View Single Post
  #10   Report Post  
Posted to rec.audio.tubes
Patrick Turner Patrick Turner is offline
external usenet poster
 
Posts: 3,964
Default 6N1P conflicting heater/cathode voltage specs



Nick Gorham wrote:

Iain Churches wrote:
"Patrick Turner" wrote in message
...


Ian Thompson-Bell wrote:

I have two data sheets for the 6N1P, both Russian with English
translations, which seem to give conflicting values for maximum heater
to cathode voltages.

One specifies Ukh as +100V and -250V. As the k precedes the h I assume
these are cathode voltages relative to the heater, so this spec says the
cathode can be at most 100V positive with respect to the heater.

The other describes these parameters in words. It says: Voltage between
cathode and heater:

- with heater at positive potential 100V
- with heater at negative potential 250V

I take this to mean the heater can be up to 100V above the cathode or
250V below it, which seems to me to be the exact opposite of what the
other spec says.

So, which is right?

Cheers

Ian

Probably all is misleading.

I'd never run a cathode at a dc potential of more than +/- 70V with
respect to the heater.

So if you have a CF where the cathode swings maybe 135Vrms, as might be
the case if driving
an 845 grid in class AB2, then be prepared to have the filament and itsa
voltage supply and transformer winding
also swing as well, which may mean you also need a screened tranny
winding to prevent stray C coupling to the mains.



If you don't have a dedicated winding, you can use a pair of
resistors across one of the B+ first filter caps to total 1M and adjust
their values to give you the elevated voltage you need at the centre
of these two. You then apply this voltage to both sides of the heater
cap via 100 Ohms resistors. The heater supply must, of course,
float. It works well. Details are given in Rozenblit's book. a small
decoupling cap (I use 0.1µF) is required at the junction of the two
resistors, and also on each leg of the heater at the tube socket.



I think Patricks point was if you have a valve where the cathode moves
+- 190v then no fixed reference will work for that heater, and the only
choice is a supply for that valve that only feeds that valve and is
allowed to swing with the cathode. This of course implies that for two
chan, you need two of the supplies.

And then as mentioned, the cathode will have to drag the supply around
with all its associated capacitances.


--
Nick


Yes, you got my point OK.

In many amps the swing is less than we need to be concerned about,
eg, say a driver tube ahead of a trioded KT88 etc, and where the driver
has Ek at say +10V.

But when the output tube is an 845, or where there is a lot of CFB, say
in a McIntosh,
where the drive voltage to an output grid can be 150Vrms, and there is a
cathode follower,
then you have some concerns about exceeding specs sometimes.

McIntosh were not concerned though. They never empoyed floating heater
windings.
If McI did id, then why not everyone else?

Well, I have to fix a lot of failed and badly designed hi-end brandname
amps
and this all teaches me that McI and others are
eternal optimists, and that **** happens!!!!

Chinese tube making quality control standards leaves much to be desired.

Big Vswing + chinese tube = probable disaster.

In ARC VT100, I have seen 6922 used as direct coupled CF to drive output
tubes.
There is +450V at the CF anodes, but cathodes are at -60V for biasing
the output tubes
and wanted max swing is 65Vrms to the 6550 grids with the mild CFB with
UL.
The heaters are at 0V and the whole set up using 6922 stinks of brown
smoke sooner rather than later.
I replace the whole gain stage with with 12BH7, and **** those 6922 outa
there,
no CF and just have normal C&R coupling to output tubes with individual
biasing,
so the whole **** load of unreliability is jetisoned.

A CF will typically have an Rout = 1/gm = say 400 ohms if gm = 0.0025A/V
..

The capacitance between a 6.3V winding and 0V or some shield in a small
1:1 tranny used between an existing 6.3V winding and the CF is next to
nothing, maybe 200pF,

So the amount of C "dragged around" is SFA depending how much effort you
use to set things up.
But of course at above 100kHz the C does cause an extra phase shift so
with FB the C does no good,
but then with a good knowledge of critcal damping, a wise designer will
get over that little
problem like all the rest. 0.002uF would be a much worse problem!

In a bloomin VAC 7070 amp with 8 x 300B for both channels,
there are 8 heater supplies taken from 4 small transformers, 2 windings
per tranny.
With direct heated tubes, you MUST have floating heater windings.

So having a couple is child's play if you need them.

As I said, to avoid the C between a 6.3V winding and a nearby mains
winding, and the
ingress of switching noise, a second small 1:1 tranny may be used
with a larger gap between P&S to reduce the stray C to mains by 1/10.
Or fit a screen between P&S, which is the best.

One could then even connect the CT of the heater winding directly to a
live
indirectly heated cathode.

I have often thought of doing a 211 amp which
might run in class AB2, so CF drive is a good idea, and a high swing is
needed.
I would probably use a an EL84 in triode as the CF driver, Rout
200ohms,
so noise from stray C probably won't affect it much.



Patrick Turner.