DIY hermetic feedthroughs: Has anybody done this?

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
Posts: 628
Joined: Mon Nov 28, 2016 7:48 am
Real name: Ian Krase

DIY hermetic feedthroughs: Has anybody done this?

Post by ian_krase » Wed Aug 15, 2018 9:29 am

DIY feedthroughs are fairly common in Fusorland because you can make them with rubber or sometimes vacuum-epoxy seals. In the days of John Strong, wax seals were also used.

However, the kinds of feedthroughs you can buy, the really hermetically sealed ones, are always either glass to metal sealed or metal to ceramic brazed.

For glass sealing, the glass is melted, either bulk glass like for vacuum tubes or a smaller amount of frit or solder glass to seal between metal and bulk ceramic.

For metal to ceramic brazing, which is apparently what the good feedthroughs are using, a process is used to create a metallic coating that's bonded (by a metal-oxide interface) to the ceramic and then the metal and the ceramic are brazed together.

I'm interested in whether anybody has ever successfully done any of this. There are some big complications.

The proper solder glasses seem to be difficult to find and obtain, and most contain lead (which isn't *that* bad, but...)
I don't know of any source for the special brazing alloys needed to braze ceramics, which almost always contain some titanium.

The brazing needs to be carried out in vacuum, which seems right up our alley -- a quartz tube vacuum furnace should be trivially easy to build.

Interestingly, this one US Military report I read suggests that aluminum can be bonded directly to boron nitride simply by heating it up really hot, but it's hard to bond anything to aluminum and boron nitride is way too expensive.

User avatar
Dennis P Brown
Posts: 1840
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: DIY hermetic feedthroughs: Has anybody done this?

Post by Dennis P Brown » Wed Aug 15, 2018 11:50 am

I have sealed copper and aluminum with glass but one MUST get the coefficient of thermal expansion or CTE's close for these metals; that requires a very special glass. Ditto for steel. I did all of this in the air (but certainly there are processes that require vacuum or argon - I never bothered with those.)

I did do steel but that was at a temperature I did not like achieving (600 - 700 C.) The nice thing about the Al glass sealant is that this glass melted at a rather low temp (300 C or so.)

I know little on direct bonding of ceramic to metal but I did use the Al/Cu glasses to bond straight to ceramic (it was a machinable ceramic - the bonding process is easy since most clean ceramics are readily wetted by a liquid glass.) I would guess that one could then install the metal to the glass (at the glass's melting point.) The oxide on the metal chemically reacts with the glass phase to form a strong bond. Then the metal, through the glass phase, is then bonded to the ceramic (I didn't bother doing this part since my application was rather different - armor.)

The glasses for Al & Cu were not cheap (very special, in fact) but I would think for steel and ceramics these glasses would be available at more reasonable prices - maybe a google search. In bonding metal's and ceramics the glass, of course, is the "flux" agent that bonds the two dissimilar materials - there are papers and even books on this subject. Before the development of the high CTE's glasses, people used complex multi-material (a series of ceramic/metals/glasses) sandwiches to bond ceramic to high CTE metals - steel's CTE is fairly high compared to many metals but compared to Cu and especially Al, it is intermediate.)

User avatar
Richard Hull
Posts: 12184
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: DIY hermetic feedthroughs: Has anybody done this?

Post by Richard Hull » Wed Aug 15, 2018 10:13 pm

Every hydrogen thyratron has a metal to uranium glass to boro-silicate glass seal for the anode. Other high power tubes,(xmitter power tubes,etc.), also use the U glass intermediate glass to metal seal.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Jerry Biehler
Posts: 900
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: DIY hermetic feedthroughs: Has anybody done this?

Post by Jerry Biehler » Thu Aug 16, 2018 3:14 am

Well not every hydrogen thyratron. I have seen small ones that have dumet seals and big ones that are metal ceramic.

Its pretty easy to do a U glass seal to tungsten. Get some borosilicate U glass, it is still available now and then. Heat up the tungsten you will be using as the electrode in a flame until you get an oxide layer that is, as my instructor told me, kentucky friend chicken brown. Slip the piece of tubing over the oxidized section of the tungsten and heat it up and it will soften and adhere to the rod and form a bead around the rod. You can then take this and do a standard glass joint into a piece of tubing. Anneal when done.

You really need to use a gas-oxy torch to work this, either natural gas or propane. Acetylene is too hot and dirty.

They used to make pre-prepped U glass electrodes that save you the trouble but I have no idea if they are still available. I know you can buy dumet electrodes like this.

User avatar
Rich Feldman
Posts: 1303
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: DIY hermetic feedthroughs: Has anybody done this?

Post by Rich Feldman » Fri Aug 17, 2018 5:38 pm

Speaking of fused glass to metal, I've got a tiny jewelry project that could use a black vitreous enamel finish. The substrate, a short thin tube, can be made from copper or steel.

Colored frit is available over the counter in 8 oz jars or bigger. Online (e.g. Etsy) in 2 oz packs. But my whole project won't need more than a gram. Maybe two grams, allowing some for practice.

Any thoughts about pulverizing glass beads from a bead craft store? I have learned that some, but not all, are borosilicate.
How about soda-lime glass from bottles, windows, or laboratory tubing?

Online sites about art glass use the acronym COE for coefficient of expansion. Typically between 90 and 104, for "soft glass" in dozens of colors, e.g. in the name of a brand called System 96. I think that means 9.6E-6 per K, but have been unable to find explicit confirmation. Some borosilicate art materials are listed with COE = 33.
All models are wrong; some models are useful. -- George Box

Jerry Biehler
Posts: 900
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: DIY hermetic feedthroughs: Has anybody done this?

Post by Jerry Biehler » Sat Aug 18, 2018 2:43 am

I actually worked at bullseye glass for about 2.5 weeks before I was canned for "not being excited about glass". Sigh.


Most art glass is soda lime. We had special machines build specifically for the job but you could do a ball mill with a bunch of alumina balls which you can get off ebay. Put them in a tumbler and let er go. You will have to use sieves to separate out the size you want.

Basically all of Bullseye's glasses are designed to have the same COE so they can be fused together. I actually ran the machine that checked COE as one of the things I did when we developed a new glass.

Post Reply