[Collins] 312B-4
Dr. Gerald N. Johnson
geraldj at ispwest.com
Wed Feb 22 00:02:19 EST 2006
On Tue, 2006-02-21 at 22:13 -0500, Cam and Juli Hedrick wrote:
> Hi gang,
> I've noticed that my 312B-4 and 312B-5 read output power levels lower by
> about 12% and 20% respectively than my Bird or Drake which give identical
> readings. I've used two different plugs in the bird and gotten the same
> reading. What's up...Caps out of value, or maybe carbon resistors?
> By the way, an engineer with IRC has informed me that you can bake a carbon
> resistor and bring it back into tolerance. Has anyone ever heard such a
> thing before?
> I'll dig deeper, but then again the other meters are working fine.
> Thanks
> Cam
> WA4JKW
>
>
There are calibration resistors and maybe adjustable pots for
calibration. And there are probably carbon composition resistors for
terminating the coupler. Those termination resistors are hidden down
inside metal blocks for better shielding and heat dissipation. The
variable capacitors adjust for best reflected power match. Feed power to
a good dummy load and adjust the reflected capacitor for minimum
reflected meter reading. Reverse the coupler (RF input and output) and
adjust the other capacitor for minimum (forward scale, but now
reflected). Coupler termination resistors out of value could lead to
poor directivity (not being able to get a good null of reflected power
on a perfect load). Its possible the trimmer capacitors have loss
connection with age and corrosion.
Remember that the Collins meter is average reading, not peak reading.
And that any RF meter will show a different reading with different
amounts of SWR, especially the meter circuit isn't as good as that used
by Collins.
Also the 1N82A diodes are not the most robust of diodes but way back
then had the least forward drop and the highest frequency capability
having been made for mixer diodes in early UHF TV tuners and converters.
The variation between diodes is why the calibrating resistors had to be
chosen in production so changing the diodes may not help the calibration
accuracy.
My schematic says the calibrating resistors were to be chosen to the
nearest 1%. That means 1% resistors which are supposed to stay within 1%
over temperature, humidity, and age. E.g. to not drift much, certainly
not 20 or 30%.
Carbon composition resistors drift from heat and from humidity. Cooking
them might recover humidity effects but shouldn't compensate for the
continued aging effects of heat.
Besides unless IRC has started making real carbon composition resistors
they haven't made carbon composition resistors in more than a 60 years.
What they used to sell as carbon composition resistors were really film
on glass tubes in molded cases (you could see the mold lines running the
length of the sides of the resistor) to make them up to the size for the
competitive resistors. Collins specs in the 60s said for carbon
composition resistors, "QPL except IRC." That means the military
Qualified Parts List, but not IRC. IRC resistors tended to go down in
value with age instead of up like all other brands. That's because the
heat was concentrated in the film on the glass tube and that heat
charred the molded case. I'd not depend on one of them to act as a fuse
either.
--
73, Jerry, K0CQ, Technical Advisor to the CRA
All content copyright Dr. Gerald N. Johnson, electrical engineer
More information about the Collins
mailing list