What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Zero Force Member in Engine Mount?

Toobuilder

Toobuilder

Well Known Member
Just doing some pondering, and the lower horizontal member between the two lower engine mount ears is in the way of something I want to do. In considering a redesign of this member I cracked open some text on truss analysis and quickly came to the possibility that once the mount is attached to the fuselage it becomes a "statically indeterminate" structure, and this particular member is actually under no load. The mount ears and the tops of the gear leg sockets seem to resolve to the structure without this tube - therefore the "redesign" is irrelevant - removal is the prudent action.


34oepo6.jpg



I have not done the math on this yet, but before I do (or break out the sawzall), has anyone done the numbers on this mount? The RV-4 mount does not feature this cross tube and the load paths are the same (different magnitude, of course).

Is this tube a holdover from fabrication, perhaps required to keep the dynafocal ring true during welding?

Any thoughts on the purpose of this single tube and the load path it satisfies?
 
Last edited:
It appears to me that the cross member will go into tension with landing gear loads, although I'd need another view (from front) to be sure. As the gear sockets attempt to "pivot" under load, they put the members between their tops and the lower engine mount into compression, which in turn try to move the lower engine mounts laterally apart.

If the joints are considered to be spherical joints (meaning no bending moments on any of the members, all pure tension of compression), the thing will not be statically indeterminate. That may be a reasonable assumption, as the small amount of bending of the members is probably a very small portion of the system stiffness.

Perhaps build a somewhat to scale model with small popsicle sticks connected by blobs of rtv?
 
Mount

The Wittman gear/gear mount as originally designed has 7/8 diameter tubes from the top of the landing gear socket to the upper firewall/engine mount bushing.
The gear in the OP does not have this so the landing loads are transferred thru the short diagonal tubes from the sockets to the engine mount and then back to the firewall. This places significant tension loads on the crossmember across the bottom. Very bad idea to remove this.
if we were discussing pure engine mount loads it may or may not be ok to remove this tube. Many of the four cylinder aerobatic airplanes do not have a tube across the bottom. There is no room for a straight bottom tube on a four cylinder Lycoming. I don't remember if the six cylinder aerobatic airplanes have the crossmember.
Also back to the Wittman gear on aircraft such as the Wolf Pitts, the shallow angle of the gear truss to the firewall does not allow the bracing as Wittman intended it, so it is braced similar to the OP picture.
 
Thanks for that response. I'm still trying to picture the condition you describe.

I think wheel loads induced by "splay" are reacted by the "Z" member at the firewall. Lower member in tension, upper in compression.

Loads induced by braking are reacted by the tie to the lower longerons in compression with the lower bolts, and to the center fuselage keel in tension by the "middle" pair of bolts.
 
Truss vs Frame

Zero Force Members are characteristic to trusses, where each member terminates at a joint and there is no load on a member between joints. This structure looks a lot more like a frame (where each member does not necessarily terminate at a joint but can have other members attached to a longer continuous member) than a truss, where the tube running between the upper end of each gear socket is one continuous member with loads at both ends (provided by the gear) and at each ear. This would make the member in question a two-force member (a simplification), but not necessarily a zero force member.

The point at which each ear attaches to the fuselage could be thought of as a fixed point. Fixed points characteristically resist forces in all axes as well as moments about all axes. If the member is question is a two force member, then the points where the ears attach to the firewall would be resisting a moment equal to the axial load from the tops of the gear sockets multiplied by the distance from the ear to that member. So, the horizontal member between the ears provides a reaction moment for each ear, keeping each ear from having to resist the moment acting on it by the landing gear loads.

I'd be hesitant to remove this member because each ear attach point would have to resist a much higher moment, potentially causing buckling or bending of the firewall and surrounding material.
 
Toobuilder said:
...
I think wheel loads induced by "splay" are reacted by the "Z" member at the firewall. Lower member in tension, upper in compression.
.....
Click to expand...

Don't think of splay, think what happens if one gear leg is pushed back due to a rabbit hole...:)
 
This is a question that is best answered by the person/organisation that initially did the design engineering on the specific engine mount. Asking the question on VansAirforce is fraught with danger and relying on the advice here for making such a major structural change could potentially be disastrous for you.
 
