Seeking help with my 6146B CW xmitter please :-)

[KC9KEP]

Hello all,

And thanks again for everyone’s help :-)

Today, I am working on my 1965 ARRL 6146B transmitter again.  Right now, I only want to understand the correct operation of the CW function of the transmitter.  I am performing these adjustments with the transmitter output connected to a power meter and 50 Ohm Heahtkit Cantenna dummy load.

Following is a reproduction of the complete ARRL article.

http://www.bignick.net/6146B_Xmitter/6146B_Xmitter001.jpg
http://www.bignick.net/6146B_Xmitter/6146B_Xmitter002.jpg
http://www.bignick.net/6146B_Xmitter/6146B_Xmitter003.jpg
http://www.bignick.net/6146B_Xmitter/6146B_Xmitter004.jpg
http://www.bignick.net/6146B_Xmitter/6146B_Xmitter005.jpg

My question lies in setting up and “tuning” the transmitter.  Most of this exercise revolves around reading the DRIVE and PLATE current values while adjusting PLATE and LOAD capacitors.

Quotes are from the ARRL Handbook (note, my comments are enclosed within square brackets)

“Note that the 0-5 milliammeter meter reads directly in GRID position, in the PLATE position full-scale deflection represents 250 ma. Cathode current.“

“With 80meter crystal installed (and 50 Ohm dummy load connected) set S2 on TUNE.”

“With C1 and DRIVE control, adjust grid current to 2½ ma. “

[OK, I can adjust to 2½ ma. DRIVE current with no problem.]

“With a key plugged into J2, set S2 on OP.  Set C4 (LOAD capacitor) at three quarters meshed, and switch S4 (meter switch) to read PLATE current.  Watching the meter, close the key and swing C3 (plate capacitor) for a plate-current dip.”

[OK, no problem, the meter dips down to:  2.7 ma. which is really 135 ma. of actual Plate current]

“If the plate current dips below 170 ma. (indicated 3.4 ma.), decrease the capacitance in C4 (LOAD capacitor) and again tune C3 (PLATE capacitance) for resonance.
The objective is to set the loading capacitor so that the plate current dips down to 170 ma.
Check the grid current after the plate circuit is tuned; if it isn’t 2 ½ ma. Correct it by returning C1 (PLATE capacitor).”

(END OF ARRL HANDBOOK QUOTES)

The best I can achieve when maintaining the DRIVE at 2½ ma. and adjusting PLATE and LOAD capacitance is a plate current dip of 2.8 ma. on the meter, or 140 ma of actual Plate current.
At this setting, the transmitter doesn’t always respond well to the keying.  Sometimes I get some whoop and other times, the oscillator won’t start up which momentarily causes a large plate current draw!

BUT .. If I lower the DRIVE current to 1 ma., I can achieve 3.4 ma. of Plate current dip on the meter (170 mA of actual Plate current) right on the nose.

The transmitter behaves much better with the DRIVE current set to 1 ma.; the oscillator always starts right up, and the power out looks just as good as when the DRIVE is set to 2½ ma. (80 Watts)

My questions:

Why do you suppose the transmitter is behaving in this fashion?

Is it OK to run the 6146B when the transmitter is setup in this manner?  (I would think that less DRIVE current is better for 6146B tube life?)

73

--KC9KEP

[KA0HCP]

Try adding more load.  That should make it more stable.

[WQ9E]

How precise are the resistors you used for the 470 and 10 ohm units in the 6146B cathode circuit?  They will greatly affect the accuracy of the metering.  The same question applies to the meter itself.  First make sure you have a 1% (or at least 5%) resistor in the 10 ohm position and the transmitter meter switch in the grid position then measure the voltage drop across the 10 ohm resistor with a good external DMM or VOM and calculate the actual cathode current to make sure it agrees with the transmitter meter.

Make the same measurement/calculation across the 470 ohm metering resistor in the grid circuit.

Too much grid drive is bad for a 6146 tube but 1 mil is too low for good Class C efficiency.

[KC2ZFA]

C1 must always be tuned to resonance (peak 6146 grid current). Drive should then be backed down if grid current with C1 at resonance is higher than 2.5 mA.

Also, I would have put a low-C trimmer where the 5 pF is on the grid of the 6AG7 and would have adjusted it so that the oscillator always starts when C1 is at resonance and the 6146 is on (thus loading the oscillator).

[KC9KEP]

Thank you all for your replies!

KC2ZFA - I can offer some additional information without returning to my workshop ..

When tuning C1 to peak grid current, the meter is easily 'pegged' even with the drive pot set for
minimum drive.  The ARRL article says that this scenario "is likely" and to "de-tune C1
preferably to the low capacitance side of resonance" if this happens.

I have been forced to use this method to adjust the drive "down" far enough.

I will be experimenting more tomorrow .. thx again!

73

--KC9KEP

[N8ETQ]

   Bias voltage?

/Dan

[KA2DZT]

170 ma of cathode current seems a bit much for a single 6146.  Probably shouldn't be much over 135-140 ma.  You do need accurate metering to know what's really happening.

You should have at least 2.5 mils of grid drive.  On AM, I run my 6146B at 3 plus mils grid drive and about 135 mils of cathode current.

Fred

[KK4YY]

I'd have to agree with Rodger. Check that the metering is accurate first, as he suggests. That might be the whole problem right there.


There may be an issue with neutralization, you can check that again too.


Try loading up on a different band and see if you get the same result. Try a different xtal or different tubes if you have them on hand. These easy things may point you to the problem. Also, use of a 60 watt incandescent light bulb as a dummy load (during initial tuning) gives a quick visual indication of output without having to look at a power meter while you're tuning and trying to watch other meters.


I built the very same transmitter while in High School in 1976. My first! I later swapped the 6146B to a 2E26 and drove a pair of 813's with it.


That's going to be a solid rig for you after you work out the bugs. Good luck!


Don


BTW, that's some pretty fine looking construction on the projects on your webpages and Youtube videos. I wish mine looked that good. I like your use of octal relay sockets for tubes on this one... http://www.youtube.com/watch?v=F_0DDkXPd60 ...and the quote at the end, too funny!

[KC9KEP]

Thank you again everyone for all your replies!!

(And kind words about my construction .. blush :-)

I'm having difficulty getting bench time in order to work on the project but I wanted to let
everyone know that I have been reading the replies and haven't given up :-)

Yesterday, I had ordered some 1% resistors to replace the existing ones in the metering circuit.

I will be reporting back soon!

73

--KC9KEP

[KC9KEP]

Thanks, everyone for your replies.

I've verified the current readings, meter, and replaced the resistors in the meter
circuit with 1 percents, so I think that readings are true.

One thing that one Elmer brought up .. the basic circuit has been changed
in subsequent ARRL publications.  I've already made some "improvements"
by changing the circuit to reflect these ARRL improvements.

They made changes such as:
Moved the 150pF cap from C1 circuit to return directly to ground and not to the
1000pF cap from the cathode of the 6146.
The same 150 pF was changed to a 1500 pF.
A 10 Ohm resistor was added into the grid circuit of the 6146.
Moved the parasitic suppressor to connect directly to the 6146 plate and
not between the DC blocking capacitor and the plate loading capacitor.

I think they weren't quite finished documenting this design?

thx