|
Post by racket on Oct 21, 2020 15:55:11 GMT -5
Hi Patty
Which bearing fails , front or back ??
Which raceway , inner or outer ??
Could you post a pic of the worn raceway please .
Lubrication..........how and how much ??
Air bleed cooling to the bearings , how , where etc ??
Cheers John
|
|
|
Post by turboron on Oct 21, 2020 20:29:43 GMT -5
Patty, what is the internal clearance of the bearing? I have had SKF bearings fail that had too tight internal clearance is less than 8 hours. Are the races pressed on the shaft and into the housing? Is the outer race keyed?
Thanks, Ron
|
|
|
Post by madpatty on Oct 21, 2020 20:47:48 GMT -5
Hi Patty Which bearing fails , front or back ?? Which raceway , inner or outer ?? Could you post a pic of the worn raceway please . Lubrication..........how and how much ?? Air bleed cooling to the bearings , how , where etc ?? Cheers John Hi John. It’s the rear(Turbine side) bearing that fails always. The inner race is which fails always. Lubrication is total loss through the same restrictor that came with the engine. The line is branched off of the fuel pump and then through the restrictor goes to the lube circuit. There are 6 x 3mm holes just in front of the compressor bearing that supply air from the ‘casing’ pressure to the bearing tunnel. 2 (diametrically opposite) of these 6 holes carry lube injector tubes for front bearing. These 2 lube injector tubes(approx. 0.7mm ID)are placed in the front of compressor bearing. There is 1 lube injector tube (approx. 0.9mm ID) placed in front of the turbine side bearing. Turbine side lube tube is placed/bent so that it is squirting fuel/oil mixture almost directly on the inner raceway. Cheers.
|
|
|
Post by madpatty on Oct 21, 2020 20:54:24 GMT -5
Patty, what is the internal clearance of the bearing? I have had SKF bearings fail that had too tight internal clearance is less than 8 hours. Are the races pressed on the shaft and into the housing? Is the outer race keyed? Thanks, Ron Hi Ron. Turbine side is a ‘normal’ skf 6003 bearing with normal ‘P5??’ Clearance modified to accept full complement SiN balls. My turbine side bearing won’t even last few seconds at high rpm let alone 8 hours. Races are simple hand push fit on both compressor and turbine side. Outer races aren’t keyed. They are just pushed in place by hand with compressor side bearing slightly tighter than turbine side bearing in its seat. Cheers.
|
|
|
Post by racket on Oct 21, 2020 22:13:41 GMT -5
Hi Patty Theres very little "colouring" of the whole raceway so it appears the lube/cooling are doing their jobs, but due to the very short run times before destruction , maybe not . Have you provided bleed holes "across" the NGV so that P2 air is blown into the turb wheel hub area to prevent hot gases getting anywhere near the turb end bearing ?? .............I have 12 X ~3 mm holes postlmg.cc/GBm3d2HHIs the "wear" only on one half of the raceway ?? Have you rebalanced the rotor ?? Did you use the original spacer between bearings?? I notice you've made a new shaft tunnel, unless alignment of bearings is "perfect" and any bearing pocket bores are exactly the same diameters as the old shaft tunnel , you'll have problem. Turb end bearings were my problem with FM-1 , that big lump of Inco weighs a lot and once it starts to gyrate were in trouble. We need to make sure EVERY surface that the bearings contact are perfectly square and true , I found that one turb shaft had its OE surface at the piston ring boss out of square to the shaft , not a problem with non contacting brass bushes but a ballrace tightened up against it was another matter. Axial shaft thrust should be forward with a radial turb wheel . Cheers John
|
|
|
Post by madpatty on Oct 21, 2020 22:33:36 GMT -5
Hi Racket.
The failed bearing had some fuel/oil when I removed it so I will guess lubrication is working. Other reason being I used the exact same setup as from the original design.
Wear is usually only on 1 half of the raceway.
There wasn't any spacer between the bearing in the original design so I didn't use it in new design.
I rebalanced the rotor but it failed again in the following retest.
I will recheck the surfaces but since I got them CNC machined with more focus and tighter tolerances on the critical surfaces i was expecting it to be square. Also a normal dry spoolup with the leaf blower gets the rotor spinning up to to easy 4500 rpm without any tightness or sticking so this was another reason to believe.
That's right once the inco lump starts gyrating it sends everything out of balance and rear bearing fails almost instantly.
Regards.
|
|
|
Post by racket on Oct 21, 2020 22:57:22 GMT -5
Hi Patty
OH !! , no spacer , so the shaft is only 17 mm dia , too thin , maybe rotor dynamics :-(
I used a spacer between my bearings which were 20 mm ID same as shaft , the OD of the spacer was ~32 mm making for a shaft with no chance of "vibrating" once all tightened together .