A while back there was a video circulating here that showed the substantial movement in dynafocal mounts when an aircraft is manoeuvring. That is likely to generate significant asymmetric forces in the engine mount, so aside from any landing gear loads, I'd be willing to bet that tube takes its fair share of the bending, torsion and axial loads in the space frame, and is not a redundant member. I suggest this is a question for Vans.

Edit: Found it: https://www.youtube.com/watch?v=B0gQrDnG4oA
 
Last edited:
Captain Avgas said:
This is a question that is best answered by the person/organisation that initially did the design engineering on the specific engine mount. Asking the question on VansAirforce is fraught with danger and relying on the advice here for making such a major structural change could potentially be disastrous for you.
Click to expand...

Asking a question is rarely dangerous in itself, nor is seeking diverse opinion - even if completely unqualified. That said, I think VAF IS a good place to ask if "...anyone has run the numbers..." or "...why the -4 mount does NOT have this bar..." because not only are there plenty of engineering types here, theres a good chance that the actual designers of this mount are here.

Discussion of this type is good because unlike many found on this forum, there is only one right answer here. Black or white. It just might take a bit of digging to find it.

In any case I can assure you all that there will be no removal or redesign of this member until I have THE answer to the question. This is not subject to "experiment"... my actions will be governed by hard facts.
 
Last edited:
If you think about positive g's then the lower mounts are in compression. While the actual action from the mounts may be along the bolt axis, the actual displacement of the lower engine mounts are parallel, thus spreading the lower mount points on the frame. The horizontal tie bar accepts that separating force.

You can draw your own conclusion from that. While analyses may be perfect, 95+% of the issues are with the loading cases. You should go back and look at your assumed loads on the structure in this light.
 
Last edited:
BillL said:
If you think about positive g's then the lower mounts are in compression. While the actual action from the mounts may be along the bolt axis, the actual displacement of the lower engine mounts are parallel, thus spreading the lower mount points on the frame. The horizontal tie bar accepts that separating force...
Click to expand...

Thanks for that Bill - I am seeing that too. However, there is some hefty triangulation sends that "spreading" force into the cross bar at the firewall. A free body diagram of this load case resolves to equilibrium through the firewall cross piece. And as my structural engineering skills are limited, I'm seeking to fill any holes in my assumptions.

I know there is a lot going on with this structure, but once you eliminate known influences, anything left over is redundant.

For example, based on discussions I've had with people, there seems to be some back and forth between trying to determine if the cross piece in question satisfies landing gear loads or flight loads. It seems to me that if you look ONLY at the trapezoidal truss formed by the gear sockets and the "Z" bar that connects them (imagine cutting the rest of the mount away with a sawzall), then all landing loads are satisfied. There are four (4) 3/8 diameter bolts to handle shear loads, the beefy truss itself EASILY contains "splay" loading of the gear legs, and the fore/aft loads for braking or potholes are through the 4 bolts in tension or compression into the lower longerons and center keel structure of the fuselage.

If the above is a valid assumption, then we can ignore gear loads from here on out, right?
 
One suggestion

Toobuilder said:
In considering a redesign of this member I cracked open some text on truss analysis and quickly came to the possibility that once the mount is attached to the fuselage it becomes a "statically indeterminate" structure, and this particular member is actually under no load.
Click to expand...

This is a bad assumption. Statically indeterminant does not mean that it carries no load. The loads are shared based on their stiffness. Taking load off the bolts with the cross bar is a good thing, particularly when considering landing loads. Redundant structures are much safer than single point of failure ones.
 
keep that tube!

I have seen the investigation report of the forces that affect the F1 mount. It had a page with colors showing stress levels - there was plenty of red on that page. I can't lay my hands on that report just yet, but that tube is a necessity. We added one across the top attach points later on...

We also changed the firewall bottom/fwd cross-angle to 1x1x.125 to absorb the heavy loads from the gear.

Best to keep that tube - but you can move it up so the bottom of the tube is level with the fwd belly skin to allow for a cowling outlet fairing to be placed there.

Carry on!
Mark
 
F1Boss said:
I have seen the investigation report of the forces that affect the F1 mount. It had a page with colors showing stress levels - there was plenty of red on that page. I can't lay my hands on that report just yet, but that tube is a necessity. We added one across the top attach points later on...

