No Start After Clutch Job
11-18-2014, 08:26 PM
#166
Instructor
Thread Starter
Divil,
That sounds like a good idea. I have a slightly busy schedule in the next few days, but I will try to get to the DME harness and disconnect pin #21 from the back of the DME and see if I still have 12v at the KLR.
Perry
That sounds like a good idea. I have a slightly busy schedule in the next few days, but I will try to get to the DME harness and disconnect pin #21 from the back of the DME and see if I still have 12v at the KLR.
Perry
11-18-2014, 10:55 PM
#167
Three Wheelin'
From reading the wiring diagrams as best I can (i.e. badly) and reading a few other threads, I believe that DME #21 is also connected to the tachometer via an 8-pin plug in the footwell (DME test plan shows a 4 pin plug but the diagrams for later cars show 8 pins). That plug also has connections for various other dash instruments and something else I can't decipher (Grid reference N7 - looks to be connected to something involving the reversing light/gear shifter.). You could always try disconnecting that too and see if any of those wires could be the source of the 12v.
11-18-2014, 11:17 PM
#168
Instructor
Thread Starter
Divil,
Yes, I have disconnected the T1 plug for the tachometer--no difference. I saw the same thing that you did and thought that there might be a short in the circuit to the tach or a bad ground for the dash instruments, but that has turned out to be a dead end.
If you think of something else, please let me know....
Perry
Yes, I have disconnected the T1 plug for the tachometer--no difference. I saw the same thing that you did and thought that there might be a short in the circuit to the tach or a bad ground for the dash instruments, but that has turned out to be a dead end.
If you think of something else, please let me know....
Perry
11-19-2014, 12:09 AM
#169
Three Wheelin'
We might have to back track a bit here. I just went out and spent a few minutes checking things on my car. At key on, I have around 11v on KLR #9! Now, I do have an occasional no-start issue, so just to be certain I cranked it and it fired right up. So I have to disagree with the premise that having positive voltage there at key on is a problem. I know that is contrary to what specsalot found in his test:
I can't explain that part, unless it was a mistake. Maybe someone else could check their car also, just as a sanity check. I am racking my brains in case I made some stupid mistake checking this, but I made a point of checking pin #2 in the 8-pin connector (tach wire) and I have continuity between that and KLR #9, as expected. I had slightly less voltage there, but it was close. So I'm sure I was checking the right pin.
Do you still have that other DME you tested with? I know you don't want to open your DME because of the warranty. But it might be time to open the other one and start probing inside it. Maybe there is something you have missed regarding the speed/ref sensors. Have you read this: Crank Angle Sensors? If you could see the square wave coming out of that custom IC, that would put the sensors issue to rest completely.
Test 1
Battery connected with Key switch in Run
DME computer connected
KLR computer disconnected
KLR Connector Pin 9 = ~0.043 VDC
Battery connected with Key switch in Run
DME computer connected
KLR computer disconnected
KLR Connector Pin 9 = ~0.043 VDC
Do you still have that other DME you tested with? I know you don't want to open your DME because of the warranty. But it might be time to open the other one and start probing inside it. Maybe there is something you have missed regarding the speed/ref sensors. Have you read this: Crank Angle Sensors? If you could see the square wave coming out of that custom IC, that would put the sensors issue to rest completely.
11-19-2014, 09:51 AM
#170
Three Wheelin'
I had another one of those headaches with pictures. 2 questions:
1. did you verify the polarity of the reference sensor signal? I know you said it started with a positive spike as called for by the test plan, but I couldn't find where it tells you which wire is which. Maybe that's in the wiring diagram somewhere?
2. does anyone know what would happen if it was backwards?
I began thinking, what if the sensor was defective by being wired up backwards...then I remembered you have a replacement wiring harness, so there's a reasonable chance that either you or the maker of the harness got the wires backwards. This assumes that the reason you car didn't start back then was something different which has since been fixed. But I honestly don't know if this could cause the symptom you have. I'll try to check later.
