We need to get a small quantity of NON-fire retardant PC board material for an ion chamber project. This will be under high vacuum conditions, so fire is impossible. We need to avoid contamination of the very low-pressure gas in the chamber due to outgassing of the fire retardant. I've looked around, I can't find anybody that seems to carry this material anymore.
We also need to get some etched boards made with the same stuff, so links to PCB fabricators that work with non-FR boards would also be appreciated.
So, just to not confuse people, we will be making some of the simplest boards by manual techniques, and having the more complex ones etched by a fab, as it needs plated through holes.
Are you specifically having problems with FR-4, or with anything that doesn't burn?
There's a popular material I see mentioned for RF stuff -- "Rogers ". I believe that it's teflon, or teflon-impregnated glass. You may have a better chance getting it, and if it's compatible with your ion chamber then your problem is solved.
We don't even KNOW! My boss said we need to get PC board material without the FR stuff, because in the "old days" the PBB would outgas. PBB (polybrominated biphenyl) is no longer made, now they mostly use TBBA, but that is at least remotely similar to PBB. What we do need is a material which does not outgas in a high vacuum system.
Yup, Teflon board should be pretty vacuum-compatible.
PCB houses several but a very good one whose name I can't find right now located in Santa Clara, CA be sure to spec high vacuum so you have an 'out' incase they use the wrong flux/solder/residues, etc.
Be VERY careful in your construction. You can end up capturing outgassing materials. Most important think like being in a thermos, you have ABSOLUTELY no convection cooling. OpAmps with fairly light power into them without enough metal to pull the heat out will unsolder themselves and fall off your board. give or take 40 seconds.
sadly I can't recall how I handled the vias now. whther had to leave unplated, or filled sorry will check notes.
Yes, we are looking at Rogers materials. We don't need much, just a few square inches.
I think we want them un-tented, but plated should be fine. In this particular case, there will be NO electronics in the vacuum. We have run up to 180 W in hard vacuum in a past project, it had water cooling to a big plate and then conduction cooling from all the heat sources to that plate. Certainly a nightmare, but they HAD to do it that way!
In this particular project, there will be a very low pressure gas flowing all the time through the chamber, so I'm really not so sure why outgassing is a problem. I haven't figured out how many replacements/hour of gas are being provided, though.
Technically speaking FR4 is a glass-fibre-reinforced composite material whe re the glass-fibres are bonded with a fire-retardant (FR) epoxy resin.
You can get poly-imide bonded glass-fibre substrates, which tolerates high temperatures better (which is why we used for a specific application at Cam bridge Instruments UK back in 1984). What we got back in 1984 was just as s tiff as regular FR4-bonded glass-fibre reinforeced boards.
Try ringing around the printed circuit board manufactures you know. People like Roger supply in quite large sheets.
We had a six-layer board made with the two out layers made with isocyanate resin-bonded Teflon cloth (for low dielectric constant) and the manufacture r charged us 1000 UK pounds per tripple-extended Eurocard board, primarily to cover the cost of buying a single sheet of the substrate.
If the printed circuit board shop has to buy in a sheet of substrate for yo ur application, they'll be both unethusiastic and expensive.
ng materials. Most important, think like being in a thermos, you have ABSOL UTELY no convection cooling. OpAmps with fairly light power into themwithou t enough metal to pull the heat out will unsolder themselves and fall off y our board. give or take 40 seconds.
Think about heat-sinking paths through the board and it's mechanical suppor ts. At one point we clamped a heavy strip of copper braid to the board, and the other end of the braid to the metal wall of the vacuum chamber. It did n't look elegant, but it worked.
unplated, or filled sorry will check notes.
Unplated vias wouldn't work very well. Buried blind vias could be a menace.
cular case, there will be NO electronics in the vacuum. We have run up to 1
80 W in hard vacuum in a past project, it had water cooling to a big plate and then conduction cooling from all the heat sources to that plate. Certa inly a nightmare, but they HAD to do it that way!
all the time through the chamber, so I'm really not so sure why outgassing is a problem. I haven't figured out how many replacements/hour of gas are being provided, though.
If the requested vacuum conditions are really tight and the volume is low (i.e. the cost is not of primary concern), then maybe you should consider copper on ceramic substrate or even on glass? The thin/thick film device manufacturers may be willing to make it for you. For small quantities and reasonable routing you can even try to make them on your own or ask a nearby galvanization shop.
Something less exotic could be ISM (aluminum-insulator-copper), but you need that one with Al2O3 insulator instead of FR4. I've successfully made high-density power boards using ISM in amateur conditions:
(the left side is a 40A synchronous rectifier)
Best regards, Piotr
wego oprogramowania, poniewa? ochrona avast! Antivirus jest aktywna.
As far as I know, there are two variants available: 1.5mm of aluminum, then 100um of FR4 and then a regular 30um layer of copper. The second one has 70..150um of Al2O3 instead of FR4. I use the former, since being a hobbyist, it was the only exotica available for me in a reasonable price.
There was nothing special compared to a regular home-made PCB, except for two things:
The aluminum substrate is designed to be a good radiator for power electronics, so soldering of D(2)PAKs is a nightmare. I used a hotair soldering station set to almost full of its rated capabilities. Soldering of small parts (0603 etc.) is relatively easy, as the soldering iron has enough thermal capacity.
You really don't want to drill a hole in it, as the substrate is conductive. Proper distances must be maintained in case of mounting holes.
The only available technology here is SMD, but the process of PCB routing needs special care, because it's imposible to make a via. It must be single-sided. I've used ~5 0-Ohm 1208 "resistors" as the vias.
The board has been made using the heat transfer process, i.e. printed the maze using a laser printer on art paper and then transferred onto the board using a laminator. The etching process was surprisingly simple, because it turned out that sodium persulfate doesn't attack the aluminum substrate at all, so no special protection is required. Then it has been coated with tin and I started soldering the parts. After initial checking I've also applied two layers of urethane conformal coating to isolate the electronics from moisture and dust.
Below in the stack are two similar boards hosting 12x triac/optotriac pairs, but they are nowhere as densely populated as the top board.
All of has been done in my spare time and is a part of my home automation project. :-)
Well, the last time I worked with vacuum related stuff, the common practice was to pump the chamber down, then let leave it undisturbed for a few days to allow for out-gassing, then repeat the process 3 - 4 times to clear out all out-gassed stuff