EMRFD Message Archive 14292

Message Date From Subject
14292 2017-09-12 15:03:18 ks1u@att.net Low Power ssb/cw transmitter 1978.

I know this is not strictly on topic, but I've exhausted other sources.  I've completed a build from the 1978 ARRL Handbook, which begins on page 409.  Everything works except the power amp and I've seen a couple errors in the schematics and/or the parts list.  It involves T1 and T4.  The parts list says they are  both 1:4 transformers, but the schematics list T1 as a 4:1 transformer along with T2 and T3.  I know many of you have this HB lying around and if you could shed any light on these errors it would save me from additional hours of experimenting and possibly destroying difficult to find parts.  Thanks.

14293 2017-09-12 15:47:19 John Marshall Re: Low Power ssb/cw transmitter 1978.
I have a 1977 Handbook with an article named “A Low-Power SSB/CW Transmitter for 80 or 20 Meters”, with a 2N6367 final PA. I bet that’s the same one you’re working on.

I do see that the schematic shows T1 as 4:1 and T4 as 1:4 while the parts list calls them both 4:1. I guess the parts list is nominally in error and should show T4 as 1:4, but the difference between 1:4 and 4:1 is just a matter of which end you’re looking into. Connect them as the schematic shows and you’ll be OK. T1 steps the driver’s collector impedance down to drive the much lower PA base impedance. T4 steps the PA’s low collector impedance up to 50 ohms.

Hope you’re making RF soon,

John, KU4AF
Pittsboro, NC

14294 2017-09-12 19:36:00 ks1u@att.net Re: Low Power ssb/cw transmitter 1978.
      Thanks for confirming what I suspected.  There was never any author listed, and I ran out of options for errata listings.  It's unlikely I'd cook the final but I found an original  nos device and would like to keep it
in working  condition.

14295 2017-09-12 19:57:51 K5ESS Re: Low Power ssb/cw transmitter 1978.

John’s reply was correct but MAYBE I can add a little clarification.  I’m assuming these transformers are around ten bifilar turns wound on a type 43 (or 61) toroid core.  If you wound this with say a green and a red wire you would connect the start end of the green wire with the end of the red wire. Or the opposite: (start end of red with end of green).  Then you have three connections with two single wires and the two (red and green) tied together ( the common).  You then (usually) ground one of the single wires.  If you input a signal into the single wire with output from the common wire it will be a 4 to 1 step-down.  If you input a signal to the common with output from the single wire it will be a 1 to 4 step-up.   

It took me a bit to comprehend this. (But I’m a little slow).




14296 2017-09-13 11:47:08 AD7ZU Re: Low Power ssb/cw transmitter 1978.
just dug out my 1978 HB .. its been a while..
there are 3 transformers between the driver and output (T1, T2, T3).  The notes reference each of these as 4:1, however the winding instructions differ.  T1 is 12t of a single twisted pair which looks to me like a 1:1?  T2 consists of  6t of 2 twisted pairs the phasing of the transformer is not clear from the schematic and the instructions on winding read " the two wires at the ends of each pair are soldered together"  "each pair then comprises one winding" this is confusing and the schematic does not show the additional windings or any phase indication. and finally T3 is described as 4 turns of 4 twisted pairs. 

first thing to check out is what the drive signal on the base of Q13  (2n6367) looks like. 
if the signal at the base of Q13 is not as expected, back up to the preceding xfmr.  
my initial take is that there is likely that there is a phasing problem in the 3 transformers and my guess is the issue is T2 or T3.
looking at the schematic i believe that a single transformer could be implemented that would be simpler but it would likely need to be designed with a compatible core and some impedance calculations. 

just a quick look,


14297 2017-09-13 13:39:48 John Marshall Re: Low Power ssb/cw transmitter 1978.

I agree it would have been better if the author had shown phasing dots for the transformers. If he had, T1 would have a dot at the bottom end of the primary winding, coming from the driver. The secondary would have a dot at the top end, where it connects to the primary winding and forms a center tap. That connection makes the windings “series aiding” and the voltage at the center tap will be half the input voltage and a quarter of the impedance. That’s exactly what the red and green wire example you gave yesterday does.

If the phasing of either winding is reversed, the windings buck each other, the driver sees a short to ground, and no power goes to the PA.

As I read the winding instructions for T2 and T3 they are schematically the same. Using twisted pairs with their ends soldered together just puts two wires in parallel, presumably for lower resistance and greater current carrying capacity. T2 uses two of those “fat wires” together to make a single twisted pair and the result is a 4:1 transformer, same as T1.

T3 just doubles down on T2. Two 4-conductor wires used as a single twisted pair.

The end result is 3 transformers giving a 64:1 impedance ratio. Some illustrations would have been helpful in this project.

As you say, a single transformer might do the job. I wonder if this design might be the result of using cores that were easily available at the time.

John, KU4AF
Pittsboro, NC

14298 2017-09-13 15:44:49 AD7ZU Re: Low Power ssb/cw transmitter 1978.
John et al,
agree,  again if i were troubleshooting the circuit i would first look at the drive level.
and likely you are correct the author had the cores on hand ..and did as most of us do, used what was on hand.

after a closer look i would think something like 8 bifilar turns on FT-37-43 core would come close to working and be far simpler to implement.
if the ratio is still too low an additional winding could be added.


