• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

building vacuum tubes... some questons

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
look what ive dug up .

ive found and intresting PDF about emission coatings and how to apply them . this might help you out

http://www.cathode.com/pdf/TB-160 thru 164.pdf
normaly used for cathode ray tubes EG TV tubes .

they even quote exact chemical composition . binder . and the ammount of emission . ! prolly all you gona need to know . to make good heaters .

you could fabricate a nice small painting booth . you could use an airbrush to apply the coating . :D:D:D
or just make a small amount of the mixture and submerge the heater wire in it , pass a current through it . and thats all you need .

the binder described there can also be used to get very fine particles of some metals like aluminum , nickel , magnesium , zirconium on your plates and use them as a getter . not a bad idea at all . they will be most active during overload conditions .
this will probably turn out in some grayish colour . aiding heat dissipation

i think the best way of using the plate as the getter is to make a mix of those volatile metals in very fine powder form . take the binder recipe from that PDF and spray it on your nearly finished plates . and warm it in an oven to around 150-200 degrees to drive of the binder and moisture , weld it on your stem and there you go .


i havent found anyone that can supply getter pellets yet . but if you can coat your plate whit a active getter you will be able to make clear glass beasts .
something that comes to mind is that some tubes (not gona say witch ) have an active getter and will survive torture witch would kill your normal tubes

note: extra kudo's for whom may guess wich tube im talking about .

will post and simplified materials list later .
 
Last edited:
Great link, nice setup there. Reminded me of some additional items needed for tube manufacture. I see a tube oven for bakeout of parts. An induction heater for firing the getters. An an annealing oven for slow annealing the glass after glass blowing and sealing. Probably a small spot welder there somewhere. Vacuum gauges and pumps of course. Actually not too much equipment, just the essentials there. I have seen all of those items in surplus stores in the past, but one would have to make a very determined effort to round them all up.

I picked up a used copy of "Procedures in Experimental Physics" by John Strong as mentioned earlier. Great book! Already had "Building Scientific Apparatus" by Moore, Davis and Coplan, another great book. Then there's "Handbook of Vacuum Science and Technology" by Hoffman et all, and "Handbook of Electron Tube and Vacuum Techniques" by Rosebury, "Handbook of Materials and Techniques for Vaccum Devices" by Kohl. Then the RCA "Vacuum Tube Design".
 
Last edited:
Vacuum tube glass and turbomolecular vacuum pump

You must use borosilicate glass tubes and no they are generally not available locally for most. The tubes will have to be special ordered from a laboratory or scientific supply house. Schott is a leading manufacturer but there are others; the problem will be minimum order sizes are geared to the large scale and not small R&D or hobbyist activities. In other words, it is going to cost you for a small order unless you can locate some on the surplus market. For example, a case from Cole-Parmer in the 51 mm size is $360 see WU-34742-54 Glass Tubing, 51mm;

Glass Tubing And Glass Tubing - Cole-Parmer Catalog

Regarding vacuum pumps, you will need a turbomolecular vacuum pump and control power supply in addition to the roughing/backing pump stage and diffusion pump. You might find one on ebay that is working but it will still run about $800 on up to 3 to $5,000. Factoring in all the glass blowing tools and sundry supplies and special materials that are needed; I just don't see setting up a working lab for under $3,000 at the minimum.
 
If one can find an oil diffusion pump, then the turbo pump isn't needed. Although the diff. pump does require a lot more support in the way of valves, water cooling and a cold trap. (both types need a good roughing pump) I would say that Ebay maybe should be a last resort, since most of the specialty vendors for this stuff will be pushing the max prices the market will bear and including refurbishing costs. A surplus store in the right area (old semiconductor production equipment say), or a university surplus store, would be much better bets for finding a good price on vacuum equipment I think. One will likely have to do some cleaning and refurbing (new heater cartridge for the oil diff. pump, cleaning burnt oil deposits out). If a turbo pump and drive electronics is located for a reasonable price, then by all means go for that, so much easier to use.

Have to agree on the high expense for a full VT setup though, so many items needed for glass blowing, annealing, gettering, materials, welding.... Might as well round it out with a few more items so you have a semiconductor fab too.
 
