Are there any generic rules-of-thumb for what variable cap (tune and
load) spacing is necessary for RF amps in a certain power class?

For example, "if you're building a 80W amp with a single 6146B the
plate spacing should be x inches".

Wasn't there a table in the ARRL handbook along these lines?

I'm guessing it depends on both the DC plate voltage and the power
level and maybe how mismatched the antenna is.

Of course, I could just buy bigger and bigger spacing surplus variable
caps until they don't arc over when I key up! Sort of how like they
put the weight limit on a bridge by building the bridge, driving bigger
and bigger trucks over it until it collapses, and then rebuild it with
the weight of the last truck that made it over :-).

Tim.

"Tim Shoppa" wrote in message
oups.com...
Are there any generic rules-of-thumb for what variable cap (tune and
load) spacing is necessary for RF amps in a certain power class?

For example, "if you're building a 80W amp with a single 6146B the
plate spacing should be x inches".

Wasn't there a table in the ARRL handbook along these lines?

I'm guessing it depends on both the DC plate voltage and the power
level and maybe how mismatched the antenna is.

Of course, I could just buy bigger and bigger spacing surplus variable
caps until they don't arc over when I key up! Sort of how like they
put the weight limit on a bridge by building the bridge, driving bigger
and bigger trucks over it until it collapses, and then rebuild it with
the weight of the last truck that made it over :-).

The handbook lists peak voltages for the plate capacitor spacings.

1000 volts .015 inch spacing
2000 .05
3000 .07
4500 .125
6000 .15
9000 .25
13000 .5

Also it states that is for peak RF volts, if the DC plate voltage is across
the capacitor it must be added. Plate modulated finals take about 4 times
the plate voltage.

Ralph Mowery wrote:

The handbook lists peak voltages for the plate capacitor spacings.

1000 volts .015 inch spacing
2000 .05
3000 .07
4500 .125
6000 .15
9000 .25
13000 .5

Also it states that is for peak RF volts, if the DC plate voltage is across
the capacitor it must be added. Plate modulated finals take about 4 times
the plate voltage.

The rule of thumb I learned for HV conductor spacing is 30 KV per inch
in dry air with smooth surfaces. Rough surfaces, points, sharp edges,
etc. all reduce the arc-over voltage, as does increasing humidity.

I'd go with the handbook's table above, as it allows a safety factor
and probably will prevent embarrassing events involving smoke.

--
I, for one, do not like to damage my precious equipment, especially
when it involves repeatedly hitting it with a computer.

-- Peter Corlett, about his hammer

Also to consider is air pressure. The breakdown voltage decreases with
decreasing air pressure, and to a slight extent with increasing
temperature.

"Reference Data for Radio Engineers" has a graph of breakdown for a
needle gap and for smooth surfaces. For the needle gap (pretty much a
worst case) at sea level (760mm Hg), it's a bit over 25kV peak at 1
inch spacing, and just over 3kV at 0.1 inches. For comparison, a
couple smooth 10" diameter balls spaced with a 1" gap between them will
exhibit a breakdown greater than 80kV. Plates for an air-variable high
voltage cap should have smooth, rounded edges. Looks like the
handbook's table assumes sharp corners, and/or gets its safety factor
from an assumption of rounded edges. It's usually not much of a
problem, but the higher dielectric constant (rel. to air) of the
ceramic or other insulation in a capacitor can create locally higher
fields that cause trouble.

Cheers,
Tom

(Great quote about hammers, Mike!)

K7ITM wrote:
Also to consider is air pressure. The breakdown voltage decreases with
decreasing air pressure, and to a slight extent with increasing
temperature.

"Reference Data for Radio Engineers" has a graph of breakdown for a
needle gap and for smooth surfaces. For the needle gap (pretty much a
worst case) at sea level (760mm Hg), it's a bit over 25kV peak at 1
inch spacing, and just over 3kV at 0.1 inches. For comparison, a
couple smooth 10" diameter balls spaced with a 1" gap between them will
exhibit a breakdown greater than 80kV. Plates for an air-variable high
voltage cap should have smooth, rounded edges. Looks like the
handbook's table assumes sharp corners, and/or gets its safety factor
from an assumption of rounded edges. It's usually not much of a
problem, but the higher dielectric constant (rel. to air) of the
ceramic or other insulation in a capacitor can create locally higher
fields that cause trouble.

Cheers,
Tom

(Great quote about hammers, Mike!)

Yep - the hammer quote was great! (finally, a sensible use for
computers hi hi)

But lets look at this a bit laterally - I cant argue the figures given,
just a lowly tech. BUT consider this - you should approach it from the
viewpoint of what variable capacitors you can get hold of!

What power you are running, and thus voltage etc, is the deciding
factor - in the case of 1 (or 2 even) 6146's, its wont need all that
huge a capacitor - most surplus transmitting caps would do it no
trubble. And the traditional amateur method is, after all:-

Do the calculations, and then build it with what you can lay your hands
on !

73 de VK3BFA Andrew

and I forgot - the loading capacitor can be a 3 gang (typical) brodcast
band one - old valve radio ones are fine. Its low impedance, so you
dont need to worry too much about rf volts.

"Andrew VK3BFA" wrote in message
ups.com...
and I forgot - the loading capacitor can be a 3 gang (typical) broadcast
band one - old valve radio ones are fine. Its low impedance, so you
dont need to worry too much about rf volts.
=======================
When transmitter ouput is 100 W then @ 50 Ohms the V(peak) equals 100V
when output is 400 W then @ 50 Ohms the V(peak) equals 200 V
when output is 1500 W then @ 50 Ohms the V(peak) equals 386 V

So for 400 W output or perhaps a bit higher a good quality air spaced BC
variable capacitor should be OK

Frank GM0CSZ / KN6WH