Amps
[Top] [All Lists]

Re: [Amps] SHV Connectors and B-

To: "'Paul Christensen'" <w9ac@arrl.net>, <amps@contesting.com>
Subject: Re: [Amps] SHV Connectors and B-
From: "Jim Garland" <4cx250b@miamioh.edu>
Date: Thu, 13 Mar 2014 08:51:57 -0600
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Paul and the group,
I don't care much for Cinch-Jones connectors for amplfier control,
especially the type that cannot be locked into place. They're a very old
design and newer, better, safer connectors are available. And of course one
should never use Millen HV connectors under any circumstances. I've not used
XLR connectors, so can't speak to their benefits. I've always thought that
the military grade Amp MS connectors are the gold standard, but are feasible
only if you can find surplus male and female mating pairs with the correct
number of pins. Buying them new would cost hundreds of dollars.  What I've
been using lately are the plastic TE Connectivity/Amp version of the MS
connectors.. These come in various sizes and pin configurations and are
reasonably priced. In my latest amp, I use 14-pin size 43 connectors, Mouser
p/n 571-206043-1. By the time you buy the shell, strain relief and pins (all
sold separately), a male/female pair will run about $30.  You'd need two
pairs for a complete cable and chassis connector assembly.  These are
twist-lock connectors, a feature that is a definite plus.

I prefer to plug the power supply directly into the 240VAC mains (rather
than plugging the RF deck into the 240V circuit and routing the mains
voltage to the power supply) for several reasons. Doing so means one only
needs to route a single 120V (L1 and N) circuit up to the RF deck, rated at
only three amps or so (plus a control wire for activating the HV supply
relays). A 3A circuit (or half that for 240V countries) is plenty for
powering the filament xfmr, blower, and LV supplies in an RF deck. A heavy
240V three or four-wire cable between the RF deck and HV supply is
unnecessary and would greatly limit control connector choices. Frankly, I
don't know what connector one could use in that situation. Second, the HV
supply is typically mounted on the floor behind the operating desk, which is
convenient to the 240V wall outlet.  Third, it makes more sense to mount the
circuit breakers close to the plate transformer, rather than in the RF deck.
And finally, you'll get better RF isolation between the HV supply and the RF
deck if the 240V power line is connected only to the power supply. 

If you want to see how I wired my latest homebrew amp to its HV power
supply, you can check out the schematic diagram on the power supply design I
wrote up for QEX in 2013. (I can email you the schematics if you prefer). My
design uses one HV power supply to distribute HV and various control
functions to multiple RF decks, and so had some logic and interface
switching which would be unnecessary for your project.

Other issues to keep in mind are how you want to implement a safety plate
current trip and reset circuit, whether you want to sample the HV for
metering purposes in the RF deck or in the power supply, whether you want
the blower to run for a minute or two after the amp power is switched off,
and how you want to fuse the HV. For a triode amp, with indicator lamps,
reset circuits, multimeter switching and relay control, I'm guessing you'll
end up needing 8-10 wires in your control cable, plus a separate RG58/59 or
RG6 cable for the HV. It's a good idea to have some reduancy in the ground
wires linking the power supply and RF deck. I use four parallel 18AWG ground
wires connecting my amplifiers and power supplies, which is probably overkil
(and also increases the number of pins in the control connector).
73,
Jim W8ZR

 

> -----Original Message-----
> From: Amps [mailto:amps-bounces@contesting.com] On Behalf Of Paul
Christensen
> Sent: Thursday, March 13, 2014 7:02 AM
> To: amps@contesting.com
> Subject: Re: [Amps] SHV Connectors and B-
> 
> > "Over the past several years I have been recommending 5-pin XLR
> connectors between the amplifier controller (assumed to be part of the
> RF deck) and a remote HV PSU. The aim is to keep the HV-minus, control
> and ground connections all together, so that the HV PSU cannot be
> enabled separately without that connector plugged in."
> 
> XLR connectors are excellent for their electrical and mechanical
> reliability.  In my case, I'm thinking of using two connectors between the
> amp and PS.  The first will pass B+ through a chassis SHV connector.  The
> second must pass 240V to the HV transformer as well as relay control for
the
> HV Lo/Hi setting.  In looking at suitable connectors, I'm leaning toward a
> 20A 8-pin Cinch-Jones or possibly a Mil-Type ITT/Canon/Amphenol multi-pin
> connector.   A female chassis connector would be installed on the PS, and
a
> recessed male chassis connector placed on the RF deck.  With 8-pins, I can
> double-up on 240V and ground wiring.  In doing so, an accidental pull of
> that plug will kill the AC primary and B+ would bleed down in several
> seconds.  I could see placing the B- to chassis diode at the PS as well as
> leaving the existing diode inside the RF deck.
> 
> Since the PS will not have its own AC power plug, I would be depending on
> the double-ground pass through the Cinch-Jones connector.  I can add a
> ground bolt to the PS chassis, but I want to ensure safe operation without
> having to depend on that bolt.  The ground pass-through seems adequate to
> me.  Anyone see other issues?
> 
> One issue I'm wrestling with is the the base design of the PS cabinet.
I'm
> leaning toward designing my own cover panels in software but the Dahl 2A
CCS
> transformer weighs 86 lbs.  I can order cut steel plates or just stack two
> steel relay rack panels to help support the transformer weight.  Maybe Jim
> and/or Vic can chime in on this.  2A is overkill and I'll inquire if they
> can drop the weight down by custom winding it with a 1.2A or maybe 1.5A
spec
> at 3200V.  That will result in about 4.3KV on the plate.
> 
> Just got the quote back from Hoyt on the 250-degree panel meters.  Ready
for
> this?  USD $400 ea. x 4 = $1600 for the set.   Pricing includes custom
> legends and back-illumination.  Simpson wasn't much better at $325 ea.
They
> do look cool though...
> 
> Paul, W9AC
> 
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>