What's your favourite voltage regs?
 

Don Pearce wrote:



Well, just out of interest, I will try and put an actual voltage reg
together tonight, while the rest is on the bread board. Amd maybe check
the freq response using all three methods.

Because with respect, what we seem to have done here, is start with an
assertion that A is better than B, thought of a reason why this could be
so, then created a C that should mimic that process. Compaired A and C,
found no difference, and declared that A is not different to B. It could
just as simply be the case that the reason we have thought of isn't
valid, so C isn't the correct test to do.


Well, not really. Putting the cap across the heater terminals does
indeed force the drive into voltage mode, so test C is fair from that
point of view. If condition B creates a different set of results, then
I think we are entitled to ask if there is a problem in the
implementation of B - has it been done reasonably.

I'm looking forward to the result, though. Adding a frequency response
test is a good idea.

d


I was thinking of the best way of doing the FR test, I could do a sweep.
But maybe it would be better to just use a 400hZ square wave, I think
that might be interesting.

Any thoughts?

--
Nick

What's your favourite voltage regs?
 

On Tue, 22 Jan 2008 10:02:08 +0000, Nick Gorham
wrote:

Don Pearce wrote:



Well, just out of interest, I will try and put an actual voltage reg
together tonight, while the rest is on the bread board. Amd maybe check
the freq response using all three methods.

Because with respect, what we seem to have done here, is start with an
assertion that A is better than B, thought of a reason why this could be
so, then created a C that should mimic that process. Compaired A and C,
found no difference, and declared that A is not different to B. It could
just as simply be the case that the reason we have thought of isn't
valid, so C isn't the correct test to do.


Well, not really. Putting the cap across the heater terminals does
indeed force the drive into voltage mode, so test C is fair from that
point of view. If condition B creates a different set of results, then
I think we are entitled to ask if there is a problem in the
implementation of B - has it been done reasonably.

I'm looking forward to the result, though. Adding a frequency response
test is a good idea.

d


I was thinking of the best way of doing the FR test, I could do a sweep.
But maybe it would be better to just use a 400hZ square wave, I think
that might be interesting.

Any thoughts?

Not so easy to interpret since a great deal depends on the quality of
the wave, and of course a good few low octaves are missing. A nice
slow sweep 20 to 20k would be better. I have to say, though, that I
can't think of a mechanism by which the heater bias method might
change the FR.

What exactly is it you are measuring? Is it a single DHT stage or a
whole power amplifier implemented with a DHT in it?

d

--
Pearce Consulting
http://www.pearce.uk.com

What's your favourite voltage regs?
 

Don Pearce wrote:


I was thinking of the best way of doing the FR test, I could do a sweep.
But maybe it would be better to just use a 400hZ square wave, I think
that might be interesting.

Any thoughts?


Not so easy to interpret since a great deal depends on the quality of
the wave, and of course a good few low octaves are missing. A nice
slow sweep 20 to 20k would be better. I have to say, though, that I
can't think of a mechanism by which the heater bias method might
change the FR.

What exactly is it you are measuring? Is it a single DHT stage or a
whole power amplifier implemented with a DHT in it?

d


Fair enough, I can do both, the square from the Farnell sig gen I use
seems to ok enough looking on a scope, I would have thought it was less
important as we can still compare the two cases.

Its just some bits I put together for the test. 6sn7, 1k cathode, 40k
anode, cathode bipassed by 220uf. 1uf coupling cap, 220k grid resistor
on 2a3. 750R on 2a3 cathode, bipassed by another 220uf cap, 3K:8R 50H
transformer into 8R resistive load. Both stages powered by 350v from
bench supply. 6sn7 AC heated, 2a3 heated by a current source made from
three pin voltage reg, a sense resistor and a pot. Switched to voltage
mode by shunting a 33000uf cap across it. Cathode resistor on 2a3 taken
from -ve leg of supply.

In my subjective comparisons of the different heating methods, the
difference has been in the higher frequency ranges. I won't try to
describe what I hear.

TBH, my view on this, is I am interested if we can measure something
that I am sure that I can hear, not to determine if I can possible hear
something, your view may differ, but thats not important to what we are
looking at anyway.

--
Nick

What's your favourite voltage regs?
 

On Tue, 22 Jan 2008 10:31:06 +0000, Nick Gorham
wrote:

Don Pearce wrote:


I was thinking of the best way of doing the FR test, I could do a sweep.
But maybe it would be better to just use a 400hZ square wave, I think
that might be interesting.

Any thoughts?


Not so easy to interpret since a great deal depends on the quality of
the wave, and of course a good few low octaves are missing. A nice
slow sweep 20 to 20k would be better. I have to say, though, that I
can't think of a mechanism by which the heater bias method might
change the FR.

What exactly is it you are measuring? Is it a single DHT stage or a
whole power amplifier implemented with a DHT in it?

d


Fair enough, I can do both, the square from the Farnell sig gen I use
seems to ok enough looking on a scope, I would have thought it was less
important as we can still compare the two cases.

For comparing the two you are right in principle, but in practice it
is visually easier to compare a couple of nominally flat lines than a
row of narrow peaks.

