[DavidfromWisconsin]

For those that might not know, a somewhat common 650 problem is the starter clutch being score by the rollers due to a combination of the rollers, the extra torque require to turn over the 650 and the lack of maintaince on the starter motor and/or a weak battery. Those score marks, once they are there keep the rollers from properly engaging the flywheel until finally the starter doesn turn over the engine any longer and an engine drop and flywheel replacement is necessary.



A friend was over on Thursday and I was showing him the problems with the CX650 flywheel and starter clutch assembly. He asked for a chance to turn it down on a lathe at the school where he works part-time.



He found out that due to the size he didn’t know how to chock it up. It ended up that it took two other teachers to figure out how to do it. They didn’t really seem to want to do it so he lied to them and told them it was for a turbo and that is was special. I have the feeling that it would be extremely expensive to have a shop cut out the marks made by the rollers.



But!!! They did it and it seems to be working. I have reassembled everything and I looks like it all lines up like it should. I don’t have an engine to mount it in right now, but it looks like you can turn down the flywheel. The thing is, it probably would be cheaper to buy a used flywheel.



If you don’t have a friend with access to an extremely large jawed lathe I don’t recommend repairing the flywheel. It would be extremely expensive otherwise …



Instead find a used 500 flywheel on EBay and use that since they are identical and don't usually suffer from the same problem. Often the starter gear is damaged also, so try to get that part also with a replacement flywheel.



Those with 650s remember as soon as your bike starts to turn over slowly or hesitates when you first hit the starter, clean your starter immediately and if it still acts that way get a new battery. A 650 should turn over as easily as a 500 if everything is good with the stater, battery and starter clutch.



If you need any further advice feel free to post here or email me directly.

 

[Cobram]

If you could figure out a way to mount it to a flywheel resurfacing machine, it looks like you could do the same thing with a small stone. I was thinking about it when you showed everyone the problem area on Tony's flywheel, but even a flywheel resurface fee (IF it fit) would cost more than used CX flywheels are going for.

 

[marshallf3]

And then there's always that poorly designed ground wire on the starter some have spoken of.

 

[William J Ennis]

David

I have two photos which illustrate the problem. Regards, Bill

 

[DavidfromWisconsin]

Someone sent me an email asking how to clean a starter so I thought I would put it here, as it is so closely related to this problem.



Cleaning & servicing the starter is a 20 minute job for me with experience. You should be able to do it in 1-2 hours.



You need to:

*The day before spray down the screws with penetrating oil to give it a chance to work in to the threads.

*Keep track of where everything came from so it goes back together exactly the way it came apart.

*Disconnect the battery

*Take out the two bolts that hold the starter in place

*Disconnect the hot and ground wires to the starter

*Put a large flat blade screw driver between the starter and the engine and gently push it out of the engine. No oil will leak out.

*Take the snap ring off the gear

*Remove the screws holding the motor together. If they don't unscrew easily grab the screw shaft near the threads with a vice grip to break it loose.

*Carefully pull it apart and look for the tiny o-rings between the two end caps and the main body. They break easily.

*Make sure you don't miss the sequence of parts and washers. Scrape the ends of everything with your fingernail to make sure there aren't any washers that you missed. They tend to stick and basically become invisible.

*Clean everything spotlessly clean not using anything metal. A Scotch Brite pad works great.

*Do not clean the brushes.

*Don't use carb or brake cleaner cleaner on the armature & main body as it could damage the insulation.

*When everything is clean lightly lube the end caps so everything will spin smoothly.

*When reassembling be sure not to put any side force on the brushes. Pull brushes gently back to put the armature back in place.

*Be careful of the o-rings. If you break one you can use silicone caulk of almost any type.

*Put the starter back in and enjoy your new starter power.



If you need additional help feel free to email me directly. Do so by clicking on my avatar and go to "about me" and you will find my email address. I have it there so only members can find and use it.

