• 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.

MV Rectifier Voltage Doubler

MV = mercury vapor?

Please PLEASE PLEASE dissuade yourself from using mercury vapor rectifiers for anything. They have a nasty tendency to break, releasing astounding quantities of hot mercury vapor to your listening area; cleaning this stuff up is nigh impossible. Just don't do it, no matter how “pretty” the blue-green glow is.

I'm not even going to try to address the primary question.

⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
 
They have a nasty tendency to break, releasing astounding quantities of hot mercury vapor to your listening area


Been using quite a few of these over the last 20 years and miraculously suffered no breakages, just sound impossible to achieve with any other form of rectification. Perhaps sync rectifiers would offer a viable sonic alternative, but right now nothing else can compete.

Do you personally know of an affected listening area?

And where do the "astounding quantities of hot mercury" come from as i suspect the fluorescent lights i have in the kitchen are even richer in mercury.
 
just sound impossible to achieve with any other form of rectification
I do wonder why that would be the case: ionized-gas rectifiers have — as you noted — good forward conduction, quite a bit better than ordinary heated-cathode vacuum rectifiers. Approaching that of a so-called “ideal rectifier in series with ideal resistor”.

Specifically, around 430 Ω between 50 mA and 150 mA for a type 83 MV bulb. Not exactly linear-like-a-resistor, but pretty close.

If I recall, it also seemed that most recommendations were for choke-input filtering lines, as from Imin to Imax, there was a substantially lower 'droop' in nominal power supply output.

Still … your garden variety low VF fast-recovery Schottky exceeds a heated-cathode MV rectifier quite handily, and doesn't have a particularly sizable series-equivalent-resistance.

I'm sure there are endless fora here, that talk about the theory as to why your promotion of MV rectifier's 'sweet result' are as “sweet” as reputed to be.

⋅-=≡ GoatGuy ✓ ≡=-⋅
 
Do not use mercury vapor rectifiers for voltage multipliers. They don't like capacitor input type loads. Full wave choke input is best for mercury vapor rectifiers.

Also, it is a good idea to preheat a mercury vapor rectifier before applying HV to avoid arc back or damage to the cathode or coated/heat shielded filament. Broadcast transmitters have automatic pre-heat for mercury vapor rectifiers, whether it's an 816, 866A, 872, 575A, or 857B. Most of the tube transmitters have been replaced with solid state, but many tube units are still used for standby at radio stations...

Avoid breakage, the vapor is hazardous. Keep a box of powdered sulfur around in case of a break or spill. You may find powdered sulfur at a garden supply store (not the big box ones)..

The voltage drop across a mercury vapor rectifier is around 15 volts, regardless of current drawn. This helps with regulation of HV for supplying class AB or B type loads...
 
Agreed, 1N4007's are a LOT cheaper....and have no warm up delay.

There is a considerable amount of commercial equipment (i.e. transmitters) still out there using M-V rectifiers in the power supplies.

There really aren't all that many Hg vapor rectifiers in use in transmitters these days. Almost all of them have been replaced with solid-state solutions that are superior in almost every way. These days you can get silicon rectifiers that can take several amps and hold off 10 kV for a fairly reasonable price.

Some guys do like to experiment with them. Some hams like to use them in transmitter power supplies, often in AM transmitters using older tubes like the 810 or 833. I've been offered a whole bunch of 866s, but turned them down. There's a lot more mercury in an 816 or 866 than there is in a fluorescent tube, I want to say it's an order of magnitude more. Still not as bad as an Ignitron, but the only people who buy ignitrons these days are aluminum smelters and physics labs.

If you want the pretty glow, use a couple of 0D3s as a voltage reference. They glow purple.

There's another issue with Hg rectifiers, and that's noise. Anything that involves plasma isn't going to be quiet, and Hg rectifiers are no exception. They are just about the noisiest solution you can get. Even the 0D3s I mentioned above are noisy, but not like an 866. You will likely fight with this if you try to use them. Unfortunately, there's really no good way to shut them up, since even if you throw a cap in parallel you've still got all the noise coming from the tube itself.
 
An old 50 kW am transmitter I had my hands in (and still a standby) used six 857B mercury rectifiers for the 15 kV supply for the finals (recall they were air cooled 5671's...) The reliable Gates and RCA 1 kw early vintage AM transmitters used four 833A tubes (2 for modulator, 2 for RF) and there were 866A's and 872A's in the LV and HV power supplies. The later vintage tubed am transmitters used four 4-400A tubes (2 for AB1 modulator, 2 for RF). Power supply had 8008 merc rectifiers. The 833 ran about 2500 volts on the plates at little over half an amp for the pair; the pair of 4-400's liked around 3100 volts and 460 mA for the pair to get a kilowatt of RF at the antenna. The modulators loafed since approximately 500 watts of audio was needed for 100% modulation of the RF. In the day, it was customary to modulate asymetrical to make the station loud, which put stress on the HV components...but the managers liked a loud station...fwiw

As far as noise is concerned, at high power levels the noise from the M-V rectifiers was insignificant...There was one AM transmitter built by RCA (BTA5H) that used six 5563 mercury thyratrons for the 5 kV plate supply...and the plate voltage could be adjusted from the front panel without a large lossy rheostat...Typically, all the tubed AM and FM transmitters above 1 kW required three phase AC power in the day...Even RCA's 1960's vintage FM transmitters (BTF10 series) used six 8008 merc rectifiers for the 4CX5000A RF final....Some of the RCA BTF series transmitters still are on the air...

For what it's worth, I recall the old Gotham disk cutting amplifier (PFB150WA) for the Grampian mono cutter head used 3B28 gas rectifiers...to keep the voltage constant for the 811A output tubes...
 
Last edited:
I should have clarified a few things. I have used MV rectifiers before, and I get it that SS diodes are superior in every way. I am aware of the dangers of MV, I like the way they look and I have a lot of them in stock; I have 866,816,8008, and 83. I also like Xenon gas diodes, which I also have many of. I want to build a 6c33c/6s41s/6as7 amp. I have a nice 75v 3.5A transformer that would be perfect for doubler duty. I will probably go with 6D22s for this project.
 
Do not use mercury vapor rectifiers for voltage multipliers. They don't like capacitor input type loads. Full wave choke input is best for mercury vapor rectifiers.
I'm not advocating MV rectifiers at all, mostly due to their poor efficiency in comparison with SS diodes. But voltage doubling doesn't exclude choke input. Get a Delon doubler and insert the choke into one of the AC leads - bingo! It even should work with a Greinacher doubler.
Best regards!
 
I'm not advocating MV rectifiers at all, mostly due to their poor efficiency in comparison with SS diodes. But voltage doubling doesn't exclude choke input. Get a Delon doubler and insert the choke into one of the AC leads - bingo! It even should work with a Greinacher doubler.
Best regards!

Nice, thanks for the tip! I have actually decided on a hybrid bridge using a 3b22 gas diode. I am surprised more people haven't used them in projects, they seem easy to implement and can handle a lot of current (they do have a 6A heater requirement).