Advice on an AC/DC circuit.

[KF9CM]

I'm kind of in a quandary on whether I should design my valve type QRP transceiver as an AC/DC circuit (no transformer) or utilize 6 volt filaments and power transformer. Utilizing the AC/DC circuit I get feedback from my fellow hams and I kind of feel like the boy Ralph is the movie "Christmas story" "you'll shoot your eye out".  I've been a ham radio operator since 1965 and electronics technician for 51 years. In that time I've designed and repaired more electronic gear that I can remember. I've managed to stay alive and not electrocute myself in this period of time. Utilizing proper board insulation in a three prong plug, should render the circuit very safe. Let's face it the AA 5 lasted a long time until our society became overly litigious. The real problem I have with the circuit is a long series filament line. This might induce a lot of hum in the overall circuit. Attached is the schematic and board layout. I would appreciate any comments both pro and con on the circuit.

73 de Gary,KF9CM

[kc9pcp]

I'm a little confused about ground and 120-n in your circuit. The cathodes go to ground through a resistor or coil but the power supply neg

[WD5JKO]

I think the voltage doubler needs to be 1/2 wave when fed from an unbalanced source:

http://www.tubecad.com/january2000/page14.html

I made something similar when in high school using a 50C5. I put a two pin polarized cord on it, and due to inconsistency of old house wiring (hot and neutral used interchangeably), I later called my oscillator, 'Killer" 

Jim
WD5JKO

[KF9CM]

The circuit C1-R11 is a RF bypass to ground, but in retrospect I don't really need that if it is utilized in a standard house wiring circuit. It was used in the AA5 if you used it in an environment such as a US Naval ship. In this situation both sides of the line are above ground. If I use this circuit, I'll make neutral and ground the same.

Your right Jim, I'll go back to a half wave doubler.

This is looking more like I should go wit the transformer type circuit.

[KA3EKH]

Transformers are not that hard to come by, I would always use a transformer and six volt tubes. If you get a transformer with a center tap or use a bridge it’s easy to do full wave and messing around with a power supply that’s derived straight from the incoming line got to be hairy at best, don’t care if it’s doubler or not just hairy stuff. Back when I was in school and worked in a TV shop we use to get all those old hot chassis television sets and as a rule you always used an isolation transformer and did a cross ground check before working on the set. Also by using a transformer you can get more B+

[wa3dsp]

If you really wanted to keep it AC/DC in design you could use an isolation transformer just for safety. It could be a real 1:1 isolation transformer or just two transformers back to back - 6.3, 12.6, etc. at an amp or two.

[KF9CM]

Thanks for chiming in guys. I've decided to go with the transformer type and leave it at that. Now what to do with the 50c5's


TNX Gary, KF9CM

[VE3LYX]

For what it is worth I just changed my Hb AM transceiver from AC/Dc to Ps operated. The reason? HMMMMMMMMMMMMMMMMMMMMMMMMMMM.  A 110 volt filamet in a regen rx is a real PITA. At 6.3 volts it is a whole different deal. And I mean it, no other changes. Just got rid of the series filament ac supply (2 50l6s and 2 12sl7s) So in place of the recvr 50l6 I used a 6v6. Direct popin after putting 6 volts on the filaments. I put a 6sl7 in place of the 12sl7. In the TX I replaced the 50l6 (sometimes 50c6)with a 6l6gb and the modulator/ osc tube with a 6bl7 (also both tubes direct popin once I had the 6 volt filament supply.) I run no load voltage on the PS of 247 v. I used a separate filament transformer "liberated" from a junked battery charger. Under heavy load I am down to 200 volts B+with the 6l6.
Hum is now less then my Heath HB10R and non existant on any signal of significant strength. Transmitter power is way up with the 6bl7 osc driving a 6l6gb (almost a poor boys 6146B) . I must confess I fought this idea for a couple of years at least but Wow am I glad I finally caved in! I never expected the other benifits which came for being able to use a different variety of tubes in the same holes. I would like you to know I completely understand your fascination with 50l6s and 50c6s but would encourage you to save them for a quick CW rig (parallel 50c6s or something)
Power supply(my always in use bench supply)was made from discarded but brand new low voltag(24volts) heating industry transformers used in the heating industry. I discovered the companies almost always reuse the old ones and heave the new ones so I got a sack of them just for the asking. I hooked the pri of the first one to the line cord through a switch and soldered another back to back (24 v to 24 volt sec) which gave me 110v (whatever the line voltage is) I hooked another one in parallel to the second one, made sure to phase it properly and then fed a small pi rectifier circuit (diode to 100uf 450v cap to 15 ohm 10 watt resistor to another 100uf 450v cap )both caps neg to ground side of course. And I have no qualms whatever about adding more C after the fact for further hum killing. In fact I left all the AC DC filters and resistor dropping network in the set (visable on the extreme left.) I am now so pleased at the improved performance of this set that I almost set my DX60b station at the curb for pick up.
On the filamant supply I rectified it because I often use ancient tubes (01A , 45s etc) and also installed a 12 v ac take off , a rectified and filtered DC 6 volt output fed through an ancient heater control from a 1920s radio so I can set heater voltage for anthing 6 volts or less. I installed a automotive 2 inch round voltmeter in the circuit so I can see the heater voltage. I have used this deal for a couple of years and am more then happy with it. Cost was less then $20 so it doesnt owe me a dime. In fact yesterday I test fired my T18 Arc5 on it with no problems even on AM.
My Bare Esssentials CW rig (50c6) though is still AC? DC and shall remain so til I die. Reason? No reason son. Its just company policy! Besides the slight hum on my CW note makes it easier for my friends(both of them) to pick me out of the QRM  
Don VE3LYX

[KF9CM]

I've re-drawn the schematic and board for a transformer with 6 volt tubes. It's still one sided with one jumper.