EDIT...didn't get a chance to try swapping my DME pins. But from the wiring diagram, it looks like reference sensor pin #1 goes to DME #25, #2 to DME #26. On my car the grounded one is the one nearest the firewall, so that would be 3. So a quick continuity test should tell you if you have them hooked up correctly. Even so, I'd probably try swapping the DME pins as a last resort in case the wires are reversed in the sensor itself. I know that's very unlikely but I think it's time to accept that whatever the problem turns out to be, it's very unlikely.
1. did you verify the polarity of the reference sensor signal? I know you said it started with a positive spike as called for by the test plan, but I couldn't find where it tells you which wire is which. Maybe that's in the wiring diagram somewhere?
2. does anyone know what would happen if it was backwards?
I began thinking, what if the sensor was defective by being wired up backwards...then I remembered you have a replacement wiring harness, so there's a reasonable chance that either you or the maker of the harness got the wires backwards. This assumes that the reason you car didn't start back then was something different which has since been fixed. But I honestly don't know if this could cause the symptom you have. I'll try to check later.
EDIT...didn't get a chance to try swapping my DME pins. But from the wiring diagram, it looks like reference sensor pin #1 goes to DME #25, #2 to DME #26. On my car the grounded one is the one nearest the firewall, so that would be 3. So a quick continuity test should tell you if you have them hooked up correctly. Even so, I'd probably try swapping the DME pins as a last resort in case the wires are reversed in the sensor itself. I know that's very unlikely but I think it's time to accept that whatever the problem turns out to be, it's very unlikely.
Last edited by divil; 11-19-2014 at 11:08 PM.
11-19-2014, 11:58 PM
#171
Instructor
Thread Starter
Divil,
Thanks for your input.
I have been unable to disconnect the wire that goes to pin #21 on the DME harness, then plug the DME back in and turn on the ignition switch to see if power from any other DME wires is shorting out against this wire. The trouble is that the plastic connector that holds all the wires and female terminals is too fragile and too cramped to allow me to remove the terminal from the plastic housing, pull out the "combs" that lock the terminals in place, pull the wire out of the way, reconnect the top half of the harness connector, and finally reconnect the harness to the DME.
What I did instead was to remove both computers then test for continuity between DME female terminal #21 and ALL the other terminals in both the DME and KLR harnesses. There is NO continuity between DME harness terminal #21 and any other terminals EXCEPT KLR #9 (which is where that wire goes). I performed this test with the key in the run and off positions with no difference.
As to the polarity of the reference sensor test, the manual specifically says to put DME harness terminal #25 to the positive input on the oscilloscope and #26 to the oscilloscope's ground ternimal, then crank the engine. The initial slope of this spike should be positive. This is exactly what I get. If you reverse the polarity of the connections, that is, put #25 to oscilloscope ground and #26 to the oscilloscope's positve terminal, then the initial slope is negative. For both conditions the amplitude is greater than 2 volts.
Finally, I have no idea what voltage (if any) is supposed to be present at KLR pin #9 with the key on. I have no idea how the DME operates the coil circuit other that it converts the analog signals from the referece and speed sensors into the alternating square wave signal via a transistor. I didn't figure this out by myself, I got all this from Josh Cunningham's Rogue Tuning website--very informative. I don't know what to make of the fact that Specs got only a millivoltage but you got nearly 12 volts.
Please, don't hesitate to make suggestions.
Perry
Thanks for your input.
I have been unable to disconnect the wire that goes to pin #21 on the DME harness, then plug the DME back in and turn on the ignition switch to see if power from any other DME wires is shorting out against this wire. The trouble is that the plastic connector that holds all the wires and female terminals is too fragile and too cramped to allow me to remove the terminal from the plastic housing, pull out the "combs" that lock the terminals in place, pull the wire out of the way, reconnect the top half of the harness connector, and finally reconnect the harness to the DME.
