Standby Switches
John Galbraith
functions are not necessary on guitar or hifi amps. The concern is "cathode
poisoning", which is apparently legit for high power transmitting tubes, but
has also leaked into hifi lore where the need for it may actually be
questionable. The guitar amp gurus are saying that you really only need the
standby switch to keep the amp quiet during set breaks, not for the health of
your tubes.
Delaying the B+ sure is a PITA for sure. I've got a solid state relay setup
switching 340 volts AC that actually works to implement this delay. It's
triggered by a 555 timer that delays the B+ for 45 seconds when you turn the
amp on, and keeps the filaments warm for 45 seconds after you turn the amp
off. It's more complicated that I would like, but the main concern is that SS
relay inserting more semiconducting switches in the power supply where it
counts. If cheap and slow rectifier diodes add RF noise, then the SS relay
surely does as well. So we spend extra money and effort on fast switching
FRED diodes and bridges, and it's not clear that the SS relay would not
clobber those efforts.
So I would love to just ditch the B+ delay and that relay, and just hit the
tubes (KT88, ECC88, and 6SN7 at 450V in this case) with B+ at the same time
the filaments turn on. It would be so simple, and keep that SS relay out of
my amp.
Any opinions here?
John
Matt Wiebe
I'm a believer in delayed B+, but since I use tube rectifiers its
easy. On my silicon rectifiers I use thermistors to slow things down.
Requires a separate filament transformer however. Yes its easy to
poison transmitting tubes, but I'm not convinced you don't have to
worry about it for lesser tubes.
Try some high-voltage Schottky diodes from Cree to compare with your
FREDs. It's not the speed that's important but what happens when the
bridge shuts.
Matt
> --
>
> You received this message because you are subscribed to the Google
> Groups "audionerdz" group.
> To post to this group, send email to audio...@googlegroups.com.
> To unsubscribe from this group, send email to audionerdz+...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/audionerdz?hl=en
> .
>
>
John Galbraith
> John,
> I'm a believer in delayed B+, but since I use tube rectifiers its
> easy. On my silicon rectifiers I use thermistors to slow things down.
> Requires a separate filament transformer however. Yes its easy to
> poison transmitting tubes, but I'm not convinced you don't have to
> worry about it for lesser tubes.
OK. Last year, when I bought my two power transformers, I chose a Plitron
toroid that has the 340V winding and a 12 amp 6.3V winding. I've learned a
lot about power supplies since then, and I might have chosen a different power
transformer. In particular, I recognize that it is inconvenient to have your
filament transformer on the same core as your B+ winding, for this very
reason. If you want to try a DC filament supply, you also can't use this
transformer. Plus, I don't need 12 amps; I'm going to use more like 7. So I
understand now why you want separate filament transformers. Also, with
KT88's, I could have used a slightly higher B+ voltage...
I guess I'm ready to just abandon the filament winding that I have, and just
use the 340V winding on it and yet another transformer for the filaments.
I've read that the PT works better if you have all the windings loaded which
is why I have been hesitant to abandon the filament winding.
I purchased some thermistors to try as well, but I have been hesitant to use
them because even when warm, they have large resistance for the current that I
was not yet ready to accept. I would on the filaments, I suppose, but rather
not on the B+.
What's the justification or experience for your belief in delayed B+?
> Try some high-voltage Schottky diodes from Cree to compare with
> your FREDs. It's not the speed that's important but what happens when
> the bridge shuts.
I saw those for sale on partsconnexion, but was scared of the price.
Especially when I need 16 of them. These dang choke-input supplies ask a lot
of the diode: 2*sqrt(2)*power transformer output, when the supply actually
makes .9*PT output. So that doubled the number of 1200V parts I needed and
added other complexity besides. I am interested in this optimization,
however. I might try it on a later project that needs fewer of them (like an
RIAA preamp, if I don't choose to use a tube rectifier).
As always, thanks for the insight, Matt!
John