What's new
What's new

Machining pockets with ultra-thin (.01mm) floors: methods, fixtures, part integrity

arrickg

Plastic
Joined
May 1, 2019
Hello,

I am investigating methods for machining pockets (in brass) with thin floors... I'm talking extremely thin -- .01 mm -- like tin foil. Please see the attached drawing for details.

thin_floor_pocket.jpg

We have successfully made this part in the past by soldering brass foil onto a ring, but I'd like to eliminate the use of lead-based solder (as this project has eventual medical applications).

My first though was that this is probably not possible with milling or turning. However, maybe adhering the part to the bed, performing machining passes, and then dissolving or melting the adhesive support could work?

I've also considered using EDM. But my experience with this is limited.

Perhaps simply welding a foil on would be best, but that strikes me as difficult given the thick-to-thin nature of the joint.

Finally, even if the part were made, it's not clear that it would be robust. Perhaps thermal stress or internal material stress would cause the foil to fail or tear on its own.

Would appreciate any ideas, thoughts or criticisms on this topic.

Thanks in advance!

-Graham
 
Unless you're just using brass for development, you should probably avoid it because many brasses have lead as a constituent. An appropriate stainless may be a better choice. And you could look at other attachment methods besides solder, depending on temperature or sterilization requirements a medical-grade adhesive or laser welding may be a good choice for attachment of a foil to a support ring.

If it must be integral, then using a thinner support material and etching the center to your desired thickness may be the best way to go.
 
Hello Milland,

Thank you very much for your reply. Great to know about the lead content of brass - didn't see that one coming.

I like the idea of etching. What solvent(s) would you recommend to achieve such an etch? I was thinking Sink EDM could to the trick, but maybe that's too "violent" of an etch.

Also, I like the idea of laser welding but have little experience with it. Do you think the thick-to-thin interface would be a problem? Any recommendations for specific (micro) laser welding machines?

As for adhesives, I'd prefer to avoid them unless they have excellent barrier properties. In short, the part will need to resist gas diffusion over extended periods of time and metal-metal seals/welds/interfaces tend to do a better job at this. Having said that, if you know of any adhesives with great barrier properties, I would appreciate a rec! (Our temperature requirements are not high -- sub 60degC.)

Regards,
Graham
 
For etching, talk to these guys: Photo Etching & Photo Chemical Machining Precision Metal Parts | PEI they can advise on reasonable thickness targets for the foil and the starting stock. I would not try to DIY this, it's tricky to get uniform and consistent results.

For adhesive advise, talk to the applications guys at Masterbond.
Medical Adhesives, Sealants and Coatings | MasterBond.com

For laser welding, talk to Laser Welding of Medical Device Components | Cirtec Medical They also do electroforming, which would be another avenue to explore for your thick/thin requirements.
 
I'd machine the floor to about 1/2" thick, then make another pass working from the center out using a HSM path on a VMC to full depth.

As far as it being robust...you wont get that answered on a forum, but it sounds extremely fragile.
 
Might work with a vacuum fixture. Ive done Delrin with a .010+ or - .002 floor , on a vacuum plate and alum down to around .010 but you have to watch the vacuum gaps/runners in on the plate with alum as it will bend into the vacuum runs. Plastic was easier to do as it didnt flex as much as alum.
I'd avoid glue and tape as I am sure you will tear the part.
 
Personally I would machine the inside to a finish spec where the overall length is left up about a mm or so then fill the bore with a putty or resin or glue or something, then machine and grind the back to the finish thickness then somehow melt and remove the inner support material of choice hopefully without distortion to the thin wall.

I also think EDM would be a no go here as the severely localised conduction of heat from sparks of 8000-12000c would not let it stand a chance, but that’s just my opinion.

But, Apart from the godly skills that are required to machine such a part, what is it? Some sort of diaphragm?
 
Excellent ideas here! Thank you all so much for these recommendations! :)

I think we will try a few of these things over the next few weeks. I will keep you all posted on progress.

Regarding the specific application, unfortunately, I am not able to disclose too much at this point.I can say that it is a sort of diaphragm for fluid containment in a medical device.

Best,
Graham
 
Use gold. I've heard it can be made extremely thin. You might be able to press it to final thickness, if you can find a place for the excess metal to flow to.
 
Perhaps simply welding a foil on would be best, but that strikes me as difficult given the thick-to-thin nature of the joint.

Finally, even if the part were made, it's not clear that it would be robust. Perhaps thermal stress or internal material stress would cause the foil to fail or tear on its own.

This is a job for vacuum brazing using one of the silver solder pastes to attach the membrane to the support ring. You are not limited to metals for the diaphragm. For example, capacitance type pressure sensors are available with ceramic diaphragms. Ceramics are used when corrosion is a problem. A very old reference for this type of construction is " Handbook of Electron Tube and Vacuum Techniques" by Fred Rosebury published in 1965.


There is one problem specific to brass. The zinc in the brass will vaporize in a vacuum if the brazing temperature is too high. Brass would be a poor choice for a diaphragm for this and several other reasons


There is another approach that is used by metal bellows manufacturers. The membrane is formed by electroless nickel plating onto a mandrel. In this instance the support ring is placed on a flat plastic plate. The chemical plating process forms the membrane. This is roughly how the cheap plastic faucets receive their metallic coating. The chemical activation of the plastic substrate prior to plating would not be required. The activated surface is needed when the plating needs to adhere to the plastic substrate.
 
