homebrew am tx from drive in

[N1BCG]

I redrew the circuit slightly based on the "8BK" pin arrangement and it seems as though a 6SG7 tube would work. Note that pin #1 is "shield" on this tube.

On the master diagram, the 6SN7 on the lower right should show the 47k resistor connected directly to pin 5 (plate) just like the other half is wired. The 0.047 coupling cap should come off that connection.

The plate circuits (pin 3) of the 6V6 tubes should be double checked. It looks like you have all the parts shown, but the position of the 0.001 capacitors is dubious.

Also, on the oscillator, the B+ lead should connect to the intersection of the screen dropping resistor and the RFC (unless I was seeing it wrong).

You can probably label the rectifier as 5U4 with confidence given its pinout.

Getting there!

[kd1nw]

Thanks! I have a 6SG7 but Id have to pull it from a signal generator. Can I use a 6AC7 or a 6AB7 ? They look similar and the builder tied Pins 3 and 5 together at the socket. I guess 6AC7 and 6AB7 take .45 amp fil and the 6SG7 takes .3.

Transformer provides 6.3v at 4.5 amps. All the 6 volts filaments add up to 2.85 amps, so it seems like it shoul dbe okay.

[Opcom]

Is the leftmost 6SN7 section driving the rightmost 6V6 grid? It didn't look right in the previous drawing.

[kd1nw]

Thanks for catching it, no it isn't. I made a number of errors and an omission. It should be fixed in the schematic dated 20150418.

[N1BCG]

Can I use a 6AC7 or a 6AB7 ? They look similar and the builder tied Pins 3 and 5 together at the socket. I guess 6AC7 and 6AB7 take .45 amp fil and the 6SG7 takes .3.

You probably could! The 6AB7 and 6AC7 have an "8N" pin out which, as you pointed out, should also work because of the suppressor (3) being externally tied to the cathode (5).

BTW, I think the plate and screen resistors on the 6SG7/6AB7/6AC7 are reversed. The 470k should be on the plate and the 2M on the screen (or does it look different in person?)

[WA1LGQ]

I was looking at that thing myself at the museum fest. It was tempting for the iron and chassis. Lots of potential there. Have fun!
Larry

[kd1nw]

You're right Clark, I had them reversed on the schematic. Thanks for the catch. I'll post the fix later.

And yes Larry, I've been have a great time with this thing. Perfect for getting a kickstart after the long winter...

I have the all the tubes now.. getting so close it might be time to hang up a bunch of sheets at the end of the driveway, grab a projector, make some popcorn and head for the car...

[DMOD]

So this is a series modulated oscillator?

Phil - AC0OB

[w8khk]

After looking at the first photo you posted, I would like to make a safety suggestion.  It appears the white wire from the power cord is attached to the fuse.  Normally, the white is the neutral, black is hot, and green is the safety ground.  The hot should always be fused, the neutral should never be fused.  If the fuse expires, this could leave you with a hot chassis, posing a shock and safety hazard.

First, I would suggest you verify that the neutral pin on the plug is connected to the white wire, and the hot is connected to the black.  This should be the case if the plug is molded to the cable, but if an aftermarket plug was attached to the line cord, there could be an error here too.  After verifying the correct plug-to-wire connections, exchange the wires under the chassis so that the black wire is connected to the fuse.  I hope you find this information helpful.  Good luck on your project.

[Opcom]

So this is a series modulated oscillator?

Phil - AC0OB

It looks modulated by a class A 6V6 audio amp in parallel with the 6L6 oscillator. The tubes both get their plate supply through the same 10H choke. Just thinking about that, it is a lot to ask a 6V6 to modulate a 6L6, would it make as much sense to use a smaller oscillator tube there?

[kd1nw]

Thanks Rick, I saw that too when I started tracing the circuit. Thanks for the reminder and advice.

And, you guys know more than me but I was thinking the 6L6 was screen modulated. Im guessing the carrier should be reduced compared to plate modulation, 5-10 watts carrier? If he used a plain 6L6 maybe a 6V6 would be enough in this case? 

[DMOD]

I think I would just modulate the plate and provide either zener or gas tube regulation for the screen.

[N1BCG]

Unless you modify the circuit, the oscillator tube will need to have a "7AC" pin out. Both the 6L6 and 6V6 meet that requirement and will probably provide the same output power for the same input power. The circuit shows that the plate and screen are both modulated, which is common for linearity, although modulating an oscillator is also going to result in an FM component.

I doubt this is going to generate much power given the untuned output and single 6V6 Heising modulator but you can always make changes at any time. For example, adding a matching circuit for more output or a resistor in the oscillator plate circuit for more modulation.

If you have all the tubes then you're in a good position to give it a try. A Variac would be helpful for bringing it up slowly and a frequency counter will quickly identify the operating frequency.

Sure seems like a lot of audio stages, doesn't it?

[kd1nw]

yeah i was wondering why so many too. anyway, i brought it up on a variac without any drama. all the filaments light up, all the transformers stay cool. The multisection can is cool.  But, it isnt working right. for awhile the meter looked like it was working and following the music. But its not working now. I plugged some headphones into the monitor jack and I can hear the music i was playing but its weak and scratchy / distorted. i was never able to find its signal  anywhere on the AM band. I checked above and below the band too. Ill look into it more later. Thanks again everyone

[N1BCG]

What are you getting for B+ from the power supply? Caps can go bad, particularly electrolytics, so it would be good to parallel each one with known good ones of the same value to see if the performance changes.

