[Collins] 75S3 S-Meter Problem

antqradio at sbcglobal.net antqradio at sbcglobal.net
Tue Jan 3 00:31:44 EST 2017 

JerryThanks for your reply.
I can understand the issues that faced design and production engineering back in 1962 WRT film and composition resistors.  But what was an issue then no longer applies.  I suspect that by the early 1970's, the issues you mention about film resistors were a thing of the past.
Yes, the small bit of inductance caused by the spiral trimming of resistance value adds very little to the total inductance of the resistor.  Actually the added spiral inductance is not much more then an additional 3/8th of an inch or so of resistor lead length.  The tighter tolerance, immunity to resistance change with temperature and voltage, smaller size, lower internal noise and the cost advantage of high volume production make film resistors an obvious choice.  If one applies the CPI calculator to the 1966 cost of the 5 cent carbon composition resistor, the current film resistor cost of 11 cents is a true bargain!
I have always enjoyed your posts and I appreciate all of the help, directly or indirectly, that you have given me over the years.  Happy New Year from Arkansas,Jim

      From: Dr. Gerald N. Johnson <geraldj at netins.net>
 To: collins at mailman.qth.net 
 Sent: Monday, January 2, 2017 9:07 PM
 Subject: Re: [Collins] 75S3 S-Meter Problem
   


On 1/2/2017 2:36 AM, antqradio at sbcglobal.net wrote:
> If carbon composition resistors are better performers then either
> carbon film or metal film resistors then they would still be in high
> volume production and would be cost competitive with film.  But that
> is just not the case.  Carbon Comp, when one can find them, are at
> least six to seven times the cost of carbon film resistors.

In 1963 the situation was drastically different. Carbon composition 
resistors were the lowest cost resistors in use. With wire leads. The 
only visible film resistors were expensive 1% resistors. The IRC molded 
resistors did have a carbon or metal film on a glass tube that was 
completely enclosed in a phenolic casting to make them the same size as 
carbon composition resistors that inhibited heat leaving so they didn't 
act exactly like carbon composition, especially in their failure modes. 
True carbon composition resistors would go up in value when overheated 
and could burn open. IRC went down in value and went closer to a short 
as the heat turned the phenolic casing to carbon.

The Collins Design Standards book said carbon composition resistors 
could be depended on as fuses. The Collins specifications for carbon 
composition resistors said, "QPL except IRC."

  As for
> RF performance, just how many CC resistors will be found in cell
> phones or cable TV boxes? The only niche that CC resistors may have
> any claim on is high power pulse survival.  This is due to the higher
> mass of CC resistors.  To get similar performance from film resistors
> all one needs to do is increase the film resistor wattage rating.

The wire leads of carbon composition resistors limit their UHF 
predictability. The helical cuts in metal or carbon film resistors can 
add some inductance, though the low value film resistors that I measured 
years ago with a Boonton RX Meter showed the same inductance as a solid 
rod the dimensions of the resistor. I had that setup without wire leads.

Pulse ratings of carbon composition resistors was limited by their 
inherent voltage rating that was different for different power, or sizes 
and the internal gaps between the wire ends inside the composition.

I introduced the high power transmitter department to the use of 3 watt 
metal oxide film resistors in 1964. They didn't have much of a coating 
outside the film it was kind of like epoxy. There was concern about 
survivability when overloaded. I don't remember the vendor I used, but I 
took a 10 ohm 3 or 5 watt resistor and checked its resistance with a 5 
digit digital DC resistance bridge. Then I hooked it to a power supply 
and ran enough current to dissipate ten times its power rating. The wire 
leads were hot enough to cut paper and the body was glowing. After a 
while, I let it cool and back on the bridge the resistance had changed 
only in the least significant digit. Its not certain that wasn't from a 
different length of leads in the bridge connection. Those resistors got 
put into the phase discriminators of the 821A-1 and maybe into the 
existing linear high power amps. Those resistors are now known as flame 
proof. I have seen them survive a house fire in a TV set, though the 
heat from one of those resistors may have ignited the plastic TV case.

> This seems to contradict the "burn open as fuses" comment if CC
> resistors are indeed more robust then film resistors of equal wattage
> rating. It is just silly to perpetuate the myth that CC resistors are
> superior or even equal to film resistors.  That battle was lost in
> the 1960's when carbon composition resistor production started to
> decline and the higher volume of film resistors made them less
> expensive.  As far as I know, film resistors have always used ceramic
> cores.  I don't remember ever seeing a film resistor with a glass
> tube unless one is referring to multi megohm precision resistors.
> Regards,Jim

I'm not saying carbon composition resistors are superior to film 
resistors, but in 1963 the film resistor option was probably 10 to 20 
times as expensive as carbon composition resistors because they were 
only available as precision resistors. Digging into a 1966 Allied 
catalog, 1/2 watt 10% carbon composition resistors were 5 cents each and 
1% molded film resistors were 85 cents and up. Prices did vary between 
the several makers. Newark 1962 catalog put 10% carbon comps at 10 to 11 
cents and 1% resistors at $1 to $2.24 each.

Carbon composition resistors do drift with heat, time, and humidity. I'm 
sure the film tube in the IRC was glass, but I haven't opened many of 
them. Many of the early precision film resistors were encased in clear 
glass, but with enough film to mostly hide the core.

