[R-390] Carrier Level meter R-390 A

[Leigh Sedgwick]

All I know is that I wish I had the insight you guys have when viewing these 
circuits....I've got a long way to go!  I always learn alot from these 
technical discussions...thanks to both of you.

73 de Mike
K7PI
Mercer Island, WA

----- Original Message ----- 
From: "David Wise" <David_Wise at Phoenix.com>
To: <r-390 at mailman.qth.net>
Sent: Wednesday, May 24, 2006 3:17 PM
Subject: RE: [R-390] Carrier Level meter R-390 A


Now you did it :)  My comments below.

- Dave

> -----Original Message-----
> From: r-390-bounces at mailman.qth.net
> [mailto:r-390-bounces at mailman.qth.net]On Behalf Of Gary Gitzen
> Sent: Wednesday, May 24, 2006 2:20 PM
> To: r-390 at mailman.qth.net
> Subject: RE: [R-390] Carrier Level meter R-390 A
>
>
> Hi R-390,
>
> David Wise gave his comments and insights
> regarding a possible modification of the R-390A
> carrier meter circuit to accommodate 100 ohm
> replacement meters. It prompted some additional
> thoughts and comments.
>
>
> > Since the DC operating point, voltage gain, and
> > plate resistance of V506A are changed, I'd expect
> > some change in the LONG AGC action (MEDIUM too
> > if you are using my AGC mod), but I believe
> > you would not notice.
>
> I pretty much agree on all points, especially that
> the difference would be hard to see. V506A at this
> point is acting pretty much as an AGC controlled
> current source developing voltage across a small
> R548. The cathode voltage, thus effective grid
> voltage, will change by maybe 150-250mv if R548 is
> changed from 27 to 160 ohms. That's an almost
> trivial grid voltage change.

I was thinking of dynamic, not static, behavior.
When AGC is set to SLOW, there's a 2uF cap between
the plate and grid of V506A, which makes it an integrator,
provided the gain is high enough.  As the gain drops, it
becomes less an integrator and more... something
else.  Something not wanted, no doubt, but I can't
say what, or how much it will degrade the AGC action.

> The specific action of the AGC-variable suppressor
> grid voltage on V504's (6AK6) cathode current is
> unknown. Can anyone provide insights? Will it result
> in any significant cathode current change, thus
> meter action; or is V504 merely a convenient current
> source?

I hadn't thought of this.  I've always believed that the
suppressors were being used as cheap diodes to keep the
1.8M(IIRC) B+ bleed into the AGC line from driving it
positive under no-signal conditions.

> > getting rid of R537 eliminates the problem with
> > the stock carrier pot where the whole adjustment
> > range is squished into the last 5% of the pot,
>
> Yea!!!!! And please see below for original meters.
>
> > Because the pot is a larger percentage of the
> > 6AK6's cathode bias resistor, IF gain is more
> > sensitive to the carrier pot setting than before.
> > This could be ameliorated by rewiring the pot
> > as a pot instead of a rheostat.
>
> Excellent suggestion! Please consider my earlier
> thoughts on the subject so modified.
> [Translation: Why didn't _I_ think of that?
> Programmers note: it's called "stepwise refinement"
> & peer review.]

Unfortunately, where my mod (applicable to 17-ohm
meters only) gave a lot while taking little, you
now have an uncomfortable choice: wire it like
a standard pot or a backwards pot.  The former
gives constant IF gain and variable meter scale;
the latter gives variable IF gain and constant
meter scale.  Since in the "pot" configuration
the meter scale is no more variable than it
used to be, I lean that way, but it disappoints
me that you can't have both.  That's where I
started thinking about dual pots, but as I said,
it's probably not worth it.  Anyway, the real
test is to just settle down at the bench and
try it.

