WTB: Looking for 6DA6/EF89 tubes -- Need 4

[karchiba]

Good afternoon all,
I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.

Any leads?

Thanks,
_kevin

[Big Bad Bob]

On 03/28/11 14:52, flipper so witilly quipped:
> On Mon, 28 Mar 2011 13:21:46 -0700 (PDT), karchiba
> > wrote:
>
>> Good afternoon all,
>> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>>
>> Any leads?
>>
>> Thanks,
>> _kevin
>
> http://www.abcvacuumtubes.com/vacuumtubes_tubelist_tubes_6d4-6jz8.html

hey, nice link, thanks (high voltage capacitors and tube sockets also)

[Patrick Turner]

On Mar 29, 7:21*am, karchiba > wrote:
> Good afternoon all,
> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>
> Any leads?
>
> Thanks,
> _kevin

I've never seen a 6DA6.

Its equivalent is EF89, a remote cut off RF pentode. ( Remote cut off
= variable µ, which means grid1 bias voltage is used to alter gain, as
in a compressor-expander or it is used for RF or IF amps with
transformer coupling and altered gain by AGC bias control to g1. )

Maybe a 6BH5 would do the same job.

There could be a few other remote cut off RF pentodes you could use
but with different pin out for the mini 9 pin base.
Check the RCA receiving tube data books.

Patrick Turner.

[Big Bad Bob]

On 03/29/11 01:38, Patrick Turner so witilly quipped:
> On Mar 29, 7:21 am, > wrote:
>> Good afternoon all,
>> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>>
>> Any leads?
>>
>> Thanks,
>> _kevin
>
> I've never seen a 6DA6.
>
> Its equivalent is EF89, a remote cut off RF pentode. ( Remote cut off
> = variable µ, which means grid1 bias voltage is used to alter gain, as
> in a compressor-expander or it is used for RF or IF amps with
> transformer coupling and altered gain by AGC bias control to g1. )

functionally correct, but there's a bit more to it than that. sharp
cutoff pentodes (aka 'regular pentodes') have too much nonlinearity at
the cutoff point, but higher gain. Remote cutoff pentodes, on the other
hand, have specially designed grids that 'smooth out' the cutoff curve
(more linear) but lower the overall gain, making them suitable for AGC
and 'electronic volume control' circuits, and low distortion amplifiers.

For maximum ideal open loop gain, a pentode will operate very close to
its cutoff point, but this also causes distortion due to high
nonlinearity in this range. So for high open loop gain with lower
distortion, you would use remote cutoff pentodes. In cases where low
distortion is less important (like FM IF stages) you can use sharp
cutoff and bias them ideally.

Similar implications for pentode audio amplifiers and pentode/beam
tetrode output stages (typically sharp cutoff pentodes), biasing them A
or AB instead of B (to stay in the linear range more often), and the
effect it has on harmonic and IM distortion.

I think wikipedia has a good description of sharp vs remote cutoff
pentodes on their 'pentode' page.

http://en.wikipedia.org/wiki/Pentode

'Nuff probably. the horse is now 'undead'

[Patrick Turner]

>
> >http://www.abcvacuumtubes.com/vacuumtubes_tubelist_tubes_6d4-6jz8.html
>
> hey, nice link, thanks (high voltage capacitors and tube sockets also)

The site at http://www.abcvacuumtubes.com
looks good.

They have
6FW5, $7 line output beam tetrode just like 6CM5/EL36 but no top cap
- makes an excellent triode for hi-fi, Ra 600 ohms, µ = 5, Pda 18W.
6FQ7, 6CG7, $9,
6EH7, $3, hi gm frame grid pentode, makes triode with µ =60 and Ra =
12k, good for phono amp,
6BX6, $3 also good for Quad-II, Mullard 520 etc EF86 input
replacement with different pin out, but more gain, good for gain in
triode.
6AS7, $6,
6AW8 $3 good triode-pentode for 1,001 uses,
6BK8, $4, drop in substitution for EF86
6BL7GTA $9, gutsy low µ octal twin triode when you want low Ra and
high Ia and high Vo, much better than 12AU7, 6SN7 as a driver tube.

Patrick Turner.