 

[DavidfromWisconsin]

David

I have two photos which illustrate the problem. Regards, Bill

Bill,



The photos are great, but could you post a clickable thumbnail to a larger photo so everyone can see the detail better???



Thanks in advance

 

[Joel]

Bill if you have the ability to draw an arrow pointing to the damaged area. As I have never opened a 500 or 650 yet.



Not hi jacking the post but I have seen pictures of the 650 cam chain tensioner and not been able to see what people were describing. I do see the damage that the cam chain on the engine block. But I can't tell if my tensioner is fully extended. So arrows would be great for those of us who just don't know.

 

[William J Ennis]

Bill,



The photos are great, but could you post a clickable thumbnail to a larger photo so everyone can see the detail better???



Thanks in advance

Done, go back to my first posting

Thanks for the help to get the thumbnail photo in the post

Regards,

 

[William J Ennis]

Bill if you have the ability to draw an arrow pointing to the damaged area. As I have never opened a 500 or 650 yet.



Not hi jacking the post but I have seen pictures of the 650 cam chain tensioner and not been able to see what people were describing. I do see the damage that the cam chain on the engine block. But I can't tell if my tensioner is fully extended. So arrows would be great for those of us who just don't know.

Joel,

I think the software I have will allow some graphic editing, but I have not yet mastered it. I will try to do what you ask because I would like to learn the steps to editing the photo. It may take me a while . . . .

Regards,

 

[richnct]

The 500 flywheel to buy for replacement needs to be a TI flywheel, not the CDI version. Makes a difference, yah it does.

 

[Cobram]

Reading a recent thread on there, it looks like the reason for the starter clutch problems on the 650's may have less to do with more mass to move around, and much more to do with ignition parameters.



May be the factory timing is the culprit.

 

[DaveF]

There was a discussion on one of the CX/GL forums years ago where this was addressed. I found it interesting so I copied the posts to a text file for future reference. The discussion is pasted in below:



----------------------------------------------------



10/7/03 6:20 PM | | GL650 starting problem solved forever Len Member 18 Posts Aylmer

The GL650 engine, with transistorised ignition and centrifugal advance, is well-known to be hard to start especially when temperature drops below 10C. There has been an ongoing discussion on this forum about certain models of starter motor being better than others, when the root cause of the problem is ignition timing, at least on the 650; I have no experience on the 500. My bike was just such an exercise in frustration in the five years that I have been riding it, so I decided to fix it once and for all. Initial timing, static and at 1100 rpm, is approximately 15 to 17 degrees before top centre (the FSFI mark on the flywheel), increasing to a full advance of 37 degrees at 3100 rpm (the double line mark); separation between cylinders is 80 degrees. I measured it with a degree wheel at the front bolt while turning over the engine with a socket wrench.



This initial timing is too far advanced at the slow cranking speed of the starter motor, allowing combustion to occur before top centre. The starter motor has to overcome the rising combustion pressure in the cylinder, almost stalling in the process, draining the battery and even causing the solenoid contacts to stick because of high peak currents. It has even been known to fire backward, making painful noises at the roller clutch. Honda probably advanced the timing for reasons of pollution control, reducing unburnt hydrocarbons at idle. However, what is efficient at 1100 rpm is deadly at cranking speed.



The solution is to reduce initial advance to 5 degrees, more or less, while retaining full centrifugal advance; here is how it is done.



The base plate that carries the pickup coils (page 197 of the Clymer manual), is adjustable by loosening the two screws and rotating it; however there is not enough travel as it is, so you must widen the two slots approximately 5mm (0.200 inch) to the right in order to allow the base plate to rotate farther to the left, counter-clockwise. This is easily done with a grinder or round file. After grinding, you can try it on the engine; you should be able to retard it nearly to 0 degrees or top centre (TL and TR marks).



