All rights of photos and text reserved. Usage of photos or text from my blog on other websites or for any other purpose only with prior permission. If you want to use any material from my blog please contact me by email.

Wednesday, March 25, 2015

Tube of the Month : The 816


In the previous post about the latest build of the Octal Preamplifier, I showed a power supply with a hybrid bridge which uses two 816 mercury vapour rectifiers. Since this was the first time I used these, it is a good opportunity to write a post about the tube.

The 816 is a smaller version of the 866A with less voltage and current handling capability. The 866A is way oversized for most audio projects so the 816 is a good alternative if you must have that magic blue mercury glow.

The 816 has the same pinout as the 866A. An UX4 base, of which only two pins are used for the filament. The plate is wired to a cap on the top. While the 866 has a medium size cap, the 816 has a small cap. A pair 866A can deliver 500mA when used up to it's peak inverse voltage limit and a whopping 1A at peak inverse voltages up to 2500V. A pair of 816 'only' handles up to 250mA (125mA per tube). Still plenty and more than enough for the preamp it has been used in. The peak inverse voltage rating of the 816 is an impressive 7500V. The big advantage of the 816 over the 866A is the filament current. It 'only' needs 2A at 2.5V while the 866A uses 5A.
This makes the use of the 816 a lot easier compared to the 866A. You get the same beautiful blue glow from the 816 but in a much smaller bottle. Which in turn can be an advantage since it does not need so much chassis space as the impressive 866A. The ST12 bulb of the 816 is about 10cm high and has a diameter of about 4cm. The dimensions of the 866A's ST19 bulb are 50% bigger. See the photo below for a size comparison between these two mercury vapour rectifier tubes. For further technical specs of the 816, please refer to the datasheet.

Both the tubes above have been made by RCA. They used to package their mercury tubes very careful with soft padding inside the box:

This one contains quite a lot of mercury.

The mercury is plastered all over the inside of the tube. This is the reason why they require a special conditioning when used the first time in a power supply or after they have been transported. Running only the filament for 30 minutes vaporises the mercury which then condenses at the base. After such preconditioning the tubes can be used with high voltage. This procedure prevents that any mercury drops are still inside the electrodes and could create a short.

The base of an RCA 816:

Another RCA 816, labelled as JAN type for the military:

General Electric GL-816:

The base:

There seems to be much less mercury in this one compared to the RCA shown above.

The mercury deposited on the metal stems:

Close Up:

More GE 816s:

These have a warning label on the base regarding X-rays. However such tubes only produce X-rays when operated close to their peak inverse voltage rating. At the typical voltages in audio amps this is not of concern.

Raytheon RK816:


United Electric:

CBS Hytron:


Due to the mercury content, I am not going to open a tube to show the internals. Here a close up showing some mercury condensed on the inside of the glass:

Instead some photos showing the beautiful glow:

The intensity of the blue glow depends on the current drawn. At low current it is very dim:

Becoming more visible as the current goes up:

At medium current the glow starts to fill the entire bulb:

Full current:

It should go without saying that such tubes ought to be handled with utmost care. Mercury is very toxic. Only use these tubes if you know what you are doing.

Best regards



  1. Thomas, do you see an advantage in your mixed full wave rectifier scheme (mixing in 2 6ax4's) over using 4 identical tubes? 4 816's could run off a single 10a fil trans. Also thinking about four 1616's vs 2&2. Thx, Jason

    1. Hi! Although some people claim to hear differences between rectifiers, I find them rather small. So this was done for cosmetic reasons. You can't heat 4 816 or 1616 in a bridge from a single winding. You need at least 3

  2. I'm looking to buy a B&W L 1000A linear that uses 4 816 tubes. Are there any (safer?) replacements for them in case they need to be replaced??

    1. It's a different socket, so it would require an adapter of some form.. but the 3B28 is a xenon-based gazotron rectifier meant as a straight substitute for the 866A. So it should have higher than needed ratings to replace the smaller 816s.

  3. I saw a Barker & Williamson L 1000A linear that uses 4 816 mercury tubes. Are there any (safer?) replacements for them?