[Patrick Turner]

On Mar 30, 5:08*am, Big Bad Bob <BigBadBob-at-mrp3-
> wrote:
> On 03/29/11 01:38, Patrick Turner so witilly quipped:
>
> > On Mar 29, 7:21 am, > *wrote:
> >> Good afternoon all,
> >> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>
> >> Any leads?
>
> >> Thanks,
> >> _kevin
>
> > I've never seen a 6DA6.
>
> > Its equivalent is EF89, a remote cut off RF pentode. ( Remote cut off
> > = variable , which means grid1 bias voltage is used to alter gain, as
> > in a compressor-expander or it is used for RF or IF amps with
> > transformer coupling and altered gain by AGC bias control to g1. )
>
> functionally correct, but there's a bit more to it than that. *sharp
> cutoff pentodes (aka 'regular pentodes') have too much nonlinearity at
> the cutoff point, but higher gain. *Remote cutoff pentodes, on the other
> hand, have specially designed grids that 'smooth out' the cutoff curve
> (more linear) but lower the overall gain, making them suitable for AGC
> and 'electronic volume control' circuits, and low distortion amplifiers.
>
> For maximum ideal open loop gain, a pentode will operate very close to
> its cutoff point, but this also causes distortion due to high
> nonlinearity in this range. *So for high open loop gain with lower
> distortion, you would use remote cutoff pentodes. *In cases where low
> distortion is less important (like FM IF stages) you can use sharp
> cutoff and bias them ideally.

Maybe your painting too good a picture for remote cut off pentodes.
You just said " For maximum ideal open loop gain, a pentode will
operate very close to its cutoff point,.."

But the gain is NOT maximum ideal gain near the low Ia level near cut
off. In 6AU6, 6J7, EF86, 6BX6, 6EJ7.their most linear gain region
is well above low Ia levels where gm is low hence gain into a given
load is low.
Usually with RL is low giving a high load current change the sharp cut
off pentode distortion is higher than many triodes driving the same
load
and higher than any remote cut off tube driving the same load, so the
secret to linear sharp cut off pentode use if to bias at higher Ia
than in most circuits you see with higher RL and higher B+ than
****ing bean counters in commercially designed gear where they can't
afford to optimise anything.
Then you can get much higher with the higher gm, about proportional to
Ia, and for where RL is lowish use an unbypassed Rk for local
lineariity if gm and gain is very high, like in 6EJ7.

There is a remote cut off brother of 6EJ7, the 6EH7, which I recall in
every way is more non linear than 6EJ7 when well away from cut off
regions when the gain is high because gm is high. However, when Ia
change is kept very low by keeping RL high, then linearity gets better
for both tubes.
But I have not compared the 2 tubes with say Ia for both = 4mA, RL =
200k, when gain for either should be fairly high
and for 70Vrms out the Ia change is only 0.35mArms.

I don't like using such pentodes in SE config much because they are
less linear than triode, and because such high gain isn't needed,
or I can get better overall performance with 2 cascaded triodes, say
6SN7.
But I recently used a pair of sharp cut off 6BX6/EF80 to replace EF86
in Quad-II and with revised PSU with higher B+ and these gave better
performance than EF86. I also tried 6EJ7 which gave more gain than
6BX6 and too much, so I settled for 6BX6.
The PP LTP config I used cancels even order H so THD was low, so use
of remote cut off tubes in balanced circuitry should also
work very well to improve linearity without NFB.

With very small Va change say under 1Vrms and Ia change of 10uA for a
100k load, the linearity is usually good enough generally for either
remote or sharp cut off.

But with high Ia change linearity becomes poor.

Near cut off for sharp cut off pentodes, sure, linearity may be worse
than for remote cut off pentodes. Remote cut off were invented to get
around this problem so you'd still get some RF and IF gain and
adequate linearity when Ia was 0.5mA.

>
> Similar implications for pentode audio amplifiers and pentode/beam
> tetrode output stages (typically sharp cutoff pentodes), biasing them A
> or AB instead of B (to stay in the linear range more often), and the
> effect it has on harmonic and IM distortion.

The class AB working of remote cut off pentodes may give better
linearity if the Ia reduction per volt of input is complementary with
Ia increase per volt of input. One guy in Electronics World, a Mr
Sandman in the 1990s deliberately tried to drive output transistors
with logarithmic op amps to avoid shrap cut off and switching
artifacts - class S it was called, but AFAIK, all manufacturers and
everyone else ignored him. But class AB triodes in PP output stage
beat pentodes because the triodes have a more gradual cut off.

