MOTOBRICK.COM
TECHNICAL MOTOBRICK WRENCHING In Remembrance of Inge K. => The Motobrick Workshop => Topic started by: december454 on June 06, 2024, 11:35:32 AM
-
Howdy, there. I've been a lurker here for a good while, but I've now encountered an issue that needs it's own post.
I'm about 9k miles into a road trip on my 1990 K75S, and after some very rough dirt roads in Colorado, my speedometer and odometer failed. Before this, I had no issues with my speedo or odo. The issue is not intermittent; the gauges seem quite dead.
Symptoms:
- Speedo shows no signs of life, stays pegged at 0 at all times.
- Odometer has not moved.
- Trip meter HAS advanced a tenth of a mile over the course of about 1,500 miles.
- Self-Cancelling Signals Act Oddly. While moving at highway speed, they do not self-cancel. With the bike completely stationary on the center stand, they do not self-cancel, but moving the rear wheel just a few degrees causes them to cancel. Running the bike in first gear on the center stand, they self-cancel.
What I've Checked:
- The three pins on the ribbon cable on the back of the speedo are making good contact, checked with multimeter. Wiggling the connector while the wheel was spinning on the center stand had no effect.
- Pins 21, 22, and 23 on the back of the gauge cluster have continuity to the turn signal unit 'I' pin, speed sensor yellow wire, and speed sensor brown wire, respectively.
- The tone ring for the speed sensor is still firmly attached inside the final drive.
- Speed sensor and speedometer show no signs of physical damage.
- No corrosion on any of the speed sensor, turn signals relay, or gauge cluster pins.
Given the odd behavior of the turn signals, I feel that the electronics in the speedo that take in the speed sensor VR signal have failed. But, before splurging on a donor set of gauges, I'm looking for some advice.
Also, I'm currently on the road, so my tools are limited. I have a multimeter and some basic hand tools.
Looking forward to hearing your responses. Take care.
-
check the square two pin connector under the right battery cover that connects the rear road speed sensor to the bike's wiring harness. Reseat it several times to wipe off any corrosion. Double check continuity between pins to ensure it is a solid electrical connection. The signal from the speed sensor is very low level and any corrosion or other electrical resistance will kill the signal to the gauge cluster.
-
check the square two pin connector under the right battery cover that connects the rear road speed sensor to the bike's wiring harness. Reseat it several times to wipe off any corrosion. Double check continuity between pins to ensure it is a solid electrical connection. The signal from the speed sensor is very low level and any corrosion or other electrical resistance will kill the signal to the gauge cluster.
Thanks for the advice Robert; unfortunately, the issue remains.
The speed sensor connector appears free of corrosion, but I reseated and wiggled the plug a few times just to be safe, no change. I did this with the bike in gear, on the center stand, watching the speedo; the needle never moved.
Thanks for the suggestion regardless.
-
Could it be a fuse?
I ask because on my other Beemer I suddenly lost clock, tach, speedo, odo and after lots of disassembly and examination I thought to check the fuses and one was blown. It ran fine but gave me no information.
Not sure if the K is the same as the F in this regard.
-
Power to the gauge cluster is through fuse 1. If it was blown, then most indicators and gear indicator on the gauge would not work either. Neither would brake lights work. PO doesn't mention that but checking the fuse is a no-brainer.
-
Indeed, I checked all the fuses. Everything is good.
Thanks for the suggestion though.
-
Re-reading your first post, am I correct in reading that you've verified that the road speed sensor actually sends a signal to the speedometer board inside the gauge? You wrote:
- The three pins on the ribbon cable on the back of the speedo are making good contact, checked with multimeter. Wiggling the connector while the wheel was spinning on the center stand had no effect.
- Pins 21, 22, and 23 on the back of the gauge cluster have continuity to the turn signal unit 'I' pin, speed sensor yellow wire, and speed sensor brown wire, respectively.