> > Reducing R524 will preserve the original IF gain.
>
> Yes, and if R523 is rewired as a pot instead of
> as a rheostat when R537 is removed, R524 (6AK6
> cath res) definitely needs to be reduced to around
> 600 ohms. A 5.1K shunt across it will do this.
>
>
> Carrying David's "pot thought" forward, replacing
> the R523 & R537 combination with a 20 ohm pot
> wired as a pot while using an original 17 ohm meter
> would appear to be a Very Good Thing. [Please see
> caveat #2.]
> This raises the following question: were 20 ohm
> AB pots with 10% tolerance available in 1950-55
> for a reasonable price? This circuit appears to
> beg for one, instead of the R523/R537 kludge.

20-ohm carbon pots are (or at least were) just about
unavailable; 20-ohm WW's (used in the R-390) are
(and were) expensive, which is why they went to carbon.
They didn't do it right, though, hence the squish.
I believe they assigned that corner of the redesign
to a naive junior engineer.  My carrier pot mod
(posted once or twice in the archives) uses the
original pot, wired as a pot, but backwards; that is,
the 6AK6 cathode current goes to the wiper, and
the meter goes to the top and to a shunt resistor
to ground.  At the expense of imperceptible
decreases in IF gain, and meter sensitivity,
and an equally imperceptible increase in IF gain
sensitivity*, this gives a smooth, linear control
over the zero setting, and as a bonus, the meter
sensitivity is made constant.  I installed this
on my radio without removing the IF deck.  It's
just change a resistor and move a couple of wires.
Meter zero is in the center third of the pot.

* That is, the gain changes depending on the
carrier pot setting.  Was 680-698, now 680-709.

> > The meter's scale factor (as opposed to its
> > zero point) is as sensitive to the carrier pot
> > setting as ever.
>
> This implies that the AGC-variable suppressor
> grid voltage to V504 _does_ significantly affect
> cathode current. Thanks for that insight.

Sorry, that wasn't my point at all.  The scale
factor depends on the series resistance seen by
the meter - which changes as you move the carrier
pot.  The original meter could theoretically see
anywhere between 27 and 45 ohms.  Add the meter's
own resistance, and the circuit has anywhere from
44 and 62 ohms, or a 40% change.  The 100-ohm meter
sees between 160 and 260 ohms, for a total range
of 260 to 360 or... still about 40%.

> > The main objection, in my opinion, is that your
> > deck is now incompatible with the stock meter.
>
> Absolutely agreed. Please see caveat #2.
>  >> Caveat #2: If you do this, _please_ document it
>  >> on both the chasis and the IF module.
>
> > If I had exactly one radio, no spares, and no
> > intention of acquiring more, I'd do it.  It's
> > very much worth a try, and anyone who does it,
> > please post your findings.
>
> Thanks to David for his comments and insights.
> It's obvious they come from someone who has "been
> there and done that" as opposed to my theoretical
> musings.
>
> If I had even one R-390A without an original
> carrier meter, and had on hand a 100 ohm replacement,
> I'd give it a try. And I'd carefully document exactly
> what was done, and why, on both chasis and IF module.
>
> I second David's request for results from anyone
> who makes this mod to hopefully allow use of
> 100 ohm carrier level meters in place of the original
> 17 ohm meters.
>
> Just so it will all be in one place:
>   Install 100 ohm replacement carrier level meter.
>   Replace R548, 27 ohms, with 160 ohms (150 ohms if
>    it's all that you have).
>   Remove R537, 22 ohms.
>   Replace R524 with approx 600 ohms, or shunt the
>    existing 680 ohm R524 with 5.1K.
>   Rewire R523 as a pot, with the meter lead coming
>    off the wiper.
>    If needed to get adequate adjustment range, add
>    22-47 ohms in series with the ground side of R523
>    and reduce R524 by the same amount. (39 ohms with
>    R524=560 ohms should work nice.)
>   Document what you have done, and why, on both
>    the chasis and IF module.
>
> All relevant constructive comments welcome.
>
> Gary Gitzen
> 5/24/2006
_____________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/faq.htm
Post: mailto:R-390 at mailman.qth.net
Unsubscribe: http://mailman.qth.net/mailman/options/r-390