Without a spacer how are your bits tightened together ??
Cheers John
|
|
|
Post by madpatty on Oct 22, 2020 0:15:40 GMT -5
Hi Racket.
Was your turbine shaft 20mm OD?
My shaft is 16mm OD. Rear bearing sits on a 17mm OD SS bushing and so is front bearing.
Shaft tunnel is such that outer race of the bearings sit in their respective seats machined in the shaft tunnel.
The axial clearance is controlled using precision washers on either side bearings.
The compressor wheel is tightened against the front bearing and step on the shaft. The outer race of the front bearings is held with bearing seat(in shaft tunnel) on one side and seal plate(bolted to diffuser) on other side.
The rear bearings outer ring is restricted by its bearing seat on one side(there's a wave washer and some washers here for rear acting preload).
I hope it all makes sense.
Cheers. Patty
|
|
|
Post by racket on Oct 22, 2020 1:47:55 GMT -5
Hi Patty
The Garrett shaft is ~21.5 mm dia as standard with the brass bushes , we ground the shaft to 20 mm for the bearings , 20 mm has 56% more metal than a 16mm shaft.
When I was contemplating building the FM-1 engine I corresponded with Mike Early on the old Yahoo Site , he was a Garrett aero engineer who knew a lot of the turbocharger engineers in Phoenix and he warned me about the fact that a turbo shaft which works OK in brass bushes with lotsa oil to keep gyrations under control wouldn't be as "stable" when running in "hard" balls .
The fact that your wear is on only half the inner raceway sorta indicates the shaft is gyrating around a "false" centre and loading one side of the raceway only , hence the outer raceway being OK as the forces are spread over a greater area .
When I eventually went to the Allison C20 #8 ball bearings installed in bronze cups with O'rings on their ODs and pressurised lube between the O'rings in the few thou of radial gap , I ended up with a "soft" mounting similar to the original brass bushes , the lube passageways in the cup allowed oil to be injected onto the inner raceway .
What sort of rpm causes the failure ??
Cheers John
|
|
|
Post by madpatty on Oct 22, 2020 2:06:08 GMT -5
Hi Racket.
The failure is mostly around 24k ‘ish rpm.
Regards. Patty
|
|
|
Post by racket on Oct 22, 2020 2:39:15 GMT -5
|
|
|
Post by madpatty on Oct 22, 2020 3:25:25 GMT -5
Hi Racket. According to calculations given by Kamps and approximated according to my rotor weight and dimensions, the critical speed of the rotor comes put to be 24500 rpm. Does adding a greater OD spacer add stiffness and hence increase the critical speed of the rotor? Does adding a spacer or using an actual greater diameter shaft have same effect. According to same calculations increasing the shaft diameter (adding a spacer in my case between bearings) to 22mm increases the critical speed to 54000rpm. I think this is an experiment worth trying. What do you think? Regards. Patty
|
|
|
Post by racket on Oct 22, 2020 4:19:58 GMT -5
Hi Patty
Yep , make the spacer as large a diameter as possible, probably a tad larger than the inner raceway , maybe ~25 mm , as long as theres room left for the air and oil to pass through to the bearing .
My spacers are a finger force fit over the shaft, slides on easily but without rattle , ends MUST be square and true to prevent tilting the bearing when the whole rotor system is locked together .
The extra diameter should increase your critical speed and hopefully prevent your bearing failures.
Its a simple job to fit one and try , well worth doing before any other mods :-)
Please check that the turbine seal boss face that the bearing locates against is square to the shaft .
Cheers John
|
|
|
Post by madpatty on Oct 22, 2020 10:21:22 GMT -5
Hi Racket.
Another bearing failure at almost the similar point. I used a 24mm OD bearing spacer in a 28mm ID shaft tunnel.
What do you think is the unbalance tolerance for ball bearings. Maybe there’s some residual unbalance in the turbine wheel after repeated balancing which is causing this.
I used a pretty old method of balancing the rotor by placing in on bearings on a flat surface and then letting the heavy spot rotate down due to gravity. Normally I use a proper production grade turbo balancing machine but that machine went out of calibration and I couldn’t use it on this rotor.
I suspect there’s stil some unbalance left which couldn’t be picked up using that rudimentary method.
Regards.
|
|
|
Post by turboron on Oct 22, 2020 15:17:22 GMT -5
Patty, as John said he used a "poor man's squeeze film damper" on the outer race of his bearings. The best squeeze film dampers are sealed on both sides with oil injected between the seals and drained from the opposite side. The seals can be O-rings preferably or piston rings. The O-rings give a square pressure profile while the piston rings only have a peak pressure in the middle due to leakage on both sides. The unsealed damper on the other race does help to get through the critical speeds.
Thanks, Ron
|
|