Its just some bits I put together for the test. 6sn7, 1k cathode, 40k
anode, cathode bipassed by 220uf. 1uf coupling cap, 220k grid resistor
on 2a3. 750R on 2a3 cathode, bipassed by another 220uf cap, 3K:8R 50H
transformer into 8R resistive load. Both stages powered by 350v from
bench supply. 6sn7 AC heated, 2a3 heated by a current source made from
three pin voltage reg, a sense resistor and a pot. Switched to voltage
mode by shunting a 33000uf cap across it. Cathode resistor on 2a3 taken
from -ve leg of supply.

OK, got that. It all sounds perfectly reasonable for a test setup. My
only thought would be that by including things like transformers which
have distortion properties of their own you may be masking some subtle
effects we are trying to see.

In my subjective comparisons of the different heating methods, the
difference has been in the higher frequency ranges. I won't try to
describe what I hear.

TBH, my view on this, is I am interested if we can measure something
that I am sure that I can hear, not to determine if I can possible hear
something, your view may differ, but thats not important to what we are
looking at anyway.

Well, of course we differ there, but as you say it isn't important at
the moment. If at the end of all this we find that there isn't an
identifiable difference between the two situations then the question
of expectation effects and so on will rear its ugly head. But we can
deal with that as and when.

d

--
Pearce Consulting
http://www.pearce.uk.com

What's your favourite voltage regs?
 

Don Pearce wrote:


Fair enough, I can do both, the square from the Farnell sig gen I use
seems to ok enough looking on a scope, I would have thought it was less
important as we can still compare the two cases.


For comparing the two you are right in principle, but in practice it
is visually easier to compare a couple of nominally flat lines than a
row of narrow peaks.


Ok, but as we know, it won't be a perfect square after its been through
the amp anyway.


Its just some bits I put together for the test. 6sn7, 1k cathode, 40k
anode, cathode bipassed by 220uf. 1uf coupling cap, 220k grid resistor
on 2a3. 750R on 2a3 cathode, bipassed by another 220uf cap, 3K:8R 50H
transformer into 8R resistive load. Both stages powered by 350v from
bench supply. 6sn7 AC heated, 2a3 heated by a current source made from
three pin voltage reg, a sense resistor and a pot. Switched to voltage
mode by shunting a 33000uf cap across it. Cathode resistor on 2a3 taken

from -ve leg of supply.

OK, got that. It all sounds perfectly reasonable for a test setup. My
only thought would be that by including things like transformers which
have distortion properties of their own you may be masking some subtle
effects we are trying to see.


Yes, I considered that, but I also thought that by starting with what is
effectivly a complete single ended amplifier we potentially avoid
missing the presense on a mechanism that may cause a difference but we
don't as yet know how. For example, I could replace the transformer with
a resistive load, and increase the B+ to match, but its possible that
there is a interaction with an inductive load and the fill that we don't
know about.

--
Nick

What's your favourite voltage regs?
 

http://www.diyaudio.com/forums/showt...=&pagenumber=1

Can I ask you guys - David, Jim, Nick, Don etc. to have a look at this
thread on www.diyaudio.com entitled "New DHT heater".

The circuit designed by Rod Coleman is the best sounding filament
supply I've heard - this introduces a "choke" as well as a current
source. Can you take a look - this circuit works in practice and so
must be doing something right. I apologise for not having any
measuring equipment - I'd love to help you work out what's happening
in measured terms. Andy

What's your favourite voltage regs?
 

"Andy Evans" wrote in message
...
http://www.diyaudio.com/forums/showt...=&pagenumber=1

Can I ask you guys - David, Jim, Nick, Don etc. to have a look at this
thread on www.diyaudio.com entitled "New DHT heater".

I couldn't find it, where exactly is it?

The circuit designed by Rod Coleman is the best sounding filament
supply I've heard

How did you establish that?

David.

What's your favourite voltage regs?
 

On Tue, 22 Jan 2008 13:43:04 -0000, "David Looser"
wrote:

"Andy Evans" wrote in message
...
http://www.diyaudio.com/forums/showt...=&pagenumber=1

Can I ask you guys - David, Jim, Nick, Don etc. to have a look at this
thread on www.diyaudio.com entitled "New DHT heater".

I couldn't find it, where exactly is it?


I found the thread using the search button, but the links in the first
article are just 404s, and I can find no reference to Rod Coleman's
work.

A working link would be good.

d

--
Pearce Consulting
http://www.pearce.uk.com

What's your favourite voltage regs?
 

Don Pearce wrote:
On Tue, 22 Jan 2008 13:43:04 -0000, "David Looser"
wrote:


"Andy Evans" wrote in message
...

http://www.diyaudio.com/forums/showt...=&pagenumber=1

Can I ask you guys - David, Jim, Nick, Don etc. to have a look at this
thread on www.diyaudio.com entitled "New DHT heater".

I couldn't find it, where exactly is it?


I found the thread using the search button, but the links in the first
article are just 404s, and I can find no reference to Rod Coleman's
work.

A working link would be good.

d


Worked for me, this is the diag in the thread.

http://img65.exs.cx/img65/702/BUFFEREDDHTFILAMENT.png

--
Nick

What's your favourite voltage regs?
 

"Nick Gorham" wrote in message
...

Worked for me, this is the diag in the thread.

http://img65.exs.cx/img65/702/BUFFEREDDHTFILAMENT.png

Thanks.

My first reaction is that it is overkill for the application, especially the
two transformers and the large number of paralleled reservoir capacitors,
but then I wouldn't be using DHTs in the first place :-)

David.