"Mikkel C. Simonsen" wrote:

Jeff Goldsmith wrote:

6106 or 6087? 75mA current rating on these.
GZ-37? I've heard they heat slow. Limitation to input cap size with
these.

No, not really. The peak current limit is the same on both the GZ34 and
37 (750mA), so the cap. size limit should be the same also. There's no
size limit mentioned in the data sheet.

You can find the GZ37 datasheet he
http://frank.pocnet.net/sheets/129/g/GZ37.pdf

Best regards,

Mikkel C. Simonsen

Jeff Goldsmith

Armand wrote:

Is there a drop in replacement for the 5V4 that has a slower voltage rise?
Thanks.

The GZ34 could be used as a series diode, with both diodes within
paralleled, and in series with the flow of DC OUT of the first
cap from a silicon rectifier, say before the choke which feeds the second cap in a
CLC filter.
This way the size of C1 can be 470 uF, no worries, and the tube diodes
are slow to turn on, yet when they are on the series R would be about 15 ohms,
and current ability would be high.
Thus the initial surge of current at turn on is controlled.

However surge currents in the tubes are not a problem from a cold start.
Its the restart when everything is hot that causes a current surge.
Directly heated cathode tube rectifiers generate a high peak voltage well before
the output tubes start conducting, and yet this doesn't seem to cause problems,
except that the supply caps must be able to take the voltage peak surge.

Brief power interruptions with the diode cathode still hot will
cause a current surge when power comes back on.
In my big amps I use a 20 watt R in series with the HT winding, and this
is shunted by a relay after 4 seconds.
If there is a turn off then turn on of the power over a period longer than
0.5 seconds, then the relay shunting the series R must go through the 4 second
cycle.
Then the R limits the current fine.
The control circuit is a couple of diodes and transistors, and an RC time constant

circuit.

In normal cold start conditions, the series R limits inrush current from the
mains,
and after 4 seconds, B+ goes near to its final value, then a lesser surge occurs
when the
R is shunted.
The same sort of cascaded smaller than otherwise current surges occur
during brief power interuptions.
The benefit of all this is that the mains fuse doesn't have to be such a high
value,
as it would if you have high inrush currents, especially with silicon rectifiers,
and the typically monster values of capacitors I like to use.

If you have a high value of mains fuse, your protection against
catastrophe is reduced, and it may cost you your house, since a transformer
might get a fault, overheat, catch on fire, and set the house alight,
before the bleedin fuse blows.

The mains fuse needs to be a value just above that which causes nuisance blows
under normal conditions.

If an amp draws say 220 VA from the mains, and the mains is 110v,
then you have 2 amps of input flow.
If the fuse was 4 amps, then there has to be 440VA of input before it might blow.
If it was 6 amps, then 660 VA of input is needed, but in practice, a fuse
might not blow until the current goes 20% above the rated amperage,
and stays there for awhile.
So if an amp draws 800 watts before the fuse blows,
it could be because the bias has failed, and serious collateral damage could occur

before a the damage eventually causes a direct short circuit.
Fuses on the B+, between C2 and the centre tap of the OPT are a must,
if there is no protection circuits operating off a 10 ohm R in the cathode
circuits
of each and every outout tube.

In the old days of tube amps, few makers fitted protection against
excessive cathode currents, or had any indication that the balance of
output tube current was correct.
Its these very old bangers that are most prone to failure, and some form of
fault indication using a few SS parts on a small board can now be easily
fitted.

Current surges in power interupted output stages can be reduced by the
size of large values of C&L in CLC type filters, and I'd say the series
tube diodes would help, and be quite OK for most class A or AB amps,
where serious regulation isn't really necessary.
I know a guy who uses a GZ34 in series with his SE 845 amp supply caps,
and he says it works quite nicely.

Patrick Turner.