What I did instead was to remove both computers then test for continuity between DME female terminal #21 and ALL the other terminals in both the DME and KLR harnesses. There is NO continuity between DME harness terminal #21 and any other terminals EXCEPT KLR #9 (which is where that wire goes). I performed this test with the key in the run and off positions with no difference.
As to the polarity of the reference sensor test, the manual specifically says to put DME harness terminal #25 to the positive input on the oscilloscope and #26 to the oscilloscope's ground ternimal, then crank the engine. The initial slope of this spike should be positive. This is exactly what I get. If you reverse the polarity of the connections, that is, put #25 to oscilloscope ground and #26 to the oscilloscope's positve terminal, then the initial slope is negative. For both conditions the amplitude is greater than 2 volts.
Finally, I have no idea what voltage (if any) is supposed to be present at KLR pin #9 with the key on. I have no idea how the DME operates the coil circuit other that it converts the analog signals from the referece and speed sensors into the alternating square wave signal via a transistor. I didn't figure this out by myself, I got all this from Josh Cunningham's Rogue Tuning website--very informative. I don't know what to make of the fact that Specs got only a millivoltage but you got nearly 12 volts.
Please, don't hesitate to make suggestions.
Perry
11-20-2014, 02:08 AM
#172
Three Wheelin'
Wow ok. What about the other DME?
I did a little reading to try to better understand why the DME would or would not send the ignition signal. I don't know if you have the stomach for this kind of thing, but I came across this interesting topic on the BoschDME Yahoo group.
Now, the speed/reference sensor signals are converted to a square wave signal by a custom IC that there doesn't seem to be much info on. Those digitized signals go to the 8051's interrupt pins. Then the DME decides whether to output anything. But what this guy^ found is that the custom converter will only send the digitized signals if it gets a sequence of pulses from the DME. These pulses start to appear going from the DME to the converter when the speed sensor signal arrives. Another poster stated that from studying the code, he believes that the DME software checks for inconsistencies with the speed signal, and if it detects any, it stops sending these pulses to the converter, so the converter stops sending the speed/ref signals.
What's not clear from the above thread (even if it's accurate) is, when does the DME start sending the pulses again to re-activate that converter. But I think if I were in your shoes, I would now be probing the inside of that spare DME with the scope. I would want to see if the DME is sending those pulses, and if it's receiving the interrupts. If it's not sending out the pulses to the converter, that would suggest that maybe the software is not happy with the speed signal for some reason. It could be some issue that you miss on the scope. Maybe it's too fast to see. Again though, this is assuming the above info is accurate. There are a few people on that list who seem to know exactly how the software works but they don't give away much.
At the very least, hooking up an open DME and probing the inside could help to shed some light on it. You'd at least be able to see if signals were making it part of the way in or out.
I did a little reading to try to better understand why the DME would or would not send the ignition signal. I don't know if you have the stomach for this kind of thing, but I came across this interesting topic on the BoschDME Yahoo group.
Now, the speed/reference sensor signals are converted to a square wave signal by a custom IC that there doesn't seem to be much info on. Those digitized signals go to the 8051's interrupt pins. Then the DME decides whether to output anything. But what this guy^ found is that the custom converter will only send the digitized signals if it gets a sequence of pulses from the DME. These pulses start to appear going from the DME to the converter when the speed sensor signal arrives. Another poster stated that from studying the code, he believes that the DME software checks for inconsistencies with the speed signal, and if it detects any, it stops sending these pulses to the converter, so the converter stops sending the speed/ref signals.
What's not clear from the above thread (even if it's accurate) is, when does the DME start sending the pulses again to re-activate that converter. But I think if I were in your shoes, I would now be probing the inside of that spare DME with the scope. I would want to see if the DME is sending those pulses, and if it's receiving the interrupts. If it's not sending out the pulses to the converter, that would suggest that maybe the software is not happy with the speed signal for some reason. It could be some issue that you miss on the scope. Maybe it's too fast to see. Again though, this is assuming the above info is accurate. There are a few people on that list who seem to know exactly how the software works but they don't give away much.
