Transmission Gear Position Switch Troubleshooting ~ author brian curry

[Scott_]

Transmission Gear Position Switch Troubleshooting ~ author brian curry

[John Lang]

My GPI display blanks out periodically, especially on a warm day. The gear numbers and the green neutral light either all work or do not. The display usually comes back on after the bike has been parked in the shade or when cooling breezes have worked their magic. I have tested the combinations of the connector under the seat (as instructed by Brian Curry on this site) and the mechanical unit appears to be OK. I cleaned the under-seat connector and the connector pins on the rear of the instrument cluster. Blanking still occurs. I suppose the next step is to clean the 6 terminals on the circuit board, reluctant tho' I am to open it up. Anyone have other ideas?

[rbm]

John, Those symptoms indicate corrosion somewhere between the GPI switch and the GPI board in the instrument cluster is causing ground to disappear.  The display blanks and the neutral light goes out because all three inputs loose their ground connection and the indicated gear jumps from "0" to "7".  My guess is the connector on the harness into which the GPI switch plugs.  Do the symptoms appear suddenly (like neutral is working normally and then suddenly the display blanks and neutral goes out)?

[The Mighty Gryphon]

If everything else on the bike, especially in the instrument cluster, works properly, you may have a bad gear indicator circuit board.  The possible problem on that board may be a cold solder joint, probably on one of the pins that feeds power or ground to the board.
Confirm everything else is okay before going into the cluster.  It's a fiddly job to get to the circuit board, but if you take your time you can do it successfully.  It is possible to test that board while it is out of the cluster.

[John Lang]

Thank you, gents. Sensible advice.


Yes, Robert, it is sudden and, now you mention it, is possibly related to the connector under the seat. That connector needs manhandling to open and could have been damaged by efforts (mine!) to clean it. I will repair/change that connector before tampering with the instrument cluster. Princess Auto (Canada's Harbor Freight?) should have a 4-wire connector that will do the job.

[rbm]

Yes, the connector retaining brackets do make it hard to disengage the connector.  I would hesitate to replace the connector.  Rather, see if you can close up the female contacts on the harness side of the connector by inserting a very thin jeweller's screwdriver between the connector housing and the contact.  Just a little is needed, and only on the ground connector (brown wire).

[John Lang]

Thanks Robert. The gear indicator was working when I parked the bike an hour ago but is not working now. It is still rather warm where that connector lives. As you suggested, I inserted a tiny screwdriver between the brown wire female side and its housing in order to re-shape it. I think I managed to do so. Not fixed. I cannot cause the fault to appear/disappear when I squeeze the closed connector or tug gently at its wires so perhaps the connector is not at fault after all. I am confident the indicator will be working in the morning or even after a few minutes ride right now. This slight annoyance has occurred over several years but is getting worse.


The wiring diagram shows the brown GPI wire ending in a group of brown wires at a central grounding point that appears to be separate from the one on the center frame under the tank. Do you know where this second major grounding point is?. I would like to ensure the GPI ground problem is not located there.

[rbm]

I believe that cluster of grounding wires is embedded within the harness.  If the ground connections under the tank are secure, then the ground at the TGPI connector should also be intact.  Confirm this by measuring the continuity between the brown wire on the harness and that grounding point under the tank.  It should be solid and wiggling wires should not alter that.

Barring a bad ground at the connector, it is possible that the connector pins to the GPI board inside the OEM instrument cluster are corroded and going intermittent.  Specifically, it will either be power or ground pins that are being troublesome.  You might have to open the cluster and correct that problem by cleaning the pins and sockets on the GPI board.  I doubt very much that it is problems with the main connector on the instrument cluster or the electronics on the actual GPI board itself going bad.

The reference schematic for the GPI board is here:  http://www.motobrick.com/index.php/topic,10566.0.html

[John Lang]

Thank you, Robert. I will check the ground continuity from the connector and then go after the board, if necessary.


This is definitely temp-related as it corrected itself overnight. Will revert.

[John Lang]

Against my hopes and parayers, sporadic blanking of the GPI display has failed to correct itself. Following RBM's advice, I tweaked the ground receptacle on the harness side of the 4-pin connector. No difference. Readings taken from the switch side of the connector while in 3rd and 5th gear indicate no problem there. I removed the instrument pod and cleaned the multi-pin connectors, then I partially disassembled the pod to check on the pins of the GPI board. During my visual inspection, they gleamed at me with shiny nickel-chrome cleanliness so I decided I could only make things worse by getting at them and I retreated. Now the GPI display does not even flicker -- it is off all the time. I doubt I broke anything. I only sprayed Deoxit into the female side of the main plugs.

I looked again at the 4-pin connector. On the harness side (female), with the ignition switch on, I get continuity and 11.45 volts between the brown ground wire and any of the other three sockets. However, shorting them jointly and severally with the ground has no result on the display.

Do I need to overcome trepidation and get at the shiny pins on the GPI board? Is it possible I dropped my chewing gum into the female side of the main connectors? Any ideas?

[rbm]

Since you are reading ~12V between any of the Yellow wires and Ground on the harness side, my guess is that there is continuity from the connector right to the TGPI board inside the cluster.  Also, since grounding any of the Yellow wires on the harness side does not cause the display to indicate a number, leads me to guess that you have a circuit problem on the TGPI board itself.  My best guess given the symptoms (i.e.  you used to have a flickering display and now you have nothing) is that either the LCD is broken, the connections between the LCD display and TGPI PCB are poor or corroded, or one of the logic chips on the card has failed.  Cleaning the connector will not bring you joy.  I would venture that you have to replace the TGPI board.  Send me a PM please.