NB MS2: CMP, CKP, WTF? (more TLAs inside)
 

So, as I'm in the process of modifying Chris' new MS2 for use in the NB-engined '92, I've been spending some time reading the MS2Extra documentation. One thing puzzles me, and it has to do with their recommendation as to the input circuits for the cam and crank.

First off, let me state that we will be starting out using the NA CAS which Chris already has installed on the engine, as it's known to work and requires no extra wiring. I like to reduce the total number of variables on a first time Megasquirt startup as much as possible. So I built the CMP and CKP input circuits to 4G63 spec, using the stock "OptoIn" circuit for the Ne signal, and a second 4N25 circuit for the G signal. Same as an MS1 NA, basically.

That said, I am puzzled by the information given at the MS2-Extra site regarding the input configuration for the 99-00 Miata. Specifically, I am looking at This Page. The information given there suggests that the NB crank sensor should be fed into the VrIn section of the MS's stock input circuit, and that a second input circuit should be build for the cam sensor using the LM1815N IC, which is a special-purpose chip built by National Semiconductor specifically for the purpose of decoding VR sensors.

VR sensors, for those not familiar, are basically just a coil of wire wrapped around a magnet. They have two terminals, and emit an AC waveform which must be decoded into a DC squarewave before it can be used. I posted a fair bit of info about this, including 'scope pictures, in my Wheel of Timing thread, as I used a Ford VR sensor to read my new 36-1 crankwheel. Among OEMs, I honestly don't know how prevalent VR sensors are relative to optical and hell-effect sensors. I know Ford uses VR sensors on damn near everything- they're the heart of the EDIS system. But so far as I know, the sensors on the NB Miata are most definately NOT traditional VR sensors.

Note the qualification there. I honestly have no idea how they're constructed internally, however unlike a "raw" VR sensor, they do not emit an AC waveform. Just like the NA CAS, they contain a circuit which drives an open-collector output, the result of which (once a pullup is applied) is a DC squarewave. It's obvious merely from looking at the wiring diagrams- the crank and cam sensors are three terminal devices, connected to +12 and GND on two of the pins, with the third traveling via a single unshielded wire to the ECU. True VR sensors only have two terminals, both of which go to the decoder, and they almost always use shielded wire. The crank sensor on the '96-'97 IS a raw VR sensor, but not the ones on the NB.

So the question is: Whiskey Tango Foxtrot?

Have the good people at Extra fallen under the control of an alien brain slug, or did they somehow figure out that it is possible (and for some reason preferable) to run a DC squarewave through a zero-crossing detector designed to rectify AC sinewaves?

On paper it should work. It's set up for a VR reluctor but it's I gave the VR circuit a try around the 1st of the year. It worked once but I changed the POTs and was never able to get it to work again. I don't know of anyone who has got it to work consistently but I would be very interested to hear the details if someone has.

Hmm... Still no bites over at MSExtra.com, which is hardly surprising.

Arga- I honestly don't see how you got it to work. I'm not saying I don't believe it, it just doesn't make sense. If there was a pullup in the circuit and the zero-crossing threshold set to some point above ground, then yes, it might conceivably work. But that circuit, as designed, is expecting a voltage coming into it, and there's nothing in the implementation or wiring that would produce a voltage- the NB sensors (like the NA CAS) give an open-collector closure to ground, and require an external pullup voltage.

I wish I still had the jpg I made of the input mods I made back when I first made an attempt at ms2 on the 1.6. I had inputs fully working but ran into an issue getting it to work with the toyota cops. At the time the miata was my only car so I had no choice but to switch back to ms1 and get it running.

I now have a ms2 kit sitting in my bedroom waiting for assembly. I am going to give it another go now that it is not my dd anymore.

I did arga's circuit and my car started after about 2 secs of cranking. Ran ok but had a random miss where it would drop to "0" RPM's for a split second sometimes. But it would NEVER miss if the A/C was on for unknown reasons (maybe the added inductance somehow affected something?). Anyway it did work but I had noise issues or something. Going standalone with EDIS now.