[kd1nw]

I measured B+ of the tubes, I'll measure the other voltages later but this is what I got

6L6                      300
6AC7                   235
6SN7                   (pin2 360)(pin5 248)
6V6GT (monitor) 390-400
6V6GTA (to 6L6)  335

i think there is bad oscillation going on almost like regen. in the headphones you can hear on and off hiss / silence

[Opcom]

Do you have audio on the modulator 6V6 grid and plate?

A scope should show oscillation, have you looked for it around the 6L6 cathode and plate?

[N1BCG]

Assuming this worked properly at one point and that the tubes that are in now are correct, three items come to mind that would cause a weak and scratchy sound:

1) Check the 100uF cathode bypass cap across the 2k resistor on the 6V6 as well as the 100uF cathode bypass cap across the 18k resistor on the 6SG7. You'll lose a lot of gain if either is bad.

2) The audio output transformer "works against" the 20uF capacitor in the power supply. Another cap to check by paralleling a known good one.

3) The input may be overdriven resulting in distortion. With so many stages of amplification, it might not take much input level.

[KA2DZT]

Not too sure about the 18K cathode resistor.  Saw that when he first posted the schematic, didn't think it was correct then and still don't.

Fred

[kd1nw]

Hi Fred, thanks for pointing it out. What value should it be? The resistor looks marked Brown Grey Orange and I measure about 19 kOhms.
Clark, yes the distorted audio was me overdriving it. I bypassed the various electrolytics with know good and it didn't stop the oscillation, maybe some hum was lowered. I pulled the 6L6GC and it stopped oscillating out of control. The audio actually sounds pretty good through the spreaker mon jack. So next up is to do what Patrick suggested and start looking what is going on in the 6L6GC area. Hopefully my scope is working..  thanks everyone.

[KA2DZT]

Well it seems 18K may be right or at least that's what the original builder thought.  It a simple Class A low level audio amp, usually the cathode resistor wouldn't be that high.  Could be the cause of some distortion you mentioned.  I'll take a look at the 6SG7 charts to see what it should be.

I took another look at the schematic,  I see a lot of resistors that don't look right.  Very high cathode resistors for the 6SN7.  The 470K B+ resistor feeding the audio stages seems way to high.  The 2m plate load resistor on the 6SG7 seems high, 470K screen resistor??  I not seeing the purpose of 1/2 of the 6SN7 in the circuit.  6SN7s can draw 10ma per unit.  Do the math, going through a 470K resistor.  Seems the builder was trying to limit the overall current draw using very high value resistors for some reason, makes no sense.  Also the 100mfd cathode bypass cap across the 18K cathode resistor is way to much.  A few mfds will bypass 18K.

Seems the original builder lacked some designing skills or I'm missing something here.

I'm afraid to look close at the rest of this jigsaw puzzle, but you're having fun working on it and that's all that counts.

Fred

[KA2DZT]

OK. I took a chance and looked at the two 6V6s.  Very strange circuit.  You have two negative feedbacks going to each grid from the plate.  Problem is there is a .05 cap to ground on one plate.  But, you have positive feedback from that plate to the grid of the first 6V6.  The wiper on the 2m pot goes through WHAT to ground, another thing that doesn't make any sense.  The 10hy choke is used as a plate load for the second 6V6 while the speaker xfmr is the plate load for the first 6V6.  The second 6V6 has no signal to its grid.  The only circuit that looks almost right is the 6L6 osc, except for the 5K grid resistor and there is no ground return to the power supply anywhere.  The oscillator is a ONE terminal black box, usually one terminal black boxes don't do anything.

Fred

Maybe there is a ground return at the output connector.  What size is the cap that is across the output connector??

[Pete, WA2CWA]

If I was doing this, I would ignore the entire audio circuitry for now and just concentrate on getting the oscillator to work. Once you get the oscillator to get some RF into a wire so that a receiver can pick it up, then go back and diddle with the audio stages. Why analyze the heck out of the audio stages now when the oscillator isn't running.

[Opcom]

I thought he original 6SN7 input tube drove the modulator grid.

[kd1nw]

Hi Fred, the 2M pot goes through a vu meter to ground. I forgot to draw the ground connections around the 6L6, there is a ground connection at the output connector and the cap across the output is 1000 pF

Hi Patrick, I made a number of errors when drawing the schematic. The one attached to the thread that is dated 20150418 has it as 6SG7 but I have a 6AC7 in there right now. I have 6SG7, 6AC7 and a 6AB7 available, they all worked the same when I tried them. There's no way a 6SN7 could be in that socket, it isn't wired for it. Plus when I pulled the unmarked twin triode from that socket it only had 5 pins still attached. Looks like someone through a broken 6SN7 in there and labeled it..

Hi Pete, I've started to concentrate only on the osc circuit. I found out that my scope isn't working so I've been studying 6L6 oscillator circuits to see if I can figure out what is wrong with the one I have.

Thanks