But around cut off, SE triodes are pretty damn awful.

> I think wikipedia has a good description of sharp vs remote cutoff
> pentodes on their 'pentode' page.
>
> http://en.wikipedia.org/wiki/Pentode
>
> 'Nuff probably. *the horse is now 'undead'

You can always catch the horse after it has un-bolted.

Patrick Turner.

[karchiba]

On Mar 29, 6:49*pm, Patrick Turner > wrote:
> On Mar 30, 5:08*am, Big Bad Bob <BigBadBob-at-mrp3-
>
>
>
>
>
> > wrote:
> > On 03/29/11 01:38, Patrick Turner so witilly quipped:
>
> > > On Mar 29, 7:21 am, > *wrote:
> > >> Good afternoon all,
> > >> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>
> > >> Any leads?
>
> > >> Thanks,
> > >> _kevin
>
> > > I've never seen a 6DA6.
>
> > > Its equivalent is EF89, a remote cut off RF pentode. ( Remote cut off
> > > = variable , which means grid1 bias voltage is used to alter gain, as
> > > in a compressor-expander or it is used for RF or IF amps with
> > > transformer coupling and altered gain by AGC bias control to g1. )
>
> > functionally correct, but there's a bit more to it than that. *sharp
> > cutoff pentodes (aka 'regular pentodes') have too much nonlinearity at
> > the cutoff point, but higher gain. *Remote cutoff pentodes, on the other
> > hand, have specially designed grids that 'smooth out' the cutoff curve
> > (more linear) but lower the overall gain, making them suitable for AGC
> > and 'electronic volume control' circuits, and low distortion amplifiers..
>
> > For maximum ideal open loop gain, a pentode will operate very close to
> > its cutoff point, but this also causes distortion due to high
> > nonlinearity in this range. *So for high open loop gain with lower
> > distortion, you would use remote cutoff pentodes. *In cases where low
> > distortion is less important (like FM IF stages) you can use sharp
> > cutoff and bias them ideally.
>
> Maybe your painting too good a picture for remote cut off pentodes.
> You just said " For maximum ideal open loop gain, a pentode will
> operate very close to its cutoff point,.."
>
> But the gain is NOT maximum ideal gain near the low Ia level near cut
> off. In 6AU6, 6J7, EF86, 6BX6, 6EJ7.their most linear gain region
> is well above low Ia levels where gm is low hence gain into a given
> load is low.
> Usually with RL is low giving a high load current change the sharp cut
> off pentode distortion is higher than many triodes driving the same
> load
> and higher than any remote cut off tube driving the same load, so the
> secret to linear sharp cut off pentode use if to bias at higher Ia
> than in most circuits you see with higher RL and higher B+ *than
> ****ing bean counters in commercially designed gear where they can't
> afford to optimise anything.
> Then you can get much higher with the higher gm, about proportional to
> Ia, and for where RL is lowish use an unbypassed Rk for local
> lineariity if gm and gain is very high, like in 6EJ7.
>
> There is a remote cut off brother of 6EJ7, the 6EH7, which I recall in
> every way is more non linear than 6EJ7 when well away from cut off
> regions when the gain is high because gm is high. However, when Ia
> change is kept very low by keeping RL high, then linearity gets better
> for both tubes.
> But I have not compared the 2 tubes with say Ia for both = 4mA, RL =
> 200k, when gain for either should be fairly high
> and for 70Vrms out the Ia change is only 0.35mArms.
>
> I don't like using such pentodes in SE config much because they are
> less linear than triode, and because such high gain isn't needed,
> or I can get better overall performance with 2 cascaded triodes, say
> 6SN7.
> But I recently used a pair of sharp cut off 6BX6/EF80 to replace EF86
> in Quad-II and with revised PSU with higher B+ and these gave better
> performance than EF86. I also tried 6EJ7 which gave more gain than
> 6BX6 and too much, so I settled for 6BX6.
> The PP LTP config I used cancels even order H so THD was low, so use
> of remote cut off tubes in balanced circuitry should also
> work very well to improve linearity without NFB.
>
> With very small Va change say under 1Vrms and Ia change of 10uA for a
> 100k load, the linearity is usually good enough generally for either
> remote or sharp cut off.
>
> But with high Ia change linearity becomes poor.
>
> Near cut off for sharp cut off pentodes, sure, linearity may be worse
> than for remote cut off pentodes. Remote cut off were invented to get
> around this problem so you'd still get some RF and IF gain and
> adequate linearity when Ia was 0.5mA.
>
>
>
> > Similar implications for pentode audio amplifiers and pentode/beam
> > tetrode output stages (typically sharp cutoff pentodes), biasing them A
> > or AB instead of B (to stay in the linear range more often), and the
> > effect it has on harmonic and IM distortion.
>
> The class AB working of remote cut off pentodes may give better
> linearity if the Ia reduction per volt of input is complementary with
> Ia increase per volt of input. One guy in Electronics World, a Mr
> Sandman in the 1990s deliberately tried to drive output transistors
> with logarithmic op amps to avoid shrap cut off and switching
> artifacts - class S it was called, but AFAIK, all manufacturers and
> everyone else ignored him. But class AB triodes in PP output stage
> beat pentodes because the triodes have a more gradual cut off.
>
> But around cut off, SE triodes are pretty damn awful.
>
> > I think wikipedia has a good description of sharp vs remote cutoff
> > pentodes on their 'pentode' page.
>
> >http://en.wikipedia.org/wiki/Pentode
>
> > 'Nuff probably. *the horse is now 'undead'
>
> You can always catch the horse after it has un-bolted.
>
> Patrick Turner.- Hide quoted text -
>
> - Show quoted text -

