Wrathall Mark wrote:
Actually I just cut and pasted your recommendation directly... " Perry had
the 35 pound flywheel in there which did not even slow down. Just bent the
bolt over at 90 degrees when it came out of the hole and contacted the
flywheel. Could have seized the engine if it had been an auto transmission
flex plate. Best to drill the heads of those bolts and safety wire them.
This is a well known problem with early necked down tension bolts. Racing
Beat recommends coating them with RTV before installation to damp the
vibratory modes"
Paul thought the flywheel weighed 35 pounds, but we weighed it
after taking it off the engine, and it is actually 23 pounds. I have
the 18 pound aluminum flywheel that will eventually be installed.
This is an aside comment, nothing to do with the failures.
I will be updating this page to reflect that the overheating occured after
shut down, and as more information comes to light..
Actually the overheating occurred during the long taxi runs from the
runway to the ramp area at PSP, where I was under ground control.
It would be interesting to find root cause here, but it is unlikely we will.
My theory is the bolt failure had nothing or little to do with the leak. I think
it occurred a couple months ago.
I think the root cause may have been a leak we found in the water pump,
or a leak we found in the hose clamp near the heater port. But these
could also have been caused by the first overheating/overpressurization.
So there is a little uncertainty there.
Perry:
Did you strip the engine before installation? (from the discussions it
sounds like no)
No, this engine was run in a car for 120,000 miles. Was running in the
car when it was removed and installed in the plane. Was run for 200
hours in the plane, 165 of those in the air.
What was the history of the engine?
see above.
Had the engine been previously split? (this would open the possiblity that
the tension bolts were incorrectly torqued - Mazda probably has robust
procedures in place to control this).
No it had not. There was consistent torque on all the bolts, we checked
them all after removing the flywheel, to see if any others were broken.
I think the broken bolt is a resonant condition and not related to incorrect
torque.
Has or will Mazdatrix return the rejected hardware from this engine? (you
indicted the rotor housings have shrunk - please provide exact measurements,
secondary damage, and photo's of condition. This would be interesting).
Paul has the shrunk rotor housings. They have not been measured yet
but I'm sure Paul will do that soon.
Is your engine still rigidly mounted, or have you added rubber damping?
Engine is still rigidly mounted.
Is your fan still hung via an extension directly on the e-shaft, or is it
now supported by bearings?
Fan is mounted directly to the flywheel on a prop extension.
What was the coolant consumption , if any before this failure?
I usually had to top off the coolant, just a few ounces, before each
flight.
What was your typical stable coolant temp in climb, on the ground at idle?
I'm just using the Mazda gauge, no temperature reading. On ground
at idle it barely came off the bottom stop. In climb it would go to the
mid range. It wouldn't be too hard to use a thermometer and a
Mazda sensor and gauge to convert these to numbers.
Had you bled the cooling jacket after you last broke into the system?
No, but it was working fine on the 5.5 hour flight the day before the
incident, the longest flight to date.
What was the coolant consumption during the trouble shooting flights
performed in Palm Springs?
Here is the sequence of events as I remember them:
1. Took off from Van Nuys Wed. morning and left full power on
longer than normal climbing out. I noticed coolant coming out the
vent of the overflow bottle, throttled back, continued climbing to
7500 ft. By the time I got to the Banning pass, I again noticed
some coolant coming out, reduced power a little more. I was
watching the coolant temp and overflow very closely at this point.
I was right over Palm Springs when it looked like it was continuing
and was going to run away, so I told PSP approach I needed to land.
Oil temp was normal all the way down, water temp got close
to redline but did not overheat, until taxiing to parking.
2. After it cooled down I topped off the coolant, idled the engine
through a few cycles of warm up and cool down to purge air out of
the system. It was hot and I was tired so I got a hotel and decided
to try it in the morning. I thought maybe I had just got some air
in the system from the long flight the day before, and it had not
been purged when I topped off the coolant before leaving for
Phoenix.
3. Took off Thursday morning early and climbed to 8000 feet.
It looked like the same thing was occurring again, so returned
to PSP. Same thing again, oil temp ok, coolant temp close
to redline, overheating during taxi.
4. So I called Paul, told him I thought I had internal o-ring
failure.
5. Took the cowlings off, noticed water coming from the
water pump while topping off the coolant. So Paul and
I got a new water pump and installed it. Did another test
flight Friday morning. It overheated much faster this time,
considering I could climb to 8000 feet the day before.
I think two cycles of ground overheating did in the internal
o-rings good at this point.
6. Went back to LA and got the engine hoist and configured
the Schrader valve test. Found a big leak at the heater
port, and a smaller leak at the lower water pump port.
Fixed these and thought we had it solved. Did another test
flight but it was worse than ever, had to stay in the pattern.
7. Everything looked normal at low power. Engine continued to start
and run normally. The last test we did was a run at 4000 RPM for
several minutes. Everything continued to look normal, but the coolant
pressure eventually got to 24 psi and started blowing coolant out.
It was finally sinking in that it had to be an internal failure.
Took the engine off that evening.
Were you getting blow down into the coolant during the troubleshooting?
This was hard to tell, only the eventual high pressure was an indicator.
What was the leakage rate you and Paul observed from the hose connection?
It did not leak at all at ambient pressure, but when pressurized to 25 psi
with the air compressor would leak at a rate that the 25 psi bled off fairly fast.
When was this connection last tightened?
I tightened several clamps before I left for CA but may have missed that one.
--
Perry Mick
Duckt N7XR
http://www.ductedfan.com