Sadudu, I don't recall ever seeing a non-rf alumina circuit made with
adhesively applied copper like is done when making an ordinary composite
pcb. Like gregebert, I recall a screening process (conceptually much like
solder paste) precisely applying an uncured slurry of silver and palladium
that was then heat bonded to the alumina. There are many ways of
incorporating mechanical fasteners as part of a substrate. Have a
discussion with a custom alumina house. Their application engineers
usually love finding solutions for unusual applications like yours. Since
you don't care about dielectric consistency or value, the production parts
should be pretty inexpensive.
Rogers is a great company with a great niche suite of products for use in
high end, low production, dielectrically sensitive devices. In your case,
I don't believe they have the product, expertise, or manufacturing you are
looking for.
On Monday, September 29, 2025 at 11:07:27 PM UTC-5 Sadudu wrote:
> Thank you all for your attention. I’d like to address some questions here:
>
> 1.
>
> *Why not directly use mica sheets, but instead try to find another
> structure?*
> In the original B7971, a large amount of mica sheets were used as
> structural components, mainly to insulate and support the electrodes. In
> other Nixie tubes, it has also been proven that mica has extremely low
> outgassing and outstanding stability, which is why it has been widely used
> in vacuum tubes. So, mica is indeed a very ideal material.
>
> Then why did I try to look for other materials? Due to the special
> structure of the B7971, each cathode is not directly connected to the base
> but instead uses some adapter wires. From the disassembled B7971, it can be
> seen that it did not use spot welding for hard connections. Instead, the
> B7971 used a kind of spring-clip-like structure that fastened the cathode
> and the adapter wire together. This design is very ingenious, as it made
> use of the elasticity of mica—you can refer to this article for details.
>
> https://www.tube-tester.com/sites/nixie/data/b-7971/b7971.htm
>
> This approach requires a rather precise assembly process. The mica sheets
> need to be riveted with a metal baseplate, and suitable-sized connectors
> must be made for the backside. As a result, the process is very complicated.
>
> If Rogers circuit boards had worked, they could have directly replaced the
> combined role of mica sheets + baseplate + connectors. Furthermore, I would
> have been able to use spot welding to achieve hard connections between the
> adapter wires and the cathodes. This was the main reason I sought out
> alternative materials—to simplify the internal structure and improve the
> yield rate of the product.
>
> 2.
>
> *Next plan*
> I am very grateful to some friends who sent me emails suggesting other
> possible materials. Our next plan is to try using ceramic-based PCBs for
> this purpose. Although it may also fail, I believe it is much better than
> Rogers high-frequency boards. Ceramic has also been widely used as an
> insulating material in vacuum tubes, but I need to figure out how the
> copper layer is bonded to the ceramic substrate. If it is applied through
> a
> plating process, then it probably won’t be a big problem. However, if some
> kind of adhesive is used to bond the copper foil to the substrate, then
> this material cannot be used in a vacuum tube.
>
> If this plan doesn’t work, then we will move forward with reproducing the
> original B7971 structure for production.
>
> Finally, thank you again for your attention.
> On Monday, September 29, 2025 at 11:45:52 PM UTC+8 J Forbes wrote:
>
>> you made me look.
>>
>> [image: mica.jpg]
>>
>> On Monday, September 29, 2025 at 7:56:08 AM UTC-7 Neil QQ wrote:
>>
>>> In my former career job (rf engineer) Rogers material was used for
>>> primarily for prototyping because it's dielectric constant is uniform and
>>> tightly controlled and because it can be easily "machined" with common lab
>>> tools. For consumer goods, though, it is too expensive. Once the circuit
>>> was proven, fired alumina substrate was used. In quantity, alumina is much
>>> cheaper, ridged, and has excellent electrical and mechanical temperature
>>> stability. There are a number of custom and prototype alumina vendors.
>>> Here's one of the biggest: https://ceramic-pcb.com.
>>>
>>> As others mentioned, mica is still widely available. I've used it too
>>> for certain high voltage, low volume applications. It's brittleness can be
>>> troublesome in high vibration/shock applications. Environmental
>>> considerations over the last few decades have driven mica prices higher.
>>>
>>>
>>>
>>>
>>>
>>> On Sunday, September 28, 2025 at 1:35:59 PM UTC-5 Mac Doktor wrote:
>>>
>>>>
>>>> On Sep 27, 2025, at 12:01 PM, dudu sa <[email protected]> wrote:
>>>>
>>>> About half a year ago, a few friends came to me and asked if I could
>>>> make a Nixie tube similar to the B7971. As one of the few Nixie tube
>>>> manufacturers, I thought about it for a while and decided to give it a
>>>> try.
>>>>
>>>>
>>>> I forwarded this to Eric Barbour, the administrator of the Tube
>>>> Collector's Association list. The fact that he isn't ripping you to shreds
>>>> is a sign that he thinks this has possibilities and that you're not an
>>>> idiot for trying. He's impressed with the repros that have been made so
>>>> far.
>>>>
>>>>
>>>> On Sep 27, 2025, at 10:28 PM, metasonix wrote:
>>>>
>>>> Well....I could have told him (for free) this wouldn't work. Only
>>>> materials that can be heated and outgassed FULLY can be put inside a
>>>> high-vacuum or gas tube.
>>>>
>>>> This is apparently the stuff he tried to use. It has great dielectric
>>>> characteristics, but there is nothing about using it in a vacuum.
>>>>
>>>>
>>>> https://rogerscorp.com/advanced-electronics-solutions/ro4000-series-laminates
>>>>
>>>> Maybe talk to a company that makes thin film hybrid circuits. They can
>>>> probably make a degassable board out of alumina with traces that can be
>>>> spot-welded, since they have been making such things for 60+ years. It
>>>> will
>>>> NOT be a bargain. Feel free to repost this to the group.
>>>>
>>>> https://www.thinfilm.com/substrates.html
>>>>
>>>>
>>>> https://www.coorstek.com/en/industries/electronics/microelectronics/thin-film-electronic-substrates/
>>>>
>>>> All modern electronic components are made with some kind of plastics,
>>>> epoxies, or phenolics or whatever. None of which can be processed in a
>>>> tube. Even Teflon will eventually disintegrate into hydrogen fluoride and
>>>> other things.
>>>>
>>>> If they manage to build a working B7971, they are well within their
>>>> rights to charge a very stiff price, at least $500 retail and probably
>>>> more. Making such things in the 1950s was easier because mica was cheaper,
>>>> one could buy parts and envelopes from third parties, and you could pay
>>>> bored housewives 50c/hour to assemble the damn things.
>>>>
>>>> That price may seem slightly high but Eric knows all about actually
>>>> recouping an investment and earning a living. As well as dealing with
>>>> customers who can't figure out how to use his products.
>>>>
>>>> Terry Bowman, KA4HJH
>>>> "The Mac Doctor"
>>>>
>>>> "Tape machines ought to be big and cumbersome and difficult to use, if
>>>> only to keep the riff-raff out."—Steve Albini, 1993
>>>>
>>>>
--
You received this message because you are subscribed to the Google Groups
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion, visit
https://groups.google.com/d/msgid/neonixie-l/7cc23ecb-81e8-4e03-9139-05eafb45f679n%40googlegroups.com.
