To close tolerance, or to not close tolerance, that is the question..... 1

[Feynman81]

Hello everybody,
I'm designing a medium size drone for maritime applications and having a bit of trouble with designing the bolted connection between main frame and engine....

The requirement from the customer is that:
- the engine should be easy and quick to remove.
- any engine must be able to easily assembled on every main frame (interchangeability).

The engine itself is about 25 kg (50 pounds), and there are 6 1/4 inch bolts that secure it to the main frame.

The problem is that if I use the NAS 618 -close reamed fit- standard we have problems in the assembly/disassembly because we need a mallet to drive the bolt shank in and out and I'm afraid that after few installation/removal of the engine we will ruin the holes. On top of that having the requirement of interchangeability means no codrilling allowed so machining the parts is pretty expensive (jigs etc..).

So I'm wondering if I can just use regular clearance fit for this connection given that the 6 bolts and definetely oversized.

I guess the more general question is "what kind of connection do you use when you have to join parts subjected to vibrations that needs frequent assembly/disassembly?"

I would very much appreciate anybody sharing their experience.

Thanks,

 

[rb1957]

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[CWB1]

Somebody did a static analysis considering that the connection has to work by transmitting shear only through the shank, no friction contribution allowed, since then this has become a requirement within the company.

That makes no sense. Clearance has effectively zero impact on a frictionless joint's shear strength. I would push for a standard BJA and standard clearances by pointing out the current requirement needlessly increases manufacturing and backend costs.

The engine is bolted down directly onto the frame and the transmission has a semirigid coupling (of the elastomer type).

Rigidly mounted automotive powertrains have a lot of issues with fatigue, not only bc of engine motion and chassis flex but also internal stresses created during assembly by tolerancing. Even on a fully isolated powertrain best practice when replacing an engine is to loosen transmission mounts and allow bolted joints to slip so everything reaches a "stress-free" state to prevent overstressing isolators. Location and distance between mounts, and attention to detail otherwise are important.

 

[TugboatEng]

The structure is aluminum, no? Put it in the oven at 250°F prior to assembly or disassembly. Aluminum has a very high thermal expansion coefficient so a few degrees of heating will put clearance on the fasteners for removal and installation.

Or, let's go back to the dowels. If you don't want to make extra holes use a hollow dowel that fits with tight tolerance and run the bolt through the dowel with clearance.

 

[3DDave]

The motor has thermal lag, so make it about 4-8 hours to soak at that temp.

 

[rb1957]

but what about all the non metallic bits associated with an engine ?

Rigid attachment of an engine makes, IMHO, no sense. Maybe use rubber bushings as isolators ?

Are these engines supposed to be interchangeable ? That'll be hard to achieve with tight tolerances.

As I said before, any joint that needs a mallet to take it apart, isn't meant to be taken apart. But, this could be a design philosophy, which would make the tight tolerance, permanent joint "sensible", or at least "rational".

Again, how do other drone manufacturers do it ?

And remember this is only a 50 lbs engine, 8 lbs of dead weight on each attachment point (yes, I know, manoeuvre load factors and engine torque are important to the loads). But also, I suspect you're using 1/4" bolts (or maybe 3/8" ?) which have an enormous strength for your application.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[TugboatEng]

Non-metallic components and lubricants in engine applications should be able to tolerate short-term exposure to temperatures exceeding 250°F.

Traditionally, doweled components would require re-doweling if one of the pair is replaced but it is currently possible to achieve sufficient precision during manufacturing where this shouldn't be the case.

 

[verymadmac]

Its a 50 lb engine, depending on the space frame geometry (principly the bracing goes to one bolt) of the engine mount you could preload the legs of the engine mount, by requiring the slightly deflection to the tube for the 2nd though 4th bolts to line up. Thus giving you clearance bolt hole and limited actual play in service.

Whether its a good idea depends on the actual loads and geometry.

 

[Ng2020]

Vmm: given the mount material potentially has poor stress-corrosion resistance (apparently 7075), and it operates in a maritime environment, preload might not be a good idea.

 

[verymadmac]

Ng2020 i had forgot that bit, was thinking of the usual spindly 4130 tube. 7075 is likely problamatic regardless of attachment method.