[Collins] 872a Lamp, revisited

[Zap (Chris)]

    Wow, I can't believe the stir that I started!!  I'm not sure how many
watts the jeweled bulbs are, but I was planning on putting one of the big
jewel lamps in series with the switches for both the heater and the plate
sections.

    I guess I should go pick up a ballast at the hardware store, I know
they're not very expensive.  As far as the filament, I still need a filament
transformer, correct?  I am doubting that I am going to find one of those
without having to dig somewhere.  Any suggestions on where to pick one up
cheaply?  Would it be smaller/easier to build a power supply for the heater?
I would like to be able to fit the guts for this lamp into a small (square
with ~8"-10" sides) enclosure with the tube sticking out the top.  Since it
would be Collins A-Line style, I'd like it to be metal with the louvres like
the A line has... anyone know how I could get pieces with these molded or
make them somehow?  Additionally, I thought it might be a cool idea to have
somebody frost/engrave a glass plate with a "On Air" or "RF Heater On"
message to place in front of the tube :)

73,
Chris
----- Original Message ----- 
From: <ah7i@atl.org>
To: "Dr. Gerald N. Johnson, electrical engineer" <geraldj@isunet.net>
Cc: <ah7i@atl.org>; "Jim Brannigan" <jbrannig@optonline.net>; "Zap (Chris)"
<zap@xzapx.com>; <collins@mailman.qth.net>
Sent: Tuesday, September 02, 2003 3:01 PM
Subject: Re: [Collins] 872a Lamp, revisited


>
> Jerry, I do understand how ballasts work. The resistor is a KISS approach
> to limit the current as you said may be necessary when I first suggested
> using a ballast...
>
> > Since a fluorescent light is based on mercury vapor, the ballast idea is
> > a good one. There are ballasts rated at 1/2 amp through the tube.
> > Voltage drop in the 872 should be lower than in a long fluorescent
> > though which might fry the ballast if its current limiting isn't all in
> > the coil and core.
>
> So, I don't understand what you're getting at. A cap is fine for AC
> current limiting as well. You said it yourself in your example of raising
> and lowering the voltage across a fan motor. One just has to understand
> what is going on in the particular setup.
>
> I've been giving some additional thought to this and have an idea that the
> 872A shoudl work great with a ballast designed for two 20W tubes (40W
> ballast). Even though the 872 is rated for about 1.5A continous , this is
> as a rectifier and as a rectifier, current will only be flowing during a
> half cycle, so the tube should be fine. OTOH, perhaps the 872A could be
> operated more like a lamp with the full cycle used for excitation. Would
> we have to limit the current to 20Watts? I don't think so, because once
> fired, the heater could be shut off and the total heat being generated in
> the tube will be abotu the same as the half wave rectifying case.
>
> I'm also thinking that a 20 W or so incandescent lamp in series with the
> tubes heater would make it a better match for preheating with the ballast.
>
> Guess it's time to root around and see if I have any magnetic ballasts in
> the junk pile...
>
> -Bob
>
>
>
>
> On Mon, 1 Sep 2003, Dr. Gerald N. Johnson, electrical engineer wrote:
>
> > Sorry Bob, coffee or no, you don't understand the details of the
> > ballast. Its all inductive and magnetic. Its complex, the high voltage
> > output of the fluorescent ballast is generated by a transformer with a
> > saturating section between the primary and the secondary that makes it
> > very much a constant current no matter whether the load terminals are
> > short circuited or nearly open. Like a welding transformer that shows
> > the same current whether shorted or arcing. Some welding transformers
> > move the position of the magnetic shunt to change the current. In the
> > old two lamp trigger start ballast the current through the four
> > filaments is practically the same when the starter has them all hooked
> > in series across the output as the current is through the fired lamps.
> > In that system, when the trigger starter opens, there is an inductive
> > kick that gets plenty of voltage to the lamps for starting the arc. One
> > need not add R for making the 872 glow, the ballast will take care of
> > that by the core saturation.
> >
> > A cap may well resonate with some condition of the ballast's inductance
> > and INCREASE the current. I've used that effect with a 220 volt muffin
> > fan on 120 volts. Just the right C approached resonance enough to give
> > 220 volts on the fan coil terminals. In another case I used a much
> > different C to lower the voltage on the fan terminals to make it run
> > slower. Both are possible, sometimes with disastrous effects when the
> > resonance isn't needed.
> >
> > 73, Jerry, K0CQ
> >
> >
>
>