Re: Building a valve amp

"Xtrchessreal" <markgharrison@xxxxxxxxxxx> wrote in message
news:1134624004.720318.176380@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> Ant:
>
> Do you remember Andy, the guy that started this post, the one looking
> for some feedback about building an amp? He is the person I wrote my
> advice to. The guy sounded rather green in the electronics world so
> maybe I was talking a bit down to him so he could more easily
> understand what he was up against. My assumption is that he is a young
> person with a lot of enthusiasm for this stuff but could easily step in
> some real *** quickly if he just jumps into it. I'll stand by what I
> said in that regard.

Yes I do - and with regard to Andy,the jist of your advice is not bad.

Your explanation *why* he shouldn't dive straight into a design,
however, was IMO deeply flawed.


> I wouldn't know how to build a tube amplifier from scratch because I
> have never learned it. My education in amplifier design was with
> transistor MOSFET designs.

Understood. So was mine. However, I have a problem with using
that as a line of reasoning.

More importantly than anything else, a college education should
give you the ability to learn acquire new knowledge and
techniques - the capacity to read technical articles, experiment,
understand - in short - LEARN. You don't stop learning the day
you walk out of college with your engineering degree (or
whatever) - that's when the *real* learning starts.

I this weren't so, then a degree would need to have an "expiry
date".


> As you may know there are many
> characteristics of each component in that area. Many of them are basic
> at first until you start looking at your Higher Fequencies and the
> various EMI associated with each component in the board. Since this is
> my background; I was assuming tube amplifiers had many of the same
> characteristics if not even more difficult ones to deal with. The tube
> to me is a large unknown so I assumed it to be harder. In the same way
> I would associate skiing or skydiving. Just sounds scary and dark.

That's another bit of faulty reasoning. "I don't know it, so
I assume it's harder, so I advise against it."

It's all stuff that can be self-taught. (Even HF techniques.)

If you haven't already, I suggest you look at some tube amp circuits
some time. You'll recognise quite a lot of the elements there, and
you might teach yourself something too.


> Looking at this now I realize that a tube amp in general is built with
> a great deal more power at the output in fact at every stage when
> compared to a high frequency amplifier.

Higher *voltages*, yes.
Higher *power*, no (at least, not inter-stage).
The impedances in tube circuitry are higher than solid state circuits
using op-amps, transistors, etc. (Not MOSFETs, tho; they have
roughly similar characteristics to a pentode. In fact, a MOSFET
can be used to replace a tube in many circuits with little/no
changes. You might not like the sound, tho...)

> The Harmonics are completely
> different. This is mostly due to the fact that it is for an Audio
> signal which is much lower in Freq range. At audio frequencies RLC
> combinations are not as counter productive as they are in higher
> frequencies.

Not sure why you think different frequency multiples are relevant
to this. That they are different is *completely* due to the fact that
the audio signal is in a lower frequency range. Anyway...

RLC (actually RC; L is seldom used in the signal path of audio
amps) are just as important. It's how you tailor the frequency
response of the various elements. And parasitic coupling by
stray C or L are indeed problems, especially as the impedances
are higher in a tube amp.

Google for "Lead Dress".


> This leads me to say that an audio amplifier should be
> relatively easy to build and possibly as easy as a radio kit from Radio
> shack - given no new design and only following the circuit schematic
> and the instructions.

Agreed. (With the caveat that higher-gain amplifiers will pose
problems because of parasitics; and even a relatively low gain
clean amp will suffer if you get the lead dress wrong.)


> Designing one from scratch is a different animal. I still would argu
> that a first timer should not be trying to design the circuit, making
> the chassis look cool and placement of parts is fine to tinker with. A
> kid just putting a bunch of parts together without understanding that
> each peice has a place in the circuit is looking for a disaster.

Agreed.


> Personally I would not design one until I understood all of the tube
> characteristics and then even some of the more subtle differences
> between one mfr to another.

As you state - that's your choice.


> But I am speaking of designing from
> scratch, not following a schematic. I would use Pspice or something to
> get the basic idea before I would start putting anything together on a
> board.

Pspice is a good tool for getting a circuit to a (theoretically) functioning
state, without spending soldering time.

But it won't tell you anything about how that circuit will *sound* when you
do build it, hook it up to a speaker cabinet and a guitar.


Ant. (Tamworth, UK)

--
"Make me one with everything." The Dalai Lama at a hot dog stand.


.

Loading