Rotax 912 Movement

[Blueadventures]

2 hours ago, rodgerc said:

If it were me, I’d initially keep life simple by sticking with all 22001 dimensions except the OD. 
 

Some (very) basic stuff assuming a 70kg engine and prop combo….Call it 700 Newtons for simplicity. 
 

Original mount net area 790mm^2 x 4off = 3,160mm^2. 
 

Compressive stress = 790N/3,160mm^2 = 250kPa

 

 

With a 1-1/2” OD the total bearing area increases to 4,216mm^2 or 33% more. So compressive stress decreases to 790N/4,216mm^2 = 187kPa.
 

By implication, the compressive strain on the mounts will also be decreased by about 1/3 and along the way, you’ll have increased the surface area to dissipate all that vibration (kinetic energy) heat into the aluminium mounting rails.
 

Naturally, the back-up washers will also need to be increased in diameter.

 

It’s pretty easy to cast urethane and there’s a bunch of videos on YouTube showing people doing it at home….Search “casting polyurethane bushings” and pour yourself a drink. 
 

If you’re still unsure of your ability (it’s easy) then pop over to 33 Holloway Dr Bayswater and talk to Melbourne Polyurethane. 

How would some purpose made cup washers go at retaining the rubber shape and compressibility better?

[onetrack]

I've tried casting urethane rubber mounts and despite the initial result looking good, I found that within 12 mths, the urethane deteriorated to the point where I could pull the mount apart easily - it tore like old, rotten material.

I was extremely disappointed in the result, and I don't know what I did wrong, I followed all the mixing and casting instructions to the letter.

I had no problems making the silicone molds, it was the urethane casting that was a total failure - but only after it had set for the 12 mths.

[skippydiesel]

13 hours ago, Blueadventures said:

How would some purpose made cup washers go at retaining the rubber shape and compressibility better?

You have made me think again - My font rubbers (as shown above) have stayed in place, as assembled - My rear bottom rubbers have not (see below). Bit hard to see but the bottom rubbers have slipped to the left/rear, with the right side being the worst. The only difference between the two set ups is, an extra washer in the fronts (to give a very small change of prop thrust angle). Could the small increase in compression caused by the addition of a single washer, to the back mounts,  be at least a partial answer to engine movement?

 

 

[Blueadventures]

37 minutes ago, skippydiesel said:

You have made me think again - My font rubbers (as shown above) have stayed in place, as assembled - My rear bottom rubbers have not (see below). Bit hard to see but the bottom rubbers have slipped to the left/rear, with the right side being the worst. The only difference between the two set ups is, an extra washer in the fronts (to give a very small change of prop thrust angle). Could the small increase in compression caused by the addition of a single washer, to the back mounts,  be at least a partial answer to engine movement?

 

 

Could even TIG some thin wall tube to the appropriate size washer to make a cup washer.  Worth a try will retain rubber insitsu.

[skippydiesel]

Hi BlueA.

 

The Barry mount washers are flat, most automotive applications, similar to this, have dished/concave retaining washers  - presumably to address  any potential for slipping out of alignment.  Could go with your idea or try and source concave washer of the same dimensions - will Google.

[Area-51]

The purpose of the mount is to "isolate" vibration; its referred to as the "Modal" factor/s in design engineering. The Modal is a resonant frequency of the individual component in each assembly; and assembly as a whole;  there are more than one and generally the first four are targeted for identification.

Determining the Modal specification should not be taken lightly because it is the determining factor in component fatigue that decides safe operating ranges across the entire allowable operational envelope. e.g 3500rpm max 5.0min rated, or 3300rpm continuous rated for same component. This is what engineers are paid to do. It is the same reason certain materials are specified for certain environments speeds loads and stress cycles. This is the scientific side of design and fabrication.

 

That being said, have seen similar mount setup used in several 5XX and 9XX aircraft without long term issue using 3.0mm carbon steel angle iron.

 

Both sides of the mount need to be kept rigid and in respective position; Cups and crush tubes are required, and the isolator needs to be the only flexing part; softest possible combination that does not self destruct in service or lose form will work best. The isolator material needs to be chemical resistant, and the entire frame assembly should not be able to seperate from each other should all rubbers perish.

 

Correct grade of fixings is another Modal related factor for consideration.

 

The FAA Acceptable Method publications are full of practical and relevant Aircraft related information that is fully explained and easy to understand and available free in PDF format off the website. Everybody should have and use them as reference documents.

[skippydiesel]

As an experiment - I have this day added a 1.5 mm thick washer, to the rear mounts,  between the top rubber & engine mount (as I had on the front).

