[R-390] 2 diode agc ssb mod c/l meter linearity

[Larry H]

Hi Jacques,  Thanks for keeping me on the straight and narrow about the
forward voltage of diodes.

To all,  I ran some measurements on germanium and silicone diodes at low
currents and find that the silicone ones are pretty close to each other and
the germanium ones are lower, but also close to each other.
Silicone diodes such as 1n4004, 1n1694, 1n4148, 1n543 and 1n540 have about
.31 Vf at 3.3 uA, .36 Vf at 12 uA, .45 Vf at 100 uA, and .49 Vf at 270 uA.
So, any rectifier diode for 200V at .25A or better should be fine.  IE: the
1n4004 should work.

Sorry for the confusion.

Regards, Larry

On Fri, Aug 27, 2021 at 11:14 PM Larry H <larry41gm at gmail.com> wrote:

> Hi again, Jacques.  Sorry I did not have more time to properly respond to
> your thoughtful post earlier.
>
> In addition to what I said, the main factor in picking out a diode was
> it's forward voltage.  I found a modern diode stth102 at Digikey that met
> the requirements, *except* it's forward voltage is .68 Vf instead of .6
> Vf.  I did not obtain one for testing as I had a .6 forward V that I used.
> So, the stth102 may be ok, but I don't know.  It's a 200 V - 1 A, which
> meets the minimum voltage and the amperage is much more than necessary.
>
> I can say that I tried the 1N4004 and its Vf was not quite low enough.
> It's Vf is 1.1 V.
>
> Unfortunately, I looked a lot online and most of what I found in new
> diodes that have a low Vf were around 70 V and lower.  Since the diode must
> be 200 V, there are no correct new ones available.
>
> The ones I mentioned in my announcement are available to purchase, but are
> not cheap.  The 1N1695 is available from Electronic Surplus and ePay.
>
> However, there is an exception - if you have the 'moment of silence' (MOS)
> fix on your agc speed switch, then a lower voltage and current diode can be
> used.    A *big caution* goes with doing this because if you install the
> lower voltage diode and swap the IF deck to a unit that does not have the
> 'MOS' fix on it, it won't last very long -1 switch will do it.  In the case
> where you have the MOS fix on, you can use a 1N63, which is readily
> available.
>
> Happy listening.
>
> Regards, Larry
>
> On Fri, Aug 27, 2021 at 4:29 PM Larry H <larry41gm at gmail.com> wrote:
>
>> Hi Jacques, The 8 M ohm res is not critical and can be 7.5 to 8.5 M
>> ohms.  The diode was chosen for it's low forward voltage, .6v or less.  And
>> it needs to stand up to the surge when the agc speed is  switched.  Most
>> standard rectifier diodes have a forward voltage of 1.2 V, such as the
>> 1n4004.  As for a transistor, it could work if it meets the 3 criteria.
>>
>> Sorry for the rushed answers, I have a Honey Do chore to get  to  asap.
>>
>> Regards, Larry
>>
>> On Fri, Aug 27, 2021 at 7:28 AM Jacques Fortin <jacques.f at videotron.ca>
>> wrote:
>>
>>> Hello Larry,
>>>
>>> Thank you again for providing improvement ideas for our beloved
>>> receivers.
>>> I have two questions, however.
>>> I understand the purpose of the Mod. 3: there is a similar arrangement
>>> used
>>> in some other receivers, the Racal RA17C12, for example.
>>> Question 1: a 8Meg. ohms resistor is not a part of "standard" value, but
>>> 8.2Meg. ohms is.
>>> I then presume that there will be no difference in the Mod. 3
>>> implementation
>>> if a 8.2Meg. ohms part is used. Please confirm.
>>> Question 2: did you recommend using those 200 PIV, 400ma rectifiers for a
>>> special reason, like a super-low reverse leakage ??
>>> They are all Silicon rectifiers of old manufacture, so they will be hard
>>> to
>>> find.
>>> Could a "good old" 1N4004 be used instead ???
>>> Thinking about another trick, does the B-C junction of a high voltage
>>> silicon transistor can be used (like a 2N3440, or similar) ??
>>>
>>> 73, Jacques, VE2JFE
>>>
>>> -----Message d'origine-----
>>> De : r-390-bounces at mailman.qth.net <r-390-bounces at mailman.qth.net> De la
>>> part de Larry H
>>> Envoyé : 26 août 2021 23:43
>>> À : R-390 Forum <r-390 at mailman.qth.net>
>>> Objet : [R-390] 2 diode agc ssb mod c/l meter linearity
>>>
>>> Hi all,  The c/l meter linearity may be skewed if you have the Lankford 2
>>> diode agc mod on for ssb reception.  The reason is that the original
>>> 'suppressor grid' clamping at the junction of R544, R546 and R547 is not
>>> effective enough to overcome the increased voltage from internal noise,
>>> an
>>> RF signal or the antenna noise.
>>>
>>> The clamping circuit provided by this fix (Mod 3), improves the situation
>>> considerably, but does not completely solve it for a high noise antenna.
>>> This problem has been there all along, but with the original design, it
>>> was
>>> not very noticable. With Mod 2 and Mod 3 on, it looks like the c/l meter
>>> noise tolerance operation is a little better than with none of these AGC
>>> mods on. Mod 3 also improves the c/l meter linearity close to what it was
>>> without these mods on. How does this affect the c/l meter linearity, you
>>> say? Well, with mod 3 on, the meter 0 set point is at a better place.
>>> This
>>> is what improves the linearity. The diode needs to be 200 volts at 400 ma
>>> with a .6 volt forward drop or less because of the high voltage pulses
>>> created when switching the AGC speed.  There were a number of diodes made
>>> that meet the requirements in the 60's and 70's, but very few available
>>> now.
>>> Ones I found are: 1N1693, 1N1694, 1N1695, or 1N2092.
>>>
>>> However, if you have the 'moment of silence' fix on for the AGC speed,
>>> then
>>> a lower voltage one (such as a 1N63) can be used.
>>>
>>> I have attached the updated schematic for this fix.  If you don't
>>> receive it
>>> from the list server and would like a copy, please let me know.
>>>
>>> Regards, Larry
>>>
>>>