At the very least, hooking up an open DME and probing the inside could help to shed some light on it. You'd at least be able to see if signals were making it part of the way in or out.
11-20-2014, 05:23 AM
#173
Instructor
Join Date: Oct 2010
Location: Sunny Florida
Posts: 172
Likes: 0
Received 0 Likes
on
0 Posts
We might have to back track a bit here. I just went out and spent a few minutes checking things on my car. At key on, I have around 11v on KLR #9! Now, I do have an occasional no-start issue, so just to be certain I cranked it and it fired right up. So I have to disagree with the premise that having positive voltage there at key on is a problem. I know that is contrary to what specsalot found in his test:
I can't explain that part, unless it was a mistake. Maybe someone else could check their car also, just as a sanity check. I am racking my brains in case I made some stupid mistake checking this, but I made a point of checking pin #2 in the 8-pin connector (tach wire) and I have continuity between that and KLR #9, as expected. I had slightly less voltage there, but it was close. So I'm sure I was checking the right pin.
Do you still have that other DME you tested with? I know you don't want to open your DME because of the warranty. But it might be time to open the other one and start probing inside it. Maybe there is something you have missed regarding the speed/ref sensors. Have you read this: Crank Angle Sensors? If you could see the square wave coming out of that custom IC, that would put the sensors issue to rest completely.
I can't explain that part, unless it was a mistake. Maybe someone else could check their car also, just as a sanity check. I am racking my brains in case I made some stupid mistake checking this, but I made a point of checking pin #2 in the 8-pin connector (tach wire) and I have continuity between that and KLR #9, as expected. I had slightly less voltage there, but it was close. So I'm sure I was checking the right pin.
Do you still have that other DME you tested with? I know you don't want to open your DME because of the warranty. But it might be time to open the other one and start probing inside it. Maybe there is something you have missed regarding the speed/ref sensors. Have you read this: Crank Angle Sensors? If you could see the square wave coming out of that custom IC, that would put the sensors issue to rest completely.
11-20-2014, 02:28 PM
#174
Three Wheelin'
I am fairly certain this is correct - tested it more than once, but will also admit that standing on my head in the footwell, trying to probe disconnected computer plugs with standard fluke test leads is a challenge. If my results are inconsistant with other owners - I will gladly label Test 1 above a mistake. One thing I know is my car starts first time / every time.
11-20-2014, 08:23 PM
#175
Three Wheelin'
Just checked again. 100% certain there should be +v on KLR #9 with the key on. I had 11.45. The DME pulls it down to create the signal. I could see that on my scope but couldn't capture a good picture of it because the frequency is so low.
11-21-2014, 12:49 AM
#176
Instructor
Thread Starter
Specs, Divil,
Thanks for all your help. I will be on vacation until December 1st and will not be near a computer. Please continue to post any thoughts, tests, suggestions that you think might help--I will get back to this thread when I return.
Perry
Thanks for all your help. I will be on vacation until December 1st and will not be near a computer. Please continue to post any thoughts, tests, suggestions that you think might help--I will get back to this thread when I return.
Perry
12-07-2014, 09:38 PM
#177
Instructor
Thread Starter
For anyone still following...
As per the most recent suggestion I have loosened the reference/speed sensor bracket and adjusted it to its maximum (farthest away from the ring gear and flywheel) position and performed the cranking voltage test at DME harness pins #8 and #27 for the speed sensor as well as #25 and #26 for the reference sensor. At this maximum position the speed sensor produces about 2V peak to peak during cranking, the reference sensor produces about the same.
I hooked up the DME and still had a no start. I then slowly reduced the clearance bringing the peak to peak voltage up to 4V then 6V then about 10V. At each new, increased voltage I would plug in the DME and test for spark. Nothing.
Again, I am open to suggestions and more testing.
Divil, I have sent you a PM.
