Triumphrat

cave1 · 09-21-2009, 08:27 PM

I've been riding my other bike in the wet the last couple of weeks (has a fender and shaft drive), and just went to start my 1997 T509 today. For the first time in three years, no joy. The fuel pump primes, the starter spins, but that is it. This happens on the kickstand or off, in neutral with the clutch in or in gear with the clutch in. If I pop the clutch back out when I'm trying to start in neutral, the starter stops spinning, which seems to indicate the clutch switch is working as it should.

What is bugging me is that I think I ran into this symptom before (pump primes, starter spins, but no start) by doing something out of sequence, but I can't remember what. It was something dumb like trying to start the bike with the bike in gear, the kickstand down, and the clutch engaged. Can anyone think of a combination of safety switch interactions on the T509 that would generate this symptom (pump priming, starter spinning, but no starting)?

I thought I would try to rule this out as a preliminary step. My next step will be to charge the battery fully with a trickle charger, and replace the infamous grounding block (if I can find the damned thing), something I learned about here on this forum and have been putting off. At that point, if no start, I thought I'd pull the plugs and look for spark and work backwards from there.

Thanks all.

DEcosse · 09-21-2009, 08:42 PM

http://www.triumphrat.net/speed-trip...ml#post1268991

If it hasn't been done, do it!!!

If you have bike in neutral it does not care if side stand up or down.
So as long as the light is on, you have no interlock.

Only thing the clutch switch interlocks is the starter motor itself.

cave1 · 09-25-2009, 10:28 PM

So I have a ground block update on order, and I have charged the battery. I finally got time today to dig into the bike. I pulled the tank, then the airbox, then the #2 plug, and no big surprise, no spark at all. The fuel pump was whining, but I checked current to the #2 coil anyway, all good. Looking at the wiring diagram for the bike, the problem had to be either the wiring to the coils or that the ECU was not sending a triggering signal (for whatever reason). I went for the obvious possibility first, and started looking for things that might disturb the ECU enough so that it would not trigger the coils. Several possibilities in, I found the two wire connection to the fan was nearly disconnected. I reconnected it, and got a nice blue-white spark at the plug. Note to self, disconnecting the fan might be a good move in a really bad neighborhood. Too bad I didn't run the "test fan" diagnostic on Tuneboy before getting my hands dirty; it would have saved me some digging...

Arenales · 10-05-2009, 12:31 AM

Hi cave1,

you've been replying to a different post of mine and I've been searching for different solutions to my issue. Came upon your post.

you mention that "I went for the obvious possibility first, and started looking for things that might disturb the ECU enough so that it would not trigger the coils."

what are some other obvious things aside from the cooling fan connection?

cave1 · 10-05-2009, 02:09 AM

Hi again Arenales,

From my troubleshooting, I knew that no trigger signal from the ECU was getting to the coils. The probem could thus have been in the ECU, the wiring from the ECU to the coils, or in some connection to the ECU from which a signal might cause the ECU to cut out spark: fan, crank sensor, in gear sensor, kickstand down sensor, etc. I regarded sudden failure of the ECU or the wiring to the coils as unlikely, so I went for what seemed to me to be the "obvious" choice.

I had to make an educated guess about what the relevant sensors might be, and I have a wiring schematic for the T509 in my shop manual, so I just started hunting down likely connections to the ECU and verifying that they were sending the appropriate signal for the bike to start. The fan connector wasn't, and I hit it fairly early in the process.

But note that I started the whole process by pulling the tank and the airbox and looking for spark. Finding no spark, I checked power to the coil primary windings (shared with the fuel pump) and found it present. This isolated my problem to an ECU input or the ECU or the ECU trigger signal to the coils. I credit this divide and conquor process with giving me a solution. Playing hunches rarely has.

If you are really not getting power to the purple/white on your fuel pump connector for the first three seconds after your key is turned, your problem is not the ECU cutting out spark because of some incorrect input. It is something else.

DEcosse · 10-05-2009, 02:26 AM

Quote:

Originally Posted by cave1

.... Several possibilities in, I found the two wire connection to the fan was nearly disconnected. I reconnected it, and got a nice blue-white spark at the plug. ...

I don't see any connection (no pun intended! - but literally!) between the fan and the ability to make spark - the fan is isolated from the ECM by a relay which switches power to the fan.
I'm struggling to see anything other than happy coincidence there - unless for some reason connecting the plug to the fan is actually completing a ground circuit that is otherwise missing ........ (which could take you back to a ground block issue .....)
I am reminded by similar situation in the TL circuits, where I recall a bike that was grounding itself through the fuel pump connection! Not a happy situation!

cave1 · 10-05-2009, 02:12 PM

Jeez, DeCosse, that is not good news. This is my first ECU driven bike, and I just assumed that the ECU would send a signal to critical components at start up and cut spark to prevent damage in cases where the critical component didn't respond. If this is not so, then I do indeed have to go digging for that ground block sooner rather than later. Have you by any chance ever tried to start a T509 with the connector to the fan disconnected? So much for educated guessing, I guess...

Hey, I just thought of a way to test your rogue ground hypothesis. If I disconnect the fan now and your hypothesis is correct, then I should get no spark. And if I leave the fan disconnected and jumper a good ground to the ground on the fan connector, this should restore spark. I'll give it a try when I get a chance...

DEcosse · 10-05-2009, 04:36 PM

There is no active sense from the ECM that knows whether or not the fan is actually connected, powered or running (or not).
Only thing the ECM does with respect to the fan, is to supply the output which energizes the relay which powers the fan.

On the T509, the way the fan motor is energized is as follows:
(see extracted schematic below)

There is a constant ground (or should be!) on the return side of the fan motor (from the common grounding point);
the positive power is supplied via the Fan Relay;
The fan relay has 4 terminals:
The contacts are the input (blue/black) & output blue/white - the input comes directly from a dedicated fuse, fuse 5.
For the relay coil, it has power (green/red) which is ignition power;
To energize the relay, a 'grounding' (sink) signal is supplied by the ECM via the black/yellow wire.

So you see what I mean about unplugging cannot stop the bike's main control system?
As I suggested, potentially the fan plug could be what is supplying the ground connection to the block, as opposed to the other way around! (supposition)
But you should certainly be able to unplug the fan with no affect on the bike's operation.

cave1 · 10-05-2009, 06:43 PM

Yes, I see exactly what you mean. The diagram and expanation you supplied make it very easy to visualize. Your supposition that the fan is now supplying the ground for the bike's control system seems very likely. I am obviously going to have to find that ground block, replace it, and establish a proper ground for it by means of the supplied new harness. Too bad the block has proved so elusive so far. I am hoping that once I start pulling off electrical wrap, the black wires will converge in an obvious way to somewhere...

DEcosse · 10-05-2009, 07:12 PM

Within the fan itself, I would have actually expected the -ve return to be electrically isolated from the actual body - but it's possible I guess (or a double problem!).
Unfortunately mine is completely 02+ system now so can't make comparisons for you

The first thing is just to unplug it & see if/what effect it has.

.