would be ideal. If I weren't so lazy, I'd design a little JB-style daughterboard and offer them for sale.

I don't have any pictures of the actual circuit built, apart from this one, which is worthless for your purposes:


(It's the circuit that the red/wht and blk/wht wires go into.)


In all seriousness, would you guys actually buy a little board containing this dual-input conditioning circuit?

 

It won't. (sorry)

 

And your evidence for this?

 

Dictum meum pactum.

 

Sorry, you need to try harder.

 

I tried really hard several years ago when I was doing the initial install on my '92. Wound up saying "**** it" and building a 36-1 crankwheel setup. (This was, admittedly, a rather disproportionate response to the situation.)

The 40106 solution has excellent filtering and hysteresis, and doesn't rely upon current-mode devices which offend my sensibilities.

 

I'm willing to try this. I added my inputs/outputs on. Is this correct?

 

Do you have a composite log showing loss of sync?

 

That's correct. Depending on how you have your cam input wired now, you may need to flip the polarity setting (in software) for that input after you build this circuit. I can't remember what it's called on the MS2. Also, you'll need to re-calibrate the hardware latency. This circuit has a very long time-constant by design.

The value and stability of C120 and C125 are highly critical. Use low-tolerance parts here (eg: 5%, C0G).

 

I do not, How do I do a composite log in Tunerstudio?

 

Under the diagnostics tab.
You're chasing your tail if you haven't taken a composite log first. It will show you (if) you have a sync problem, and if so, which sensor is not coming through.

I am generally amazed at when I see a report of a sync problem on a Miata. I personally have numerous Miata installs under my belt, as in I can't count/remember them all, and have never had a sync problem. As a company, Miata is one of our top selling applications, and I don't recall any sync problems with any of our prebuilt units if correctly installed.

 

What makes the CD40106 trigger superior to an opto-isolator?