[Leigh Orf]

Hiya folks...

I posted a few weeks ago with an intermittent noise problem in the
right channel of my ST-70. This had been plaguing me for years and had
kept me from listening as much as I would have liked. After some poking
around I narrowed it down to the circuitry surrounding the ECC83 phase
splitter tube (Triode Electronics driver board mod). To the gentleman
who suggested the shotgun approach: Thanks, that's what did the trick!
For the right channel, I replaced the plate resistors (2), feedback loop
resistors (2), coupling capacitors (2) and that little mica job and the
noise is completely gone.

I considered, for my own interest, to replace each part one at a time to
trace which one was bad (so I could smash it with a hammer but that
proved to be a bit silly and unrealistic.

Thanks to Triode for sending out the parts gratis after a
misunderstanding led to a delay in shipping. I've always been happy with
their service and parts quality.

So now I've been playing just a bit. I have a Squeezebox DAP which has
an external volume control on the line out, and for grins I removed my
old NAD 1000 preamp from the audio chain and was pleasantly surprised
with the improvement of the sound through my monitors (ProAc Super
Tablettes); but I hooked things up back the way they were because in
order to feed my amplified subwoofer I had to run straight off the ST-70
(rather than the second preamp line out) and I really didn't like what I
heard with the bass. This is even with the power supply upgrade on the
ST-70. I wonder if the low bass problem is common with tube audio, even
the expensive stuff?

Thanks again to everyone who offered suggestions.

Leigh

--
Leigh Orf http://orf.cx
SDF Public Access UNIX System - http://sdf.lonestar.org

[Ian Iveson]

Leigh Orf wrote

I posted a few weeks ago with an intermittent noise problem in the
right channel of my ST-70. This had been plaguing me for years and
had
kept me from listening as much as I would have liked. After some
poking
around I narrowed it down to the circuitry surrounding the ECC83
phase
splitter tube (Triode Electronics driver board mod). To the
gentleman
who suggested the shotgun approach: Thanks, that's what did the
trick!
For the right channel, I replaced the plate resistors (2), feedback
loop
resistors (2), coupling capacitors (2) and that little mica job and
the
noise is completely gone.


Thanks for reporting back, and glad it turned out well.

I considered, for my own interest, to replace each part one at a
time to
trace which one was bad (so I could smash it with a hammer but
that
proved to be a bit silly and unrealistic.

Now, since you don't know which is bad, you can smash them all with
the hammer. I like to string them all together and hit them with 10kV
or so.

Thanks to Triode for sending out the parts gratis after a
misunderstanding led to a delay in shipping. I've always been happy
with
their service and parts quality.

Always and now may not be the same.

So now I've been playing just a bit. I have a Squeezebox DAP which
has
an external volume control on the line out, and for grins I removed
my
old NAD 1000 preamp from the audio chain and was pleasantly
surprised
with the improvement of the sound through my monitors (ProAc Super
Tablettes); but I hooked things up back the way they were because in
order to feed my amplified subwoofer I had to run straight off the
ST-70
(rather than the second preamp line out) and I really didn't like
what I
heard with the bass. This is even with the power supply upgrade on
the
ST-70. I wonder if the low bass problem is common with tube audio,
even
the expensive stuff?

Basically, they go as low as they go, and that depends on the quality,
size and winding ratio of the output transformers. If your main
speakers go as low as the amps, that's as good as you will get. What
you *don't* want to do is boost and flatten that part of the bass that
the amp has rolled off.

Consider that the output transformer places a big steel-cored
inductance across the load. At low frequencies, the impedance of this
inductance is low, and becomes significant in comparison with the load
itself. Consequently, lower frequencies would become progressively
attenuated and shifted forward in phase. At very low frequencies, so
much current would pass through the inductor that it could saturate,
as the steel reaches the limit of magnetisation and the impedance
drops off suddenly.

To avoid all this, and to prevent low frequency oscillation due to the
combination of phase shift and feedback, valve amps are generally
rolled off somewhere else in the circuit. This is very often rather
higher than modern tastes expect.

Now, you might conceivably, using a powered sub with equalisation,
retrieve more bass by boosting and the dying embers of the amp's
bandwidth. You may even be able to correct for the phase error.

But because the steel-cored inductor is non-linear (it varies with
amplitude as it follows the BH curve of the steel), these low
frequencies will be increasingly distorted. They are also beyond the
correcting influence of the amp's feedback. Hence the signal isn't
worth retrieving.

There is a limit to how big the transformers can be because they also
place a small inductance in series with the load, and a small
capacitance across it. These elements, which roll off the high
frequencies, tend to increase with transformer size.

Finding enough clean bandwidth between these limits requires clever
winding with the best materials. Even then, traditional high quality
gives you adequate bandwidth for traditional sources. These days many
folk want more.

The only real way to get low enough now is to reduce the winding
ratio. That means using output valves with very low anode resistance,
or many valves in parallel, or both.

But it is said by some that such valves and circuits lose the
essential delicacy and poise of valve audio. You may as well use solid
state.

How about a line-level x-over?

cheers, Ian

[Jon Yaeger]

Basically, they go as low as they go, and that depends on the quality,
size and winding ratio of the output transformers. If your main
speakers go as low as the amps, that's as good as you will get. What
you *don't* want to do is boost and flatten that part of the bass that
the amp has rolled off.

Consider that the output transformer places a big steel-cored
inductance across the load. At low frequencies, the impedance of this
inductance is low, and becomes significant in comparison with the load
itself. Consequently, lower frequencies would become progressively
attenuated and shifted forward in phase. At very low frequencies, so
much current would pass through the inductor that it could saturate,
as the steel reaches the limit of magnetisation and the impedance
drops off suddenly.

