MS2 V3.0 lost tacho signal. Help needed urgently

[brad91]

http://www.msextra.com/forums/viewtopic ... 01&t=34060

have read this as well.

I have pulled the cpu back out and checked voltages and they are seriously different to the values given in the link. If need be I can record them all?

[DaveEFI]

What happens if you go to TS communications, settings and detect, with your MS powered up and plugged in?

[brad91]

occasionally it will come up with connected, other times it just comes up with failed.It will drop in and out. Last night I changed from RS232 to usb in the top box with RS232 in the new box that appears under the usb selection and it found 2 different megasquirts???

Im hoping now that its just a fried cpu.

A mate of mine with a PNP said something about did I hook 12v to where 5v is supposed to be hooked but I dont think there should be 5v anywhere? I am running the constant 12v from the battery and the 12v from the 'ON' wire out of the back of the key to supply the switched 12v power. Or is that supposed to be 5v?

The only other thing that could have gone wrong would be when I was cranking the car and the earth had fallen off so I grabbed the earth from the relay block and help against the steering column while still cranking which caused a couple sparks until it grounded. Dad said that could have created enough of a spike to fry the cpu chip? Is there anyway of testing if the cpu is good without another board?

[Matt Cramer]

occasionally it will come up with connected, other times it just comes up with failed.It will drop in and out. Last night I changed from RS232 to usb in the top box with RS232 in the new box that appears under the usb selection and it found 2 different megasquirts???

This one can be confusing. What actually happened was it found two options for ways of communicating with the same MegaSquirt.

A mate of mine with a PNP said something about did I hook 12v to where 5v is supposed to be hooked but I dont think there should be 5v anywhere? I am running the constant 12v from the battery and the 12v from the 'ON' wire out of the back of the key to supply the switched 12v power. Or is that supposed to be 5v?

The MegaSquirt is supplied with only 12V, but includes an internal 5V source. Shorting that to 12 volts can cause severe damage.

The only other thing that could have gone wrong would be when I was cranking the car and the earth had fallen off so I grabbed the earth from the relay block and help against the steering column while still cranking which caused a couple sparks until it grounded. Dad said that could have created enough of a spike to fry the cpu chip? Is there anyway of testing if the cpu is good without another board?

A flakey ground can cause corrupted firmware or hardware damage. It is possible the serial chip is damaged instead of the processor.

[brad91]

hmm so should I be assuming that the board is fried? What would be my best point of attack from here?

Just go order a new complete ecu, order a new cpu chip?

Just want it done and running asap as I need to move the car off my hoist to get the race car on and im not keen on trying to push this thing with a welded diff and the 10 point turn required to get onto my hoist haha.

[Matt Cramer]

Assume nothing, test everything. First confirm you have 5 volt power on the board, then try a serial loopback test.

[brad91]

serial loopback test passes on all 3 tests.

Where should I have 5v on the board?

Im going through doing a continuity test currently as advised by a mate who works at an electrical store. I found a link outlining the tacho circuit. I lost continuity over the first resistor however all resistors in the circuit seem to test out fine so I will look at getting a cheap oscilloscope to run some further tests.

Ive ordered your book as well Matt. Hopefully that will be here soon so I can have a read and hopefully it will help me out some more.

[Matt Cramer]

You should have 5 volts at the following points: S5, the two +5V holes in the proto area, the right (non-banded) end of D9, the right leg of U5, the left (banded) end of D19, and pins 1, 20, and 31 of the CPU.

[brad91]

awesome thanks! will check that tonight. My oscilloscope turned up yesterday. Had a play last night trying to sort it out but have a feeling being a cheap 12v handheld ebay one it may have a faulty probe or something as it is constantly showing a pulse and even when hooked to the 50hz square wave test thats built in nothing changes :S Have emailed them to find out. Hopefully I dont have to wait another week for another one!

[brad91]

ok so I checked the 5v. I have 5v at all those points. Only one I couldn't check is S5? Which pin is that? Did you mean JS5 in the cpu?

I got my oscilloscope setup right and have a nice wave length with setting it to 2v and 2ms at all pins on the tacho circuit up until U7. Do you know what I should be seeing on my scope at R45 so I can set a bench mark?

Q22, what should be coming into this and what should be going out?

U7 pin 8 has 4.9/5v, pin 2 has the incoming signal fine, pin 3 and pin 1 show a voltage but my scope only shows it as a flat line??? Would this indicate a problem with this component or is it supposed to flatten out completely/more than my basic scope can read?

I have tried changing the hz all the way from 10hz to 10mhz and it stays flat and also played with the time and voltage settings to see if I can zoom in and find a wave.

If its only that piece that should be cheap enough to get over here. If its not then im assuming that the board should be all good and I can go on to assume its the cpu chip??

