"Fixing" a bad dog clutch.

[sclay115]

Okay, so here is the skinny, have a 72 25 horse Sportster, and I have infamous slipping gear box, caused by a badly worn dog clutch. Now, these bad boys are pricey, new from the dealer, about 110 bones.

I have another lower unit, so I pull it apart, that clutch looks just as bad, but I think to myself, why in the world can I not just get rid of these worn in ramps on the mating side of the pawl?

So, why not? I worked a bit on it today, filing, by hand, the perpendicular edges of the pawl, but before I continued on, I wanted to ask. Is there any reason I cannot do this? Seems like it is a very common problem, and yet, everybody is going the route of the 110$ dogs, and possibly even the 300$ gearset. I cannot seem to find any information on people just simply fixing their dogs.

Now I know it is not the best solution, as the engine was designed to feed the power through two pawls of a certain thickness, but there is no way I can imagine that filing the edges of the pawl down a mere amount will cause the pawl to shear off.

I come from a vintage motorcycle world, and still ride a 71 CB350 very often, and sometimes, parts just don't exist, so we have to make do. After diagnosing the problem on my 72 outboard, my first thought was, how can I repair this, here, now. But, is there a reason that I should not even attempt to do so? These motors have been around for long enough, someone must have tried this at one point in time...

Steve

 

[bktheking]

Re: "Fixing" a bad dog clutch.

I'm going to be tearing into the same motor later today with the same issue, i'd be curious to know how you make out on this topic.

 

[blimp]

Re: "Fixing" a bad dog clutch.

there was a thread recently where a guy machined his and checked his work using machinist dye to make sure the surfaces mated perfectly. Really interesting thread with pics, I'll see if i can find it. Sounds like it is very feasible with the right equipment and attention to detail.

 

[blimp]

Re: "Fixing" a bad dog clutch.

here it is:

http://forums.iboats.com/showthread.php?t=407172

 

[wilde1j]

Re: "Fixing" a bad dog clutch.

Well, I guess if you can hold a perfect 90? with a file. harden the metal when you're done and do likewise with the recess in the gear face, you've got it made. If you could make these pieces out of pressed horse dung, someone would do that as well. You need more than a file or a Dremel to do this one. BTW, the clutch dog is the easier part here.

 

[AlTn]

Re: "Fixing" a bad dog clutch.

post #9 of Blimp's link might be worth noting in this case

 

[ditchen]

Re: "Fixing" a bad dog clutch.

done a budget fix to two already and still going strong.

swapped reverse and forward around, then squared up the edges. Yes, they were not perfectly squared, and yes there is a little larger clearance then before, but one thing is for sure, it is still working and not clunking and popping out of gear anymore.

 

[ebkolove]

Re: "Fixing" a bad dog clutch.

I took mine off and fipped it round to the other side. Works fine

 

[sclay115]

Re: "Fixing" a bad dog clutch.

Yes, after some very casual "Hey this seems much better" testing in the garage with the forward gear in one hand and the dog in the other, I can say for sure that me just knocking the ramp out of the dog helps the situation tremendously. My forward gear in my box however has a fairly apparent angle to it, so after mating the gear with the other forward gear, it is much, much better.

And yes, I understand this is nothing like it's operation in the gearbox, but if I can FEEL the difference after the filing, then, well, there must be some change.

I will work on this some more, and post back with results, waiting on some seals, got the girl apart, might as well seal the whole thing up.

Steve

 

[bob johnson]

Re: "Fixing" a bad dog clutch.

the metal in all those parts is like 62 rockwell C!!!!!!

no file will touch it...

even stones will be laborious!!!!!!!!

a machine shop could reface the parts, because they can do it symetrically so both drives surfaces touch at the same time.. and they are square!!!

I was thinking abou touching up my V4 clutch dogs surfaces here at work with a surface grinder... they are not bad, they are tricky because the surfaces are not on a plane that goes through the center!!! they are off center!!! so you cant push them back in a parralel motion.... they have to be cut angularly, because the ANGLE on the gear has to be matched doesnt matter you ahve 1 or 2 more degrees of rotattion....but they ahve to meet flat to each other!!!

the OPs clutch dog and gears has radii which makes it even trickier!!!!

