Re: Remote Power Supply Connection

Trevor Wilson wrote:
> 
> "Ian Thompson-Bell" <ruffrecords@[EMAIL PROTECTED]
> wrote in message
> news:g13o17$10ph$1@[EMAIL PROTECTED]
> > Trevor Wilson wrote:
> >> "Ian Thompson-Bell" <ruffrecords@[EMAIL PROTECTED]
> wrote in message
> >> news:g0up1a$1ap1$1@[EMAIL PROTECTED]
> >>> Trevor Wilson wrote:
> >>>> "Ian Thompson-Bell" <ruffrecords@[EMAIL PROTECTED]
> wrote in message
> >>>> news:g0rmo0$27co$1@[EMAIL PROTECTED]
> >>>>>
> >>>>> I agree when the rectifiers are close to the transformer supplying
> >>>>> them but not when they are separated from it my a couple of metres
of
> >>>>> cable.
> >>>> **And again, it is not difficult, nor expensive to deal with such
> >>>> issues. A handful of capacitors will effectively deal with any
> >>>> problems.
> >>>>
> >>>>
> >>> Definitely not. The point is that the caps normally used to suppress
> >>> reverse spikes in semiconductor diodes do just that. The have no
effect
> >>> on the forward current spikes which are significant.
> >>
> >> **No.
> >>
> >
> > Denial is not much of an argument.
> 
> **How about: Not necessarily? Better?
> 
> >
> >>  With remote ac
> >>> supplies these current spikes must pass through the cable and they
will
> >>> cause serioes radiation. Any attempt to stop them happening will
simply
> >>> stop the rectifiers from rectifying. That is why is is essential to
have
> >>> the rectifiers close to the transformer.
> >>
> >> **No.
> >>
> >
> >
> > Denial is not much of an argument.
> 
> **OK. It is not necessary to have rectifiers close to the transformer. I
set
> up a test last night, with 2 Metres of cable, connected to a
transformer,
> thence to a rectifier, thence to an amplifier. An AM radio place VERY
CLOSE
> to the cable picked up some radiation. Covering the cable with an
earthed
> sheild completely eliminated the noise.

Trevor, you learnt nothing im****tant.
And YOU deny you are wrong.

In many cases the switching noise is MAGNETICALLY INDUCED, not 
transferred electrostically, or by capacitance. 

In many cases diode switching noise in so many power amps is caused by
both capacititive and or magnetic stray coupling.

Therefore some makers move the whole PSU to a separate chassis and ONLY
the DC is brought to the amp.


You stubbornly refuse to accept their good design practices.

In tube amps the weight of the amp becomes so high if the PSU is on the
same chassis
as the amp channel, so a separate chassis is necessary.

For good isolation of the PSU artifacts from the audio signal where the
PSU IS on
the amp chassis its good practice to have the PT and diodes all tightly
wired inside a steel sheet enclosure.

But you never design and build anything.

When you have successfully build your first tube anything, you might
have 
some credibility at r.a.t.

Patrick Turner.

> 
> >>>>>>>>>>> http://www.arcadvisor.com/faq/dc_direct_current_arc.html
> >>>>>>>>>>>
> >>>>>>> This clearly shows that at low currents (e.g HT currents) and
high
> >>>>>>> voltages (e.g.HT voltages) dc arcing is very unlikely to occur.
> >>>>> **Remind me: How much energy is able to be stored by large
(unstated
> >>>>> values)
> >>>>> capacitors?
> >>>>>
> >>>>> 0.5*C*V^2
> >>>>>
> >>>>> In a 100uF capacitor at 300V that's a mere 4.5 joules. Not worth
> >>>>> worrying about.
> >>>> **Points:
> >>>>
> >>>> * 4.5 joules is a substantial amount of energy.
> >>> No it isn't. it is 4.5 watts for one second.
> >>
> >> **Yes, it is. It depends on how long it takes to charge/discharge
> >> whatever capacitance is on the other side of the connector. In 0.1
Sec,
> >> that is 45 Watts. In 0.01 sec, that is 450 watts. Certainly plenty of
> >> energy to damage surface treatment of most connectors. And, as I
stated
> >> earlier, lots of builders use considerably more capacitance than
100uF
> >> and more than 300 Volts for their power amps. 1,000uF and 700 Volts
is
> >> not unheard of. And THAT is a considerable amount of energy.
> >>
> >>> My small rotary hammer drill
> >>>> is rated at 2.5 joules per strike and can pulverise bricks.
> >>> And there you have hit the nail on the head. It is not the *amount*
of
> >>> energy but the *rate* at which it is consumed that matters.
> >>
> >> **EXACTLY my point. 4.5 joules consumed in 10msec is a *lot* of
energy
> >> for thinly plated connector to cope with.
> >
> > No, 4.5 Joules consumed in 10mS is not a lot of energy, it is a lot of
> > power.
> 
> **And it is enough to damage any plating on connectors. Imagine the
damage
> which could be caused by real-world power amps, employing (say) 1,000uF
at
> 700 Volts DC.
> 
> --
> Trevor Wilson
> www.rageaudio.com.au