Porsche 911 VSS not working

[djb25]

Hi everyone,

I'm trying to get the a VSS input into my ms3/ms3x.

The sensor is on the transmission and is triggered by a wheel with eight magnets mounted in it. I guess it is a reed switch, but I'm not positive on that.

I've tried using all of the ground signal inputs - datalog in, table switch in, and launch in. I've also tried the nitrous in 12v. No signal from any of them. I've used the digital input in the tuner studio settings under advanced engine. I've tried wheel, driveline, and pulse settings. Nothing on any of them. I'm really at a loss here.

What am I missing? I must be overlooking something but I don't know what.

[djb25]

Anyone have any thoughts or suggestions on this?

Perhaps there is a jumper or something that I have to configure on the megasquirt itself? I'm running an MS3 with MS3x from DIYAUTOTUNE. I'm also using DIYAUTOTUNE wiring harnesses for both the MS3 and the MS3X expansion.

I know there are people that have gotten this to work with the factory sensor. The car is a 1977 911S with the original 915 transmission.

Here's a screenshot of how I currently have megasquirt configured for the vehicle speed sensor (VSS):

[ducman82]

here is where i am.... same boat as you

http://www.msextra.com/forums/viewtopic ... 31&t=58177

[Matt Cramer]

Are you able to get an oscilloscope on the sensor to see what waveform it puts out?

[ducman82]

i am in the process of trying to find a scope.... one of the science geeks i know must have one.

[djb25]

Are you able to get an oscilloscope on the sensor to see what waveform it puts out?

I don't have an oscilloscope, but if I have to get one, I will.

Here's where I am (copied from my thread on Pelican Parts):

I checked the signal coming into the ms3x. I'm now using the "tableswitch" input - which is Pin 28 on the MS3X. I was trying all of the inputs the other day, and that was the last input I messed with, so I just left it. According to the manual, any of the ground switching inputs should work (datalog, tableswitch, and launch).

Using my multimeter on the diode/continuity setting, I was able to get the signal at the ms3x plug on Pin 28. I then checked the ms3x board; I was also able to find the signal there. I tested the signal by connecting the black lead to the sensor plug. I then touched touched the red lead to Pin 28 on the MS3x connector. I then spun the rear tire, and the meter would beep as the magnet wheel passed by the sensor. I could also get the signal at the ground connections on the main megasquirt connector (as the sensor passed by the magnets on the wheel). It certainly seems like it is a reed switch, but I'm no electrical engineer.

Anyway, I then took off the top of the megasquirt case and checked the signal at the actual pin on the MS3X board. Same result, beeping as the magnet passed by the sensor. So the signal is definitely reaching the MS3X board.

So... I'm all out of ideas now. Is there some way I can check the MS3X board?

Is it possible something is wrong with the MS3X board? The USB connection on the MS3X has never worked, I've always had to connect to the MS using the serial port.

However, the MS3X is feeding the signal to my tachometer, and that works perfectly. That's really the only thing the MS3X is doing so far, the rest of the connections go to the MS3.

I was supposed to send the MS3X board back into DIYAUTOTUNE so they could troubleshoot the USB problem, but I'm in PA and my time to play with my car is limited. We only have a few days of nice weather up here! Anyway, I was planning on sending it in this winter, after I pull the engine to set up the ignition controls.

Is there some jumper or something I need to change on the MS3X?

[ducman82]

its not an issue with jumpers, its the "type" of signal that the processors at the inputs are made to read. i thought the signal would be "just wire it " to an input... like you, but not the case. i was under the impression it was a VR type sensor, (like a cam sensor), so i build the ZEAL daughterboard and used that as a signal changer... and it "works" but just barley. SO, it looks more like the speed sensor is a Reed switch, so another signal conditioner is needed... I'm working on trying to figure that out.

[grom_e30]

were you power cycling the ecu after making changes to the settings? you could enable port settings in the project properties then you could assign a dash indicator to the input you are using to see if the input state is changing eg if it is dropping from high to low when you manually ground the input wire and also when the speed sensor does when you spin a wheel.

edit: if you enable port_status in project properties. http://www.msextra.com/doc/pdf/html/Meg ... 4-315.html from that page onward covers the indicators.