I think years ago i redid a clutch dog and gear like those from a late 50's 5.5..

worked fine!!! i cut them with a carbide end mill


bob

 

[bktheking]

Re: "Fixing" a bad dog clutch.

Not hijacking, 69 20hp that I just took apart.

same shiz

 

[sclay115]

Re: "Fixing" a bad dog clutch.

Bob, you can actually file these. It just takes time, and a decent file. And yes, they are radius cuts, however, if you take the time to examine what you are doing, and observing how it mates up to the gear as you go, while I know you cannot get it perfect by eye, well, it'll be a heck of a lot better than when you started.

These are expensive parts, and simply, I am not going to put a couple hundred dollars worth of gears into the motor. I'd rather take those funds and put it towards a newer mill.

King, that is interesting that yours is like that. Seems like the only point in contact with the gear would be that sharp edge on the facing surface of the pawl, is that correct? Heck, if that is how "precise" that setup is, I could use a gazillion RPM angle grinder on mine and get close. The gears on my 72 look just like yours, but the dog has a radius starting from the outer edge all the way to the inner splines. With no sharp angles along the mating surface.

If that is the sort of "precise" fit that needs to be done to make all this work, I do not believe it will be an issue to attain a fit that exceeds OEM standards.

Steve

 

[bob johnson]

Re: "Fixing" a bad dog clutch.

Bob, you can actually file these. It just takes time, and a decent file. And yes, they are radius cuts, however, if you take the time to examine what you are doing, and observing how it mates up to the gear as you go, while I know you cannot get it perfect by eye, well, it'll be a heck of a lot better than when you started.

These are expensive parts, and simply, I am not going to put a couple hundred dollars worth of gears into the motor. I'd rather take those funds and put it towards a newer mill.

King, that is interesting that yours is like that. Seems like the only point in contact with the gear would be that sharp edge on the facing surface of the pawl, is that correct? Heck, if that is how "precise" that setup is, I could use a gazillion RPM angle grinder on mine and get close. The gears on my 72 look just like yours, but the dog has a radius starting from the outer edge all the way to the inner splines. With no sharp angles along the mating surface.

If that is the sort of "precise" fit that needs to be done to make all this work, I do not believe it will be an issue to attain a fit that exceeds OEM standards.

Steve

you made me go do it... ha ha

I just took the gears out of the box...the ones that are waiting to go back into my 1997 90 hp evinrude.

they are 60 rc on the nose!!!

which is pretty standard for heat treating hard stuff.....lots of parts were made with 4140 and similar tools steels and the stuff hardens to like 62-623 and they draw it back to 60.....

I am of the mindset that filing 60 rc material is futile.

I might try to stone a nice round edge on a sharp outer corner...but i wouldnt attempt any filing


I hope your gears were cooked to the point they are got soft!

bob

 

[bktheking]

Re: "Fixing" a bad dog clutch.

The edge in contact with the dog (the gear) is gone, i'm sure with some precision machining it could be fixed, the clutch dog is toast.

 

[bob johnson]

Re: "Fixing" a bad dog clutch.

The edge in contact with the dog (the gear) is gone, i'm sure with some precision machining it could be fixed, the clutch dog is toast.

I THINK the wall has to be flat and square... other wise the parts want to seperate!!!!!!!!!!!!!!! any angle and they are trying to move away from each other!!...ie thats the slipping out of gear act.

a square surface and they will not be trying to get apart...

but the machining must be symetrical..otherwise all your load in on one face
thats why spare motors are nice to have lying around
good luck

bob

 

[bktheking]

Re: "Fixing" a bad dog clutch.

It does indeed, it slips just by trying it in my hand.

 

[sclay115]

Re: "Fixing" a bad dog clutch.

I wanted to clarify something, about what I was saying before, about how the mating surface of the clutch King posted, and what is in my gearcase. Here is the one out of my case:

See how it is curved all the way to the center splines? While the one a few posts up has a flat point? It does not look like it was worn onto the surface, it is as if the flat spot was machined into the piece. Just an interesting design, as it seems they opted out of much more contact surface with the face of the gear.

I worked on this a bit more today, and have the seal kit on the way. It really feels quite solid now. Had the dog in the vice and the gear in some vise-grips and the grip of the tool gave out before the connection did. I feel I am getting there.

Steve