We also changed the firewall bottom/fwd cross-angle to 1x1x.125 to absorb the heavy loads from the gear.

Best to keep that tube - but you can move it up so the bottom of the tube is level with the fwd belly skin to allow for a cowling outlet fairing to be placed there.

Carry on!
Mark
Click to expand...

Thanks Mark.

I would love to see that report. Are you suggesting that the cross tube in question satisfies landing loads? Because it appears that the "Z" bar trapezoid truss does that all by itself, so I'd like to see what I'm missing.
 
Here is an example of what happens to this tube when excessive side loads are applied - enough to put a permanent bend in the titanium gear leg. Not sure if this adds anything to the discussion, but another data point.

Bent%20Motor%20Mount%20(800x480).jpg
 
BJohnson said:
This is a bad assumption. Statically indeterminant does not mean that it carries no load. The loads are shared based on their stiffness. Taking load off the bolts with the cross bar is a good thing, particularly when considering landing loads. Redundant structures are much safer than single point of failure ones.
Click to expand...

I'm assuming the "classic" frictionless pin or socket joints.

I'm also seeing the cross bar there... But it's at the firewall. I do have a hard time imagining a scenario where a designer would use two members to share the same load case when a single, properly sized member will work.

Redundant structures do have their place (combat, for instance), but I'm not seeing the need here.
 
SMO said:
Here is an example of what happens to this tube when excessive side loads are applied...
Click to expand...

Wow. Did you shear one of the lower longeron bolts? If not, then I can assume that you have some deformation in the fuselage as those two lower bolts got squeezed toward each other.
 
F1Boss said:
...Best to keep that tube - but you can move it up so the bottom of the tube is level with the fwd belly skin to allow for a cowling outlet fairing to be placed there...
Click to expand...

Just reread this in a new light...

Just for clarification, the structural tube I'm talking about is NOT the lower tube at the firewall (i.e. the kinked tube in Mr. Olson's example above). It's the lower tube between the engine mount (engine side) ears. It is the one tube that is NOT used on the RV-4 mount.
 
Michael you are not going to be happy until you do it yourself. Make a structure of coat hangar wire and solder it together, then bend it to satisfaction. You might be surprised at how much can be learned so easily.

I think you may have underestimated the spreading forces of that engine mount due to vertical loading. Overhung moment, assume frictionless wedges and do the free body diagram.

With a 300 lb engine, 18" moment arm (less for the spread from upper to lower mounts) and 6 g's of loading the loads would have to be transferred to the lower structure in bending, not tension/compression. There are NO center bolts, only those at the corners of the fuse. So any load that does not triangulate at those points will be bending.

Good luck.

Edit:I should add, it might not fail outright but due to fatigue, but if these spread the engine "droop" and prop drop with deflection could be significant with g loading. Hopefully there would not be a vertical resonant vibration due to prop forces/balance.
 
Last edited:
BillL said:
...There are NO center bolts, only those at the corners of the fuse. So any load that does not triangulate at those points will be bending...
Click to expand...

Thanks Bill, I will take your comments to heart and build a model.

However, I'm unclear on your comment above... There clearly are "center" bolts on the cross bar between the upper gear sockets, and it is exactly this tie to the fuselage structure that my entire hypothesis hangs.

Are you suggesting that these bolts are irrelevant to the load case, or something else?

Just looking for clarity, not argument.
 
Toobuilder said:
Thanks Bill, I will take your comments to heart and build a model.

However, I'm unclear on your comment above... There clearly are "center" bolts on the cross bar between the upper gear sockets, and it is exactly this tie to the fuselage structure that my entire hypothesis hangs.

Are you suggesting that these bolts are irrelevant to the load case, or something else?

Just looking for clarity, not argument.
Click to expand...

Nah - No need for argument, we are all good here. Hmmm,Yes I was not seeing the bolts on the top gear socket cross bar, I was seeing across the bottom bar, sorry. Nevertheless - I don't know what structure is behind those would guess they are for stability or vibration stops, not major structural load carrier. That is a pure guess, though as I am ignorant with the aft FW structure of the 4. Does it form a channel/tunnel back to the spar?