So, that makes me think you've taken a resistance / continuity measurement between Pins 22 (Yellow) and 23 (Brown) on the inside of the gauge. Is that right?
-
Thanks for the reply Robert.
When checking pins 21, 22, and 23 I verified that they had continuity to the speed sensor plug and the turn signal unit.
I did not take any measurements on the speed sensor itself; I cannot confirm that it is outputting the correct signal. Do you know what an expected value would be to verify that the sensor is working?
Thanks.
-
The road speed sensor in the rear drive outputs a sine wave of about 1 -2 volts peak-to-peak. If you have an analog voltmeter or oscilloscope you could see the output. A digital voltmeter may be too slow to respond. When Flyingbrick existed, there was a great page on the electronics of the K-bike. Unfortunately, Flyingbrick is no more but Wayback captured the information:
https://web.archive.org/web/20211204114440/https://www.flyingbrick.de/2021/index.php?elektronik/
-
When Flyingbrick existed, there was a great page on the electronics of the K-bike. Unfortunately, Flyingbrick is no more but Wayback captured the information:
https://web.archive.org/web/20211204114440/https://www.flyingbrick.de/2021/index.php?elektronik/ (https://web.archive.org/web/20211204114440/https://www.flyingbrick.de/2021/index.php?elektronik/)
English translation. (https://www-flyingbrick-de.translate.goog/2021/index.php?elektronik/&_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp)
-
112350 Cheers Laitch
-
Years ago I had a problem with the Speedo not working. I put a soldering iron near the sensor in the final drive (engine off, ignition on) and the Speedo needle moved, to about 60 km/h. We have 240 Volts here, and 50 Hertz.
Just checked, 60 Hertz in USA.
-
Here is a link to a downloadable PDF copy of the Flyingbrick.de article The Electronics of the K Models (https://www.motobrick.com/index.php?topic=15659.0) mentioned in Replies #8 and #9. It is now in the Motobrick Technical Directory Shop Manual along with Brick workshop manuals, Jetronic and Motronic training manuals, schematics and other reference resources.
-
If you get a good reading with a soldering iron, shake things up a little bit. For example if you get a reading but it drops to 0 when you shake the instrument cluster, that would be a pointer.
After a while we get used to the little quirks on our bikes. If you're anything like me, the first time you rode a K75S, you said to yourself, this instrument cluster is shaking to the shit house.
If you don't get a reading, shake things up anyway, to see if it comes good.
-
Thank y'all for the replies. I had some time today to do some more troubleshooting.
I tried the soldering iron trick; unfortunately, that had no effect. The speedo did not move when holding a soldering near the sensor.
I was able to get some readings with my (digital) multimeter from the sensor itself. With the bike idling in second gear, I placed the leads on the two pins from the sensor plug.
Measuring the frequency, I got a signal of ~10 - 13 hz. Revving up the engine caused the frequency to increase accordingly (20 - 25 hz at 2000 RPM, second gear).
Unfortunately, the voltage measurement was inconclusive. It measured at about 0.12 v. Revving the engine did not change this. I'm unsure if my digital meter was at fault or if there's an issue with the sensor.
I'm on the road for a couple more weeks. If nothing else, I'll do some more serious testing once I'm back home. Thank y'all again for the help.
-
FYI: Not all soldering irons generate a magnetic field strong enough to register on the K speedo. I have a few soldering irons and not all of them register on a K speedo.
-
Forgot to mention, I used an old Weller 8100 soldering gun for the test, really basic, AC powered. I'm pretty sure it produces a magnetic field.
-
You never know. I have a couple of old cheapo orange ones that look almost identical. One emits a field, the other does not. Somebody here explained it to me once.
If you a have a laptop you can use the headphone output to calibrate the speedo.
There's a Windows app called Karamba that you can use that some German developed for K bikes.
-
I'm unsure if my digital meter was at fault or if there's an issue with the sensor.