Gentlemen,

I sure appreciate your help. I'm in the process of restoring a 1954
Grundig radio for my uncle, and this was the only tube that I didn't
have on hand.

I'll be placing an order with ABC Vacuum Tubes today.

My uncle will be thrilled...

Thanks again,
_kevin

[Big Bad Bob]

On 03/29/11 16:49, Patrick Turner so witilly quipped:
> There is a remote cut off brother of 6EJ7, the 6EH7, which I recall in
> every way is more non linear than 6EJ7 when well away from cut off
> regions when the gain is high because gm is high. However, when Ia
> change is kept very low by keeping RL high, then linearity gets better
> for both tubes.
> But I have not compared the 2 tubes with say Ia for both = 4mA, RL =
> 200k, when gain for either should be fairly high
> and for 70Vrms out the Ia change is only 0.35mArms.

that's sort of where I was going with this - there is a 'sweet spot'
somewhat near cutoff (not exactly at, but low current) where you can
have a high RL and get really high gain from a pentode with relatively
low plate current, maybe in the 1-2 ma range with 200k load and 450V
power supply. As I understand it, voltage gain could be 1000 or more
depending on the tube. Bad part, there's more nonlinearity than with a
triode, and of course, a remote cutoff pentode, both of which have lower
gain. I'm in favor of using large load resistance and high voltages
like this in my own amplifier designs (but with triodes, like 12AX7),
with a 100K Rl, 0.47 capacitor coupled to 1M Rg for the subsequent
stage[s], self-biased to keep Vp somewhere close to 1/2 B+. 2 of these
gives you overall gain of near 10,000, and open loop distortion is very
very low.

[Patrick Turner]

On Mar 31, 10:31*am, Big Bad Bob <BigBadBob-at-mrp3-
> wrote:
> On 03/29/11 16:49, Patrick Turner so witilly quipped:
>
> > There is a remote cut off brother of 6EJ7, the 6EH7, which I recall in
> > every way is more non linear than 6EJ7 when well away from cut off
> > regions when the gain is high because gm is high. However, when Ia
> > change is kept very low by keeping RL high, then linearity gets better
> > for both tubes.
> > But I have not compared the 2 tubes with say Ia for both = 4mA, RL =
> > 200k, when gain for either should be fairly high
> > and for 70Vrms out the Ia change is only 0.35mArms.
>
> that's sort of where I was going with this - there is a 'sweet spot'
> somewhat near cutoff (not exactly at, but low current) where you can
> have a high RL and get really high gain from a pentode with relatively
> low plate current, maybe in the 1-2 ma range with 200k load and 450V
> power supply. *As I understand it, voltage gain could be 1000 or more
> depending on the tube. *

Fo ALL tubes, µ = gm x Ra.

if you look at pentode curves, as you get down to low Ia the Ra curve
for each value of Eg1 becomes more like a flat horizontal line
but in fact there is always a finite value of Ra which seldom goes
over more than 1Meg.
If Ra = 1M, then over 500Vdc the Ia change = 0.5mA, and plotted on
graphs it appears as a flat line.