Off topic. When my son was a sophomore in HS, there was a toothpick bridge contest. His older sister held the record for weight. He used an online statics program (with very little guidance) and made a design. He tested scarf joints and compressive stability. The bridge, with 1/8" assembled beams held 318 lbs over a 14" distance. (~3.5" tall X 3" wide)

A tension/compression triangulated structure can produce a very high load/weigh structure. Maybe, the cross tube could be greatly reduced (as it is in tension) to reduce the weight. Need numbers to verify precisely.
 
BillL said:
Nah - No need for argument, we are all good here. Hmmm,Yes I was not seeing the bolts on the top gear socket cross bar, I was seeing across the bottom bar, sorry. Nevertheless - I don't know what structure is behind those would guess they are for stability or vibration stops, not major structural load carrier. That is a pure guess, though as I am ignorant with the aft FW structure of the 4. Does it form a channel/tunnel back to the spar?...
Click to expand...

Those center 3/8 dia bolts tie to a substantial "V' shaped keel which terminates at the elevator bellcrank. It is a very significant structural element of the system, and one can consider those attach points fixed in every plane, but particularly in longitudinal loading.
 
Last edited:
34rww47.jpg


As far as I can determine, looking at the photo, you're wanting to remove tube "C." If so, then....

Tubes marked "A" are all in a plane.

Tube "B" is the tube that takes forward/aft loads from the top of the landing gear socket and reacts them.

Tube "C" is the only tube that is out of the plane of the "A" tubes, and is therefore necessary.

At least that's how it appears to me from this photo.

Dave
 
Last edited:
David Paule said:
2l91gxw.jpg


David Paule said:
[As far as I can determine, looking at the photo, you're wanting to remove tube "C."
Click to expand...

I'm not seeing your demarcation of tube "C" in the above picture, but...

David Paule said:
[Tubes marked "A" are all in a plane.
Click to expand...

Agreed

David Paule said:
[Tube "B" is the tube that takes forward/aft loads from the top of the landing gear socket and reacts them.
Click to expand...

I can sure see that, yes, but I have an alternative perspective. I think "B" is redundant (or plays a minor role) to landing gear socket loads because the top of the socket reacts into the fuselage keel structure behind the firewall through that cross tube and bolts.

That said, "B" is a primary load path (in compression) to keep the lower engine mount pads from closing together (working in concert with middle "A" and lower "A", both in tension). If the lower mount ears move inboard, compression force drives from ear to "B" to upper socket, to cross tube to bolt to "keel" (and resolves further to spar, skins, etc).

In short, what I "see" is a classic pyramid structure formed from the firewall attachments of "lower A", "Middle A" and "B" all fixing an engine mount ear right at the apex. If the engine mount ear is fixed by those three members, then the cross tube spanning between the ears is not required, right? After all, a tie between two pyramid structures does not keep either one from falling over.
Click to expand...
 
Last edited:
Toobuilder said:
....In short, what I "see" is a classic triangulated structure formed from the firewall attachments of "lower A", "Middle A" and "B" all fixing an engine mount ear right at the apex. If the engine mount ear is fixed by those three members, then the cross tube spanning between the ears is not required, right?....
Click to expand...

It's not the stability of the lower engine mount that's at risk, it's the strength of the landing gear mount.

You could download the trial version of the LISA finite element analysis program and verify that.

Dave
 
The easiest / best way to reduce drag to have a look at Tom Martins cowl exit arrangement.

The Rocket 540 mount is just barely strong enough as it is. Just imagine standing with a leg in two canoes and 2 feet space between them on water. Then add some waves. IMHO You are just begging expensive disaster by reducing anything from what you have.
 
David Paule said:
It's not the stability of the lower engine mount that's at risk, it's the strength of the landing gear mount...
Click to expand...

Copy that Dave, yes, "C" is the member in question.

And you bring up a good point which needs to be resolved. If we assume the upper socket support is not adequate with only that trapezoidal truss and its 4 attachment points to the fuselage, then I agree that "C" is required.

That said, that trapezoid should be very easy to analyze.

Clearly the four 3/8 attachment bolts are overkill in shear with a hefty safety margin, so no worries there.

landing gear splay should easily be reacted through the hefty cross members of that structure, so TLAR indicates a bunch of margin there...

