[Collins] 312B-4 wattmeter query

Tue May 8 00:09:18 EDT 2012

Many thanks to Jerry and others who responded to my call for counsel.  

Following a test setting up the output as input and recording results, I
will be diving into each directional coupler checking resistors, diodes, and
connections.  

Per Dave , Schottky diodes to be used as replacement diodes.  Any
suggestions as to a specific Schottky diode as a better choice?

Per Jerry, the #1 meter will be exercised.  

Will post results in June (or sooner).

Tnx es 73,
Chris
W7JPG

-----Original Message-----
From: collins-bounces at mailman.qth.net
[mailto:collins-bounces at mailman.qth.net] On Behalf Of Dr. Gerald N. Johnson
Sent: Sunday, May 06, 2012 6:19 PM
To: collins at mailman.qth.net
Subject: Re: [Collins] 312B-4 wattmeter query

The RF wattmeter is symmetrical. The ranges are achieved with different
values of resistance, set by selected resistors so each is individually
calibrated.

Since a wattmeter can introduce some SWR, be sure to compare them with each
driving and each driven. Even a Bird will introduce some impedance change
(or especially).

The schematics tend to show a 290 microamp meter. That's very odd and a
draftsman's error where the scale on the Leroy missed and he put a 9 where
there should have been a zero. The meter is really 200 microamps which is
not very common but has been a standard movement.

The sticky meter is inside the meter itself and takes careful finesse to
repair. Finesse that is about gone from meter repair shops. I have tried but
the gains have been limited. Its possible its been open and there's been
some iron filings that have snuck in its also possible that the pivots are
blunted from it being dropped or the pointy pivot bearings have cracked. The
mate for the sharp pivots are often doughnut shaped jewels. And they can be
cracked by too much shock. A good meter man should be able to correct them.

You can exercise the sticky meter with a suitable variable voltage supply
and series resistor. 5K ohms per volt series resistor. So for a 0
- 10 volt supply, a 50K resistor will keep the meter from pegging. And can
be a check of the basic meter's calibration. When its most moving in
response to knuckle raps, its unlikely to be accurate.

I'm not sure why the meter that pegs on all ranges is doing that, could be
its a replacement meter movement too sensitive, or the four factor selected
resistors have been shorted or replaced with lower values to make it a handy
meter for legal CB powers of 5 watts. Maybe the variable capaictors have
(one or the other) have been shorted by lightning. That meter can be tested
for sensitivity the same as the sticky meter.

The basic circuit was discussed in QST more than 50 years ago by Warren
Bruene, then probably W0TTK, now W5OLY. ARRL's W1FB "simiplfied" the circuit
which reduced its performance and that simplified circuit has been in
handbooks for decades. The original article is titled: "An Inside Picture of
Directional Wattmeters" April 1959 beginning on page 24 available to ARRL
members on line or in good libraries and on CD-ROM.

There are two pages on how it works in the Collins Yellow Book titled
"Fundamentsls of Single Side Band."

The two variable capacitors are set to minimize reflected power on a fine
load each one set with the RF flowing the opposite direction. They could
have been damaged or shorred by lightning. An ohmmeter check should show the
coax line center conductor an open circuit to ground when not connected.

The coupler can be reversed with no problems, just reverses the switch
indications of forward vs reflected. As far as I know, that's how the
reverse power scales were calibrated by feeding it backwards, and the two
variable capacitors were calibrated by feeding RF forwards and adjusting the
one on the reflected circuit side for a minimum meter indication, then with
the sensor reversed the "forward" was adjusted for minimum indication with
the RF flow backwards.

Changing diodes would require new calibration resistors.

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.

On 5/6/2012 4:33 PM, Chris Kepus wrote:
>
> I have 2 B-4 units: one winged, one meatball.
>
> I haven't used either for some time so decided to test their wattmeter 
> function before putting them "on-line".  Used a Valiant transmitter 
> for RF power into a Bird into a 312B-4 into a Heathkit Cantenna. The 
> results were very disappointing.
>
> The CCA site had nothing in the way of info to help.  The Collins list 
> archives back through 2007 were also quiet on B-4 coupler problems or 
> rebuilding. Google didn't turn anything up.
>
> Before I plunge into these units, I hope someone familiar with the B-4 
> can suggest some possible causes of the symptonms and/or the most 
> efficient way to trouble shoot my #1 and #2 B-4's problems.
>
> First up to test was #1, the winged B-4.  The meter pointer at rest 
> (no power in) is a cat whisker above zero.  With power applied, the 
> meter pointer moved to indicate 75 watts (200 W mode) while the Bird 
> indicated 110 watts. Rapping the meter with a knuckle caused the meter 
> pointer to advance to roughly 140ish watts. When power was removed, 
> the pointer did not return to zero but stopped at about 30.  Hmmmm.  
> Rapping with a knuckle caused the pointer to return pointing near 
> zero.  Repeated this about six times hoping the meter exercise would 
> help but no joy. It appears the meter has a "sticky" movement.
>
> Second to test was #2, the meatball B-4.  At rest (no power in), the 
> meter pointer is at zero.  With power applied (110 watts on the Bird), 
> the meter pointer pinned above 200 watts (200 W mode).  Same result in
2000 W mode.
> Since the coupler is bi-directional with 200 and 2000 watt ranges for 
> reflected power, I was tempted to input the power on the antenna side 
> to see what the meter would read. Nothing on the schematic looked to 
> be in harm's way if the input/output were flipped.  Anyone?
>
> As Al, W8UT, said in a very similar circumstance in 2008, "Any help 
> greatly appreciated."
>
> Thanks es 73,
> Chris
> W7JPG
>
>
______________________________________________________________
Collins mailing list
Home: http://mailman.qth.net/mailman/listinfo/collins
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Collins at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html