[Patrick Turner]

On Aug 7, 11:57*pm, Bret L wrote:
On Aug 7, 6:07*am, Patrick Turner wrote:





On Aug 7, 1:21*am, Bret L wrote:

Anyone have a schematic?

I typed in the heading of your post to Google and found lots of hits
but no schematic, but then I only tried for 5 minutes.

Maybe you will find the schematic is something Prima Luna wish to keep
secret to stop anyone here, there, or any damn where from copying what
must be a "trade secret".

But I did see pictures of the auto adjust board and it is crammed with
a much more complex circuit than the whole of the audio amp circuit
and it has transistors and a multi-pin chip of goodness knows what
kind.

Unfortunately, I cannot condone the use of such gee whiz complex solid
state junk included in any tube amp, although it is becoming more
common as makers try to add in crap that really does not belong, or is
terribly fragile, like the Prima L board or is needed if you wish to
use an i-pod or computer digital signal feed.

The design of biasing methods in many tube amps is the one thing that
is most commonly treated with utter contempt by many makers including
top brands like ARC, and its almost as though the designer wishes the
owner to get all muddled up up and get the biasing wrong and then have
have the amp fail and so the owner buys another. Or else the designer
wishes the person adjusting bias risk his life measuring voltages
across a hard to get to resistor in the anode circuits which are at
+500Vdc. Or else the designer uses utter crap quality pots, etc, etc,
etc. In the vast majority of tube amps there is no attempt to include
a circuit which detects excessive Idc in any one or more output tubes
and then turn off the amp automatically so smoke and a blown OPT is
avoided.

Servo circuits which adjust grid bias voltage to keep the Idc constant
are rather useless like tits on a bull when a tube decides to become a
bit gassy and conduct huge currents for awhile until melt down and a
short circuit occurs.

And the Idc varies hugely in class AB amps so adjusting grid bias
voltage for constant Ik in AB op is utterly pointless, so whatever a
maker tries in any attempt for intelligent bias voltage control/
adjustment it MUST allow Idc in tubes to increase with AB operation
and the condition of fixed bias MUST be maintained where the amount of
class AB much exceeds the first few pure class A watts, which is
usually the case in most tube amps made now to get high output power
for sales reasons.

There are some simple ways to monitor bias condition and protect an
amp at my website. NO opamps are ever used.

Patrick Turner.

*Good design uses a resistor specifically intended to burn up in a
high current condition in the cathode circuit. With transmitting tubes
a light bulb is sometimes used.- Hide quoted text -

- Show quoted text -

Yes, indeed the sacrificial resistor is one solution to protection
rather than have fuses which may be difficult to obtain,
eg, typically 200mA rated which offer some protection against
catastrophic faults. But owners of hi-fi systems do not want to have
to change a fuse where they will often choose the wrong value, say 2
amps instead of 0.2 amps. Nor do any of them ever want to pay a tech
to solder in a replacement resistor. So the most convivial and
civilised solution for the average hi-fi amp owner is active
protection where nothing ever fuses, but the amp is turned off from
the mains when the Ik rises above say 150mA for longer than 4 seconds.
Simply turning off the amp and then back on again resets the amp. Such
a scheme offers protection against turningup the volume with a shorted
speaker cable, or when obe tube of a PP pair has an open heater
winding leaving just one OP tube to struggle on as a very badly set up
single ended tube which soon begins to conduct too much Idc.

Quite a few of my customers have much appreciated the protection
measures I offer in all my amps - no huge bills for OPT or PT.

There are more ways than one to achieve protection. I like to limit
inrush current needed to charge up big values of B+ filter caps by use
of a resistor in series with mains and once B+ has risen to 3/4 of
full working value the R is shunted by a relay. One will find that the
mains fuse value can be much smaller then if no such timed inrush
circuit is present. Thus the mains fuse will more likely blow if say
one OP tube of say a total of 4 conducts 300mA in addition to the
other 3 conducting 50mA each. But tubes can conduct 100mA for awhile
until they lose all bias control and become saturated. The excessive
Ia needs to be monitored, hence my use of applying ONE average value
of bias to say 2 output tubes and then having a balance pot to trim dc
current equaity plus a BJT long tail pair to make to LEDs glow equally
when the Ik iof the two OP tubes is equal.

For protection I have a 5VA auxilliary power trans giving +/- 12V
rails for operating all the protect & delay & LTP circuits.

One other simpler way is to monitor Ik and have an SCR which latches
on with excessive Ik and pulls the B+ to ground through a rugged 220R
which will overload the B+ rail causing the mains fuse to blow. This
is the notorious Crow Bar solution, and the circuits can be operated
by a +12V rail generated off the filament voltages. The mains fuse can
easily be accessed and replaced.

Usually there is no room to put an incandescent light bulb in series
with a cathode to ground, and then it needs to be a high voltage rated
type; it is a primitive non-solution, and ppl cannot buy old fashioned
light bulbs in Oz now because they have been deemed to be greenhouse
incorrect and the only globes are now compact flurescent types.

Patrick Turner.