36-1 VR cam trigger 65,000 rpm problem

[sans-pareil]

Hello everyone, I am running a 36-1 cam trigger wheel on a Citroen CX M25/666 engine. It's a wasted spark, 4 cylinder 2.5 litre turbo engine. OEM, it's triggered with a VR sensor off a 145 tooth crank trigger wheel of which a single tooth is 0.3mm shorter.

Obviously, this won't work with the VEMS, so I've fabricated a 36-1 trigger wheel for the cam shaft (distribution chain). Now, actual issues with the  trigger wheel aside, as soon as I crank the engine and the sensor is within 20cm of the engine (anywhere within 20cm), I get a 65,534rpm reading from the VEMS. Mounting the sensor so that it's right near the trigger wheel makes no difference.

What I've done:
1. messed with the settings
2. changed the ground setup in various permutations
3. shortened the wires to the sensor
4. tried a different sensor of the same type
5. tried another VR sensor of a different type
6. added a mantle to the sensor, grounded at the VEMS end
6. Used a sensitive volt meter to verify there;s a signal. There is. From all three sensors.

None of this has helped. Does anyone have any ideas? If anyone has a some screenshots from megatune for a 36-1 cam trigger wasted spark setup, that would be really helpful, as I can then verify that it's not settings based.

[rob@vems.co.uk]

Is this your trigger:


In which case I'd not expect particularly good results.  When making your own you're in a world of potential problems, and I'd never expect a sensor at 90 deg to the wheel to work particularly well.
As the wheel is running at half engine speed to get 36-1 to work properly you'd need the pattern repeated twice.
In short, you're going to have endless problems doing things this way.
Making a custom trigger really requires an oscilloscope, you'd make your life a hell of a lot easier getting a known good 36-1 crank trigger and using it.

[sans-pareil]

Yes, this is the current setup, I have a different wheel already. But to be honest, I'm not at the stage where what's on this image is working or not working.

I have access to an oscilloscope, which I will setup as soon as I get beyond the issue of the vems reading 65,534 rpm the moment the sensor gets within 20cm of the rotating engine. Even so, I used a very sensitive volt meter and kan see nice regular variation from the sensor on the trigger wheel in the setup below. My assumption was that if this works:

Then what I'm doing ought to be able to work too, but I'm more than happy to throw everything around, but I feel I have to solve the 65,000 rpm issue first. Here's v2 of the trigger wheel:

v3 is in the making, but I'm trying to decide if missing teeth or missing gaps would be better in this 90 degree setup.

[Bat]

Hi,
I'd say your getting interference if the RPM reads when the sensor is 20cm away from the engine. Are you using factory fitted wiring to the VR sensor? If so fit a new piece of screened cable and try that....
I'd also have to pitch in and say you could run into more problems when you get it running with the sensor that close to the alternator, there will be a massive amount of magnetic fields generated when the alternator excites!
I'd be looking to fit a trigger wheel on the crank pulley or even machined into it if its a damper type pulley.
Why do you say it won't work with a 145 tooth trigger?
I'm running 36 with 4 odd spaced missing teeth and I'm fairly sure the Audi's have 135 teeth...
Cheers,
Gavin

[sans-pareil]

I've made new wiring, and added a 3rd shielding wire which is grounded on the VEMS end.

As far as thwe Alternator, I'm really out of options there. The chain distribution is on the other side of the enging and it has its own cover and it's a wet environment in there. Other than that I have only the cam shaft there. Also, it's not all as bad as it seems on that image, the engine block is between the primary trigger setup and the alternator. The rpm reading happens even when I hold the VR sensor 20cm away from the rotating engine on the non alternator side.

The fellows on VEMS irc said that they thought 145 teeth was even outside of the range of auditrigger, plus the tooth it needs to react is only 0.3mm lower than the surrounding teeth. So far the consensus seems to be that that's not going to work. If someone can think of a bunch of settings or auditrigger hardware which would make the original trigger off the 145-1 x 0.3mm work, I'm all ears.

[Bat]

Hi,
I'm not an expert on this, but surely if the factory ECU worked on this then VEMS should be able too.
Have you tried using the factory setup?
Cheers,
Gavin

[sans-pareil]

Yes, I've tried the factory setup in various configurations, no joy unfortunately. Also, a 0.3mm variation is tiny, which, while increasing my respect for the factory ignition computer, does not seem to be something that VEMS can handle.

[rob@vems.co.uk]

Lets attack this in a different way.
This setup:
http://www.vems.co.uk/VEMSInstalls/VEMSFordRST.zip

Is tried and tested on a number of 36-1 installs.  First things first, try this on your trigger and you should get a nice clean RPM signal at half engine speed.

[sans-pareil]

Excellent, this looks like it will be a big help, and in any case exclude any possibility of my misconfiguring the VEMS. Thanks very much.

[sans-pareil]

Well, the new config definitely helped solve the 65,000rpm problem. I now have a working RPM signal, so thanks for all the help, I will post here to let everyone know of further progress (and undoubtedly problems).

[Bat]

Hi,
Congratualations
Your car looks just like mine at the moment... Vems on top of the wing and wire everywhere!
Cheers,
Gavin

[sans-pareil]

Thanks,

Now to find out if this is a good enough signal for VEMS

[sans-pareil]

We moved the wheel of schockingly bad design closer to the VR sensor, using a 20 feeler. It now gives 1-1.2v at cranking speeds and the signal has cleaned up, the amplitude is now not any bigger than of the surrounding teeth. The VEMS reacts nicely to it.

[sans-pareil]

It's been a while since I had time to continue with my VEMS, but as of today all the wiring is in place, and the OEM system has been completely dismantled. So, no way back.

I've been looking at that setup on the fords with the 36-1 wheels and that gives me a nice rpm signal. However, as soon as I switch over to multi-tooth, I just get gibberish. Not to mention megatune has a habit of crashing, but that's probably unrelated.

I am going to hook up the scope again and see what the signal looks like at the EC36 connector. Last check on the tdc VR sensor (which was in june or july), I had a nice 1-1.2v at cranking speeds and a nice clean rising line with an amplitude well inside the amplitude of the rest of the teeth. I will check if that's still the case tomorrow afternoon. It's just strange to me that a signal which looks all within spec like that doesn't give me a multitooth rpm of any kind...