Last edited:
On surplus turbos:
Be aware of possible damage from items having fallen into the turbo while in operation (break a glass ion gauge and it all heads for the turbo inlet, or say a melt down of stuff in the vacuum...), bent or out of balance turbo vanes from store visitors or disassembly and subsequent handling issues, contamination from toxic chemicals (semiconductor use especially), failed bearings, seized rotors, burnt motors (some can run at 20,000 RPM and up, there is no such thing as a slightly out of balance working turbo). I wouldn't buy one off Ebay without some guarantees and pre-testing and right of return.
 
Last edited:
hmm there is a indian company making the nickel dumet and NI48/FE52 wire . JLC Electromet - Product Range

they can provide nearly everything

plate material nickel sheet strips ni200 /205 probably 40mm strips 0.1 to .25 thick for anodes
suport wires 1 and .5 mm NI48/fe52
dumet wire .3 mm


your going to need to source glass from a good supplier
20-22 mil is normaly used as stems i got a broken rca DHT on my desk right here whit a 20mm stem thickness
30-40 mil to make the bulb
4-6 mill for the pumping stem

for grids you can make nice grid supports from straightened 1mm ni/fe wire . and use some good conducting wire wrapped around it . to make it weld . or take bad conducting wire like nickel .25 wire and use copper grid posts


next is tungsten . wich comes whit its own problems as i have not the slightest clue on what thickness wire would be right at wich lengt to make a good wire heated to 700-800 degrees celcius at normal operating voltages

but the rca vacuum tube design booklet gives the formula to calculate the wire lenght its on page 23
 
Here's a PDF of the Varian diffusion pump brochure (no affiliation...). Check out the AX-65 and HS-2 pumps on pages 4,5 and 6,7. Stainless steel construction, 3 stage jets and ejector stage design. Heater power at 200 Watt and air cooled operation for the AX-65, or 450 Watt and water cooled op. for the HS-2 model. Perfect for a home lab setup. Anything bigger gets into kilowatt power consumption. There are even smaller diff. pump models around for vacuum instrumentation like leak detectors or mass spectrometers that would work fine for Vac. tube evacuation.

I've used an HS-2 system (with an Airco-Temescal 3 inch slide valve, Varian water cooled inlet baffle, a bronze inlet flange machined to fit Corning Pyrex 3 inch glass beaded pipe, a Varian 2 stage mechanical backing pump, gauges...) for many years. It's in storage now I think, I haven't seen it for a long time. (got replaced by a Balzers Turbo) A great little oil diff. pump for hobbiests, indestructable, reliable, high performance.


http://www.mhzelectronics.com/ebay/manuals/varian_diffusion_pumps_datasheet.pdf
 
Interesting DIY diffusion pump! Somehow missed that in the earlier thread. Have you made any performance measurements? How many Watts to operate the heater? I was shocked to see the restored prices for a Varian HS-2 diff. pump recently, several times what I paid for a new one 30+ years ago. I used to see old diff. pumps at the scrap metal yard years ago, could buy rusty ones for $0.50 a pound scrap metal price. Maybe just local supply and demand before Ebay came along.
 
The heater power is about 120 watts. As far as testing goes, I've chosen to make a useful chamber arrangement to accommodate the gauges and flow meters needed to qualify the pump. This pump and chamber should be quite universal for an assortment of small vacuum experiments including the pumping of homemade vacuum tubes.

The point of the whole exercise is to demonstrate that a working high vacuum system can be fairly easily constructed without having to plug in any power tools or use anything exotic in the way hand tools. This is a truly handmade pump and chamber.

Here's a look at where things stand so far: Fusor Construction & Operation - Download complete thread
 
I came across a US patent (#5137429) for reducing power consumption in diffusion pumps. It uses a thermally insulating ceramic collar between the heater and the cooled outside cylinder of the diffusion pump. Seems that one could just arrange for a Viton "O" ring with some flanges to do that. Fiberglass bolts to hold the flange together. Or a T shaped Viton circular gasket with some hose clamps around the outside.

