[TR] 'A' type overdrive question

[Triumph]

When I rebuilt my Type A overdrive a few years ago I remember spending
about 20 minutes meshing my gears and I found that when you get it
right there is no question about. They literally fall together. There is no
binding.

Bob Maassel
Fort Wayne, IN
1959 TR3A
tr3abobm77 at verizon.net
--------------------------------------------------
From: "Randall" <tr3driver at ca.rr.com>
Sent: Monday, April 19, 2010 5:00 PM
To: <Triumphs at autox.team.net>
Subject: Re: [TR] 'A' type overdrive question

>> The 3 gears on the planetary set are marked so that they are in a
>> specific
>> position relative to one another when the sunwheel is engaged.
>>
>> Why is this so? In other words, why is their position critical?
>
> I'm not sure I can explain this clearly, so bear with me.
>
> The number of teeth on the annulus is not a multiple of the number of 
> teeth
> on the sun gear.  Thus, if you draw an imaginary line from the center of
> rotation of the sunwheel, through the center of each planetary gear, it 
> will
> hit both the sunwheel tooth and the annulus tooth in different positions.
> The two gears that are joined together to create each planetary gear are
> offset to match this "phase shift", but only in a certain position (since
> they also have different numbers of teeth).
>
> Some positions of the marks are simply impossible (the teeth won't mesh);
> but in some cases there is enough accumulated clearance to allow them to
> slide into engagement.  That will pull the planetary carrier and sun gear
> off-center to some extent, leading to vibration in operation.  It also
> causes most of the load to be carried by one planetary gear instead of
> evenly divided between 3 of them, so increased wear and even potentially
> tooth breakage could ensue.
>
>> If the gears were not marked for some reason, how could proper
>> alignment be  determined?
>
> If the gears were not marked for some reason, I think I'd want to look for
> another set of gears.  At any rate, I can't think of a guaranteed method 
> of
> lining them up, short of cutting apart an annulus so the tooth engagement
> could be examined in situ.  However, you might be able to do a setup 
> without
> the housing, and check that everything is running on the same center, 
> which
> _might_ be good enough.
>
> -- Randall