Great comments, Jim and Roy. Two points. First, my 8877 Amp, described in
several editions of Orr's book, was designed and built in the late 1970s. In
it, as Jim noted, I used a vacuum relay to interrupt the HV under
overcurrent conditions. That was a bad idea, which I learned years later the
hard way when I had a flashover and ruined the relay and other circuit
components. I redesigned the HV overload circuit in the mid-80s to
eliminate the vacuum relay. Now the overcurrent circuit monitors cathode
current and trips the amp off-line for currents in excess of 1.3A. It uses
an optically isolated relay to do this and has worked well for many years.
Unfortunately, there was no way to correct the published design. (I also
upgraded other aspects of that 8877 anp: it now has QSK capability and
covers all WARC bands. I can send a revised schematic to anybody who want
it.)
Roy questions whether it is appropriate to use a current-limiting resistor
(in my case, 25 ohms) between the transformer/rectifers and the filter
capacitor (50uF in my 8877 amp). His point is that doing so is equivalent to
using a cheap transformer with a 25 ohm "Equivalent Series Resistance" and
thus negates the benefits of using a stiff (and expensive) transformer. This
is a reasonable objection, but I don't fully agree. Here is the
counterargument. First, there are two benefits of the resistor; it limits
the inrush current when the HV supply is switched on, since without it the
rectifiers would see a dead short from the uncharged capacitor. (Of course,
a step-start circuit in the xfmr primary is another way of limiting the
inrush current.) Second, it protects the transformer and rectifiers from a
shorted filter capacitor. Many amps use banks of electrolytics for the HV
capacitor and a short circuit in one can lead to a cascade of failure along
the entire bank. (My 8877 amp now uses oil caps, but the original design
used electrolytics.).
The second point is that the effect of the resistor on power supply voltage
regulation is negligible, On PEP peaks, the current is supplied by the
charged filter capacitor, which is not limited by the resistor. Under key
down conditions, when the power supply is delivering, say, 1 Amp of current,
the drop across the resistance is only 25V, or less than a one percent drop
in power supply output. By contrast, an inexpensive tranformer, with an
underrated core, will not only overheat in sustained use, but will have very
poor voltage regulation -- in effect an ESR much greater than 25 ohms. Note
also that the resistor has a negligible effect on the transformer's ability
to recharge the filter capacitor. A 25 ohm resistor in front of a 50uF
capacitor has a time constant of only 1.25 mSec, which is much faster than
the pulsed DC from the diode rectifiers that replenish charge on the filter
capacitor.
73,
Jim Garland W8ZR
> ------------------------------
> Date: Sun, 7 Feb 2010 03:45:43 -0800
> From: "Jim Thomson" <Jim.thom@telus.net>
> Subject: [Amps] QRO-RG-142
> To: <amps@contesting.com>
> Message-ID: <030B78EE98BC44B3A4EBCA14168AB551@JimboPC>
> Content-Type: text/plain; charset="iso-8859-1"
>
> ## I just looked at the W8ZR 8877 amp in Orr's last book [23rd ed]
> ## I was also going to use the vac relay as intended... to open off
> B+ UNDER LOAD... . In that case... it was part of the W8ZR
> plate over current setup. IE: If plate current exceeds say 1000 ma
> during tune up, vac relay opens off the B+. Problem is, during
> a fault condx.. and only a 25 ohm WW for a glitch R... fault current
> is 4000 V /25 = 160A !! Which will FRY vac relay contacts !
>
> ------------------------------
>
> Date: Sun, 7 Feb 2010 06:37:02 -0600
> From: "Roy" <royanjoy@ncn.net>
> Subject: [Amps] B+ relays
> To: <amps@contesting.com>
> Message-ID: <6F3629C315AD4D3B88B36541561ABD81@RoyPC>
> Content-Type: text/plain; format=flowed; charset="iso-8859-1";
> reply-type=original
>
> If you really mean the resistor is located between the rectifier and the
> filter cap, that defeats all efforts to design and build a plate
transformer
> with low ESR. Just go buy a cheap transformer instead.
>
> 73, Roy K6XK
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|