 

The bottom rubbers were quite distorted/ID ovallised due to slipping sideways - seem to have gone back OK/ not quite like the front as you can see.

 

My hope is the small increase in compression will "stiffen" the back mounts  ie match the front.

 

[Blueadventures]

You mention ed a tube or something about being added up the center of the rubber mounts?  Have you added these.  I replaced the engine mount rubbers on a friends Skyranger and they have a steel tube in the centre of the rubber mount and the rubbers slide inside themselves.  May assist preventing distortion.

Edited January 2 by Blueadventures

[skippydiesel]

Hi Blue,

 

Yes the Hutchinson/Barry Control Mounts have the same sort of arrangement.  There appears to be a supplier in Australia "SVT" Milperra, NSW 2214 Phone 02 97710444.

 

 

 

There are other makes eg https://www.mackayrubber.com.au/vibrationisolators

 

[facthunter]

All of these would more appropriately be  for a fuel tank  or oil tank, not an engine. A cantilevered bolt straight into the aluminium crankcase? It's got bending loads. Not really acceptable. .   Nev

[skippydiesel]

4 hours ago, facthunter said:

All of these would more appropriately be  for a fuel tank  or oil tank, not an engine. A cantilevered bolt straight into the aluminium crankcase? It's got bending loads. Not really acceptable. .   Nev

I don't disagree with you Nev - but could do with some constructive criticism ie working with what I have.

 

FYI - Rotax Engine Mounting ; General consensus is, that a ring mount is the best. There is a OM Rotax ring mount, however Sonex has yet to adopt it (I live in hope). My last Rotax/Aircraft used an 8 point "rubber" isolator system that worked (very fiddly to get engine in/out).

[facthunter]

I have been trying to think about it for sure. I suggest  something "off the hook" would suit you Best...  Nev

[pluessy]

Unfortunatley, you are trying to put a bandaid on a very poor design. This will stay a very poor design not matter how many bandaids you are trying. The root cause of the issue is the very close mounting of the rubber isolators. Stiffening the current mounts will have other effects as other people have pointed out, many will not show up until much later (loosening rivets from high-frequency vibrations etc).

The only way out of this is a new mount/adaptor that moves the rubber mounts further out.

[facthunter]

Tony Bingellis has a book. Firewall forward Try that.. It's not as if you have a unique motor and they all have to be mounted A firmly direct bolted cradle onto the motor would be the start of it. 

[skippydiesel]

29 minutes ago, pluessy said:

Unfortunatley, you are trying to put a bandaid on a very poor design. This will stay a very poor design not matter how many bandaids you are trying. The root cause of the issue is the very close mounting of the rubber isolators. Stiffening the current mounts will have other effects as other people have pointed out, many will not show up until much later (loosening rivets from high-frequency vibrations etc).

The only way out of this is a new mount/adaptor that moves the rubber mounts further out.

Understood well before purchasing this almost complete project - Now that I have it and it is actually flying, with the odd teething problem, as this is. I am committed to making the best of what I have at least until my fiances have recovered & Sonex comes up with a better design (preferably from scratch, rather than an adapter).

 

The Sonex factories belated recognition of Rotax, is at best a cheap & nasty engine adapter. I can only hope that this is an interim measure, that will soon  be followed by a more "airworthy" engine mount  not too far down the track. If they go this route, I will seriously consider purchasing. In the meantime I would like, if possible, to improve what I have ie stop the engine from rubbing/striking the cowling & engine frame.

 

The cowling is not such an issue - some careful grinding/cutting, application of fiberglass & a little more space is made. The striking of the original Sonex engine frame, can not be so easily corrected and must be stopped.

 

The bump stop (Thruster 88) is a good idea ... but how to mount it?

 

Cupped/concave washers  - following up - yet to find exactly what I am after.

 

Wider "shoulders" on the mounts to resist movement without impacting vibration damping - great idea - will follow up with custom makers (doesn't sound like something I want/can fabricate for an aircraft application)

 

Other than doom & gloom, negative criticism, I have heard that there will actually be a problem with, stepping up one level (14), to a stiffer mount - may be just replacing the bottom 4 rubbers, that principally resist lateral movement, with stiffer 14 versions, leaving the 13s on top, could be the way to go??

 

[facthunter]

You know. you do this every time, Skip and we get called doom and gloom and negative etc and I ask myself "why do  I continue to fall for it? Every time I put in a lot of effort as do others.   Nev

[Thruster88]

I think my solution of bump stops has some merit. With a piece of 50x10 or 12mm aluminium bar under the head of longer bolts it can work in both directions at each mount. The bump stops are only needed for starting and stopping. The existing mounts remain unchanged if they are doing an acceptable job in flight.