[W3RSW]

Some serious and not so serious but nice for real operation errors.
1.)  you have the plate of the final 6aq5 AF grounded right to the B+ line with its huge electrolytic filter caps.  Need to run B+ to the AF output. trans. first and then to the 6aq5 plate on other side of primary winding.

2.) you show 6vdc to filaments at one point but 12 volts at all the tube sockets. I assume you mean 6vdc.

3.) on the rcvr input you show a 100pf from ant. to grid input of the regen circuit. this should be 3-30 pf variable if taken directly from a high impedance antenna.  If a low impedance antenna I'd wind a proper few turns coupling winding on the regn/grid coil.  You need a way to make the regeneration variable, e.g. variable screen voltage, or variable resistor across coil windings, etc.

4.) and up.  
 Not sure I understand your interstage audio coupling transformer if that is what is intended without standard schematic notation.

You might want to research a lot of older regen receiver circuits to see how to make them much better.
 A little more research and feedback from the guys will help.  I ,of course, may have misinterpreted some of what you've already solved.

[KF9CM]

Thanks for your input Rick,

I am working with Eagle Cad and the maximum size is 100mm X 80mm. The schematic provides the auto-routing for the PC board. I find you have to use a lot of poetic license in order to make work.  
I hope the attached schematic satisfies some of your questions.

[W3RSW]

Ok, understand the pots now.
L3 might not be necessary since you have a 100k resistor behind it.  You'd br better using the 2.5 mh choke in an RF filter right after the .01 coupling cap. And before the audio gain pot ...Along with a couple of 100 pf caps to ground in the traditional PI filter circuit used in regen receivers for many years to prevent the final 6aq5 from amplifying RF.

Also the screen resistor on the xmitter might better be run directly to B+ rather than to the plate. Might also make the screen voltage from a voltage divider so you have a reverse current path for the transmitter's screen.

More stuff I see comes from a lot of my and others' experience, e.g. Stuff like don't put tube radios on circuit boards unless spacing between components is large, components are rated for elevated temperatures, etc.  Point to point wiring with shortest paths and adequate shielding, very solid variable cap mounting necessary to get non warbling CW from a regen.
You're not receiving AM short wave with a CW xmitter attached unless for casual listening. At the least you need electrical band spread, add' l padding caps, and mechanical veneers to achieve decent CW reception over a limited band of, say 100 to 500khz max.

I can think of a lot more, including cabinet/front panel design, shielding and bracing necessary.  Good CW reception from a regen is not trivial.  Your approx. 5 watt electron coupled Osc. Is pretty solid on the transmitter side.  T.G. For crystal control or you'd have a lot of the same problems there.  
But your on track and I wish you well.

[VE3LYX]

Regens with interstage coupling transformers tend to have a special howling problem. I try to avoid them. The one in mine is for speaker output from the audio final.
Best regen design ever I would say from experience is the old twinplex Doerle.(I just built a vintage one for my older brother two weeks ago. Pair of 01A tubes)
Easiest to build is the hartley based regen.
The more complicated they become the less sensitive they tend to be in most cases.
To be honest I didnt realize we were both working on regen based transceivers.
I have built three. One cw /regen transceiver and two AM Regen /transceivers one of which is pictured above. Great projects. Whatever it is when you first fire it do not despair. Just keep rubbin on it and it will pay you back big time.  
I wish you all the best on your project. My hat is off to you. Be careful , these things are adictive!
Don VE3LYX
PS use a good pot for regen control. I use 2 watters always(100K 2W)
BTW
" Utilizing proper board insulation in a three prong plug, should render the circuit very safe"
First off the people who talk about electrocution from ac dc circuits are idiots. 110 volts would have a hard time doing you in unless you are QSOing from the bathtub. And these are the same bohunks who are running a pair of 300Z5s with a gazillion volts on the plate. The concept that because it came through a transformer it is safer is highly questionable at best. These are the same weiners who are whining about having full plate voltage on their earphones. In fact if you touched both bare wires on the old 2000 ohm phones the potential accross them is only 8 volts, tops!
The ONLY reason I went away from ac dc was the hum factor to which a real good working regen is very sensitive to.

[W3RSW]

A wiser man than I,
Named Murphy Bohunk I believe,
Sought not meek differential voltages for fun,
But full groundings from phones alone.

Electroylitics sustained his joy,
Right across his crispy chest.
He's no longer with us, of course.
But enjoying new wings in remorse.

Sorry, urk...  Well at least broken prose...

So if you go the AC/DC route.
Wrap all protected and stout.
You never know who will operate your show,
In ignorance of of your too late shout.

---It takes only a teacup to simulate a bathtub especially in little hands.---

Oh, and don't we all know the brighter of us can make a transformerless voltage quadrupler?
Why stop at a simple doubler for our  6aq5 when a good four times shot will be perfect for our quad PP 6L6 's ?  No transformer current limits for us.