Re: The KISS AMP: a progress report

That's excellent timing, Peter. I unkillfiled everyone to see who wants
to work and who is a permanent flame merchant -- and there you are with
a useful question. Give that boy's young lady the pink fluffy rabbit.

pfjw@xxxxxxx wrote:
Why do you still have some 100 watts of dropping resistors on the
secondary side of the transformer?

http://members.lycos.co.uk/fiultra/t68mzwe417acircuit.jpg

Peter Wieck
Wyncote, PA

"still" ? Always been there.

The resistors I used are overrated (see below) and my associate who
reverse-engineered the little amp for the schematic wrote in the rating
they are rather than the ratings they should have been probably because
I forgot to tell him to recalculate everything. Now that you've pointed
it out to me, I've put the right ratings in the posted version.
http://members.lycos.co.uk/fiultra/t68mzwe417acircuit.jpg
Thanks for the heads-up. While we're here, we may as well discuss
resistors in secondaries, and dropping resistors in general.

I made the T68bis potato amp by stripping out a pair of 300B on a proto
and using my junkbox resistors to drop the voltages. When the little
amp goes out of the door or is stripped down, they go back in the
junkbox. My only concern at the time was that they were the right value
and overrated for the job.

NB: A 50W resistor doesn't dissipate 50W just because it is rated 50W.
The heat it dissipates is related to current flowing through it and the
resistance value. Colonel Ohm saves the day again:
P = I*I*R -- says nothing about the rating of the resistor which by
tradition is 2 or 3 times the dissipation arrived at for a safety
margin.

There are many reasons for using a resistor of the right value and a
higher rating.

1. You have it in junkbox already and don't need to wait for a delivery
from your component pusher.

2. The manufacturer doesn't make all the values in all the ratings, or
your supplier doesn't stock the full range, so you buy the value you
need in the next higher rating.

3. I like big chunky aliminium-cased resistors because they're
cool-running with all that area and fins besides.

4. In the case of at least one of the resistors you think over-rated,
the alternative was a nasty cement-cased pluggable resistor of the
right rating. I like neither the looks nor the sound of those.

5. The standard resistor I keep in all values and both the available
ratings is Kiwame. It gets a bit expensive sticking them in where they
won't make a difference to the sound.

6. I'm designing for DIYers. The great thing is not to give Joe Doe a
chance to say near enough is good enough and put an underrated part in.
Better safe than sorry.

7. Those particular resistors derate to 50 per cent (IIRC, look it up
if you don't know -- *that* is why I overrate them!) or less if they
aren't on a heatsink, if they aren't in free air, etc. What you see
isn't a 100W of power capability but only 50W inside a case, maybe
less. It is for tricky little subtleties like this that Patrick wants
people to hit the books rather than learn only by soldering.

8. It isn't so widely known that resistors have voltage ratings. It is
not uncommon, though less so these days of CE marks over here, to see
amps built with resistors that at least theoretically could burst into
flames any moment. Those ali-cased resistors I like have very
attractive voltage ratings too, so they live naturally in the junkbox
of someone who moves casually from a little 175V 417A potato amp to a
kilovolt-plus 845 or SV572.xx amp. The ali-cased resistors save the
bother of having to check the voltage spec of some orphan resistor from
the junkbox; I've long since cleared all the low-voltage stuff from my
junkbox.

****

Returning now to the particular resistors you mention, the two 330R in
the mains secondary are ballasts. In a tube rectifier circuit you're
supposed to have those, usually about 75R minimum. Check the spec
sheets for various tube rectifiers, on some of which is an explanation
and schematic of the equivalent circuit for various impedances in and
around a tube rectifier (for each type of filter, cap input and choke
input) and their effects.

I have long since found that a pair of stiff ballasts smooths the power
an agreeable fraction when you have a hefty bleed already.

*******

Remember this: the idea isn't a perfect schematic but to build
something, to listen, to develop it. I used to waste lots of time
agonising over the perfect paper design and collecting all the perfect
parts. Now as often as not I build the prototype on the fly with a few
calculations on the logarithmic bezels of my watch (why is a rotary
slide rule still called a "rule" -- tell me that and I'll think you
clever) to check my mental arithmetic, then develop it right, and only
then reverse engineer the amp to draw the schematic of the circuit.
There are a few traps in that, as we have seen, when sharp eyes catch
out your overrated junkbox parts but the amp gets done a lot sooner.

The secret of designing good-sounding amps and still have time left to
listen to music is to fail fast. Anyone who hasn't failed often, never
mind "never" as too many claim, probably can't distinguish mud from
music.

Andre Jute
Visit Jute on Amps at http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review

.

Quantcast