Last edited:
Might work with a vacuum fixture. Ive done Delrin with a .010+ or - .002 floor , on a vacuum plate and alum down to around .010 but you have to watch the vacuum gaps/runners in on the plate with alum as it will bend into the vacuum runs. Plastic was easier to do as it didnt flex as much as alum.
I'd avoid glue and tape as I am sure you will tear the part.

You machined Delrin to .01mm +/-.002mm thickness? That would be impressive.

Regards.

Mike
 
There are lots of brass grades to pick from, yes the free machining alloys contain lead but reduced lead or other copper based alloys could be used.
Why not bore the inside and place the part on a spigot with a small hole in the centre and an o ring on its outside, a vacuum will hold the part in place while you face it to thickness
Take care with heat during machining.
 
Good application for electron beam welding if part done in stainless or similar. Prepare a 0.01mm pocket at the bottom of your part and make the foil disc same diameter as the pocket. I assume a similar results can be achieved with laser welding, but I have experience only with EBW and this works very well.
 
PM implex (sp) on here hes into the medical world and has a laser welder, which should be a easy way of attaching a foil over such a ring. Like this you can use harder sheet, which may well give you a nice strength benefit.
 
There is another way to accomplish the thin bottom added to the ring.

We buy bellows from Servometer, they are .001 thick nickel even in the convolutions. Doing things their way, machine the ring from whatever material as long as it is not aluminum, press in aluminum plug face the bottom then plate it .04mm thick on the bottom, and soak it in caustic till the aluminum is gone.
Servometer can check for leaks on their bellows we buy nonchecked to save a little as leaking will not affect our end product, for the OP's need leak checking will probably be necessary. In their bellows they do machine both ends off the plated blank to leave both ends of the finished bellows open. In the OP's case he may need o machine off one end to expose the aluminum if the entire part is plated. A nice blended radius on the bottom would prevent the plating from pealing. Gold was mentioned as a material for medical use .04mm thick plating would not cost much as it is so thin.
 
Hi All:
This is a fairly difficult part to even think about trying to make as one piece.
For those who have not done the conversion yet, and still think mostly in Imperial measurements; the bottom surface is 4 tenths...yes, that's 0.0004" thick.

You'll never turn it or mill it or EDM burn it successfully in one go, so a reasonable alternative to consider, is to attach a foil to a ring (as you've been doing) and ignore the fillet that's drawn in the corner of the pocket if you are allowed to.
At this thickness, there is a possible benefit to preserving the physical properties of the foil that it gets when it's rolled to thickness, so that excludes low temperature processes like furnace brazing or soldering unless it's also OK to lose those physical properties when you cook it.
You also may need to consider potential galvanic corrosion effects of dissimilar metals in the joint.

Trying to laser weld very thin foils to thick pieces has not been very successful for me in the past...I have had far better luck when a fin can be provided that I can fusion weld to that is the same thickness as the foil so I can do an edge weld which allows me to drop the weld energy right down so I don't blow a hole into the foil.

It is also theoretically possible to do a keyhole weld through the foil but your fitup must be perfect at these foil thicknesses, whether the energy source is going to be an electron beam or a laser beam.
As you can imagine, you cannot easily squeeze the joint surfaces together and still get access to the joint without some kind of indexable fixture that exposes a small segment of the joint surface at a time while acting simultaneously as a pressure plate.
You also need more filler metal than the foil alone can provide, so it's technically not really keyhole welding.

Another alternative to consider is resistance welding.
With an annular electrode, you can get a continuous weld and "may" be able to achieve a hermetic seal, but the weld engineering and process control necessary to do it reliably will not be trivial.
Since the electrode acts as the pressure plate, this approach is simple in terms of component fitup; just lay the foil on the ring and zap it.

For laser, EB or plasma welding approaches, a tiny machined fin on the outside corner of the ring, a pressure plate on the fixture that holds the foil against the surface of the ring, and then a fusion edge weld to melt the corner together is how I would probably approach it initially; assuming that this kind of re-design meets the other requirements of the part.
Implied in this approach, of course, is a crevice between the foil and the under surface of the ring, and this may be unacceptable for a variety of reasons.
Also be aware there is a stress riser at the apex of the crevice inherent in this kind of joint, so if it sees pressure differentials, it will be vulnerable to failure from cracking of the weld.
This will be true for any type of welded joint.

If you must make it one piece, the best way forward (although it will be a production nightmare), is to machine the pocket in a thicker block, then wire EDM slice it so the floor is as thin as you dare to go, then fill it with something like Cerrobend, then face the front dead flat, then flip it and surface grind and/or lap the underside to final thickness, finally melting out the Cerrobend.

If you can make resistance welding work, it will be the fastest and most production friendly, but you WILL have a crevice, even if it's just a small one, and you WILL have leakers so there will be 100% leak testing inspection in your future if you care about a hermetic seal.
You may also choose to resistance weld, then over-weld with EB or laser to improve your chances of a hermetic seal.

You will lose the physical properties of the foil in the HAZ with any welding process you choose; even EB welding.
You may or may not care.

On a last note, I concur with those who advise not using brass, but choose a grade of stainless steel suitable for your application.
The better physical properties of hard drawn stainless steel foil will help you in production in a whole bunch of ways, as will the wider range of joining options.

On a very last note; have you considered exploring diffusion bonding?
A customer of mine uses it in a filter application but I know virtually nothing about it.
They are bonding very thin stainless steel foil to itself... the process involves heat, pressure and controlled atmosphere but that's the limits of my knowledge about it.

On a very VERY last note, have you looked at the foil seal on your jar of peanuts that you peel open when you sit down with them in front of the TV?

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 








 
Back
Top