Perry
As per the most recent suggestion I have loosened the reference/speed sensor bracket and adjusted it to its maximum (farthest away from the ring gear and flywheel) position and performed the cranking voltage test at DME harness pins #8 and #27 for the speed sensor as well as #25 and #26 for the reference sensor. At this maximum position the speed sensor produces about 2V peak to peak during cranking, the reference sensor produces about the same.
I hooked up the DME and still had a no start. I then slowly reduced the clearance bringing the peak to peak voltage up to 4V then 6V then about 10V. At each new, increased voltage I would plug in the DME and test for spark. Nothing.
Again, I am open to suggestions and more testing.
Divil, I have sent you a PM.
Perry
12-07-2014, 09:45 PM
#178
Instructor
Thread Starter
Oh yeah, I almost forgot: As I would turn the ignition key from the off to the run positions, with the DME computer installed and KLR harmess pins #9 and #16 jumped, the coil would sometimes fire--I have the coil HT wire hooked to a grounded and exposed sparkplug. I don't know what this means for the overall picture, but it does prove that the ignition system can function.
Perry
Perry
01-17-2015, 05:56 PM
#180
Three Wheelin'
The DMEs and KLR worked in my car!
Just to bring everyone else up to speed, I received the original, Specialized/ECU Doctors repaired DME, KLR, and the eBay DME from Perry during the week, to test in my car. I just now got around to testing them (sorry for the delay Perry, it was a busy week).
My car started just fine with all 3 units. Here's the sequence of tests I did:
1. Went for a drive, warmed up my car and made sure everything was ok
2. my DME, KLR removed, jumper installed - this resulted in no start at first. There was fuel and tach bounce, but no start. I reinstalled my KLR and the car started just fine. I decided to try once more - this time, the car made an odd cranking noise, as if it couldn't turn over, but then suddenly fired up and ran ok. So I assume I just didn't get a good connection with the jumper before.
3. eBay DME, KLR removed, jumper installed - fired right up, no issues.
4. Specialized repaired DME, KLR removed, jumper installed - fired right up, no issues.
5. Specialized repaired DME, Specialized repaired KLR - fired right up, no issues.
I should qualify "no issues" by saying that the odd cranking noise I mentioned in step 2 happened occasionally, but never seemed make a difference.
I took video footage of each test, taking care to show which units were attached, so I'll get that uploaded as soon as I can.
I think this is good news...I realize it means there must be something wrong with your car, but this result is more conclusive than a no-start would have been, since there would always be some remote possibility that it could be due to me doing something wrong, or my car picking that moment to misbehave for some unrelated reason. The initial trouble I had even with my DME proves my point. But since they all worked fine, they must be good.
Video to follow as soon as possible
My car started just fine with all 3 units. Here's the sequence of tests I did:
1. Went for a drive, warmed up my car and made sure everything was ok
2. my DME, KLR removed, jumper installed - this resulted in no start at first. There was fuel and tach bounce, but no start. I reinstalled my KLR and the car started just fine. I decided to try once more - this time, the car made an odd cranking noise, as if it couldn't turn over, but then suddenly fired up and ran ok. So I assume I just didn't get a good connection with the jumper before.
3. eBay DME, KLR removed, jumper installed - fired right up, no issues.
4. Specialized repaired DME, KLR removed, jumper installed - fired right up, no issues.
5. Specialized repaired DME, Specialized repaired KLR - fired right up, no issues.
I should qualify "no issues" by saying that the odd cranking noise I mentioned in step 2 happened occasionally, but never seemed make a difference.
I took video footage of each test, taking care to show which units were attached, so I'll get that uploaded as soon as I can.
I think this is good news...I realize it means there must be something wrong with your car, but this result is more conclusive than a no-start would have been, since there would always be some remote possibility that it could be due to me doing something wrong, or my car picking that moment to misbehave for some unrelated reason. The initial trouble I had even with my DME proves my point. But since they all worked fine, they must be good.
Video to follow as soon as possible