To avoid all this, and to prevent low frequency oscillation due to the
combination of phase shift and feedback, valve amps are generally
rolled off somewhere else in the circuit. This is very often rather
higher than modern tastes expect.

Now, you might conceivably, using a powered sub with equalisation,
retrieve more bass by boosting and the dying embers of the amp's
bandwidth. You may even be able to correct for the phase error.

But because the steel-cored inductor is non-linear (it varies with
amplitude as it follows the BH curve of the steel), these low
frequencies will be increasingly distorted. They are also beyond the
correcting influence of the amp's feedback. Hence the signal isn't
worth retrieving.

There is a limit to how big the transformers can be because they also
place a small inductance in series with the load, and a small
capacitance across it. These elements, which roll off the high
frequencies, tend to increase with transformer size.

Finding enough clean bandwidth between these limits requires clever
winding with the best materials. Even then, traditional high quality
gives you adequate bandwidth for traditional sources. These days many
folk want more.

The only real way to get low enough now is to reduce the winding
ratio. That means using output valves with very low anode resistance,
or many valves in parallel, or both.

But it is said by some that such valves and circuits lose the
essential delicacy and poise of valve audio. You may as well use solid
state.

How about a line-level x-over?

cheers, Ian


One of Van Alstine's recommended mods for the Dyna ST-70 is a network that
rolls off the low frequencies even more than the stock Dyna amp was intended
to do. His theory was that there wasn't much audio information at those low
frequencies (e.g turntable rumble & such) and what is there only robs the
amp of its power.

I built his input circuit and added it to my Dyna, which I had already
modified with a Welbourne Labs circuit. However, I didn't like what I heard
-- it seemed to rob the amp of some of its fidelity (at the expense of
conserving power? and adding complexity) and so I removed it.

IMHO tube amplifiers rarely do well with the lowest frequencies, which can
sound flabby yet anemic. SS amps can provide really high damping factors
for controlling the woofer cones.

One practical solution is to use a line crossover, like Ian suggests, and
biamp the signal, feeding both the bass speaker (and subwoofer, if present)
with a powerful SS amp like the Ampzilla, Hafler, etc. The "sweetness" of
the tube sound is reserved for where it shines best, in the middle and upper
registers.

Jon

[Ned Carlson]

Jon Yaeger wrote:

One of Van Alstine's recommended mods for the Dyna ST-70 is a network that
rolls off the low frequencies even more than the stock Dyna amp was intended
to do. His theory was that there wasn't much audio information at those low
frequencies (e.g turntable rumble & such) and what is there only robs the
amp of its power.

The original Van Alstine mod deleted the HF compensation capacitors and
increased the value of the coupling capacitors (up to 1 uf IIRC, and
thus broadening the open-loop bandwidth considerably). adding this
extra mod to an original ST70 circuit or anything that's a clone of it,
isn't necessary and merely duplicates the frequency compensation
that was in there to begin with. It's only useful if you have a Van
Alstine board.

While there were certainly some compromises made to get ST70's
to sell at a $99.95 price point (or 79.95 if you hit a sale at
EJ Korvette's audio department), the frequency compensation
elements were and are specific to getting decently flat
response and low distortion from the audio transformer.
All audio transformers have limitations that can be addressed
in some form of frequency compensation, this is not unique
to Dynaco transformers.

--
Ned Carlson
SW side of Chicago, USA
www.tubezone.net

[Bob H.]

On Jan 27, 10:47 pm, Leigh Orf wrote:
Hiya folks...

I posted a few weeks ago with an intermittent noise problem in the
right channel of my ST-70. This had been plaguing me for years and had
kept me from listening as much as I would have liked. After some poking
around I narrowed it down to the circuitry surrounding the ECC83 phase
splitter tube (Triode Electronics driver board mod). To the gentleman
who suggested the shotgun approach: Thanks, that's what did the trick!
For the right channel, I replaced the plate resistors (2), feedback loop
resistors (2), coupling capacitors (2) and that little mica job and the
noise is completely gone.

I considered, for my own interest, to replace each part one at a time to
trace which one was bad (so I could smash it with a hammer but that
proved to be a bit silly and unrealistic.

Thanks to Triode for sending out the parts gratis after a
misunderstanding led to a delay in shipping. I've always been happy with
their service and parts quality.

So now I've been playing just a bit. I have a Squeezebox DAP which has
an external volume control on the line out, and for grins I removed my
old NAD 1000 preamp from the audio chain and was pleasantly surprised
with the improvement of the sound through my monitors (ProAc Super
Tablettes); but I hooked things up back the way they were because in
order to feed my amplified subwoofer I had to run straight off the ST-70
(rather than the second preamp line out) and I really didn't like what I
heard with the bass. This is even with the power supply upgrade on the
ST-70. I wonder if the low bass problem is common with tube audio, even
the expensive stuff?

Thanks again to everyone who offered suggestions.

Leigh

--
Leigh Orfhttp://orf.cx
SDF Public Access UNIX System -http://sdf.lonestar.org

Bass performance also depends on speaker efficiency and the over all
impedance picture. In my opinion, both my Cornwalls and my Hammer
12's (45hz at best) have more satisfying bass than my Boston Acoustic
VR's which are rated flat to 33 hz or so, and high at 20hz. I wish my
BA's were better, as they fit in a room and have high WAF, but sadly
this is not the truth. 91dB/m is often stated as minimum sensitivity,
but I don't start considering until at least 93 dB/m or better. My
opinion only.

Bob H.