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

[David Wise]

Roger,

There are two problems with using a LED (or any semiconductor diode)
as a reference regulator.  One is the 25mV/C temperature coefficient
of the forward-biased PN junction.  This could be overcome by using
an actual Voltage Reference; they are two- or three- terminal devices
in the same package as a diode or transistor, and available down to at
least 1.22V .  I have one, which I was going to put into my Fluke
801 Differential Voltmeter to replace its dead Standard Cell, until
the power transformer shorted; but that's another story.

The other is that a separate reference cripples a feature, namely,
when you overload in MGC the meter reads upscale.  This occurs because
the gains and operating points are arranged such that the 6AK6 overloads
first, and cuts off rather than saturating.  Once the tube is cut off
its current can't drop any further, but on the positive grid swing it
stays in the linear region.  The result is an increase in average
cathode current, and since the meter reference is derived from it, the
meter reads up.  The designers specifically intended this, and that
discourages me from overriding it.

Anyway, if you were trying to reduce the reference impedance as a
way of retaining the normal full-scale reading, there are other
ways to get it done.  R549 affects it strongly; the same 12AU7
that reads 100dB with -12V AGC, reads 100 at -9 instead if you
reduce R549 from 82K back to its R-390 value of 56K.  This gives
you a lot of control over full-scale cal.  I'm looking into this
right now.  I was just hoping I could get away without it, because
it would then be so easy there'd be no excuse not to do it.

Note: V506A Ik idles at about 2.2mA not 22, bet that was a typo.

Gary Gitzen proposed exactly what you lay out below (changing R548
to enable substitute meters).  Pete Williams tried it out with
encouraging results, and I've been refining it, trying to hit
the right values to work for the average receiver.  It's taking
me a long time because I'm slow and I don't get much time at it.
My main contribution, however, is the realization that with any
increase in R548, it becomes necessary to supplement the V504/V508
suppressor-grid diodes with a 1N914; otherwise you idle with
several tenths of a volt of positive AGC and cook your 6DC6.

Regards,
Dave Wise

> -----Original Message-----
> From: r-390-bounces at mailman.qth.net
> [mailto:r-390-bounces at mailman.qth.net]On Behalf Of
> Flowertime01 at wmconnect.com
> Sent: Tuesday, August 28, 2007 12:18 PM
> To: R-390 at mailman.qth.net
> Subject: Re: [R-390] Call for measurements - 100dB carrier reading
> 
> 
> David,
> 
> Could a diode and cap be used to make DC from the 6.3 volt 
> filament line?
> Then a resistor and LED with its voltage drop from this new 
> DC source be used 
> as a reference voltage to R523 the Carr Meter Adjust. I'm 
> thinking about 1.2 
> volts and 10 - 20 ma through the LED as a reference point for 
> one end of the 
> meter circuit. 
> 
> Carrier Meter Current is in parallel with R548 a 27 ohm 
> resistor in the 
> cathode of V506A a 5814. Cathode current is under 22 Ma.
> 
> I am open to us measuring the current of V506A for various 
> Carrier Meter 
> readings.
> 
> Then we could do a new table of substitute meters, with the 
> meter full scale 
> current and the value of resistor we should use with that 
> meter for R548.
> 
> On the other end of the meter we would have a voltage set 
> point source to be 
> used as a reference voltage to R523 the Carr Meter Adjust.
> 
> If we measure and publish the voltage and current of R548 
> with an open meter 
> circuit, we could then pick substitute meters and calculate 
> the value needed 
> for R548 to fit with the new meter current.
> 
> Dave, I think any thing you do is good and useful to those 
> that want to or 
> need to go down this path. I like the stuff we are learning 
> about the AGC 
> circuits. I think this is one of the best threads going in some time.
> 
> Roger L. Ruszkowski AI4NI   </HTML>
> _____________________________________________________________
> 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
>