But at low Ia, say 1mA, gm as fallem rather low, maybe 1mA/V, so µ =
0.001 x 1,000,000 = 1,000.

Gain cannot be over 1,000, and if RL = 200k including any following
Rg, then gain = 1,000 x 200 / ( 200 + 1,000 ) = 166.

But with higher gm pentodes, say 6BX6 at 6mA/V when Ra = 350k, µ =
2,100, and when RL = 200k you get gain = 736.

But to get gm up to 6mA/V, Ia is quite heavy at maybe 5mA, so 200k x
5mA = 1,000Vdc and with Ea = 200Vdc you'd need a 1.2kV B+ supply.
So the sensible answer is to use a CCS anode supply, with B+ at say
+350V. Then you may directly couple to a CF buffer output and the gain
can exceed 736,
and maybe reach 1,500 or more as nothing else loads the anode.
Linearity will be marginal at best, but its quite legitimate feasible
to run a shunt FB network from the CF back to pentode input using a
900k and 100k shunt FB network so that Rin with a biasing R of 100k
ahead of the FB network gives a net 50k Rin.
Gain will be close to 10, and the 5% THD at say 50Vrms without NFB
will become 0.03%, and at 1Vrms output will be 0.00067%, not bad for a
1 tube using say a 6U8 or 6AW8 triode-pentode tube plus a 2SK1924
mosfet as a bootstrapped SS load for the pentode, with the gate dc
feed via a fixed B+ of about +180V.

> Bad part, there's more nonlinearity than with a
> triode, and of course, a remote cutoff pentode, both of which have lower
> gain. *I'm in favor of using large load resistance and high voltages
> like this in my own amplifier designs (but with triodes, like 12AX7),
> with a 100K Rl, 0.47 capacitor coupled to 1M Rg for the subsequent
> stage[s], self-biased to keep Vp somewhere close to 1/2 B+. *2 of these
> gives you overall gain of near 10,000, and open loop distortion is very
> very low.

Indeed the cascaded pair of 12AX7 triodes does give high gain. Such
amps were used as phono stages in hundreds of brandname amps in the
50s and 60s.
But you'll never get open loop gain of 10,000. Usually the 12AX7
loading is 220k RLdc plus following 1M Rg giving total ac load = 180.
A = 100 x 180 / ( 180 + 65 ) = 73, and two such stages give 5,300
approx, but if a phono amp has a NFB network load as well as other RL
then OLG at 1 kHz may be only 4,000. But that's a lot and 12AX7 even
with RL = 100k can make 35Vrms at 0.35mArms and open loop THD might be
2%,
so at 1V the THD = 0.057%, and in a phono amp stage where gain is
reduced to 100 at 1kHz, THD = 0.0015% - IN THEORY.
Gain at 30Hz is reduced by about only 4, not so hot.

Better results are obtained with a 12AU7 buffer CF after second 12AX7
gain stage so the NFB network is powered by stage with 5mA, and easily
able to drive the phono NFB network which has a declining load value
as F rises. Good phono amplification really needs 3 triodes at least,
but the noise is still determined by the EINR of the first triode, so
a MM cart OR MC with step up.tranny is required.

But if you do everything right with passive circuitry the THD with MM
cart need never exceed 0.03% and there is NO NEED for any loop NFB
and ppl say the sound is better.


Many MM phono preamps of the past used just one EF86 as the phono amp
tube. Quad-22 and Leak did this. They ain't regarded as very good.
Getting it to be better with a cascode j-fet plus triode input stage,
RIAA passive filter plus µ-follower gain stage IS BETTER.

Patrick Turner.

[John L Stewart]

But if you do everything right with passive circuitry the THD with MM
cart need never exceed 0.03% and there is NO NEED for any loop NFB
and ppl say the sound is better.


Many MM phono preamps of the past used just one EF86 as the phono amp
tube. Quad-22 and Leak did this. They ain't regarded as very good.
Getting it to be better with a cascode j-fet plus triode input stage,
RIAA passive filter plus µ-follower gain stage IS BETTER.

Patrick Turner.[/QUOTE]

The phono arm tracking error will completley mask the 0.03% attained above. So no need to worry.

