'83' mercury vapor rectifiers

[Shannon Parks]

Tube Gurus,

OK - I realize these need a warm up, but used properly will last awhile. Relays, HV secondary tap switches and other methods I don't care for. Won't a simple HV center tap work as good as any solution?

Shannon

[erichayes]

Hi All,

Far be it from me to throw cold water on experimentation, but the 83 has at least two things going against it in audio work:

1. It has a directly heated cathode. This creates a zero-cross twilight zone similar to a silicon rectifier where the tube conducts less when the filament voltage approaches zero. This isn't a problem with indirectly heated rectifiers due to the thermal inertia of the cathode.

2. It's noisy. All ionic tubes, whether hot or cold cathode, generate hash that has to be dealt with. That's why you don't see VR tubes in supplies upstream of output tubes (you will see them as screen regulators on outputs that aren't being run UL), and why 0Z4s are found only in car radios. Lots of filtering/shielding might be needed to get an acceptable noise floor.

The 82 and 83 were developed to provide constant output voltage over a wide range of load currents. That's why they were almost universally used in tube testers even after silicon diodes were available. They have a forward voltage drop of fifteen volts, whether you're pulling 1.5mA or 150mA out of them. But that's all they were really good for.

The 83-V was a drop-in indirectly heated cathode substitute for the 83, within its limitations (it could only handle 175mA, as opposed to 225 for the 83), and the forward voltage drop wasn't as zener-like as the 83, but it was a lot better than the 80 and 5Z3 directly heated rectifiers of the time. The 83-V begat the 5V4, which begat our ol' buddy, the 5AR4/GZ34. This, to me, is the paradigm of high current vacuum tube rectifiers; it just has a problem of blowing up in rather spectacular fashion.

All that having been said, Shannon, the 83 is happy with the standard center-tapped , choke or cap input configuration. Just keep the tube vertical.

[EWBrown]

Some of the tube testers I've seen that have 83s (and 5U4s) in them, the rectifiers are mounted horizontally, and they are enclosed within the bowels of the equipment. If I remember right the Hickok 600 and 6000 series do this. But then, they aren't being run anywhere near full current capacity. HgV tubes like teh 83 and the 866 do generate a lot of RF "hash" which may require the use of inductors in series between the power trannie secondary and the rectifier anode(s), and may further need to have the tube inside a screen / shield to further contain the emanations.

Here's another point: Do these mercury vapor tubes pose a UV radiation hazard, as do some of the UV "sterilizer" and e-prom eraser bulbs do? This probably wasn't a concern back in the 1940s and 1950s, but it is now.

I have a couple of 83s, I figure on building up a variable regulated supply that has some more ballz than my old trusty Heath IP-32, (400VDC@100mA). I need something like 0-450V @ 200 mA) these and a quad or Russian 6L6s should do the job, I already have a huge honking 900VCT 250 mA plate trannie just looking for permanent employment, it was salvaged out of some old mil surplus green behemoth years ago, and is just taking up valuable space right now. I also had it's big brother (2KVCT @ 500 mA, but passed it on to Gary Kaufman for use in a future GM70 SE amp).

NH doesn't use the electric chair anymore, so it had to find a good home :twisted: :shockingzap: :onfire: :ahhhhhhhhhhhhhhhhh:

/ed B

[EWBrown]

There is a discussions on the merits and hazards of mercury vapor rectifiers located here:

http://antiqueradios.com/forums/Forum13 ... 02825.html

It is also a good idea to use a fuse of about 150% of maximum secondary current capacity for further protection when using these tubes.

/ed b

[erichayes]

Hi All,

In my first reply, I said "...at least two..." because I knew there was another biggie, but couldn't recall it. Thanks, Ed.

The third: Even though the 366 nM primary wavelength is considered longwave UV. you can still get a really nasty set of sunburned eyes if you stare at a mercury vapor discharge. The secondary spectral lines are much lower in amplitude, but are shortwave and definitely harmful. That's one of the reasons black lights appear so dark--there's a low-pass filter to trap what SW UV makes it through the silica glass.

Back in my spectrophotometer days, I was accidentally exposed to an unshielded deuterium light source for no more than a second. It felt like someone had pushed a couple of icepicks into my eyes, and took a week for them to heal. I was lucky.

BTW, I went out to the shop and measured the applied plate voltages of my Hickok 750 and B&K 700 tube testers, both of which use horizontal 83s, and measured 173 and 200 VDC, respectively. Since the 83 is rated at 500 V and 250 mA per plate, I think it's safe to say it's not breaking much of a sweat.