[skippydiesel]

28 minutes ago, facthunter said:

You know. you do this every time, Skip and we get called doom and gloom and negative etc and I ask myself "why do  I continue to fall for it? Every time I put in a lot of effort as do others.   Nev

Nev me old mate - I respect your wealth of knowledge/experince in the aviation world I am but an ignorant farmer. No offence intended friend BUT pointing out the obvious deficiencies in a design, that is up & working/ flying like very fast bird, not unsafe (for the moment),  without giving workable (as in cost/time efficient) solutions/suggestions is in my book, not helpful.

 

Sure a better design would  avoided my need for advice in the first instance and may in the future be the only long term (& costly) solution - this I acknowledged right at/near the start of this conversation.

 

What I need now, is some way of improving what I have.

 

I have referred this problem to Sonex (with some of the suggestions from this Forum) - I await their considered response.

[facthunter]

There isn't a simple fix. I didn't realise how awful the original is till I had a good look at it. It's not just a matter of something hitting the cowl.  Nev

[LoonyBob]

4 hours ago, skippydiesel said:

Nev me old mate - I respect your wealth of knowledge/experince in the aviation world I am but an ignorant farmer. No offence intended friend BUT pointing out the obvious deficiencies in a design, that is up & working/ flying like very fast bird, not unsafe (for the moment),  without giving workable (as in cost/time efficient) solutions/suggestions is in my book, not helpful.

 

Sure a better design would  avoided my need for advice in the first instance and may in the future be the only long term (& costly) solution - this I acknowledged right at/near the start of this conversation.

 

What I need now, is some way of improving what I have.

 

I have referred this problem to Sonex (with some of the suggestions from this Forum) - I await their considered response.

Ahem (sweats visibly...) there are several principles at play, with regard to shockmounting an aircraft engine.

 

Before getting to that, is your installation passing through a resonance event on startup or shutdown? Watch, say, a PA28-140* as the ignition is cut. A s the revs drop, the engine shakes the whole aircraft, and looks about the escape. This is the engine/mount/rubber system passing through a resonance; but because the resonant frequency is below the idle rpm, the energy involved is not disasterous... if your are resonating, a lighter prop - especially a fixed-pitch wood prop - should help a lot.

The 9XX typically has a lower prop RPM that a Jabiru or VW, which means larger torque pulses at a lower frequency...

 

*or anything with an O-320 on a ringmount... the Cherokee is just an outstanding example!

 

Now, back to those principles...

 

A basic and very important one is to separate the rubber in shear and the rubber in compression cases... rubber has a much lower stiffness in shear than in compression. This is handy, if one uses the rubber in compression to support the weight of the engine, but in shear to allow the torsional oscillations to take place without being transmitted to the airframe. That's right, we want to allow the engine to oscillate in response to the torque pulses without restraint, but to restrain it against the average torque what's pushing the prop around... (One of the advantages of the 'flat" or "boxer' engine layout is that it will have a small amplitude oscillation in response to the torque pulses (more so at higher revs)).

 

The nifty enginemount pictured in Rogerc's post, shows the traditional approach of aiming the lines of action of the compression axes through the powerplant CG (or close to it, depending on prop weight...); this way, the engine wobbling about the crankshaft only loads the (theoretically perfect) rubbers in shear, but the awesome weight is carried in compression.

 

From the point of view of the rubbers doing their bit, that Sonex enginemount is actually not too awful; because the shockmounts (rubbers) are close to the centreline, most of the rotational vibration is seen as shear, and because they're vertical, all of the weight is seen as compression They look a tad squashed, actuallyu, which means a bit small for the weight). Iff'n you were to replace them with slightly largerer shockmounts of the same Durometer rating, the load passed through the rubber due to weight would be exactly the same. The rotational vibration would be more restrained in amplitude - less hitting things! - at the expence of transmitting more vibration / load into the engine mount metalwork & airframe, AND slightly increasing the resonant frequency (see above - if one pushes the resonant frequency into the operating range of the engine, bad things can happed...).

 

If you were to pull out your magic wand and move the holes in the aluminium further outboard, so the rubbers were a tad further apart, you would also reduce the rotational freedom, and reduce the bending moment on those crude lumps of angle. Again, the reduced angular (rotational) freedom comes at the cost of higher forces transmitted into the metalwork. (If you increase the load through the rubbers, but reduce the leverage on the flange, you could end up with null effect on the most obvious stress concentration...)

