WestfieldMX5

2nd opto .
I just seem to remember that you use both stock input circuits (VR and Opto). Not sure though, I seem to have developed 'youth alzheimer' .

Ben

iTrader: (33)

I've set them up using the mainboard VR conditioner for CKP input and the mainboard opto input for CMP input. That works well, too.

duffbuster243

iTrader: (1)

I'm going to try the 2nd optoisolator and I am ordering the parts now. Please make sure these parts number are appropriate for my needs:

Index Quantity Part Number Description Customer Reference Available Quantity Backorder Quantity Unit Price Extended Price
1 2 4N25VS-ND OPTOCOUPLER PHOTOTRANS 20% 6DIP 2
Immediate 0 0.57000 $1.14
2 5 470QBK-ND RES 470 OHM 1/4W 5% CARBON FILM 5
Immediate 0 0.10000 $0.50
3 5 4.7KQBK-ND RES 4.7K OHM 1/4W 5% CARBON FILM 5
Immediate 0 0.10000 $0.50
4 5 1.0KQBK-ND RES 1.0K OHM 1/4W 5% CARBON FILM 5
Immediate 0 0.10000 $0.50


I also need the 0.01 uF Capacitor, what is a good digikey part number for this?

I'm ordering some extras as I may also do this for a friend that is having the same issues with his MS1.

Braineack

iTrader: (62)

399-4326-nd

duffbuster243

iTrader: (1)

Great, Thanks! I'll post again if I'm successful, Probably won't be for a few weeks as I have my cracking weekly ebay manifold off getting duplicated by a professional welder.

Braineack

iTrader: (62)

please post about how shitty it is and what you had to do to solve it so people stop buying them.

duffbuster243

iTrader: (1)

Success at last! The 2nd optoisolator for the CMP input has solved my problem. I drove around for and hour or so yesterday and not one sync loss.

Braineack

iTrader: (62)

glad to hear. I wish I knew why on some installs it doesn't like the simple pull-up in, but on others there's no issue.

Joe Perez

iTrader: (8)

The traditional circuit, with a pullup on JS8 and nothing else, has virtually no noise immunity and zero hysteresis. In other words, it's extremely sensitive to every form of noise and distortion imaginable.

The opto converts the circuit from being a voltage-mode device to a current-mode device. This grants it some immunity from induced noise. (eg: voltages induced into the long wire between the ECU and the sensor have relatively little effect on this topology.)

And, to the extent that optoisolators do have a miniscule amount of hysteresis, that also helps.

duffbuster243

iTrader: (1)

OK so after I did the second optoisolator I didn't touch the timing which I had set prior. I went to check the timing tonight as it seems to be running sluggish, I set the timing in megasquirt to fixed at 10 degrees and couldn't find the mark. Well as I looked around the pulley I found it on the alternator side as show in the picture. Before I did the 2nd optoisolator I set the timing at 10 degrees and it worked fine. it now looks like I'm 180 degrees out If I put my timing light on the 2nd plug the mark shows up on the opposite side closer to where I want it to be. How is this even running?

duffbuster243

iTrader: (1)

No suggestions? Is there something in the software I need to change such as the polarity of the CMP input?

Braineack

iTrader: (62)

the crank and cam polarity have to be the same, else you wont sync.

since you did change the input method it's possible the trigger angle needs to be readjusted.

Reverant

iTrader: (10)

You need to set it to falling edge, going high.

duffbuster243

iTrader: (1)

I'll give that a try this weekend. How is this thing running right now as it is though?

JasonC SBB

I would shrink them to 10 nF each, and 10% X7R type would be fine. In this thread
https://www.miataturbo.net/ecus-tuni...-sensor-71726/
10 nF worked well.

0.1 uF would produce a delay of around 100 us, (4° at 7000 RPM) which would vary if the input trigger thresholds of the CDxxxx IC varied, like due to temperature.

BTW one thing the opto-isolator circuit does, is be slow and not respond to very thin spikes of noise.

GAMO

iTrader: (1)

With a .1uF cap, the ECU wouldn't even sync since the RC constant was so high.

As for the 40106 discussion, it's an excellent trigger due to it's delay (lack thereof) and hysteresis. I think for 10V, it has about 3.5V of Hysteresis, and the response time is sub-100ns.

Glad the optoisolator worked.

Joe Perez

iTrader: (8)

I have built that exact circuit a number of times, and it syncs just fine. Probably wouldn't be a good thing to use with high-toothcount wheels or wheels with very small teeth (NBs), but on NAs it's the best thing since sliced bread.

Although it's nice to be able to converse with people who actually understand the concept of an RC filter and how it applies to the real world.

duffbuster243

iTrader: (1)

Right now its at rising edge going high, I set it to Falling edge going high last night and it died and would not start back up until I changed back to rising edge.

Braineack

iTrader: (62)

what does the composite look like?

duffbuster243

iTrader: (1)

I didn't have time last night to take a composite log or any log for that matter, I probably won't be able to get a log until sunday evening. My weekend is booked up with a bunch of things, none of which I want to do. It sucks.