I want to replace my 6D22S tubes with the 816. I've heard there MAY be some hum with the 816. any other tips or hints I should know
Hey Dennis, I've used the 83. I would suggest you treat it as being delicate, ie warm it through before applying HT and watch the series resistance in their loop as they pass current readily.
They are supposed to be useful for class AB stages with non constant current draw, maybe closer to SS diodes. They can be a little noisy.
They are supposed to be useful for class AB stages with non constant current draw, maybe closer to SS diodes. They can be a little noisy.
Not necessarily hum. Since this diode uses a glow discharge, it can develop negative resistance that can cause RF noise.
You also gave to keep an eye on the Isurge, since this can be quite limited for Hg diodes as well. It's seldom that you'll see these used with a capacitor-input ripple filter. Choke input is almost always used.The 816 has an Isurge= 500mA -- half the per-plate Isurge of the 5U4GB, and way lower than the Isurge of most Si diodes.
The main virtue of Hg diodes isn't current sourcing, but rather the nearly constant cathode fall that remains pretty much fixed for widely varying current demands, more like an SS diode.
As for how to use, make certain you separate the heater power from the HV DC.
When installing or any time you disturb the Hg pool, run heaters only for 30 mins to make certain you don't have any droplets making plate-to-cathode shorts.
Before powering up the HV DC, prewarm for ~60 secs to get some Hg vapor pressure developed.
Performance can be erratic since the Hg vapor pressure depends on ambient temps.
I use 83 in a preamp and 866 in power amps. Would not seriously consider changing to anything else.
Why does everyone think they have to treat the 83 with kid gloves? I have a tube tester that uses a 83 and it applies HT immediately and its still living.
Have used 83's with preamps and my preamps are dead quiet.
I use full wave AX50 mercury vapour rectifiers with separated filament trafo (2x 4V, 3.75A) in my -stacked- 300+300V/200mA PSU (requiring 47R/5W resistors to HV secondary).
If you use it daily, necessary preheating time is about 30-60s. Few days unused state requiring 5-30 min preheating.
If you use it daily, necessary preheating time is about 30-60s. Few days unused state requiring 5-30 min preheating.
In my experience, the 816 does not like cap input, choke only. 60 seconds preheating before you apply HT, and you should be OK.
I've used 83's in both pre and power amplifiers without any issues. One suggestion would be to try it, and if it isn't working you can always swap in an 80.
My I-177B manual specifies 30 second warm up before testing. During the warm up period, there is no current demand on the #83 rectifier.
Does the 83 work ok with a cap input supply or should it only be used in choke input? I would like to try the 83 with 50uf for the first cap in a CLC configuration, and I am wondering whether that will present any problems.
What are issues with using a choke first instead of a cap? For my the 816 is almost to big for my space and the 83 is bigger yet. Is there a merq tube smaller than the 816?
From what the data sheets say, 40uF would be the max cap input size, and 3H the minimum choke input inductance.
I built an EL84 ST35 clone using an 83 into a 22uF cap and its still working just fine 10 years later.
I built an EL84 ST35 clone using an 83 into a 22uF cap and its still working just fine 10 years later.
The spec sheet I found states "When a filter-input capacitor larger than 40 microfarads is used, it may be necessary to use more plate-supply impedance than the minimum value shown to limit the peak plate current to the rated value." 50 ohms is the stated impedance.
Think as a street light tubes, chokes are must.
Cap input is like short to them. I use lclc. 4h+100u' s , not even delay relays
Bright flash at start, then it horks happily. After a months, same performance.
Dead quiet. Less than a minute, and the mercury fog on glass is gone.
Last edited:
- Status
- Not open for further replies.