hybrid power supply: diode bridge into diode tube

[engels]

This is a very basic question; What are the pros and cons for the following setup:


(The diode tube is, of course, indirectly heated)

It is intended to delay the HV supply and eliminate standby switch, as well as find some use for all those damper and other TV diodes that are sooo cheap.

[Fuling]

Perhaps you could put a capacitor between the bridge and the tube diode to decrease the current peaks trough the tube.

[Frank Berry]

engels,

I have done this on a couple of amplifiers. It will give you a B+ delay and additionally will protect the electrolytic capacitors in the event of a rectifier tube short.

[ray_moth]

My PP EL34 triode-strapped amp uses a bridge of MUR4100E diodes and a filter consisting of C - L - 6D22S TV damper diode - C.
This gives the required slow start but it doesn't reduce SS diode hash (which I can't hear anyway).

I would use a hybrid bridge (which does reduce hash) but I can't afford the voltage drop (it's about twice the voltage drop I get with my present arrangement). Heck, I'd use a choke-input filter, too, if I had the voltage to spare, and get rid of the spiky current in the first cap! (Amplimo obviously designs its power transformers for cap-input filters. What else can you do with a 340v secondary?)

[Merlinb]

Some explanation of that circuit here:
http://www.freewebs.com/valvewizard/bridge.html

[engels]

Wow, thanks! That's a very usefull article although there's one thing I don't get. It's the MUST resistance between the bridge and the tube:



Minimum Limiting Resistance: A valve rectifiers must have a resistance in series with each anode. The data sheet will provide a Minimum Limiting Resistance (Rlim(min)) for different supply voltages.
For example, when supplied by a transformer rated at 300-0-300Vrms, the EZ80 specifies Rlim(min) = 215 ohms per anode. However, the quoted resistance assumes zero source impedance, when in fact we're using a transformer which has a finite resistance. In which case, the following formula can be used:

Rlim(min)per anode = Rs + (n * n)* Rp
Where:
Rs = DC resistance of the transformer secondary winding. (If using a centre tapped transformer you should take the resistance of one half of the secondary winding.)
Rp = DC resistance of the transformer primary winding.
n = Secondary to primary turns ratio (equal to the secondary voltage divided by the primary supply voltage).

For example, if we were using a mains transformer with a 240V; 100R primary and a 310-0-310V; 60R half secondary:
Rlim(min)per anode = 60 + (0.77 * 0.77)* 100 = 119R
So we would use one 120R resistor per anode.
Because they will have to carry the ripple current of the reservoir capacitor it is best to use high wattage resistors. Even 7W resistors will usually get quite warm. Remember that the voltage drop across the limiting resistors will cause the HT fall proportionately.
If the valve rectifier is being used in series with the HT after a silicon bridge as mentioned above, a single resistor can be used [as shown above], although its power dissipation will be doubled.


Anyone can imagine what this resistance does, how it heats up and loooks like half the power of the transformer will probably be dissipated in that part. So why is the MUST???? what exactly happens if there's no series resistance?

[Miles Prower]

Originally posted by engels
Anyone can imagine what this resistance does, how it heats up and loooks like half the power of the transformer will probably be dissipated in that part. So why is the MUST???? what exactly happens if there's no series resistance? [/B]

You may, or may not, need a series resistor. The A Number One problem with using VTs is that this is a high voltage, low current device. Even taking one of the beefiest VT rectifiers, the 5U4GB, it has a surge current rating of 1.0A (2.0 if you parallel plates). Now, that may look impressive until you consider that even a small silicon diode might have a surge rating of 10A or more. The VT diode won't be able to drive a 100uF filter capacitor, at least not for very long.

If you connect that VT diode between a SS bridge, and it uses the typically large first ripple filter capacitor, you will exceed the VT diode's surge rating, and it won't last for very long. You have to treat that as if it were a VT rectifier all the way through. You need to use a smaller filter capacitor, and add a ripple filter to clean up the mess. That also means that the final DC voltage will be lower.

If you don't do that, then you need a resistor to reduce the surge current.

[engels]

Originally posted by Miles Prower


The VT diode won't be able to drive a 100uF filter capacitor, at least not for very long.

Who on earth would put a 100uF after a tube rectifier? Are you trying to say that if I'm looking at this as if there's no bridge and it's a normal tube rectifier and I treat it as it is supposed to be treated there's no need for the limiting resistor?

[Miles Prower]

Originally posted by engels
Who on earth would put a 100uF after a tube rectifier?

I've seen it done before. I took apart a 6V6 amp from an old record player that had a 5Y3GT connected to a 100uF / 450Vdc filter capacitor as the only filter element. There wasn't very much that was useable there, but that's what they did. It was a cost-cutting measure, I'm sure, and the manufacturer was counting on an atechnological public to not notice that having to replace those 5Y3GTs (did they even know what a 5Y3GT did?) so often was an abnormal situation. Then, again, what difference did it make when you could get new ones at any corner 7 -- 11, etc.?

Are you trying to say that if I'm looking at this as if there's no bridge and it's a normal tube rectifier and I treat it as it is supposed to be treated there's no need for the limiting resistor?

That's sure how it looks to me. My guess is that they're recommending this as a mod to a pre-existing design. SS power supplies tend to include much larger input filter capacitors. (I keep that capacitor small even with Si diodes since most of the HV power xfmrs were designed assuming VT rectification and its lower surge currents so as not to over stress the xfmr.)