Another thing about carbon composition resistors was that the 
manufacturing process was not controllable to make a specific 
resistance. The process resulted in a considerable spread of resistances 
and then they were selected and the color code added. The RETMA standard 
values were chosen so that making resistors with 20% tolerance that even 
with a continuous distribution of resistances there were no rejects. If 
it didn't fit the limit for 47K on the high side it fit the limit for 
56K on the low side. Depending on demand sometimes there was a gap near 
the nominal value for 10% resistors taken from the selection so that 20% 
resistors were closer than 20% error but greater than 10%. The demand 
for 5% was not so great so 10% resistors tended to have some close to 
the nominal value. The standrd values for 5 and 10% resistors also 
allowed selecting resistors from a continuous distribution without rejects.

I'm not sure but its probable that the resistance stability barely 
justified 5% tolerances.

One 32S and KWM-2 application where the original carbon composition 
resistors need to be replaced with metal oxide film power resistors is 
the resistors from the B+ to the cathode of the VOX relay driver tube. 
Every one of the carbon composition resistors has drifted high in value 
from running hot and has lowered the cathode positive bias sometimes 
causing the relay driver to never turn off.

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.
>
> From: Dr. Gerald N. Johnson<geraldj at netins.net> To:
> collins at mailman.qth.net Sent: Sunday, January 1, 2017 7:55 PM
> Subject: Re: [Collins] 75S3 S-Meter Problem
>
> In my schematics the S-meter zero pot for the 75S-1 and 75S-3 is
> R13, 250 ohms. Has a 68 ohm resistor R21 to the cathode pin of V5.
> The other end of the meter is connected to a voltage divider made of
> R16, 47K and R17 100 ohms. The circuit is different in the 75S-3B.
>
> Carbon composition resistors are known to drift high in value with
> age, heat, and humidity so long as they are not IRC. IRC resistors
> were not allowed at Collins because they actually are metal or carbon
> film and don't burn open as fuses as reliably as pure carbon
> composition resistors and because over heating tends to char the
> insulation cast over the glass tube with the film they tend to go
> down with heat and age and to have different RF characteristics too.
> IRC resistors are detectable by the lengthwise mold line on the
> resistor and sometimes slightly rounded ends.
>
> The pot has been known to go bad. Moving its range with a series
> resistor change can work, shunting with a 470 ohm wiper to ground
> will also change its position. It shouldn't be hard to find a
> suitable pot to replace it at Mouser or Digi-Key or Nebraska Surplus.
> It may be fixable with some magic fluids made for pots from Cramolin
> makers of DeOxit.
>
> The S-meter uses the varying DC current of the AVC controlled stage
> as a DC voltmeter for AGC voltage. The RF gain in these receivers
> applies negative voltage to the AVC line to lower the gain manually.
> Many other receivers use a pot in the IF and RF stage cathodes to do
> that.
>
> One other way the s-meter will kick backwards is when an AVC
> controlled tube develops grid emission from cathode material
> splattering onto the control grid. Then the reverse grid current
> makes the grid go positive and that is opposite to the normal AVC
> voltage. There are specialized tube testers for grid emission. When
> an IF tube is replaced, the receiver gain may need adjusting as well
> as the S-meter zero and the peaking of the adjacent IF transformers.
> Usually if its grid emission the receiver will power up with the
> S-meter zeroed and in a minute or less of operation the S-meter will
> drift negative as the grid heats slower than the cathode, purely by
> the heat radiated from the cathode glowing dull red.
>
> I know my 75S-3B had the grid emission problem some 40 years ago and
> I think I've also replaced the S-meter zero pot. The S-line receivers
> get most of their gain in the two 455 kHz IF stages and so work those
> 6BA6 fairly hard. These remote cutoff pentodes have more gain the
> greater the plate current.
>
> 73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.
>
> On 1/1/2017 5:42 PM, Al Parker wrote:
>> Hi Bill, I don't have my skem close by, but I've just been thru
>> this on 2 75S-1's. There is a zero adj. on the top of the chassis,
>> with a series resistor to gnd. It's in the cathode of the S-meter
>> tube ckt. In both my cases the pot was scratchy, and the series
>> resistor had drifted high. I put in some lower res. resistors, and
>> got the pot so at least it was in a better spot. I'm sure someone
>> else here will come up with the res. nrs, but you should be able to
>> find them from my description. 73, cu in the next CX,
>>
>> Al, W8UT www.boatanchors.org www.hammarlund.info
>>
>> "There is nothing -- absolutely nothing -- half so much worth doing
>> as simply messing about in boats" Ratty, to Mole
>>
>> On 1/1/2017 5:39 PM, Bill Stewart wrote:
>>> The s-meter on my 75S3 worked ok after I did some minor repairs
>>> on the rcvr...that's been several years ago. But now, set up for
>>> CW and RF gain at max.(full CW position), the meter pegs to the
>>> left pretty hard and I stopped using the rcvr fearing damage to
>>> the meter. The RF gain has to be turned back CCW a good bit
>>> before the meter will come back, off the LH peg, to the zero
>>> mark. Now I wanna get that rcvr back on line. Before I go into
>>> it, I wonder if anyone can point me toward a possible fix..could
>>> it be just an internal adjustment? Tnx for any guidance. 73 and
>>> HNY de Bill K4JYS
>>> ______________________________________________________________
>>> 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
>>>
>> ______________________________________________________________
>> 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
> ______________________________________________________________
> 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
>
>
>
> ______________________________________________________________
> 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
______________________________________________________________
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

More information about the Collins mailing list