"Li'l 7" AM Transmitter
 

A few weeks ago I completed a "Li'l 7" transmitter from the schematic and instructions on Phil's Old Radios, http://antiqueradio.org/transmitter.htm
https://lh4.googleusercontent.com/-u...%252520010.jpg
I designed my variant of the Li'l 7 so that the whole chassis lifts from the enclosure without even removing any knobs.
https://lh3.googleusercontent.com/-q...%252520020.JPG
It looks great...but it doesn't work! I've gone over every connection and every wire three times, tested continuity wherever I could, tracked the input from my CD player to the tube, and tried multiple frequencies on two different radios. I also learned that it's not a good idea to discharge a capacitor into your headphones.

The only variation from the original design is that I used two 68 mfd caps in place of the specified 22 mfd caps, but I don't think that should make much difference. I'm hoping there's something obvious that I've overlooked.

The link above to Phil's website shows the schematic and description. I don't know what to do next. Any suggestions?

- Winky

Are you getting B+ voltage?

The most likely problem may be that you have to swap the connections on L1, but only the primary OR the secondary. Get it wrong and the oscillation won't happen.

From Phil's website: http://antiqueradio.org/art/schemat.gif

Check that you wired to the correct tube socket pins. You might have shifted by one by mistake. And I'm sure you know this, but tube pinout diagrams are viewed from the bottom of the socket, where you'd be looking when wiring the socket below the chassis, unlike IC pinout diagrams.

I am getting proper voltages as far as I can determine, I have rechecked the L1 contacts and the 117L7 pins numerous times. In fact, I let it sit two weeks and then rechecked everything again before I decided to post this question. This weekend I'll do it again and chart voltage readings. I'll probably record a lot of worthless values, but maybe I'll hit on something significant.
Thanks.

- Winky.

Good suggestion. I built mine nearly fifteen years ago and I don't know exactly what you get when you order that part today (or exactly how it's marked). There aren't many parts in this rudimentary project, so the solution should be straightforward.

Nice job of construction, by the way. Once you get your transmitter working, send me a PM if you would like me to include a photo or two at the end of that article.

Phil Nelson
http://antiqueradio.org/index.html

P.S. As a last resort when debugging a project, I put away the original schematic and draw a new one by hand, looking only at what I have built. This forces you to look at things with fresh eyes. It's still maddeningly easy (for me, at least) to overlook a small mistake.

Freshly Drawn Schematic
 

Rechecked every connection against Phil's schematic with special attention to the tube and the RF coil. The coil is exactly as originally described by Phil, and it included a diagram showing the red spot and the terminal numbering. Recorded voltages at pins 1-8 and rechecked the continuity across every accessible connection. Also verified that each component is the correct value. Then I followed Phil's suggestion--made my own schematic directly from my own appliance. This is what I came up with:

https://lh6.googleusercontent.com/-w...%252520030.jpg

Variations from original schematic are (1) C3 and C4 68mfd, (2) fuse, and (3) rotary switch for C6a/b/c.

Any glaring errors?

- Winky

I'd find it helpful to draw the internal parts of the tube on the diagram. It's easier to trace that way than to stop and think: hmm..pin 5 goes to ..what? Forgive me for this really obvious suggestion, but do both parts of the tube test good?

I know you've gone over the wiring time after time, but something I've found helpful is to trace each circuit and then mark them off one by one with a highlighter on a copy of the schematic. I have a 1920's TRF on the bench right now that somebody had got into and this is helping me get the wiring back to original.

Wink, as a helpful suggestion, a schematic should always show the internal elements of a tube (or other active device like a transistor as the case may be). A quick Google under 'Vacuum tube symbols' and 'Electronic symbols' would probably yield a lot of good info.

BTW, that's a gorgeous resto you did on the Stromberg Carlson radio. :)

If i may, being a cantankerous old fuddyduddy, i'd like to air a perennial gripe (again) about vacuum tube nomenclature. That's in the usage of the terms "filament" and "heater". Back in the day, for a techie to use those terms incorrectly earned a demerit and severe reprimand.:nono: :bash:
A tube with an indirectly heated cathode uses a heater, not a 'filament'. Whereas when the filament is the cathode and the direct emitter of electrons, is simply called the filament. Off soapbox now. :o

At the very least, L1 has to be drawn incorrectly in the above.. From the info provided, I'd suspect one winding goes from pin 1 to 2, the other 3 to 4. Not having the 117L7 diagram handy, I'm guessing on the interconnects. It looks (from the voltage readings) like it's not conducting.. (ref to the 8V "drop" across the 10K filter resistor.. (Less than 1mA) I'm also assuming that the heater is working? (Either visual inspection, or does the tube envelope heat up after a bit?) I also agree with the above suggestion, does the tube test good?

I'm working on it.
 

I apologize for my gaffe in failing the schematize the tube. I was going to do that, but that's where I would likely make an error in the schematic. I can figure out cathodes, heaters, plates, but I only have vague concepts of screens, grids, and bias. I give it a try, though. My hope was that we could assume that Phil's original diagram is accurate and see if my schematic matched his pin numbers.

