Making a crank stronger

[Matty F]

Here's a diagram of a crank driven by a gearbox. The existing arrangement is on the left:

crank pushes a half ton weight backwards and forwards a distance of three inches, five times per second. After a few years the bolt keeps loosening or breaking, and the bearing surround is cracking. Is that a surprise?

The gearbox and shaft and mandrel cannot be altered. I propose bolting a steel pipe around the mandrel and welding that to a plate with an offset shaft, and using a much bigger roller bearing.

Or if there was some existing product that would do the job, I'd consider that. Any other brilliant ideas?

[Jules Richardson]

I always look forward to your questions :-) What you propose should work, I'd think - although I'd question whether bolting the pipe to the mandrel falls foul of "The gearbox and shaft and mandrel cannot be altered"? Or are there convenient bolt-holes in the perimeter of the mandrel already?

Presumably there's enough clearance between the crank and the mandrel for the plate that you mention to fit without needing to offset* the crank somehow? My only other suggestion - assuming you'd not be making use of existing holes in the mandrel - would be to bolt the plate to the face of the mandrel and forget about the welding and steel tubing, but that does assume there's enough space to work with (given that you'd have the combined width of the steel plate and mounting bolts to deal with, rather than just the plate in your version)

• or are you planning on this? I note it's not on the list of things which can't be altered :-)

cheers

Jules

[Matty F]

The mandrel is immovably attached to the shaft and the gearbox and motor are too huge to lift and take into the workshop. I can drill and tap into the mandrel. There are already a number of holes in it (and broken bolts) from failed repair attempts.

There's plenty of room. The face of the mandrel doesn't have a lot of space on it, what with the large shaft going through it and a couple of holes with damaged thread. I'll make the pipe a tight fit and put three or four bolts through to the side of the mandrel. At the moment there are two set screws going against the thread of the old bolt to stop it turning. That's horrible!

[Robin]

Another idiot boy question but if I don't ask.....

Is it possible that the bolt/bearing acts as a "damper" for vibration transmitted through the crank which might otherwise do some harm to the shaft/mandrel/gearbox with a more rigid coupling?

[Andrew Mawson]

bearing.

Is it not a case of the crank pin (bolt) being under specified? Is the weight guided in any way or only by the crank pin? If so maybe the alignment is not perfect. I'd want the bolt to be replaced by one significantly fatter and made from a good medium carbon or alloy steel, with good provision for lubrication. could the bearing be replaced by a spherical self aligning one?

AWEM

[Matty F]

Well, I suppose I could add a spring to the crank. But it otherwise has been running OK for some 40 years. Just the bolt needs replacing often, and the hole it is screwed to in the mandrel.

[Matty F]

Yes the bolt is vastly under specified even though the bolt that has been put in is high tensile. There's no room in the mandrel for a fatter bolt. And the roller bearing is way too small. I thought someone would have asked by now, but the crank is pushing a whole room full of people and furniture backwards and forwards around five times per second! I think it's mounted on roller bearings.

[polygonum]

Is it a figment of my imagination? Or the drawing? Or your proposal?

But the distance from mandrel to crank shaft appears shorter in the proposal. And that is in fact one of the things that sprang to mind. The shorter that is the less leverage applied to the bolt where it enters the mandrel.

I think part of the advantage of your proposal is the reduction in effective length of the bolt. But I'd also take on board the other comments here about acting as a shock absorber. If that is the case, maybe some absorbent material inside the new pipe and plate would be advantageous?

Rod

[Matty F]

The sideways cranking distance is supposed to be the same. But it's just a sketch made from memory.

I don't think a shock absorber is necessary. But I'll find out if it breaks!

[Tim Lamb]

In message , Matty F writes

Ah! The earthquake machine!

The purpose of the *pipe* seems to be to transfer the reciprocating load from the bolted pivot to the outside of the mandrel. An alternative might be to locate and support the spigot plate with plain pins and use recessed cap head screws to secure it to the mandrel. Presumably you are anxious to minimise the thickness of the plate.

A thought on the bearing.... SKF must still have a technical dept. within the sales group who could advise on a suitable bearing. Self-aligning roller sounds likely. The sort you see on rail wagons:-)

regards

[Matty F]

The flat end of the mandrel does not have enough space on it as there are already holes in it with broken bolts inside. The side of the mandrel is in good condition.

Presumably you are

I don't think the size of the plate matters. It's going round at a constant speed like a flywheel. If I had a block of steel big enough I would dispense with the pipe and make the whole thing in one piece. That's my specialty, and welding isn't.

The first job is to find a suitable roller bearing lying around, free of course! The bearings on each end of 50 HP tram motors are a good size.

[Thomas Prufer]

Earthquake simulator?

I'd try to lose the thing marked "bolt" on the repair, and just go with the things marked "bolts". "bolt" is just held in with setscrews, and will cause trouble aligning the cap thing...

Thomas Prufer

[Matty F]

Yes maybe. I suspect that the thread in the hole is stuffed anyway. But I though I would need to put washers on each side of the bearing to hold it on. Whatever I do would be stronger than what's there.

[Tim Lamb]

In message , Matty F writes

E -mail doesn't cost and you can ask what are the common applications as an indication where to start looking. You can estimate the mass of the room + furniture + people and you already know the rotational speed and crank throw. [e-mail]

They were very helpful people when I worked there nearly 30 years back:-)

regards

[geoff]

In message , Robin writes

I think it's more a matter of force = mass x acceleration

the bolt isn't up to the task

What's the bolt made of, steel?

Maybe splash out on a bit of titanium or summat

... summat's good

[geoff]

In message , Matty F writes

Smaller earthquake or ...

Atkins diet all round

[MuddyMike]

The plan looks good to me. The big improvement will come from the designed reduction in the gap between mandrel and bearing combined with the increased diameter of the shaft that goes through the bearing over that of the bolt alone. I assume the black area surrounding the original bolt is a piece of thick walled tube which will be welded to the backplate of the big piece of tube as well as having the bolt through it. It would be a good idea to either have that thick walled tube machined, or weld on a washer to prevent the bearing moving sideways, with a second large washer under the head of the bolt.

I would also use more than one bolt through the wall of the big tube to clamp it onto the mandrel. Welding nuts on the outside will give greater depth of thread than just tapping the tube.

Please post picture of the before and after.

Mike

[Andy Dingley]

Don't use a crank. Switch it to an eccentric. You avoid the overhung crankpin and bearing and you can also use a much bigger eccentric sheave. The downside is thatyou have a much bigger bearing surface in the eccentric sheave, so you probably need to source a scrap-price large ball race before commencing.

Otherwise don't make a crank, use an existing one. Take a small scrapyard engine (light diesel for preference) and recycle 1/4(ish) of the block, with the crankshaft, one con-rod and one piston as a crosshead.

If you do this with a wet liner engine (a Rover K is ideal, as it has a main bearing ladder) you can even throw away most of the block and just weld brackets to hold the wet liner (which is now a cylindrical trunk guide) in position with the crank bearings.

[tony sayer]

'Com on Matty fess up!, witch bit of engineering is it this time?..

[andrew]

Isn't it likely to be a fatigue problem from the bending stress?

What about an outrigger bearing on the opposite side with the crank mirrored. The bearing could be a self aligning plumber block.

AJH