How in the world is this feature machined?

[cornerRad]

I thought I've seen it all but this one is a new one for me and I have absolutely no idea how to produce this feature.

They call it a "notch" but it is a raised feature extending axially down the length of a shaft and terminates at a shoulder.

My initial thought was some type of specialty knurl but I don't see how you could get it to track properly with only 4 notches.

This is a current production part so someone somewhere knows how to make this. Anyone got any ideas?

 

[Milland]

It the "notches" are a raised feature, sort of a "dig into the bore" type thing, I'd imagine they were pressed into the part by four segments similar to how hydraulic hoses get their end fittings attached - by swaging. But not sure as the image isn't clear.

And how it would be done without an upset to the diameter that's reasonably tightly controlled would be tricky.

 

[Garwood]

Looks like a punch press op to me.

 

[BOB-OO]

+1 punch press, got a pic of the finished part?

 

[cornerRad]

+1 punch press, got a pic of the finished part?

I don't unfortunately.

For punch press, are you figuring a wedge style end pressing into the shaft? I can see how that could work. Though, can a punch press be used with a indexer/rotary? Otherwise I don't see how you'd reliably keep the notch spacing and also avoid damaging the 32Ra.

The material is heat treated ~50Rc and annual quantities in the thousands so we definitely want stable process.

 

[SeymourDumore]

Otherwise I don't see how you'd reliably keep the notch spacing and also avoid damaging the 32Ra.
The material is heat treated ~50Rc and annual quantities in the thousands so we definitely want stable process.

My vote: Cold heading:

 

[Strostkovy]

It would be really funny if those are just marks from the clamp of the press that stamps a head on, and they are completely unnecessary.

I have seen strange things from people who hire engineers straight out of college for reverse engineering.

 

[Garwood]

I don't unfortunately.

For punch press, are you figuring a wedge style end pressing into the shaft? I can see how that could work. Though, can a punch press be used with a indexer/rotary? Otherwise I don't see how you'd reliably keep the notch spacing and also avoid damaging the 32Ra.

The material is heat treated ~50Rc and annual quantities in the thousands so we definitely want stable process.

I would build a basic tool that did all four grooves in one hit.

Or, yeah, you could use an indexer on a punch, but that sounds real slow.

The tolerance on that thing is pretty wide open. Should be easy for anyone with experience to figure out.

 

[Milland]

So is this an indented trench with the side edges upset to the specified diameter? If so, then obviously a formed punch would do it, and if mass produced it would be similar to a swaging process.

 

[john.k]

The axles in toys used to have four long keys pressed up on the basic diameter to hold the pot metal wheels.

 

[DDoug]

So is this an indented trench with the side edges upset to the specified diameter? If so, then obviously a formed punch would do it, and if mass produced it would be similar to a swaging process.

I was thinking rotary swager.

 

[Ox]

That is NOT a post-opp process as the spec is only +/- .0015 over those features.
I would say that you would need to run a tubing cut-off wheel (?) down the shaft (not up in the corners quite as shown) and then go back and take another turn pass to qualify diameter.


--------------------

Think Snow Eh!
Ox

 

[Milland]

That is NOT a post-opp process as the spec is only +/- .0015 over those features.
I would say that you would need to run a tubing cut-off wheel (?) down the shaft (not up in the corners quite as shown) and then go back and take another turn pass to qualify diameter.


--------------------

Think Snow Eh!
Ox

But there's the rub - the drawing does show a "to the shoulder" feature, a wheeled element wouldn't do that. So a pic of the real part would be nice to prove this isn't somebody's version of a Klein Bottle...

 

[50BMG DUDE]

Try a Slater external rotary broach

 

[cornerRad]

That is NOT a post-opp process as the spec is only +/- .0015 over those features.
I would say that you would need to run a tubing cut-off wheel (?) down the shaft (not up in the corners quite as shown) and then go back and take another turn pass to qualify diameter.


--------------------

Think Snow Eh!
Ox

I guess you lost me. If the max shaft diameter is 0.249 and min distance over the "notches" is 0.251. How would you get that result from this method? Wouldn't a cut off wheel remove material and just leave you with a recessed feature, not raised?

I will say that this is for a large company and just one of the dozen different parts they sent us with this feature. They are 100% real parts and someone else has been making them, original print goes back to 2008.

 

[implmex]

Hi cornerRad:
Is there a chance they could have been drawn through a die and then cold headed?
Obviously they'd have to remove the ribs from the right hand end, and the cold heading would have to upset an awful lot of metal.

If you could inspect some of them under a microscope maybe you could get some clues from the orientation of the toolmarks.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com

 

[Ox]

It would essentially be a knurling type opperation.


---------------

Think Snow Eh!
Ox

 

[Fal Grunt]

If they are a item currently in production, ask them to send a sample part. I ask customers for sample parts all the time, especially because I have often found the part, and the print, are two different things. I did a job for a customer that had the thread form dimensioned to 5 decimal places. When I asked how they reached those numbers, they said they used a pair of calipers.

Having the part in hand also clues me into what the customer "thinks" is a good part. Having the part in hand should clue you in on how they are made.

In other words, to me it is apparent that they are shopping a currently in production part. Are they having issues with the current supplier? What are the issues?

Great idea on the cutoff wheel Ox, essentially swaging the material up.

It is really hard to tell from the picture what the "form" of the notch is. Is it radiused, square, a triangle, or an indentation?

 

[Ox]

Well, that's the only way that I see it happ'nin.
They are asking you to raise that feature .001" per side +/- very little.
Only way that I can think to doo that is to turn it, knurl it, and then re-kiss the OD - all w/o un-chucking it.

IDK how you would doo it any other way.


At first it appears to be a swiss job, but the end view does show a center in it - and you would need one for both the knurl and the re-kiss opps. I'm sure that a live center could be slipped down in a swiss, but just saying that - as I see it - that is a live center job.


-----------------

Think Snow Eh!
Ox

 

[SeymourDumore]

Is there a chance they could have been drawn through a die and then cold headed?
Obviously they'd have to remove the ribs from the right hand end, and the cold heading would have to upset an awful lot of metal.

Marcus, I think you'd be surprised how much force these guys are working with!
Here is a local aerospace fastener shop I do some wire EDM work for, and talking with one of the toolmakers, he says some of the carbide inserts for the dies they use
have a .015 to .020 PRESS FIT!!

That looks like a narrow width upset of .003 on a .250 dia ...
I guess I can ask him if that's possible.

The removal of the rib on the right end might be possible by just wiping it off in the die at the same time as the front is upsetted, though there like be some visible markings remaining.