DIYEFI.org Forum

In terms of priority :

NipponDenso 24/2 = first - once I have that working...
Missing tooth wheels = second - once I have that working...
Whoever will actually test = third etc :-)

Fred.

Audi/Porsche trigger wheels, the ones that use the flywheel teeth and are over 100 teeth and a second trigger. 135+1 on the crank and a third trigger for the cam would be great. Yes, three different triggers, two for crank alone.

Doing this would make a lot of Audi and Porsche guys very happy, since no one would have to fiddle around with divider circuits.

Why 3 triggers surely a flat 6 boxer engine requires a crank and cam reference for full sequential what is the third one for??

No missing tooth in the 130+ wheel, so crank sync tooth required for quick startup, and cam for COP/Seq.

Supporting it fully could be done experimentally with an ordinary interrupt pin during cranking when events are far slower than ISR times. After that, the cam could be relied on for sync UNLESS it's VVT...

If it's VVT, you are SOL as that would make it very complex.

Basically, at the very least we can support it with somewhat slow starting due to lack of crank sync OR no COP/Seq due to lack of cam sync. Either are easily achievable.

Thanks for your interest :-)

Fred.

I've always figured a wheel with too many triggers (130 in this case, 180/360 in Nissan's case) could be dealt with by ignoring 9 out of every 10 trigger to effectively divide the number of triggers by 10, turning the Porsche flywheel tooth trigger situation into a 13 tooth wheel. This would remove VVT from being a problem.

BenFenner wrote:I've always figured a wheel with too many triggers (130 in this case, 180/360 in Nissan's case) could be dealt with by ignoring 9 out of every 10 trigger to effectively divide the number of triggers by 10, turning the Porsche flywheel tooth trigger situation into a 13 tooth wheel. This would remove VVT from being a problem.

That gets you a rate/speed OK, but it doesn't get you a position. You could do it in a complex way, but it still would be a bit sketchy. Just arbitrarily starting the div/ignore process won't work because you can't guarantee that you are sufficiently far from a tooth to avoid sync from becoming an issue. Besides, you need to interrupt on or count all teeth anyway to know when you are on the ninth one.

If you meant hardware, not code then that is what these guys already do AFAIK.

Fred.

What's the position of the extra tooth relative to the rest?

I would think this would be easy to solve in hardware.

BEST case: The extra tooth exactly lines up with a regular tooth, and is wider. Then, you block the signal from the crank to the ECU when it comes around.

Workable case: You come up with a circuit to block the next signal after seeing a rising edge on the main crank signal.

Then you end up with a 300-1 or whatever the numbers are on the crank. Especially if you're building something to skip teeth anyway, you'll already have all that there.

Sound like something worth pursuing? Personally, I'd say no - without the superior engine management advantage, how will I continue to beat up on porsches in my little 1.8l miata?

The TDC trigger is just a pin in the flywheel, you can see it here on the left side. Theres two VR sensors, one picks up the flywheel, the other picks up that pin. The third trigger I mentioned isnt stock, its a hall sensor I added to my camshaft gear for future sequential injection and semi-sequential on MS (I have BIG injectors). That TDC pin is approx 58 degrees before TDC on cyl one. I cant find the exact value in my notes right now, its actually 58.something. 944 N/As have 130 teeth on the flywheel, the turbos have 132, and other Porsches vary, some have 100 (some 911s and 928s). Porsche 968s and S2's (both 3 liter N/A engines) both had 60-2 rings on their flywheels, but those are very expensive, require a clutch job, different bellhousings that I dont think are compatible with Turbos (Id have to find out), or theres Fidanza flywheels which could be used... But again require a clutch job. Best case shop hours for a clutch job is 18 hours so those two alternatives are undesirable



Audis have 135 tooth flywheels, and a TDC pin thats 62 degrees BTDC I think.

None of the Porsche engines that have these kind of flywheels have VVT or anything like that that I know of. Pretty simple in that area.

I tested the LM1815 VR chips with 132 teeth at 7000RPM and they seemed to work fine. I did it with a USB oscilliscope and that trigger generator program someone made that uses a soundcard to create pseudo VR waveforms though.... Maybe not very high tech or a proper test though

This is all a moot point for me at this stage, as I gave up and bought a 60-2 wheel and trigger mount that uses a hall sensor, as I gave up on MS+132 teeth+divider.... But LOTS of people want an affordable capable EMS that can use the stock sensors. The cheapest EMS thats plug and play in that area is the Wolf3D, and its still $3k I think.

The 60-2 wheel is probably the smartest way to go, I've thought about it for my car as well.

That said, I still like the idea of using one sensor to gate another - it's hard to say from the picture (an o-scope trace of the two if you could get it for us would be awesome!), but I think you could VERY easily make something to just mask one tooth based on the presence of another. And a single chip with a couple flip flops could probably divide down the wheel if it were needed - though with any luck, esp. the co-processor enables chip for the FreeEMS, it shouldn't be a problem at all. Certianly 80 MhZ could go "have I seen ten pulses yet?" in real time.

Anyway, I'm picturing something like a transistor inline with the signal from the teeth, and then it's gate tied to the sensor (perhaps processed) which is watching the pin. BAM, instant 310-2 wheel, or something.
-Abe.

60-2 is best for lower rpm engines, for really high spinners 36-1 is a better choice. high/low boundary = 8k or so I guess.