Delicate part holding suggestions.

[nlancaster]

I have a prototyping job coming, that could turn into a long running production order but small quantities, 10-20 a month.

The part has a cylindrical shape at the tip that gets cut into 8 segments.

And this conical tip is glued to a ceramic isolator, and gets fully separated into 8 segments.

the part is about .375 in diameter, with a thru hole of .1575 and slots all the way out to the full diameter (outer slots cut before glueup)

I have to ground this and the only way I can think of is a titanium flexture, think a cone shaped spring with 8 parts touching each segment.

I am really worried about how delicate the glue bond is going to be, we make many parts already for this customers with glued titanium segments.

Most of them have way more glue area.

So my question is, is there any material, grease etc, that I can use to increase the grounding contact without having to use too much spring pressure?

 

[implmex]

Hi nlancaster:
Do I understand correctly that your parts are pie shaped segments cut from a pipe that's 3/8" diameter with a center hole of 4 mm?
You are looking to segment the blank into 8 pieces and I presume you don't want any little wire cut exit remnants.

Do you need exquisite precision?

A way that springs to mind for me is to do something like this:

I'd pop a short semi-sacrificial pin into the cylindrical blank, then squeeze it in my block, thread through each hole in turn, and nip out my cuts.
Pinching the pie slices between the walls of the fixture and the OD of the sacrificial pin should hold everything in place so you can rough the slots and skim them if needed.
The blanks obviously need to be very consistent and the fixture needs to fit very well both ID and pin diameter.
No glue needed.

The other way is to tack them to a plate with a laser welder, and then go to town.
But if you don't just happen to have a laser welder kicking around, it's kind of useless advice.
Also you have to be able to live with a tiny weld mark in 8 places around your part.

Cheers

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

 

[nlancaster]

I have to cut the ID and slots to shape while holding it, so I can't clamp around a pin sadly. And the ID of the hole is where the .0002 profile tolerance is, on shapes that are .0118 radi(sp?)

 

[implmex]

Hi again nlancaster:
Can you post a picture or is that verboten?
Obviously there's more to it than I am picturing in my head.
At this point, it's looking like you're going to be best off taking these from a plate and expecting to do a second op in a fixture to trim the cutoff stub from the first op.
So I'd abandon the notion these are going to be cut from a tube, but there's just so much I do not know about the project it's almost impossible to offer decent advice.

Cheers

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

 

[nlancaster]

Yeah, no can do on a picture.
Picture titanium a cone with a 45d outside, od of about .625, and a 8 .035 slots cut part way thru from the outside, and a .150 id hole. Now this cone is glued to a ceramic insulator on top of a 4inch long tube of titanium. after it is glued on, this is then cut with an almost sunburst pattern with 12 .0118 radiuses, and 8 slots that go out to join the .035 slots. Creating 8 independent segments that are only attached by glue to a ceramic insulator. And those .0118 radius have to maintain a profile finish of .0002 for their entire length.

What I am thinking is a titanium spring clamp with 8 fingers that comes down on the edge of the titanium part, as close as possible to the ceramic insulator.

 

[implmex]

OK, nlancaster:
So do I understand correctly, you are going to turn a cone with a wire start hole in the center.
You're going to slot it part way through with 8 0.035" wide slots coming in from the outside but not all the way to the center bore, so it's still all one piece.
You are going to glue it to a ceramic block that has a big enough hole in the middle to allow you to wire the inside of your titanium cone while the two are glued together.
You're going to dump the whole works in the work tank and wire out the inside of the cone, then make 8 cuts to meet up with the 8 0.035" wide cuts you've already made coming in from the outside before you glued the parts together.
When you're finished, you will have 8 tiny pie shaped titanium parts glued in a radial array with a super accurate goofy profile on the inside bore.

What you are asking about is how to make the titanium cone connect to the table of the machine so it's temporarily conductive and will allow you to wire it even after it's glued to the ceramic insulator.

Do I have it right?

So what you plan to make is a little springy collar with 8 fingers that you can clamp around your titanium cone so it touches all the segments reliably.
The shape of the collar doesn't really matter, so long as you can make it touch the table (with a wire if necessary) and so long as you can still wire cut in the vicinity without carving it up in the process.

Seems like a perfectly reasonable way forward to me...just don't make the springy arms so strong they pop the little segments off once you've cut them free.

Cheers

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

 

[nlancaster]

Markus, you got it right.

It is the only way I can think to ground the little segments.
Going to setup the spring arms to touch the bottom of the segments as close to the glue/ceramic as possible to reduce side forces.

Was hoping someone else might have any other thoughts.

 

[implmex]

Hi again nlancaster:
How strong is the glue?
Can you test it...have you tested it?

I assume it is impervious to water and the ceramic does not absorb water either?

I like your idea of touching the titanium cone as close to the base where it's glued as possible, for exactly the reason you describe.
Sadly, with the outside of the cone being a 45 degree angle, you will always have equal force pushing inward as you have pushing down, and if you don't get the conducting ring centered over the cone, you will have some fingers that touch harder than others.

Also you need to be sure that the glue does not spread into areas where you want the fingers to touch.
You might have to build a little test rig to make sure you have a conductive path from each finger to each segment...that means the fingers cannot touch one another if you intend to push a test current through each to see if each finger actually makes contact with the cone and not an invisible glue film.

They've really handed you a flaming football with this project, haven't they.
I hope you have apprised all stakeholders that this is going to take some fiddling and they should be bloody grateful you're even willing to try.

I suppose there's no way to use a conductive glue is there?
One would think the whole point of the ceramic insulator is to insulate...conductive glue kind of defeats that.

Cheers

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

 

[nlancaster]

Marcus,

Yeah we met with the customers today, they are going to see about improving the glue surface area to improve strength.
All of us, the lead assembler that glues them, our other engineers and me of course, tried to impress no them just how fragile this apart is.

Apparently we are going to quote somewhere around $10k for the prototype parts for this, plus other non refundable engineering fees.

Nick

 

[implmex]

Hi again Nick:
A wacky thought...could you offer a re-design that puts 8 little posts on the underside of the cone with matching holes in the ceramic insulator.
Gives the pie segments a better chance to stay put when you cut up the cone, because you have more surface area and you also have something that will resist sideways forces.
Makes the cone more challenging to make, but it makes everything subsequent, less of a nail biter.

Cheers

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

 

[plastikdreams]

GIN me

 

[nlancaster]

Marcus, not a bad idea, we have discussed it internally. the problem being the ceramic is a ring only 1mm wide (.039) so it is pretty fragile already.

 

[implmex]

Hi again Nick:
Man, these damned engineers are really stacking up the problems for you, aren't they?
You'd almost think they want you to fail, or at least turn yourself upside down and inside out, trying to meet their needs.
If you can pull this off and still have any hair left that you haven't pulled out in desperation or frustration, you too will have my respect forever!

Please let us know how it goes...I'm cheering for your success from the peanut gallery up here in the cold snowy north.

Cheers

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