So I've had my car in the dyno for the first time. I'm running a Microsquirt ECU on a 2.0L Blacktop Zetec, but this is more of a flywheel vs crank sensor compatibility question.
All went well and successfully mapped up to ~3700rpm. Above that, it was losing crank signal sync. Swapped the sensor and connector for a brand new one - this wasn't the issue.
After some discussion with the dyno guy, we suspected the fact I'd installed a flywheel from a 1.8L engine (along with 1.8L clutch and release bearing) meant that the trigger teeth were too far away from the crank sensor and only losing sync at higher engine rpm.
He thought he'd seen this issue before on these engines, where there were many different crank sensor housings (fitted into the bell housing from the inside) depending on engine and application. It possibly needs to be a different part number in order to match the 1.8L flywheel and keep the correct distance between trigger teeth and sensor.
I measured the distance from end of sensor to trigger teeth with a depth gauge: 3.1mm. Dyno guy suspects it should be much nearer to 1.0mm or less.
Can anyone confirm this theory? I know there's a different sensor housing depending on manual or auto gearbox on the original Ford donor vehicle, but not heard of a difference between 1.8 and 2.0L sensor housings before this issue arose.
I have two ways out:
1) If the theory is correct - just fit a new sensor housing of the correct type (don't know what that is yet though)
2) Fit the original 2.0L flywheel (I still have it), but needs me to buy a new clutch and release bearing, so is the more expensive option.
(Both options require gearbox off, so similar amount of effort)
I know I've got a reasonably unique setup and with hindsight should have kept as standard as possible, but hopefully some of you zetec guys have swapped a different flywheel to the standard 2.0L and can shed some light.
Your setup is not unique.
Every one and there dog sticks a 1.8 flywheel on a 2.0 because it's lighter and cheap.
There are only two cps housings. Manual and Auto but I would expect the tip of the sensor to be nearer than 3 +mm. But I can't say ive measured one recently.
Gbs do a nice cnc cps housing for not a lot of money but I don't think it will change much. I have seen plenty of posts over the years with Ms based ecu's having sync problems due to noise.
There is a guy on you tube with a similar setup but a pinto clutch. Similar issues on an Me ecu.
Around 17.30 in.
And previous vids of it on the dyno.
[Edited on 21/9/20 by big_wasa]
The wiring for your cps is it shielded ? Earthed one end only or are you running a twisted helix or just two straight wires ?
Do you have a noise suppression capacitor near the coil ? Ford fitted a 20 micro farrad cap to a lot of early zetec's.
Interesting. Yep, I thought from the research I did at the time that it would be a straight swap from 2.0L to 1.8L flywheel.
I have now ditched the EDIS and running a VAG coil pack direct to MS, so maybe this is another factor in the mix (original EDIS unit maybe did some filtering of the crank PIP/SAW signals?).
Cable is shielded and earthed at one end, but will double check earth continuity etc. this afternoon.
But, I do not have the noise suppression capacitor fitted. I have the original Ford capacitor in my box of bits somewhere. Is it still needed if running VAG coil rather than EDIS? And if so, I assume it's still to be connected between the 12V supply at the coil and earth, just like the EDIS setup?
Another thought: I currently have the MS Noise Filtering set to Off. Could turning this On possibly resolve the issue? Engine is now out the car so I can't easily test it !
Thanks for the help.
As said, there are only 2 housings used, a manual or an automatic version, the automatic version does not align with a manual flywheel trigger wheel
and iirc is 20-30mm further away from the trigger wheel.
I recently changed from a 1.8 flywheel to a titan flywheel, the sensor was approx 1mm from the trigger wheel on both flywheels, i've not had any issues, but i am using ms-1 & edis for spark control.
The vr signal is an ac signal, with voltage & frequency increasing with rpm. If the vr conditioner is not tuned to the sensor output, you will loose signal at the ecu input.
This is not an issue when using edis as the edis unit deals with the vr sensor signal and gives ms a square wave digital signal, but is an issue when ms is directly handling the vr signal.
Tuning vr conditioner guide
Very useful link, thanks!
I have a Microsquirt unit so don't think I have the ability to tune the conditioner variable resistors as with a full Megasquirt board.
However, the 10k shunt resistor may well help. Microsquirt manual also suggests this option.
I've definitely done this the wrong way around (live and learn...), but now that the engine is out, I'll first double check the sensor housing etc. Then start looking at the composite tooth logger signal, noise suppression and shunt resistor etc.
I have always stuck with an edis to handle the ignition as it is specifically designed using automotive quality components and filtering and gives a
good spark and squarewave output to the megasquirt to handle.I know some people have noise problems when using the VR signal direct to the megasquirt
and the ignition output circuitry can also cause problems.
Confirmed that there was nothing wrong with my crank sensor housing / flywheel setup. Gap was around 1mm.
Fixed the issue with a 1 kohm shunt resistor across the sensor (i.e. between VR+ and VR-). obfripper had it spot on with that diyautotune link. The shunt resistor reduced the signal voltage and prevented the high-RPM breakup I was getting.
Revved it to 6000rpm woohooo. Now ready for the power runs at the dyno and then on to the IVA.
Cheers for all the tips guys.