For Reece, I like your method of tracing over the schematic. In fact, you can see below how I did that myself. If the adage is true, that "Great minds think alike," then you must have a truly staggering intellect.
https://lh6.googleusercontent.com/-F...%252520032.JPG

The tube is new from AES, and I assume (I think I've made too many assumptions) it's OK. Unfortunately I misplaced my tube tester, but I might have just left it in my pants pocket, so I'll check through the laundry later. Sorry, I don't mean to be facetious. Earlier this week I visited my usual 1960s-era electronic supply with the tubes from the Atwater Kent 37 and neglected to bring the 117L. However, after testing the first five tubes "bad," Alan the proprietor said, "I think my tester's bad--not the tubes." I have one other place to try when it opens next week.

I've been trying to cultivate a supportive network, but this isn't exactly like Chicago. I drove 40 miles just to have Alan tell me his tube tester was bad. I met a guy who could help me with this stuff, and I was looking forward to working with him, but he didn't answer my last email some months ago. There's a VK member in my area, and he hasn't answered my email either. Is it my breath?

Well, I'm done venting. Thanks to everyone for the tolerant, patient assistance. I still have to digest teevee's comments, so I'll get to work sorting out the mysteries of the vacuum tube.
- Winky

There has to be an error, else the B+ would get shorted to ground thru the coil. Manufacturers of such coils I think did not have an industry standard for what the individual lug connections were to be. Meaning that one coil will have a diferent pattern than another. One qay to check that you have the two windings correctly identified: the winding with the higher resistance will be the resonant inductance, with the capacitor, sets the frequency, and have a voltage close to ground. The other winding, with lower resistance, is the feedback winding, and will have B+ voltage on it. Assuming that this is correct, then The phasing of this winding is important. Try swapping the conections on this winding.

Looks like you are seeing rectifier tube action, so the tube is probably good.

"Talk about bad breath, but you could knock a buzzard off a sh** wagon" George Carlin. Don't worry about such, some of us technical types are not the greatest with the social skills.

"Knock a buzzard off a sh**wagon." Gotta remember that gem.

:lmao:

Enhanced Schematic
 

I added the innards of the tube to the schematic. I do get 125V from pin 2 to pin 7, and the singed hair on my knuckles indicates that the heater is working. Sometimes I can learn something from tube data sheets, so I downloaded RCA and Tung-Sol sheets, but I've gained little from that. I keep jumping from one question to the next, and I thought it best to post the revised schematic now instead of waiting for a revelation.
https://lh4.googleusercontent.com/-m...%252520035.jpg

There seems to be a consensus developing that I've mixed up the coil terminals. It'll take me a while to follow up on the latest advice.

Gotta cook dinner now. Thanks.
- Winky.

Hmm. Assuming the circuit is wired as shown, the first thing that jumps out is that the B+ supply is shorted to ground through the primary of the coil (lugs 2 and 3). Assuming that that coil has continuity (not open), resistor R3 would be really smokin' (assuming it's 1/2 watt rating). What voltages do you have at the top of filter caps C3 and C4?

Li'l 7 Schematic, Version 3.0
 

I corrected the schematic of the RF coil--I had placed the numbers arbitrarily. Now I think they'll make sense. Following is the P-C70-RF instruction sheet, the latest revision of my schematic, and a picture of the coil showing how the terminal numbers were determined.

https://lh4.googleusercontent.com/-Q...%252520038.jpg

https://lh4.googleusercontent.com/-_...%252520039.jpg

https://lh4.googleusercontent.com/-A...%252520036.jpg

Voltage and resistance data for the coil:
Terminal 1 to chassis ground: 0V
Terminal 2 to chassis ground: 0V
Terminal 3 to chassis ground: -156V
Terminals 4 to chassis ground: -156V
Resistance terminals 1-2: 10 ohms
Resistance terminals 3-4: 78 ohms (Curiously, when I use the continuity tester between 3 and 4
I get an "open" reading or an occasional quick flash of "short.")

Voltage at C3: -156V
Voltage at C4: -165V

I see how my error misled everyone. If I've made any other errors in the schematic, it's clearly the fault of the United States Congress and the previous Presidential administration.

I considered trying the connection swapping, but I found it too confusing to follow the advice based on the erroneous schematic.

With sincere appreciation,
- Winky

First you need to address that intermittent between lugs 3 and 4. The break is most likely right where the coil wire joins the lug.

Lacking a tube tester, you can still do some basic tests. You know the tube lights up, so that's good. Pull the tube and using your ohmmeter, make sure that there are no shorts between tube elements that should not be touching. There probably aren't, but....you know how that goes. And then we can all rest more easily...

I wanted to double check that the tube basing (pinout) shown in the schematic is correct, and indeed it is, as verified here.. http://www.bunkerofdoom.com/tubes/sy...1949/117L7.GIF

After you've gotten that break in the coil fixed, there appears to be a couple of errors in the schematic. It shows pin 8 (cathode) of the pentode section going to ground. I don't believe this is correct. Try ungrounding pin 8.

