DIYEFI.org Forum

You could write software to make it work, but it will not work with FreeEMS at the present time, and possibly not ever. Hard to say. It's such a niche thing to want to do that it's likely not worth bothering with. For the time being missing tooth gives you the sync level from the place you mount it, no more, no less. Sorry if that's a pain. Having said that, we can run an engine with 2 outputs, a dizzy, and batch injection, from one of those wheels, and I'd like to prove that out.

Fred wrote:it will not work with FreeEMS at the present time, and possibly not ever.

Ah well, shows what I know. What specifically about it won't work? I thought that if you pulled up the signal to a square wave you'd be able to write a decoder for the pattern of teeth you welded onto the back of the cam gear. Counting the 38 teeth on the cam/aux shaft would let you stop reading the goofy teeth flywheel. That said, I did see this when I was planning to use ms and I chose the CAS for the fewer screw-up opportunities it presented.

Kicking myself in the teeth for throwing away a miata CAS when I moved house last year, it looks similar enough to that DSM CAS so it just might work. Think I'll see if my brothers spare BP engine has a CAS and pull it to check if the chaft thickness is in the correct ballpark.

38 on the cam? Sorry, I must be confused or something. I thought you meant missing tooth twice per rev on crank + cam sync. If I'm wrong, then my answer is also wrong.

Those CASes are a dime a dozen. Virtually all of the B engines use them, sometimes in a dizzy format, other times in a CAS format. Ditto the 6 cylinder engines from mits and maz. There is no 6 decoder, but it will be easy to do, and I'd like to support it.

They are a dime a dozen, but average 2nd hand prices on ebay do not reflect that at all. Suppose I could try a wrecker but they usually pull the engines before adding the rest to the accessible stock.

I keep forgetting that you're in Europe! Sorry. In NZ or the US they'd be cheap as chips. Maybe get a friend to post you one from Canada or something? In Spain ALL shipments from the US got taxed to fuck.

Let me know if there's a better way to get these numbers. I fixed an old plastic cam degree wheel onto the CAS and tried wiring it up like the manual apparently says to:

But with the two grounds separate I couldn't get the voltage down from 5V and I couldn't convince myself that joining them wouldn't end in sparks and sobbing. So I ended up doing this:

I think the voltage was the opposite of what it should be because of this. When the rotor is in the gate the voltage is zero, the leading edges are falling edges, etc. I'll join the grounds and do it again if someone says it's a good idea.

I picked a random edge as a reference to read both rotors against. The values with asterisks might actually be 1 degree less, but the lines on this cam wheel are a little suspect. I'd guess it's the line printing rather than the neat 55-35 degree pattern being changed by 1 degree for just that one tooth. I can always spin the degree wheel 180 degrees and measure again, see if the same tooth is out by 1 degree.

Rotating it clockwise:
<bad data removed>

The underside of the rotors look like this.

Good research! Do the math for me, though? :-p

The outputs need a 1k pullup to 5V on them. IE, they're open collector outputs which ground your signal line during a "tooth".

I found a slightly better way to mount the cam wheel, I think the tolerance is about 0.25 degrees, so 0.5 degrees at the crank, is this accurate enough? I measured everything again, is this what you mean by the math? If not let me know.

Cool, those angles match up with what I have within a degree or so. :-) So you can run on this, as is. Accurate enough to film this video: https://www.youtube.com/watch?v=6l-KyVG-teM

The setup on my daily (#3) is even less accurate.

Fred.