Hi guys, I had really bad days dealing with what came up to be a damaged signal coil. Here is the story.


First, the damaged signal coil...

It all started with momentary loss of power on the highway, due to no spark in one cylinder, but after a few seconds everything went back to normal, but a few weeks later it was more and more frequent, then I found a broken magnet underneath one signal coil, so I went to the fridge for a replacement, and sand one magnet to make it fit, and the bike worked fine from cold start. After about 3 miles the engine and the coil warmed up and the coil failed again. So I decided I needed a new coil, but guess what, here, in México City, the Suzuki dealer would charge me more than 250 dollars for a new signal coil assembly and they'd delay more than 30 days to deliver. So I bought a brand new chinese signal coil, made for a chinese manufactured bike, 10 bucks, and adapted it to my GS, had to make a bracket and "align" the coil nipple with the good one, the left engine signal coil. Here a picture of the result.



My concern now is about the timing, 'cause the geometry of the new coil is different and therefore there is some small advance or delay of ignition on the right piston. I feel some knocking on the engine, very weak, otherwise, the bike pulls hard and fine, so, I'm not sure where the timing went , advanced or delayed, what do you think? The bike fires not perfectly even, it remainds me a HD, he he...


the disalignment...

Later, I though that might check it with a timing lamp...

Great work to adapt that signal coil, but you should fix the timing issue before you damage the motor. If its too far advanced the bearings/rod/piston could be damaged, if too retarded it could overheat on that cylinder.

Oh boy, I get to drag out my geometry.

A top down picture of the rotor would be more helpful, but from what I see it looks like the timing is retarded by some angle set by t.

Using a small-angle assumption,

Degrees = (180/3.14) * atan (t / R)

Where R is the distance from the center of the crank to the pickup.

For plug figures t =5mm and R =30mm, that's about 9.5 degrees. Which is a lot. You'll want to measure this.

All this is fine, but there are a lot of assumptions here. Chief is your ability to accurately measure t. Do you know at what point, exactly, that the new pickup is fiiring when the rotor swings by?

Bottom line: Get out your timing light and get a real number, as a difference from the unmodified cylinder. Then you can use the above equations to guide you in fixing the issue.

Hi there,
I agree with the mole.
You will damage the engine in bad and expensive ways if the errant timing continues.
You have to fix the timing problem immediately before damage is done. This can't wait.
Also, have you tried asking if anyone here on the site has a good used signal coil assembly they will sell you cheaply?
Then send them money and they will ship you the right part, solving the problem.
I do, however, admire your efforts in making the Chinese part work on your bike. It's just that the wrong timing will definitely kill your engine if you continue to run it. Detonation will destroy pistons, cylinders, etc.
Good luck,
Trwhouse

He's in Mexico. Shipping there isn't the greatest.  He'd be better off running down a used part there.

That said, his fix does work. He just needs to finish the job and get it right. Before he trashes his expensive engine.

That is amazing. You have to post what bike you got that coil from, I need to repeat this idea on an XS650 that I am rephasing to a 277 degree crank from a 360. That way, I dont have to deal with the crappy points.
Other than that, cant you jet it richer and stop it from knocking ? I mean we advance the ignition all the time without a serious problem. we also cut the nose shorter and increase the advance and cause it to peak earlier. It can take it just fine.
Cool.
Buddha.

He can play with jetting all he wants, but the root cause will still be the timing. He needs to throw a light and a degree wheel on his bike and find out where he's at.

I'm sure it was something like this. A generic pickup coil from a GY6 scooter engine, China style. $10 on Ebay, salvagable almost anywhere.



http://cgi.ebay.com/ebaymotors/Pickup-Coil-scooter-bike-parts-61292_W0QQcmdZViewItemQQhashZitem3efa907a84QQitemZ270491744900QQptZMotorcyclesQ5fPartsQ5fAccessories

If you need an entire system, you could pick up a couple of these, a couple scooter CDIs, and a GY6 wiring diagram for less than $50. You'd just need to machine the backing plate and rotor.

CDI:
http://cgi.ebay.com/ebaymotors/Digital-Performance-CDI-scooter-bike-parts-62502_W0QQitemZ270491755488QQcmdZViewItemQQptZMotorcycles_Parts_Accessories?hash=item3efa90a3e0

Great, I guess I need one that isn't that L shape. I'll look on fleabay for pics.
Cool.
Buddha.

Thanks for the input guys. I don't have a Timelight, but will borrow one soon. And a short film of it will be posted. The model of the chinese bike I don't know, but the signal coil is very similar the the one noiseguy posted, only the base plate was different, which I discarded and made my own bracket. I also though about replacing both coils and make a new plate so both new coils were perfectly simetrical, with long grooves to spin the plate 'til the timing was right. But don't have the tools for that. However will look for a used assembly for cheap, and will refurbish the bad coil. Whatever is done on a budget first wins...

I'm not sure you absolutely need a "real" timing light for this. At this point, a static timing light would be better than nothing, and better than waiting. You seem pretty proficient in electricals, so I'll just give you the rough details; you can figure out the rest. This should cost about $1 plus some time. Right up your alley. Static timing lights are used on simple engines to set timing, and should work here.

Hook up an LED to the output of one of the sensors. I think there is enough signal voltage there to light an LED; if not you can probably figure something out. I don't know if these are Normally Closed or Open; you'll either get a light most of the time or dark most of the time. Voltmeter would work too; same idea anyway.

Install a degree wheel on the crankshaft. Set a marker on it. Here's one you can print out.

http://www.tavia.com/free_degree_wheel.jpg

Remove the spark plugs. Rotate the engine over by hand.

You should find that as the rotor passes the sensors, it either lights or goes out. Note the degree mark that it happens at.

Now hook to the other sensor. Rotate again and note the degrees.

If they are not 180 apart, adjust the newly installed sensor until they are.

Post back with the results. I'm curious now.

Before I adapted the coil, made some tests to know when the rotor triggers the signal, but static measurement were not easy nor accurate cause a slow moving rotor does not generate as much voltage to notice on the multimeter, the signal is very small, like half volt if I recall right. First I mounted the coil with bigger offset to the delayed timing side and the bike idled poorly and bogged a lot on half throttle, so moved it forward, but didn't improve very much, on that same position switched wires, and it worked, I assume it was due to the positive or negative pulse the ignition box expects. I tought about drawing a line across the rotor and wire up the timelight to fire with both plugs so if there is some offset the light won't show a line but two crossed lines, if the angle is to small it will look almost like only one line. Tomorrow will post my new findings.