sr20det cranking VR2 trigger

[drumnbasslvr]

Hello, forum. I have read many of the posts within here and have gotten this far with my install. I am at a loss for what to do next, so I need your help. Many thanks to all of the moderators, helpers, and superms people. Your posts to others have helped me postpone my inquiry here and have gotten me further along. My setup is as follows.

sr20det engine
diypnp v1.5 with diybob
sequential board
wired for cop (4 jumpers including aled, wled)
ftdi usb-rs232 cable
diy 12-1 trigger disc

See files attached to review my problem. I can spin the CAS with a drill and the signal is perfect with no loss. If I install the CAS to the engine and crank it (full battery), the red bars at the bottom of the composite log appear in spots. The center blue pattern remains unaffected which is wired to opto+, opto- jumped to ground, and 1k pullup to 12v. The upper green signal in the composite log generates many triggers all over the place which shouldn't be. The VR2 has a 1k pullup in 12v.

I attempted to use a 51k in r21 and r20 with no success. One makes the signal disappear, one does not affect the signal. I have not tried r39 pot yet.
I read CRSTune stated that 1k pullup in VR2 to 5v will work. I didn't try this because the CAS is a 12v device. Staying electrically safe on this one kept me from trying it.

I read that the CAS signal should be grounded to SG. No I have not done this. The wiring harness is brand new and ran the motor just fine with Nistune previously. Furthermore, If there is a diypnp2 with sr20det that simply plugs in and works, then why bother with hacking up my harness? That ecu makes an engine run, I should be able to get this one too.

I would also like to let pin 4 get pulled low as proper to activate the ecu relay, The onboard relays are 5v, in a 240sx documented build wire 36 is wired to relay 2, is that safe if it's sending 12v in to a 5v relay?

I appreciate your consideration to help me unravel this noise problem in my VR2 circuit.

[Matt Cramer]

Don't use R39, R20, or R21 with an optical sensor. What happens if you try a 5 volt pull up instead of 12 volts?

[drumnbasslvr]

5v 1k pullup on VR2 shows lots of trigger hits all over the place again just like when I had it pulled to 12v. No change. Lots of sync loss while cranking.

[drumnbasslvr]

I have tried larger pull ups of the values 4.7k and 10k tonight without success. I believe the best course of action is to reinstall the 1k pullup to 12v, hunt down noise and eliminate it. I have two approaches in mind.

One is to unravel the first foot or so of harness from the connector inside the car and locate the shield wire for the CAS. Then clip it from ground, and send it to Signal Ground instead.

Two is to find the ground terminal wire for the CAS and send it to SG instead of ground. I can achieve this by delooming near the oil filler cap and sending the wire to the SG that goes to the TPS.

Which one do you believe will produce the best result? At this time I have 0.5 ohm max reading from ground in the ecu to engine block.

EDIT: I will replace or refresh the starter as mentioned with some grease in the gearing of it, freshen the contacts, and see what happens to the EMI and audible noise levels.

[Matt Cramer]

I have seen failed starters put out massive amounts of noise. If yours is marginal, it's the first thing I'd suspect.

[drumnbasslvr]

Matt, I am John Conti with the fast talking post on Facebook you saw. I felt a video would help people understand better which is why I did so. Here is a cranking log for consideration.

[drumnbasslvr]

I have been able to clear up the signal while cranking with the low pass filter mod on the cam signal. I have tried 10nf, 100nf, .47uf, 1uf, 10uf, 25uf, and 330uf which were in my parts bin. 330uf removes so much of the signal so I must choose a lower uf value. 25uf is not enough. I have an old 240sx automatic transmission ecu here that I can salvage a 47uf and a 100uf cap from to test. I'm unsure why it is taking so much to tame the signal in the first place but so be it unless I have other guidance. This all performed with a 1k pullup on 5v.

EDIT: just tried 47uf and things are looking better! See attached picture.

[Matt Cramer]

That's definitely progress - you might be able to clean up the rest with software noise filtering.