Thanks
Brad

[DaveEFI]

Even with a 60-1 wheel, you're only looking at frequencies of about 6000Hz at 6000 rpm. With a 36-1, about half that.

Using my JimStim, I see a near perfect square wave of about 5v peak to peak at VRin, and pretty well the same at VRout

Pin 2 of U7 shows the same approx 5 volt signal
Pin3 shows the same square wave - but 0.5v. And about 1v DC. But note the pots are simply set to the default settings given in the manual.

Have you checked you have 5v DC between pins 4&8 of U7?

If so, and you've checked the pots are on the default settings, I'd be inclined to replace U7. It is fitted the correct way round?

Don't even consider the CPU until you see the same(ish) signal at VRin and VRout.

[brad91]

ok so i should have written this up last night before i went to bed as ive forgotten exact figures now. I realised that my power adaptor was 12v AC!!! so I wasnt getting a square wave. got a 12vDC adaptor and now I get a perfect square wave out of the stim. I also replaced my tachoC pot as it was some random pot the previous owner had fitted and wasnt on great so its night and neat again now.

Pin 2 of U7 shows the same approx 5 volt signal
for the life of me I cant remember what this was showing last night but i dont think it was 5v. also this was showing a saw wave? not 100% sure how as it has been a square wave up until there? haha

Pin3 shows the same square wave - but 0.5v. And about 1v DC. But note the pots are simply set to the default settings given in the manual.
i think mine was showing up to about 1.3v? but still showing a square wave

Have you checked you have 5v DC between pins 4&8 of U7?
yep 5v between those pins

If so, and you've checked the pots are on the default settings, I'd be inclined to replace U7. It is fitted the correct way round?
i checked the pots with the scope hooked up and wound them all the way counterclockwise to get it back to 0mv reading


Don't even consider the CPU until you see the same(ish) signal at VRin and VRout.
i see the 5v square wave at vr in but dont think i could find it at vrout

also I have noticed that pin 1 of u7 which by schematics should be going into the ecu. This pin is showing nothing. by the time I get down small enough on the scope to measure it im just getting noise so im going to go with there is nothing there and im assuming this should be a good readable 5v square wave?

[DaveEFI]

Pin 1 on U7 is directly connected to VR out. I use a bare wire loop on the component side to connect VR out to Tsel which makes a convenient way of attaching a scope probe. I do the same with VRin.

With both pots set fully anti-clockwise to start with, now turn R56 6 turns clockwise. R56 is the pot furthest from U7, at the edge of the PCB.

[brad91]

ok so quick question... if VRin and VRout should be the exact same signal what would stop my bypassing all of the circuit and jumpering VRin straight to VRout?

my performance fuel injection systems book just turned up and had a quick flick through at work. Think there will be some late nights reading ahead but wow there looks like there was a lot of time and research put into this book!!!

Ill order a new u7 chip as no one in australia seems to stock it but I can get them from ebay US pretty cheap. But if its coming in as a square wave and then going into the VRout as a square wave I dont know why I cant jumper it? im sure there is probably a reason like the capacitors in the line holding a more steady signal or something but thought id ask ask even if I could do that to test my cpu works that would be awesome!

[DaveEFI]

My figures are with a JimStim. A true VR sensor will give different results at the input. For a start, it's a sine wave. The VR input's purpose it to take a varying level and frequency approximate sine wave and convert to a suitable pulse for the processor.

[brad91]

ah yep. knew there would be a reason haha.

Ok so should I start by replacing U7? and hope that it is that?

By the time it is getting to U7 it is a very small signal however from my research U7 is an amplifier to then boost the signal obviously once it has been converted to a usable pulse?

So VRinput circuit will convert from a sine wave to a saw wave (atleast I find a saw wave at the input to the U7 haha) and then back out to a pulse at the VRout which then goes into the processor? And then from the processor obviously through a different circuit back to the output to edis?

Think im starting to get it a bit more now haha

[DaveEFI]

If the signal at Pin 2 is very different level wise from the signal at the input, that is the first thing to address.

[brad91]

ok my new LM2904N chip turned up. Got 2 so I have a spare because it was $2.50 ea and $9 courier charge so thought I may aswell double up incase it is a problem and happens again.

Will change out tonight and see what difference it makes.

[brad91]

ok so i have soldered in the new U7 chip. I am seeing 5v at VR in and 5v at pin 2 (one square wave one sine/saw wave cant really zoom in enough to see on my scope). Seeing nothing at VRout. Im about to give up and bin this sytem haha. Its doing my head in and I just cant work it out and theres no one near me that can inspect it. Considering going to megajolt and twin dcoe carbs but turbo is going to suck with carbies.

[DaveEFI]

Can you do a pic of the board showing the U7 area? Don't give up.