Broaching question

[tripleZ]

We are working on developing a part that has a hex shaped cavity in the center. The hex covers the entire length of the part. The size of the hex will be 0.9825"±.002" across the flats, and the part is rougly 1.5" long.

This issue is the hardness. It's sintered P/M stainless steel. Our part will have a macro hardness in the mid-30s (Rockwell C) but will have a particle hardness in the mid-50's. So we'd essentially be looking at "hard broaching". The part has essentially no elongation properties and a high UTS.

With the high cost of broaching bars, we're a little leery of trying to use our horizontal broaching machine. We've got an inquiry in on rotary broaching, but this looks like we're pushing the limits (based on what I've read).

We are looking at other methods, but I'm curious as to what your opinions might be on the broaching option.

Thanks!

 

[MikeHalloran]

Try forming the hex in the powder mold.



Mike Halloran
Pembroke Pines, FL, USA

 

[tripleZ]

We're already going to be doing that, but it's a high temp sintered part so the shrinkage is going to be greater than the normal 0.5% you'd see in conventional P/M. So the hex will be molded in, but the geometry of the entire part suggests we won't have a statistically capable process at ±.002" after sintering. Hence the need to qualify the dimension.

There are options to coin/size the ID using a coining press, but they require a slightly different and more expensive processing cycle. I'm just trying to evaluate some alternatives.

 

[TVP]

If you are not broaching the entire hex, just the periphery of it, then it should be doable. There are some phenomenal tool materials and coating options to reduce wear and fracture.

 

[unclesyd]

Could you use sinker EDM to finish the hex after molding the net shape?

I've seen splines completed in a preformed hole. I don't recall the amount of metal that was removed.

 

[tripleZ]

EDM is out of the question due to cycle time and the quantity of parts that are needed (>50K per year).

And the whole hex has to be qualified and within the profile spec. So we're broaching every surface. =(

Rotary broaching was ruled out by the vendor due to the part length, and the broaching bar company I've used in the past for my horizontal machine has used the phrase, "Whoaaaa...hmmmm, that could be fun." I get the feeling I need to pursue another path. =)

Seriously though, if anybody has done something similar, I'd love to hear about it.

 

[MikeHalloran]

Maybe it's a crazy idea, but I'd look into the possibility of making a tapered hex diamond broach.

It would have to be pretty long to carry any reasonable chip load, given the tiny 'gullets', but I think you could do it with a cable-pulled tool.





Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

A possibility:

Here are two examples where the apex of the angle in a hex has been taken out of play while driving. If your application is using the hex for driving it may be possible to adapt something similar to these for your application. both of these forms take some of the precision out of the hex cavity. If you application is antirotation you could also use this approach to take the precision out of the hex.
Note that the Ozat tools have their profile electroformed. Their quanities are large, but the question remains how many machines do they have in production.

http://www.ozat-tools.com/

http://allproducts.com/manufacture7/dess/460mm.html

I have several sets of 3 Ozat sockets of the same size from different lots and comparing the inside dimensions with a caliper I can detect no difference in each sets' dimensions.

Do you need the full length of the hex hole?

 

[tripleZ]

Yes, they need the hole length of the hex to fit a drive shaft.

The diamond hex broach sounds like a good idea, but I don't want to guess what that would cost. The horizontal 4' T15 bar was around $3700...and no clue if I'd get 10 parts or a 1000 off of it.

I've talked to several production broach shops, one broach manufacturer, and a rotary broaching supplier. Nobody wants to touch it, unless on an R&D no-guarantee basis.

I'm still waiting to hear from the high speed broach company outside of Chicago. If that's out, I think we're going to close up the broaching avenue and look into coining (squishing it to size). We'll have added steps to harden it up, but c'est la vie at this point.

Thanks for the ideas guys! This is one of those cases where the customer's "general" quote drawing and finish spec print are deviating a bit. =)

 

[MikeHalloran]

One other alternative, which would probably still require heat treatment later in the process, is to force a hex shaped carbide drift through it.. or maybe a hex 'ball' like the tip of a balldriver. You may need to mess around a bit to find a lubricant that will prevent scuffing of the carbide.





Mike Halloran
Pembroke Pines, FL, USA

 

[tripleZ]

Mike,

That's kind of what we do when we "coin" features. With a high temp PM part, there is still some porosity. So we actually will compress the part in a coining press using a die, core rod, and punches (carbide insert die, carbide core, tool steel punches). The part smooshes inwards and outwards under compression until it is limited by the die wall and core rod surfaces. Then it's ejected from the die and off of the core. You usually see a little more densification towards the surface of the part. We pretty much can qualify the dimensions that way and get a little better surface hardness. The tolerance is a wee bit tight for this part, but the method is pretty similar in principle to what you're suggesting.

It's a neat process to work with, but can be hell sometimes when trying to anticipate which way the material will move.

 

[unclesyd]

If you are not too far from dimension you might want to look at a burnishing broach. Not as forceful as the coining process but will give imprssive results. I haven't seen one for a hex but have seen them for holes and splines. It could be either a straight or button style.

http://holdenventures.com/BD.aspx

We do a lot small hole burnishing/broaching and have found out that the lubricant is key to tool life and finish. We get very good results with a Ballizing lubricant from Ball-O-Matic, but the life saver is a flash copper plating for the tougher materials.

I've been told that the PC-10 lubriant is very good for ballizing.

http://www.usalubrications.com/mpc-pc10.html

In talking with our production guy he stated that for production run as you state he would gladly give up $3500 tool to get a better feeling of where he is at.

 

[ornerynorsk]

mandrel forge after sintering? This is a tough one, to say the least

 

[TVP]

Z³,

I have been reading this thread with interest, and your second-to-last reply provided one key detail: hex broach manufactured from T15 costs ~ $3700. I think that using PM T15 will yield a tool life (if uncoated) closer to 10 pieces than 1000 pieces. This application would need something really hard and yet still has some toughness, which PM T15 really does not have. I would be looking at grades like Cartech Micro-Melt® Maxamet® Alloy, Crucible CPM® 15V®, etc., possibly with a TD coating.

 

[tripleZ]

Thanks for the update TVP. The coating spec included TiN, but I tend to agree with you that it likely won't last long (even with the coating). I think the 15V cost might add a few hundred...and would probably yield less issues with surface cracking during re-sharpening. But I would definitely prefer the toughness. Good points to consider.

 

[EdDanzer]

Have you considered a carbide broach? You should only have to remove .002” to .005” material total so it should only take 5 teeth maximum in the cut (3 to 4” long). If a HSS shank is brazed to the carbide on each end for guiding and done as a push broach tool life should be good and cost should not be $3k.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com