I think your multimeter and your sensor are okay because I repeated your multimeter tests on my K75S and got basically the same results as you did, but with the occasional odd blip outside the range that you had.
Hertz results were about the same as yours at 1,000rpm and at 2,000rpm.
Volt's test about the same, 0.1 V, regardless of revs.
I also tested in AC Volt's and got about 0.1V at 1,000rpm, and 0.18V at 2,000rpm.
I've been trying to believe that the problem is not inside the instrument cluster, but it's getting harder to think that.
It might be worth looking inside the instrument cluster, in the hope of finding something obviously wrong.
-
Remember that you are measuring RMS voltage if you have the meter on AC. And that the meter may not give accurate information at frequency. The only test instrument that will give an accurate measurement is an oscilloscope. Even still, it seems to point to a gauge problem.
-
Thanks Rob, it's a bit over my head but I mainly thought I'd repeat his tests on my brick to see if I got the same results.
-
The only test instrument that will give an accurate measurement is an oscilloscope.
I'm drawing the line against packing an oscilloscope when I travel. :laughing4-giggles:
-
I'm on the road for a couple more weeks.
Are you using a phone or other GPS to keep you apprised of your speed and mileage, or are you traveling by feel? Despite your assiduous multimeter wrangling, I'd still pull that rear sensor and clean the sides and face of it with contact cleaner despite its appearance.
Half of my riding is pounding down municipal gravel roads with varying degrees of rills, ruts and potholes. I'm surprised my speedometer hasn't experienced a disconnection like the one you're describing. Rather than within the cluster, it seems like a plug or wire intermittent fault if the source isn't the sensor itself.
When the pressure of travel is reduced, your chances of discovering it will improve. icon_cheers
-
Thank y'all again for the replies and for repeating my tests. It means a lot.
As for the gauges, I disassembled everything when I first encountered the issue. Inside the gauge cluster, there is no obvious physical damage. The wires going to the speedometer have continuity, nothing's loose, nothing is rattling around, etc.
As for the sensor, the plug is free of corrosion, and I've cleaned the pickup portion that sticks into the final drive.
Thanks again for the help. Please don't feel too obligated to keep working on this. For the time being, I have a GPS speedo on my phone and Spotwalla for mileage tracking.
-
A bit of deoxit here and there might do it.
Maybe keep going downstream with your Volts tests and Hertz tests, for example at both sides of the instrument cluster mutiplug.
I wonder if removing the blue/green wire from the instrument cluster mutiplug (which goes to the flasher unit) could result in the speedo working.
-
Howdy there, wanted to give a quick update on this thread.
I replaced the LM-2904N op amp which amplifies the signal from the speed sensor. No dice, unfortunately. The speedo / trip meter are still inoperable. Also, there was a faint smell of burnt electronics when I cracked open the gauges; I do not recall this when I first worked on the gauges during my road trip.
I'm going to throw in the towel and pick up a used gauge cluster on eBay. If anyone with a bit more electronics prowess wants my problematic gauge cluster, PM me, and I can ship them your way. (within the U.S.)
Take care.
-
Let's say there's a problem with the blue/green wire, that might cause the Speedo to stop working. The fact that the auto cancel feature works intermittently, adds weight to this as a possibility, I think. How likely is it that this is the problem? Dunno, I'll say somewhere between 0% and 100%, but it's such an easy thing to do, it's worth checking before ordering a new cluster.
Let's say there's a short in this wire, when you install another instrument cluster, that might also end up with a burnt electronics smell. There's continuity between the ends of that wire, but that doesn't eliminate the possibility of a short.
I would unplug the flasher unit and isolate the blue/green wire from the instrument cluster to see if the Speedo then works. A quick way to isolate it from the instrument cluster is to fold a piece of paper, longer than it needs to be, and push it into the socket for the green/blue wire pin. Longer than it needs to be, so that it's easy to remove after this test.