In order to compensate for the later initial advance, centrifugal advance (page 209) must be increased. The range of travel of the spring-loaded flyweights is limited by two stops that look like ears; they must be ground on the inner faces to allow more travel. The original thickness of 2.5mm (0.100 inch) is reduced to 1.5mm (0.060 inch); a hand-held grinder is best, but be careful not to touch the central shaft. Check your work as you go, to ensure that both weights operate the same.



Install the advance mechanism and base plate, rotate the ignition cam (on the centrifugal advance mechanism) by hand to full advance, and adjust the base plate so that full advance corresponds to the timing mark on the flywheel. Let go of the advance mechanism, leaving the base plate where it is, then rotate the engine to see where initial timing occurs. It should be approximately 5 degrees BTC, the actual value is not critical. For reference, I made new marks on the flywheel with a hammer and punch.



The result: no more hard starting! Even on cold Canadian mornings when the temperature is 4C (40F) and with 15W40 diesel oil in the crankcase, starting is effortless, and so much easier on the electrical system. You may need to adjust the idle speed screw a little. There is no loss of power above 3000 rpm because full advance is unchanged, and the slight loss of efficiency below 3000 rpm is more than compensated by longer-lived starter, roller clutch, battery and solenoid!



10/9/03 9:45 AM | | Wow! Good research and developement. Panther Member 65 Posts Palm Springs CA

Thank you for your information and time. I had often wondered about this posability, but did not have the time to bother with it. I plan to do some of this to my CX650 engine when I put it into my Little Grey Lady later this year.Though I will limit the change to about 8-10 deg. and not shave the stops as much. I do have two sets, so i can mes with them some to get it right. I plan on building a bench mount so i can do all my fidling on a bench at working height before putting her back on the road. Thanks again.



10/12/03 6:28 AM | | Interesting Mercury Member 260 Posts Silver Creek NY

I just sold my GL650, but always wondered why it was so much harder to cold start than my CX500's, which start easily. I just thought it was bigger and more cold-blooded. If much below 50 degrees, it took 8-10 cranking sessions to get it running. I sold the bike to my nephew who is a mechanical engineering student, and will forward this info to him, as he is already whining about how hard it is to start the thing

-Weave-



10/12/03 6:16 PM | | Gl650 starting Len Member 18 Posts Aylmer

Panther, good idea to start gradually (8-10 deg BTC); you can always grind more later if needed. It's too bad that the layout of the GL makes experimenting with timing such a chore, so much to dismantle just to get to the parts. Let us know how it works for you.

 

[DaveF]

The 500's do not have a kickback problem. It would be interesting to do a side-by-side comparison of the GL500 and GL650 timing components.



Could the solution be as simple as installing GL500 timing components in the 650? Hmmmm....

 

[DaveF]

The GL500 and GL650 use the same flywheel/rotors, ignition modules, and ignition coils.



They also have the same static timing procedure performed upon engine assembly. This procedure calls for the mechanical advance rotor nub to be aligned with a pickup coil nub when the flywheel is positioned to the "FS" mark.



With the above in mind, it follows that the initial timing point (at zero rpm) is identical for the 500 and 650 engines, per the factory.



The timing pickup coil assembly has a different part number for the 500 and the 650. When comparing them side-by-side I did not find any differences, although it's possible I missed something. The base plate is machined the same and the various screw holes are in the same place. The coils look identical. Their electrical resistance is very close. The cable and connector appear to be the same, even in length. I would not expect there to be any electrical differences as they are used with the same ignition modules.



As far as I could tell, the mechanical advance units appear to be identical with one small but important difference. The springs are stiffer on the 650, measuring 0.024 inches in diameter versus 0.021 for the 500. This would cause the timing to be different until the stops are hit at higher rpms. At slow cranking speeds I would expect the timing to be close to the initial timing value, or nearly identical for the 500 and 650.



Based on this, I'm not convinced that factory timing is the sole reason for 650 starter clutch issues - otherwise the 500's would also have problems. It's probably a combination of things, including a sluggish starter and/or weak battery.



That said, the posted recommendation may still have merit. I might try it on the next 650 engine I dig into.