So in my feeble brain all that remains is the reaction to fore/aft movement of the tires. Braking, gopher holes, etc. Lower attachment bolts tie the compression loads into the lower longerons so we know we're safe there, so we are left with the tension loads into the keel:

Cross tube, bolt bosses, and the bolts themselves.

If this truss is "good" as a stand alone structure, then "C" is redundant to landing loads, right?
 
Maybe it's for...

...basic assembly of the part. How else might you keep those points secure while building the mount? I would think a shallow X below and behind these two lower points - welded to the lower area of the sockets - might add some rigidity, but a small tube would almost need to be tacked in place to hold the 4 points in a square while the X is welded in..but that X would use more assy hours, and add some weight.

If I could remember how to add pics - I have two mounts with this X structure in place here in my shop...but no gear leg sockets are used on those two.:eek:

The assy fixture holds that square setup in place; the firewall area is welded 1st, then the welded square is set into the fixture, then the fore-aft tubes are welded in place.

Note the short tubes welded on the lower corners in SMO's picture - those are critical to the F1 mount.

Yep - was thinking you were referring to the cross-tube between the two lower firewall attachments.

This is an interesting thread indeed. Merry Christmas to all participants and readers!

Carry on!
Mark
 
Last edited:
F1R said:
...IMHO You are just begging expensive disaster by reducing anything from what you have.
Click to expand...

I'm "begging" to develop an understanding why things are done the way they are. Many homebuilts are "designed" through direct copy, TLAR or other informal means. I'm not suggesting that the F-1 or Rocket did not receive the appropriate attention, but this kind of thing happens often enough to warrant a closer look.

Fortunately, we don't have to deal with opinion in this case. There can be only one right answer, and nothing is going to change on my airplane until that answer is found.
 
You've made a hypothesis as to why that tube is unnecessary. My professional experience leads me to think that the tube is, in fact, necessary and I've explained why I think so. It can be tested either by models or by FEA, and both are available for free or inexpensively, as noted.

As a point of reference, I'm building an RV-3B. I'm not doing anything to knowingly reduce the strength of any of it, and I check with the factory from time to time if I think there's something I might have done that's problematic.

Please, before whacking that tube, verify its purpose either by test or analysis, or at least by contacting Mr. Harmon.

Dave
 
Dear Toolbuilder, as I predicted in post #7 you are not going anywhere here.

For a number of reasons, the combined engine /landing gear mount is a complex structure and as others have suggested it really needs an FEA to accurately determine intrinsic load paths.

However, if the engine mount comes from Harmon Rocket I can't for the life of me understand why you don't go directly to them for an opinion. Is it possible that you already have their opinion but don't like what you're hearing.
 
Yes Bob, this case is beyond the level of understanding by mere mortals. You have uncovered my attempt to subvert the VAF collective with a discussion which has an actual black or white answer. I will hereby withdraw back into the shadows.

And yes, next time I see John I will ask him about the tube.
 
Toobuilder said:
Yes Bob, this case is beyond the level of understanding by mere mortals.
.
Click to expand...

Michael, you started out by asking a very complex structural question about a Harmon Rocket engine mount without even an annotated photograph or a drawing of the relevant device. And then it wasn't until your post #18 that readers even became aware of the specific member you were considering removing from the mount. In IT terms I think they call it garbage in...garbage out. ;)

At any rate I think that eventually the "mere mortals" have given you a number of good suggestions which include a) Leaving the member in place. b) Referring the matter to Harmon Rocket c) Paying a professional to perform a proper analysis.
 
Last edited:
Picture and description in the first post...

...but yes, nobody on the forum has produced a loads survey so I will take it off the board since a grass roots discussion seems to be such a source of annoyance.
 
Speaking purely for myself it was clear from the first sentence of the first post which tube was being referred to. It did not take until post 18 to work it out.
I haven't a clue as to the science involved, but there is nothing wrong with the English being used to describe it!
 
Assembly?

Toobuilder said:
That is one of my theories Mark. Its there for assembly.

Certainly a possibility.
Click to expand...

You may find that the cross-member in question is absorbing some residual weld stresses from fabrication, and removing it will cause the mount to distort at the engine interface points.

Skylor
 
You must log in or register to reply here.
Back
Top