Alfred E Newman, Mad Magazine

[Vance_Iam]

Good afternoon all,
I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.

Any leads?

Thanks,
_kevin

Dear Kevin,,,I am also looking for 6da6 tubes. I am looking for 2nd hand. If you get a proper suggestion...kindly inform me!!! I will be very thank ful to you.


Oldies Music (http://977music.com/)

[Big Bad Bob]

On 03/31/11 00:22, Patrick Turner so wittily quipped:
> On Mar 31, 10:31 am, Big Bad Bob<BigBadBob-at-mrp3-
> > wrote:
>> On 03/29/11 16:49, Patrick Turner so witilly quipped:
>>
>>> There is a remote cut off brother of 6EJ7, the 6EH7, which I recall in
>>> every way is more non linear than 6EJ7 when well away from cut off
>>> regions when the gain is high because gm is high. However, when Ia
>>> change is kept very low by keeping RL high, then linearity gets better
>>> for both tubes.
>>> But I have not compared the 2 tubes with say Ia for both = 4mA, RL =
>>> 200k, when gain for either should be fairly high
>>> and for 70Vrms out the Ia change is only 0.35mArms.
>>
>> that's sort of where I was going with this - there is a 'sweet spot'
>> somewhat near cutoff (not exactly at, but low current) where you can
>> have a high RL and get really high gain from a pentode with relatively
>> low plate current, maybe in the 1-2 ma range with 200k load and 450V
>> power supply. As I understand it, voltage gain could be 1000 or more
>> depending on the tube.
>
> Fo ALL tubes, µ = gm x Ra.

etc. - good discussion of theory of operation

> Linearity will be marginal at best

yeah, kinda where I was headed with the whole open loop pentode
amplifier stage thing.

> Indeed the cascaded pair of 12AX7 triodes does give high gain. Such
> amps were used as phono stages in hundreds of brandname amps in the
> 50s and 60s.

yep - you also find 12AX7 and 12AT7 in guitar amplifiers for the same
reason. Occasionally you'll see 12AU7 but they're usually drivers for
the final stage, or a VLF oscillator for tremolo, or the reverb driver.

> But if you do everything right with passive circuitry the THD with MM
> cart need never exceed 0.03% and there is NO NEED for any loop NFB
> and ppl say the sound is better.

maybe yes, maybe no. I typically use low stage gain in RC coupled
pre-amp circuits and rely on the active component's relatively high gain
to improve the linearity and lower distortion. So there's no "global
NFB" for the pre-amp. However I would use global NFB in the final
(power) stage. I find that global NFB has its best benefit in the power
stage due to 'the obvious'. But it's not necessarily true for a pre-amp
stage. Choosing RP/RK ratios that are a fraction of the mu factor, and
biasing for linearity, gives you adequate stage gain and very low THD.
Still global NFB might improve this even more, though I haven't seen the
need. And anyway, THAT kind of pre-amp usually has a volume control
between its output and the final power stage, so global NFB wouldn't be
possible anyway.

I guess it all depends on what you're really trying to accomplish.

[Big Bad Bob]

On 03/31/11 00:22, Patrick Turner so wittily quipped:
> But if you do everything right with passive circuitry the THD with MM
> cart need never exceed 0.03% and there is NO NEED for any loop NFB
> and ppl say the sound is better.

forgot to mention, phono amps are a special case where the filtering CAN
be done in the feedback loop in lieu of using a 'passive network' like
you mentioned. I've seen both methods used. I would prefer using the
feedback loop to create the properly shaped LPF, but limit stage gain
internally using Rp/Rk ratios and high mu triodes. The trade-off, like
you mentioned, is the potential for higher THD at the lower frequencies.
If open loop THD is already small, it won't matter which method you use.

[Big Bad Bob]

On 04/01/11 04:17, Vance_Iam so wittily quipped:
> karchiba;928575 Wrote:
>> Good afternoon all,
>> I'm looking to purchase 4, good condition or NOS, 6DA6 tubes.
>>
>> Any leads?
>>
>> Thanks,
>> _kevin
>
> Dear Kevin,,,I am also looking for 6da6 tubes. I am looking for 2nd
> hand. If you get a proper suggestion...kindly inform me!!! I will be
> very thank ful to you.

http://thetubestore.com/nos-6da6.html

google found that at the top. dunno how good a vendor they are