How to wire CA/SR Tach signal to 89-90 Cluster

[johngriff]

***DISCLAIMER: This is posted in the premie section for a reason. If you do this, and you fry your shit, dont complain to me. I will be actively testing this, and will say yes this is safe after testing. If you repost this on another forum, i'll cut you. If you install this on someones car, and charge them money, I will find you, fly to you, and cut you. Wait until this becomes part of zilvia's public domain. That is why I am taking the time to write it.**********

***Rough***

Premise:

The SOHC tach works on a Pulse Per Minute basis, like the vss. The tach signal is based on the 12v Single coil signal, creating a sine arc pulse of 4 12v pulses per 360 rotation.

The DOHC/CA/SR Tach works on the same premise of Pulse Per Minute, yet uses a signal that is less than 12v's. Because of this the pulses are not accurately read by the 89-90 SOHC Tachometer because the voltage of each pulse it too low.

***This information needs to be verified using a Scope. If anyone (russ) would like to chime in on me being incorrect in my assumptions, do not hesitate to chime in. ***

Hypothesis:

There is another place in the Engines harness where a single 12v pulse per cylinder is present, the coil pack signal wire. With all four coil pack signal wire combined down to 1 wire it can be possible to replicate the desired tachometer signal of the 89-90 SOHC Tach.

Issues:

In combining the 4 signal wires together, the possibility for the signal to get mixed back down the harness is great. Hence, the signal from wire 1 being sent to 2,3 and 4 simultaniously. The corrective measure and fix would be to attach 12v high amp diodes to each signal wire after they have been spliced off of each consecutive signal wire. Placement, temperment to heat, and exact quality of said diodes is yet to be determined. Suggestions are recommended.

In practice:

I have applied this theory to function in two applications: the nissan 200sx (s12). Once with the ca18det and once with the SR20DET. This method accurately produced a useable tach readout.

Application:

The implementation of this is questionable. The coil pack harness is a rather fragile portion of the harness. Having the inexperienced installer use incorrect parts or bad wiring can cause damage. Testing MUST be done to speak to the longevity of a tachometer (hour/lifespan) using this method. Recorded scope's must also be taken before and after to confirm pattern to anticipate failures in devices.

I am more than open to suggestions, I picked up a fluke scope(standalone) and am still making my way through it (learning). I will start trying to map the SOHC signal sometime this week, afterwards I will be changing the engine, and mapping both the tach wire output and the new wired signal to compare/contrast results and gauge safety.

[johngriff]

bump, please read and comment

[projectRDM]

Excellent idea, but as you already mention there are questionable issues. I'm not certain of the coilpack signal's output, but I'd assume you'd want to use a 6A diode to isolate each signal to protect everything. Diodes that large get very hot, and would need to be mounted somewhere in the open since they'd be working 100% of the time. That alone would suggest you'd want to build a subharness with the diodes prewired in and a breathable casing or heatsink around them, pushing up cost, if you plan to offer it to the public. I only say that because 95% of the potential buyers will NOT follow the proper instructions and wrap them up with tape or lay them across a water hose somewhere. That's a given, and then you're on the hot seat having to replace parts and kiss ass. I learned a few years ago not to offer anything unless A) you can back it up 110%, or B) you offer as is without any sort of support whatsoever and they know it.
Secondly, the longevity of a diode in that application is hard to determine. The OE harness uses diodes in a few places (the headlight circuit for one) but it's a high current sealed unit that's inside the car. Outside under a hood with water intrusion, NVH, and the added heat would play a heavy part in keeping it alive.
As far as the actual testing, you're chartering new territory. I personally have only touched SOHC engines and parts to remove them and core the parts out for scrap. I don't think I've even opened a SOHC FSM before.

But it's a great idea nonetheless and would make the above average installer eager to skip that one extra step of tracking down a DOHC cluster or tach, as they're becoming more scarce every day. To the rest of them you'd just plan to expect some hand holding to walk them through it.

[bardabe]

after you finish this make a fix for RB signal to work with KA/SR tach. =]

[johngriff]

Ok.

Only use 4 of the wires.

Since you will not pulse the same wire twice in 360, this should actually work.

uhhh... its 2am, dont hold me to that.

[Tenchuu]

after you finish this make a fix for RB signal to work with KA/SR tach. =]



** cough cough adjustment screw calibration cough cough **

[bardabe]

** cough cough adjustment screw calibration cough cough **

*cough cough* S13 tach dosent allow enough adjustment *cough cough* :bigok: