Dual Mass Flywheel

Dual Mass Flywheel Upgrade

So somewhere in the middle of 2012, it was discovered that the spline shaft used to connect the crankshaft of the engine to the gearbox was devloping a history of failure.  A bunch of people on the Subenews group attempted to ascertain which engine configurations were failing and why.  The failure was basically excessive wear and/or stripping of the splines.  In one case, the shaft actually broke into pieces.

broken spline shaft
A broken spline shaft removed from another aircraft.

Honestly I’m not sure there was a conclusion drawn, but it seemed like those engines that received a dual mass flywheel (“DMF”) were not stripping out their spline shafts, or at least not as often.  A fairly significant effort was put forth on creating a procedure to change out the solid flywheels for the DMF.  Fortunately, Jan Eggenfellner still had some stock of the spline shafts and many people, including myself purchased new ones.

This turned out to be another year in which I spent more time in the shop than I did flying.  Although in all honesty the whole upgrade could have been done in one or two full weekends, this upgrade ended up taking me about six months.  Because upgrades are never simple.

I will say that the Subenews group did an amazing job putting together enough resources to take some pain out of the process.  The repository of documents is locataed at this website:

Subenews Wiki

This upgrade was challenging.  It required custom made alignment tools to be manufactured, special parts to be ordered, and many hours of custom fit and finish.  There were several points in the process where I didn’t know I needed new or different parts and it usually meant a lost week or two until I obtained them.

Since the procedure required disassembling the entire front mounting plate of the engine, I decided at the time to upgrade to the larger H6 style radiators as well to help the cooling situation.  Fortunately at the time someone was selling their radiators and I grabbed them.

Removing the old radiators also meant losing the air filter box which was bolted on as a shroud around the right radiator.  An improvised solution with a K&N filter was made to remedy this (pictures below).

Step 1 of the process was to remove enough stuff from the front of the engine to be able to replace the flywheel.  That’s a lot of stuff.  The propeller, PSRU, starter, and oil cooler were all pretty easy since they regularly come off for various maintenance procedures anyway.

wm_DSC00001

The first surprise after pulling the prop off was a damaged brush block.  Not sure exactly what happened, but it looks like the brush may have broke a piece off causing it to wear unevenly.  Not too big of a deal since there was still plenty of “meat” left on the brush, but I was left wondering what may have caused this.  In any case, these are readily available from Aircraft Spruce, so I ordered a new one.

wm_DSC00025

Now the cooling system had to be opened up.  After draining the coolant and pulling the hoses off, the radiators could be unbolted from the front mounting plate.  This is where I discovered the second surprise, which was that the flanges on the right radiator were damaged.

wm_DSC00024

It’s not like I would have lost a radiator in flight or anything like that, but of course it made me aware that there is an issue with fatigue on these mounting flanges.  On the other radiator, there were metal strips to distribute the load across the length of the flange.  On this side was the airbox shroud which served the same purpose.  And yet, here we are.  Definitely a future inspection point.

With the radiators off, it was time to roll in the engine crane.  Since the engine is bolted to the front mounting plate, it has to be supported in place while the plate is removed.  I used four existing bolt holes on the engine to attach the load leveler chains and put a saw horse under the tail of the plane to prevent it tipping backwards.

wm_DSC00029

At this point, mounting plate removal was easy enough.  The engine bolts had some spacers in between the mounting plate and the engine, so I had to be careful to capture them as they came out.  Otherwise, the plate came off uneventfully.  No surprises underneath.

wm_DSC00044

After removing the flywheel, I inspected the entire “front” of the engine (normally in the car, this would be the “back” that mates up to the transmission).  It was very clean, and no leaks or any unusual wear or damage was present.  I wasn’t expecting any, but you never know.  This was the first time I had actually seen this part of the engine first hand, so it was kind of a relief knowing it was in good shape.

wm_DSC00046

The DMF had to be modified before installation.  In a car, a clutch friction plate is used to make the physical contact between the engine and transmission.  In this application, the clutch plate is not needed, so several rivets were drilled out and the plate removed.  The details of that procedure are in the Subenews document repository.

For my next surprise, it turns out the old bolts will not work with the new flywheel.  I had to special order them.  The originals were hex head bolts, and the news ones were… different.  After a lot of research, I found out they are something called “Ribe”.  They look similar to Torx head, but not quite.  Fortunately, Amazon had a Ribe bit set in stock which I had to order to drive the new bolts.  This sidetrack took me about two weeks to resolve between research and ordering parts.

Finally, with the new DMF in place, I was more or less half way done.  Reassembly started with the new spline adaptor plate.

wm_DSC00060

The next part was not necessarily tricky, but required some precision and slow careful process.  The problem with this particular engine configuration is that it had no alignment pins to guarantee that the engine and PSRU would be precisely aligned when reassembled.  At the factory, a special tool was used to line everything up, and the whole mess bolted in place.  Unfortunately, loosening up the assembly destroys the alignment, so re-alignment with the tool is necessary.

A fix was put into later engines, where a pin was used between the engine and the mounting plate for exact alignment.  To replicate that fix, pins were obtained to plug into the engine that were long enough to go through the mounting plate.  I then fabricated a close fit plate which was bolted to the mounting plate on the front side which precisely aligned the engine when the pins went through.  These pins were aligned using the custom tool which was manufactured by some folks on the Subenews group.  The idea here is that the tool was used to do the initial alignment of the pins.  The alignment plates I made would then be permanently mounted such that the alignment tool would not be needed again.  As long as the pins go through the plates, alignment is guaranteed.

Below is a picture of the alignment tool in place.  The PSRU did already use alignment pins, and those are used as the reference.  The outer holes of the tool are for the engine pins.  With the tool placed on all four pins, the engine can be tightened down and kept in alignment.

wm_DSC00062

Now it turns out that the DMF was a little thicker than the solid flywheel, and this presented a problem.  The original spacers between the mounting plate and the engine were not sufficient to clear the bolts on the back side of the mounting plate from the DMF.  I had to add some washers to raise the thickness a bit and make sure everything was interference free.

With the engine bolts torqued down, the DMF upgrade was officially done.  However, the full upgrade was really only half way complete.  The next step was putting in the new H6 sized radiators and trying to figure out how the new puzzle pieces would all go together.  Next step, bolt on the new rads.

wm_DSC00091

The new radiators came with the fiberglass shrouds, so I am really glad I did not have to fabricate those.  Since the coolant hoses were all original, I decided to replace everything with new even though they were still in good shape.  Not many places in town carry the good silicone heater hose, so I had to search a bit, but eventually tracked down everything I needed.

After bolting down the PSRU, I ran into yet another surprise, which was that the new alignment of the starter motor caused some interference with the PSRU housing.  I ended up having to grind just a little bit of metal in order to get everything to fit correctly.  Unfortunately there isn’t really a procedure for starter motor alignment, so it is just a “try it, loosen the bolts and move it, try again” kind of deal.  The good news is that once the PSRU and starter were back in, it started looking like I was getting close to being done.

Another unexpected problem came up with the coil pack mount.  I ended up having to fabricate a small offset bracket to mount it at the height and position that would not interfere with the mounting plate or PSRU.  Apparently those new engine spacers had widened the gap just a bit too much for the original bracket to fit.

wm_DSC00097

Finally, with all the puzzle pieces bolted on, I was in the home stretch.

wm_DSC00105

Now for the fun part.  There are numerous other pages that document my cowl modifications, so my love of fiberglass work is well known to be non-existant.  But that doesn’t mean I’m not getting good at it.

I decided to try a little trick.  The basic problem was matching up the radiator shroud to the cowl openings.  To do that, I roughly matched the cowl first, then layed up several layers of new cloth which ended up making the bottom cowl and the shroud one piece.  I used some corrugated cardboard as reinforcement to stiffen up the glass.

wm_DSC00115

Now obviously it was not my intention to permanently bond the bottom cowl to the radiator.  So the next step was to simply take a Dremel tool with a cutoff wheel and seperate the two.  This made a perfectly aligned seam.  A lot of “sand, fill, repeat”, a bit of touch up paint, and whattayaknow.

wm_DSC00365

I punched a hole in the original radiator shroud to accomodate a blast tube to run to the fuel block, so I drilled a hole and reattached the hose flange.  Bit of a tight fit, but it did fit.  The hose was re-strapped along its length to prevent it from moving around.

wm_DSC00366

The final problem to solve was the loss of the airbox filter on the right side.  Since the original radiator incorporated a shroud with the filter, I needed a completely new alternative.  I ended up simply running the intake hose straight down in back of the engine between the battery box and the timing chaing cover.

wm_DSC00362

From there, a transition was made from creative use of plumbing parts to turn 90 degrees, and open up to the correct size for the K&N filter.

wm_DSC00363

Finally, the K&N filter was strapped down to the engine mount.

wm_DSC00361

When a large effort like this is completed, it makes me, well, not nervous, but more aware of the need for carefully flight testing and measuring the result of the upgrade.  After several hours of hard flying, I can say that I am quite pleased with all of the changes.  The engine is absolutely smoother running.  And with the new radiators and use of regular antifreeze, I am convinced now that I have the cooling system dialed in.  Even in the catastrophically hot Arizona summer, I feel much better about the engine temperatures.  In fact, last winter I had to close the cowl flap half way because the engine never warmed up to 170F.  It’s all good!