[R-390] Call for measurements - 100dB carrier reading

Thu Aug 23 14:42:53 EDT 2007

I only had a moment before work.  More later.
Resistances with the RF deck unplugged.

RF resistors:
R201 281K
R233 466K
R234 1.45M
R202 229
R203 77.6K
R204 10.2K
R205 2.33K

2nd mixer resistors:
R214 28
R213+R206 1.06M
R215 2.38K
R216 2.43K

3rd mixer resistors:
R230 31
R217 450K
R215 2.33K
R219 2.38K

With the innocuous exception of R230, they're all in spec.
I'm surprised at the number that read low; I'm used to
carbon comps going high.  I checked with a second DMM,
all same.  Oh, and Z503 was 13.5 ohms.

Next I'll measure resistors on the IF deck, then power
the set up without oscillators, elevate the AGC line, and
look for current.  Finally I'll measure diode load voltage
for you, then substitute tubes in the signal chain.

Could you take an additional measurement?  With your IF
gain set to the stock value (150uV for -7V diode load
in MGC mode), please apply 100mV to the IF deck in AGC
mode and measure the AGC voltage.  This will reduce the
problem to RF or IF.  I hope :)

Say, I forgot to ask what you measured the AGC with.
A 10M DVM like me?  Just making sure.

Dave Wise

> -----Original Message-----
> From: grahambaxterdelphe at googlemail.com
> [mailto:grahambaxterdelphe at googlemail.com]On Behalf Of Graham Baxter
> Sent: Wednesday, August 22, 2007 2:56 PM
> To: David Wise
> Cc: r-390 at mailman.qth.net
> Subject: Re: [R-390] Call for measurements - 100dB carrier reading
> 
> 
> Hi Dave
> 
> I take your point about some of my suggestions appearing to have the
> opposite effect to the one you are experiencing. Perhaps I should ask
> you to just humour me and check them anyway! When I had filter leaks,
> paradoxically my S meter read very high. But my diode load voltage was
> excessive too.
> 
> I agree R201 would fit the bill nicely. I am also quite 
> interested in R544.
> 
> Years ago I had a Collins R-390A with a Motorola IF. The meter zero on
> that was all over the place. I replaced the pot and it was lots
> better. A couple of days ago I found that old pot, and just out of
> interest I measured it. It had gone up to 330 ohms. I was quite
> surprised at that.
> 
> Let me know what you find out; I am hooked now!
> 
> Graham
> 
> On 22/08/07, David Wise <David_Wise at phoenix.com> wrote:
> > Thanks for your debug tips, Graham.  I will check
> > tonight, time permitting.  Right now all I can do
> > is comment.
> >
> > 1. Any trouble in Z503 or the AGC amp in general
> > (besides runaway oscillation which would be obvious)
> > would lessen my AGC, not increase it.  I'll measure
> > it anyway, since I've seen a lot of posts about it.
> >
> > 2. Yes, if R201 (270K, top leg of V201 AGC divider)
> > were to increase, or if R234 (1.5M, bottom leg) were
> > to decrease (but that's rare), V201 would get less
> > than usual AGC, forcing the control loop to throttle
> > down the other tubes more.
> >
> > A similar effect would occur if V201 were absorbing
> > electrons at pin 1, but I tried four different ones
> > that also test good, so I can rule it out.  On the
> > other hand, V203 (2nd mixer), and V204 (3rd mixer)
> > have high-impedance grid returns and could do the
> > same thing.  (So does V202, 1st mixer, but my
> > peculiarity happens above 8MHz as well as below.)
> > These are easy to check: drive the AGC bus negative,
> > kill the oscillator, and measure between the grid
> > test point and the AGC bus.  Should be zero.
> >
> > 3. AGC leakage is unlikely to be towards
> > anything but ground, which would lessen the voltage
> > not increase it, and I have replaced C551, the usual
> > culprit, but I'll check anyway to get early warning
> > on filter trouble.
> >
> > 4. Thanks, I'll check over my own radio first.
> >
> > 5. The design is poor.  The R-390 had a 15-ohm wirewound
> > pot, and its cost and complaints about its coarse
> > adjustment granularity led to use of a carbon pot
> > in the R-390A.  The smallest practical value was
> > 100 ohms, and some unimaginative junior engineer
> > tacked a 22-ohm resistor across it to make it look
> > like 15 ohms.  But the whole R-390 adjustment range
> > is squeezed into the bottom end of the carbon pot,
> > and at the nominal setpoint, change is magnified tenfold,
> > so the slightest scratchiness (like mine) is unbearable.
> > I have a keen interest in design and low tolerance for
> > kluges, especially unnecessary ones.  This one really
> > irritates me.
> >
> > 6. I'll compare my diode load voltage to yours
> > although any discrepancy would be caused by the
> > AGC amp, 4th IF amp, IF cathode follower, AGC
> > rectifier, or detector, none of which can
> > exaggerate the AGC voltage.
> >
> > I'm betting on R201.
> >
> > Regards,
> > Dave
> >
> > > -----Original Message-----
> > > From: r-390-bounces at mailman.qth.net
> > > [mailto:r-390-bounces at mailman.qth.net]On Behalf Of Graham Baxter
> > > Sent: Wednesday, August 22, 2007 11:39 AM
> > > To: r-390 at mailman.qth.net
> > > Subject: [R-390] Call for measurements - 100dB carrier reading
> > >
> > >
> > > Hi
> > >
> > > I would urge you to test three things. One is the dc 
> resistance of the
> > > AGC amplifier anode coil, Z503, should be no more than 17 ohms and
> > > peaked for max AGC voltage.
> > >
> > > Second, with the multipin connector off the RF deck, check the DC
> > > resistance to ground from pin E, positive probe to 
> ground. This should
> > > be about 1.8 Meg at any voltage up to 250.
> > >
> > > The third thing is to test the dc leakage resistance of 
> the AGC line
> > > in the IF module. Unplug the IF multipin connector. From 
> pin 6 of the
> > > connector on the IF module to ground I see about 50 
> Megohms once all
> > > the capacitors have charged. This is with 250v test 
> potential, again
> > > with the positive potential to ground.
> > >
> > > Any significant leakage here will mess up the AGC, and sadly a
> > > potential culprit is one or more mechanical filters. Lets 
> hope not.
> > >
> > > I have a second EAC here. It is slightly less convenient 
> to check the
> > > levels on that one because it is not near the test 
> equipment. If it
> > > becomes important, let me know and I will check it. However, on
> > > broadcast signals it gives very similar S readings to the one I
> > > measured.
> > >
> > > I have to say, I have no complaints about the standard 
> R-390A metering
> > > and AGC circuit. The zero pot is a bit twitchy but once 
> set mine tend
> > > to stay put. Is it possible you have a fault?
> > >
> > > One item of interest might be the diode load voltage.
> > > With the measured 100mV unmodulated at 1.45MHz its about 14v
> > >
> > > Graham
> > > _____________________________________________________________
> > > 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
> > >
> > _____________________________________________________________
> > 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
> >
> 