 

Assuming resonance is not the culprit, and Sonex did their fatigue calculations i.a.w. MIL-HDBK-5, I would not be alarmed by the results of going up one size or  ~10 Durometer numbers with the rubbers, as far as the engine mount metalwork is concerned. I WOULD be very wary of resonance for a while, and (subject to calculation) I'd make checking for cracks in the lower Rose angle radii, a part of my DI.

 

The crude way to check for large-amplitude resonance is to sit in a PA-28-140 with a fixed pitch McHartzell during shutdown, and feel the vibe. If your Sonex does or starts doing that, you have a problem; kick Sonex...

The engineering way to check for resonances is to use a V-gauge, which in this case would simply be a shallow V lying on its side, about 30mm long, drawn on the side of the cowl about midway. Set up a video camera or observer on a tripod, about 3m away from the side of the cowl, and start up, and run through as much of the rev range as practicable. On reviewing the video, if the V starts to look like an X at any stage, that's a narrow-frequency powerful vibration, probably resonant in origin. Note that the engine will almost certainly be moving a lot more than the cowl...

 

Note that a cradle mount normally reduces the amplitude of the engine movement, by transmitting more load into the airframe. A given engine/prop combo will run quite happily hanging on the end of a rope, provided it can magically offset itself to react torque; ALL engine mounts restrain the engine, and the more a given engine is restrained, the bigger the loads (vibrational forces) transmitted...

 

Edited January 3 by LoonyBob

[turboplanner]

6 hours ago, LoonyBob said:

Watch, say, a PA28-140* as the ignition is cut. A s the revs drop, the engine shakes the whole aircraft, and looks about the escape.

 

This can be stopped by adopting this Shut Down procedure:  Mixture full lean > Master off > Mags off > Fuel off.

[facthunter]

Having the throttle fully closed reduces al that. It stops the cylinders getting air in reducing the effective compression to almost nothing.  A soft start can be done the same way. Commence with throttle closed.  Nev

[LoonyBob]

4 hours ago, facthunter said:

Having the throttle fully closed reduces al that. It stops the cylinders getting air in reducing the effective compression to almost nothing.  A soft start can be done the same way. Commence with throttle closed.  Nev

True this, and also turboplanner's suggestion; however, the primary benefit of both w.r.t. resonance is that both minimise the rpm before ignition cut, to some way below "normal" idle, thus reducing the rotational energy quite markedly, and so the drama.

 

I seem to recall that our 140 started to shake like a wet dog about 50 hours after we purchased it; a friendly LAME found the rubbers to be inside the wear limit, but advised their replacement, which reduced the effect to a shiver (I think we used full lean before switches off, after that). The local flying school had a couple of 172's, one of which had the same trick, and also a private Grumman on the same base. Dad's Auster never did it!

 

You will no doubt be familiar with a number of GA aircraft that have a "no sustained operation" (or words to that effect) rpm range between idle and full power; most of these are to avoid crankshaft or prop resonances, but some also relate to powerplant/shockmount resonances.

[turboplanner]

11 hours ago, turboplanner said:

This can be stopped by adopting this Shut Down procedure:  Mixture full lean > Master off > Mags off > Fuel off.

The Cherokee is not a Commodore where the key is used to stop the engine; the Cherokee engine is stopped by fuel cut off, not mags.

 

These are directly from the Cherokee Owner's Manual:

 

Page 5, Mixture Control: "aft position idle cut-off for stopping the engine.

 

"Section 11, Operating Instructions

To stop the engine after landing, pull the mixture control full back to idle cut-off.

After the engine stops turn magneto and master switches off."

 

Throttle open 1/8" is the start-up position, so full fuel in the system at that point, so no surprises if there is an erratitc stop if you turn the key off while the prop is rotating.

 

[LoonyBob]

7 hours ago, turboplanner said:

The Cherokee is not a Commodore where the key is used to stop the engine; the Cherokee engine is stopped by fuel cut off, not mags.

 

These are directly from the Cherokee Owner's Manual:

 

Page 5, Mixture Control: "aft position idle cut-off for stopping the engine.

 

"Section 11, Operating Instructions

To stop the engine after landing, pull the mixture control full back to idle cut-off.

After the engine stops turn magneto and master switches off."

 

Throttle open 1/8" is the start-up position, so full fuel in the system at that point, so no surprises if there is an erratitc stop if you turn the key off while the prop is rotating.

 

i wasn't flying it much myself, aged 10...