I came across the thermal collar term for a Veeco EP3AB diff pump, but it does not seem to be of reduced power consumption. But I do notice Edwards makes the TVA50/60, which uses only 80 Watts and has an obvious mod to the boiler to cooled cylinder interface. (other similar pumps use 200 to 250 Watts) No mention of how it works in their literature.

-----------------------
-----------------------

A little off topic, but I just obtained a Pfeiffer TPH060 turbo pump for cheap and it needs the bottom bearing replaced. The official tool kits to do this are like $750 and the replacement bearing kits something like $450 to $650. So I proceeded to see how far I could get without any special tools. After removing the bottom bearing cover I found the bearing had lost its plastic spacer between the rollers.

There is a long conical tapered nut that holds the bearing to the shaft. It has only a tiny 2 mm Allen Hex wrench fitting on the end. After bending up several Allen wrenches, I tried contacting several of the pump repair outfits to inquire if this was a normal right hand thread nut or a special left hand nut. (since the operating shaft rotation would tend to unscrew a normal right hand thread) The ones that did reply essentially told me to pay the $1000+ and offered no help (top secret apparently). Well if one can't get the old bearing off, there is no point in buying a new one! I would just order the bearing from a bearing manufacturer anyway for < $100.

I am thinking maybe the threading is left handed. Or the previous bearing changer used Loc-tite or something on it, or the special tool kit has a conical female taper tool that gets hammered onto the outside of the nut (instead of using the tiny Allen wrench fitting). (Pfeiffer mis-conveniently deleted the bearing change instructions from the turbo user manuals of all their products a number of years ago.) Any help on this would be greatly appreciated!! If I knew for sure which way to turn the nut, I would try heating the nut to loosen it. A big soldering gun with some #12 copper wire configured to wind around the nut should do.

I do have an old TPU050 manual in storage up in Ct., which likely had the bearing change instructions still in it. But I won't be getting up there till next Spring likely. Worst case would be to just take some Vice-Grips and remove it, but I fear that might bend the shaft (90,000 RPM!), plus I would have to obtain a new tapered nut, or machine one. Wouldn't be too hard to machine one from some threaded SS spacer rod, but if its a left handed thread....

If it does turn out to be economical to repair the usual (damaged, dead ...) Ebay turbos, then this would be a viable solution to setting up a high vacuum system such as the original poster needed.
 
Last edited:
OH, if I do get the Vac pump up and running, I plan to make some TiO2 cathode heaterless vacuum tubes for an efficient OTL amplifier. 20 or 30 amp emission triodes should work well. Seeing as how none of the current tube makers have taken up the heaterless tube technology, maybe I will start selling them too.
 
Smoking Amp, could you post a photo of the bottom bearing?

If it has been Loctited then heating it and letting it cool could do the trick. Tends to work, it expands the thread then contracts on cooling cracking the Loctite.

Some of the guys who build and repair model jet engines use ceramic ball bearings. A Google about may find a suitable size and supplier. Some of these little engines spin at 120,000RPM!

Cheers Matt.
 
Here are some pics:
Only the inner race of the bearing is still in place yet at the bottom 1/4 inch of the nut. I was able to remove the outer race and balls already. Some pics have the Allen wrench in place at the end of the tapered nut.
 

Attachments

  • turbo1.jpg
    turbo1.jpg
    16.3 KB · Views: 113
  • turbo2.jpg
    turbo2.jpg
    18.9 KB · Views: 109
  • turbo3.jpg
    turbo3.jpg
    24 KB · Views: 109
  • turbo6.jpg
    turbo6.jpg
    19.8 KB · Views: 23
  • turbo5.jpg
    turbo5.jpg
    19.1 KB · Views: 111
  • turbo4.jpg
    turbo4.jpg
    22.8 KB · Views: 112
  • turbo7.jpg
    turbo7.jpg
    22.3 KB · Views: 31
  • turbo8.jpg
    turbo8.jpg
    23 KB · Views: 34
The easiest way to "make" a vacuum tube: Take a bifilament car bulb with convenient geometry. Distroy one of the filaments (the lower power one). The result is a very basic direct heated triode (good enough for school demos).
Building high-performance vacuum tubes is a serious challenge. It requires deep knowledge of physics, chemistry and matheatics, not just some cookbook recipes. I've worked a few years in that field (SEL).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.