14299 2017-09-13 16:33:12 ks1u@att.net Re: Low Power ssb/cw transmitter 1978.
Thanks for additional comments on this.  It makes me feel better knowing the answers are not immediately obvious, as it had me wondering for quite awhile.  I have RF out of T1, but the amp is not producing any power and I have checked thoroughly the wiring and parts values.  I may have to come up with a new approach to matching the low impedance at the base of the 2n6367. I made the circuit boards to this project back int he 80s from a template supplied by the ARRL.  They sat in a box until a few years ago when I decided they needed to actually produce RF.
14300 2017-09-13 20:28:43 AD7ZU Re: Low Power ssb/cw transmitter 1978.
I still suspect there is an error in the winding description of the transformer arrangement .. its been bugging me all day!! 
just wondered if you were able to get 35ma of quiescent current through the PA after setting the bias?  if so the DC operation of that circuit is probably ok. transformer circuits i have built all required some measurement to get the wiring and phasing correct and it is NEVER obvious!  

hope you get this running, it looks like a very cool project!

good luck


14301 2017-09-14 10:37:16 ks1u@att.net Re: Low Power ssb/cw transmitter 1978.
     Before I headed to work this morning, I pulled the amp board out and applied DC.  It alone drew 440ma at 13.8 VDC, so I need to look at why that is.  
14302 2017-09-14 11:24:20 Jim Strohm Re: Low Power ssb/cw transmitter 1978.
Maybe the question is "What," not "Why."

Is that the idling current for the final(s)?  What does the output (if any) look like on a scope or SA?

If it's idling current, you should see a very low output, and the output, as well as the input current, should vary in a more-or-less linear fashion as you change the bias.

I know, these are obvious symptoms.


14303 2017-09-14 11:41:24 John Marshall Re: Low Power ssb/cw transmitter 1978.
Bias for the PA comes from a series network consisting of a 10 ohm variable resistor, 12 ohm fixed resistor, and a forward biased diode. With the variable set to max and a 13.8 volt supply the bias circuit alone should draw (13.8 - 0.6) / 22 = 0.6A.

The 0.6V drop across the diode provides base bias for the 2N6367 and the high standing current in the bias circuit keeps the bias voltage constant at any drive level. When you do a test like that you should have an ammeter in the collector supply lead, separated from the bias lead, as per the adjustment procedure in the book. A 5 ohm or so current limiting resistor in the collector lead wouldn’t hurt, either. And make sure you have a dummy load connected to the PA. If it oscillates with no load it could let the smoke out in a hurry (don’t ask me how I know).

John, KU4AF
Pittsboro, NC

14304 2017-09-14 14:51:29 AD7ZU Re: Low Power ssb/cw transmitter 1978.
the text describes setting the bias via R9 (10 ohm pot) to 35ma collector current with no drive.
start with R9 at its max resistance position and measure current into only the collector only .. ma meter in series with the collector lead... not the entire pa board.  

The text description and schematic do not match.

looking at the schematic i see a 10 ohm pot in series with a 12 ohm resistor to a forward biased diode to ground..that indicates to me the current through  R8 + R9 is about 600ma @ 13v with diode CR2  full on and essentially in parallel with a 10 ohm  resistor
there is  a note in the text that the voltage drop across R6 should be 4.4v after Ic is set.  i do not see R6 anywhere in the schematic. It's likely a misprint and actually means R8 (12 ohm @ 5W)?  that would be 360ma through R9 + R8 which doesn't add up either.

if the text is referencing R9 and misprinted to R6 then the current in the bias network is 4.4v /10 = 440ma .. which is what you measured.

It would probably be worth while to verify the board matches the schematic, as there are clearly discrepancies in the text to schematic. That may also be an issue with the input  transformer wiring/phasing?

in a few days I will spice that amp circuit if i can find a model for the transistor and check it out..
I don't have any similar parts or i would just breadboard it and measure.

thoughts of the day,

14305 2017-09-14 15:08:59 ks1u@att.net Re: Low Power ssb/cw transmitter 1978.
Thanks again.  I initially built the amp with an MRF433, but found a 2n6367 at the MIT flea market, so if anything "blows up real good", I have a replacement available.  I guess part of me figured there couldn't be more than one error in the article, but I really should look at everything more carefully before any testing.  
14306 2017-09-14 15:50:53 John Marshall Re: Low Power ssb/cw transmitter 1978.
The R6 note references Q12, the driver, rather than the PA, so I think R6 must be Q12’s 33 ohm emitter resistor. 4.4V, 133mA gives a bit over a watt so I hope Q12 has a heatsink.

No idea what the PC layout is but it would have been nice for the schematic to show exactly where to insert the meter for setting Q13’s quiescent current. From the schematic I would probably lift the +12V end of RFC 8 and connect the meter between there and +12V.

John, KU4AF
Pittsboro, NC

14307 2017-09-14 17:00:58 ks1u@att.net Re: Low Power ssb/cw transmitter 1978.
John, I agree about R6 being associated with Q12.  It isn't until after that measurement the author indicates applying power to the final.  I'll probably get some time this weekend to do some more tests.