dre7 wrote:
Hey RATs:
Can anyone recall a "classic" commercially made amp that used a choke input?
Not a guitar amp, but hifi. Been sifting through schematics but haven't
happened upon one yet.
Andrew
I have never ever seen a choke input used in any **domestic** hi-fi amp,
regardless of its class.
It would suit a class B SS amp, and I have used choke inputs in modified
SS amps, eg,
http://www.turneraudio.com.au/solidstateamps2bjts.html
The choke input PSU used in this massive amp had no better regulation
than the original
configuration. But it did bring some benefits.
Recently while perfoming singing lesson operations on a tubed CR Audio
Developments
Woodham, I placed a choke input into the PSU to drop the stupid 650v B+
rail
from a Cap input PSU down to a sensible and reliable +430V.
Most ppl have no clue as to how to build a decent choke input set up,
and bean counters always say no to the cost of the choke.
The choke input really has a benefit in class B or class C amps where
PSU drain
varies a lot, and where one wants to keep peak charge currents flowing
into capacitors
low. This is the case in class B tube amps where tube rectifiers are
used.
One finds that with silicon rectifiers and large caps without a choke,
the B+ regulation can be better than with any
choke input. Hence choke inputs are not used.
A choke input must have critical L with reactance = RL at the lowest
current, or bleed current,
so you need a lot of L. This means lots of wire and turns, and the B+
regulation always suffers from the
dc voltage drop across the winding resistance at least as well as across
winding resistances.
So a 100 ohm dcr choke will have a Vdc drop of 20V with a change in Ia
of 200mA only, and regulation is no
better than silicon and caps only.
A rule of thumb is that for where one wants really good choke input
regulation, ie,
with negligible Vdc drop caused by winding resistance, one needs to have
a choke the same weight
as the power transformer for the amp.
Charge currents in cap inputs can be reduced a lot by simply
adding series R between rectifiers and the reservoir cap until one
reduces the B+ voltage about 5%
from what it is without such R. The number of degrees if the charge
cycle are increased, and the peak chage reduced.
The amount of resistance can easily be worked out experimentally, and a
balance between peak currents
and B+ reg worked out. Hi-fi tube amps mostly don't need well regulated
PS because they
spend 98% of their lives working in class A with an unchanging I demand
on the psu.
OR one uses a low value cap as a reservoir, then a choke, then any
larger sized C for good filtering.
C1 should have reactance at least 1/20 of dc RL, the filter choke should
have reactance at least 10 times the reactance
of C1, and C2 have reactance at least 1/10 of filter choke reactance.
Regardless of what chokes and caps one uses, the resonance of the last
LC section should not be above 7Hz.
The added R can reduce peak charge I from say 10 times Idc to around 3 x
Idc,
and capacitors suffer less, because they do have finite ripple I
ratings.
Usually, the reduction of I peaks reduces the transformer noise
considerably.
The heat dissipated in the HT winding become low enough, since it
depends on I squared.
Better a hot resistance rather than a hot winding.
And in the case of a fault, the resistance fries and goes open before
anything else.
Patrick Turner.