Hi Paul, Do you have any more info on Tracy's box, like the shaft and
internals, I can't figure out where the thrust goes etc. The box is
very chunky, I would say overkill somewhat. Also I can't see any oil
supply, in or out. You wouldn't have to saw off the flange, I would
do my box and then remove it from the pattern before casting others.
From the side view it looks like the prop shaft is one piece, is this
right ? Also, is the box intended to be bolted up to a plate or a
bell housing. Chris.


Tracy used a cut down GM pick up truck rear axle and ball  bearing.
The bearing was retained by a pressed on collar with  several tons of
force. Tracy fed the box with oil from the engine. So too the Ross
but the Ross had all needle bearing so it required less oil.

Very heavy at 9 pounds, I would not do it that way.

I would use a hollow shaft like the  Ross with a separate prop
flange. No forging required. The flange would be retained by special
set screws called  "Extended Point" . A relatively new idea and much
stronger  than pins that go all the way trough the walls of the shaft
(which is what Ross used for years) or splines and much less cost.
Easy to machine with a rotary table.

About a 2 inch OD steel tube with a .190 wall is about right. Just
like the Ross box. That is about the largest needle bearing you can
buy.

Use about eight or ten of these .25 inch screws and they are almost
as strong as a one piece shaft and flange. Here are some pictures.

Do you use a dwg program Chris? If not I suggest getting a Rhino. It
will save you money in the long run.

Paul Lamar

Speaking of the Ross PSRU, if one just decided to cast his own housing [perhaps out of an
aluminum alloy], what would be the best alloy to use…decent properties and not too
expensive or hard to do?
Gene Kahn


356 is the usual casting alloy and the foundry heat treats to T6 hardness.
Paul Lamar

I have been following this discussion as this is something I was thinking of trying. At first I thought it would be easiest to simply copy Tracy's 2.85 box because it seemed easier to plug a shaft into the planet gear rather than build something to drive a planet and 2.85 seems good enough. But as I read through this I can see a definite advantage to building the 3.17, less weight and more horsepower for free. However I am having a problem with the connection of the planet flange and the prop shaft. If I am seeing the drawings and understanding correctly, there will be eight or ten .25 small screws going through the flange and into the end of the shaft with a .190 wall. This doesn't make any sense to me. Is there something welded to the .190 shaft to accept the .25 screws? If the shaft were .250 and tiny #2 machine screws were used there would be enough meat in the end of the shaft to drill and tap but would that be strong enough? I must be missing something or I am not understanding it correctly.
Also how long should the prop shaft be? On Tracy's website he says that a longer shaft is better. What is the length of the Ross and Tracy shaft?
Paul Shamlian

We edm holes in the planet carrier planet shafts. This saves the hard smooth
bearing surface for the needles bearings. The existing planet shafts are VERY hard
to get out with out destroying the planet carrier.

As I recall the holes are at least 7/16th in dia.. The bolts going through these six holes attach the planet carrier to the aft flange on the hollow prop shaft. The aft  flange is either tig welded on the hollow prop shaft or it is bolted on  with a dozen or more 10-32 screws.
Ross tig welded it on. It is then wise to normalize and re heat treat to at least
Rockwell 60.  The loads on the rear flange, unlike the front front prop flange,
are pure shear loads. Then the shaft should be ground for the main needle
bearings. The reason the prop shaft is so long is to react the gyro bending
loads from the prop when the A/C spins or stalls.

BTW Poul, your planet Jatco gears should be there in about a week or two.
Went out Monday.

I will be off line for about 48 hours until I get re-situated in Parkes NSW.

Leaving on a B747 in a few hours.

Paul Lamar

 The Rotary Engine News Letter. Powered by Linux.
 ACRE NL web site. http://www.rotaryeng.net
 You Tube http://tinyurl.com/beqqxas
 Copyright 1998-2013 All world wide rights reserved.