[Dennis930]

Damion,

There is something wrong. This should work. It works fine on the my 930 transmission. Same set up. It is a reed switch. Make sure the wire that you connected in the tunnel checks out by doing this test: connect the test lead of your meter to pin 28 (MS3X connector) and connect the other lead to vehicle ground. Spin the rear tire to verify an on/off signal to ground. Make sure that the meter reads close to 0 ohms (1 - 4 ohms) and not a higher resistance. If this fails, you need to use the other wire going to the speedometer in the tunnel.

[djb25]

Yep, I was power cycling the box each time I made a change to the settings. Tuner studio actually splashes a big notice across the dashboard after you burn the changes.

Dennis, I'm going to try your test next. I was using the meter in the diode/continuity mode so I could listen for the buzzer. I did get the signal on pin one of the large ribbon connector (JP1, I think) once I powered up the unit.

I'm also going to try get a dash indicator working. Thanks, Grom. I have wondered if I'm spinning the wheel fast enough. I do see the factory speedo jumping as I spin the wheel, so the sender is sending something out...

I also bit the bullet and ordered an oscilloscope, which should be here tomorrow.

[ducman82]

SUCESS!!!!!!!!!

built a circuit, made it work! and it works great.

[djb25]

Ok, what exactly did you do???

It is a bit odd that some people have gotten this to work without any additional hardware.

[ducman82]

not sure. i was having signal issues. lots of interference. this sucker smoothed it all out.

[grom_e30]

I'm also going to try get a dash indicator working. Thanks, Grom. I have wondered if I'm spinning the wheel fast enough. I do see the factory speedo jumping as I spin the wheel, so the sender is sending something out...

with the port_status turned on you can select a dash indicator to the input you are using (ptt6 for datalog pin) then the indicator will light up if the pin has voltage on it and turn off if you ground it so at least you can test if the ecu is working as it should. and should see the indicator turn on and off as you spin the wheel.

[djb25]

Ok, I got my oscilloscope and I tested the sensor. I have no idea what I'm doing, but I was able to get the above waveform using the automatic settings.

Does this tell anyone anything?

This was done with the car running. It was in stands with the rear wheels off the ground. Speedo was indicating around 20 mph.

[Dennis930]

Damion,

It looks like your reed switch is working properly. The MS3X is looking for a ground to trigger the circuit on the MS3X board for Datalog In, Launch IN, and Table Sw inputs. Did you check your input wire(pin 28) to ground as I posted earlier? Every time the reed switch "makes", MS3X needs to see a ground to make the VSS input work.

[djb25]

So the photo shows the appropriate waveform for a reed switch?

I'm going to check the grounding at pin 28 next.

I probably will also check the signal at connection A on the speedo, which I believe is the signal output to the cruise control (which I don't have).

[djb25]

Good news! The VSS input is now working!

Unfortunately, I have no idea why it suddenly works. The only change I made was to extend one of the signal wires so I could get a better connection for the oscilloscope.

I hooked up the scope, started the car, and popped into first gear (car is still on stands). The speedo was reading around 20 mph or so, but I couldn't get anything on the scope. I glanced at my laptop, and saw the tuner studio speedo reading 20 mph! I revved the car and the speed increased!

Even the gear indicator is working - I put the car in second and it showed second gear on the display.

No idea why it is suddenly working. I've been power cycling the box everytime I made a change. I checked the resistance between the wiring to the megasquirt and the sensor, etc.

[ducman82]

hmmm

did you change which wire from the sensor you were getting the signal from? maybe a bad connection that got fixed when you added the wire?

thats fantastic that you got it to work without a conditioner circuit. makes me wonder if i was using the wrong input wire (diff type of signal read) or the wrong wire from the sender.

[djb25]

I'm fairly certain that I didn't switch the wires today. I suppose it is possible, but I don't think so. The wires coming from the sensor are brown and green (that is, one is brown and one is green). The brown wire is always grounded, the green one only part of the time. I figured that out early on, but I tried both wires anyway.

One weird thing. I tried the test Dennis suggested, and the results were weird. It almost seemed like the magnets on the wheel opened and closed the switch. The circuit would be open, I would spin the wheel, and then it would close. It would stay closed for 1/8th of a rotation, and then it would close. Another 1/8th and it would close again. I marked the wheel, a and it was perfectly spaced out. The circuit closed in four spots and opened in four spots. Shouldn't the circuit close eight times?

Seems like there are three options:
1) the magnet wheel doesn't rotate at the same speed as the rear wheels,
2) half the magnets close the circuit, and the other half open the circuit, or
3) I'm doing something wrong.