Now connect lug#1 of the coil to pin 8 (after disconnecting lug#1 from the junction of C5/C6). Be sure lug# 2 connects to ground.

Connect the jct. of C5/C6 to lug# 4 of the coil (lug# 4 should already be connected to pin 3 {plate} of the tube, and to C1 which feeds the antenna).

If the circuit doesn't oscillate, try reversing polarity of either the primary or secondary of the coil (but not both).

Crosslinking the Diurnal Framistat
 

Thanks. I see six individual things to check out. I have to ask about the grounding of pin 8 first. That's something that actually didn't make sense to me, but I left it alone because it seemed to be consistent with source schematic:
https://lh6.googleusercontent.com/-Y...7%25252044.jpg
So I'll break the circuit as above?
I'll be back later after I try to address the other issues.

- Winky

No, you want to leave the 47K resistor (R1) connected to ground, and break between pin 8 and ground.

Big "oops." That should have been obvious to me. I'll have to retreat and repeat that last six steps later tonight.
- Winky

You want to double-check that C5 connects from pin 4 (control grid) to the jct. with C6 (which also connects to lug# 4 of the coil and pin 3 of the tube.

C5 provides the feedback to sustain oscillation so it has to be in place. Also double check that the 47K resistor R1 goes from pin 4 to ground.

Maybe post the schematic of the most recent iteration when you get it done.

:smoke:

Something else that bothers me a little is that I would not have though a pentode would work with its G2 at ground voltage. It's usually around 70 to 100% of the plate voltage. Being at 9V might cut the tube off? Or it's a way to keep the amount of RF power low?

Yeah, i thought about that too. But figured that the cathode being at `near` ground (except for the coil resistance) probably mitigates for the absent G2 voltage a little. Then with both grids at the same potential, the tube behaves electrically more like it's a triode running at lower power than a full-up pentode. Or at least that sounds sorta plausible. :nerd::D

Got to looking at the 117L7's spec sheet posted earlier, and it's classed as a beam power amplifier. If that's correct, it's a tetrode with beam-forming plates and there would be no suppressor grid (G3) between G2 and the plate. So with G1 and G2 "strapped" at the same electrical potential, the tube would be running as a "beam triode". :tongue:

Li'l 7 Version 4.x
 

The last two posts are way over my head, but I'll point out that G1 and G2 are -.8V and -.9V.

In addition to curing cancer and establishing world peace, this is what I've done this evening.

Tube:
- Pin layout is consistent with Tung-Sol, RCA, bunkerofdoom, radiomuseum diagrams.
- No continuity between any pins except the heater, 2 & 7.

Coil:
- I don't think there's a break in the coil. Lugs 1-2 read a consistent 10 ohms, lugs 3-4 read a consistent 78 ohms. Checked the continuity tester through a couple of resistors, and it reads "open" at 75 ohms and above. Visually all connections appear solid.

Changes in circuit:
- Ungrounded pin 8.
- Disconnected junction C5/C6 from lug 1.
- Connected pin 8 to lug 1.
- Connected junction C5/C6 to lug 4.
- Tested the unit.
- Reversed connections on lugs 1 and 2.
- Tested the unit.

Here are the last two versions that tested negative. Voltages are updated. Don't understand why there's 0V at pin 8. I'm going to let it sit for now and try to get the tube tested tomorrow morning.

https://lh3.googleusercontent.com/-9...7%25252048.jpg

Is this actually a hazing? Are you guys going to come back with, "Gotcha! Nothing was wrong except you forgot to plug it in!"

- Winky

Hm. Possibly a bad tube. If the RF section of the tube is conducting, there should only be a very small fraction of a volt on pin 8, readable only on the lowest V scale on the meter. That's because the low resistance of the coil, only 10 ohms, will drop very little voltage at the current the tube pulls.

When you go to have the tube tested, be sure the tester will do both sections of the tube independantly (not a "drug store" type tester).

Oh, it can sometimes seem that way... And can get quite fustrating. Thing is, one small error can be the cause, and people can be their own worst proofreaders, as they become blind to their own errors. I would avoid extensive modifications to Phil's circuit, if it never worked for you. That usually leads to additional fustration. Maybe one of us who lives nearby can pay you a visit. Not me, though, I'm in NJ.

I have one of those units. It was factory made by, IIRC Bud radio products. The oscillator coil was several turns of scc wire wrapped around a chassis mount compression trimmer. Don't remember if the coil was tapped or not. I was going to clone it, but the performance wasn't that great. My "radio dj " works a lot better.

Dang, you are right amigo. And i am wrong for coming up with an alternative circuit that obviously doesn't work. I need to apologize profusely for the wild goose chase. Upon visiting Phil's schematic, the circuit depicted is obviously the correct one, once the tube internals are added to the tube symbol. http://antiqueradio.org/transmitter.htm

Oh man. Sorry as hell.

The cathode (pin 8) should be grounded as it was originally.

The C5/C6 junction does go to lug# 1, not lug# 4.

All else remains as it was.

One thing that's still puzzling though, the resonant tank caps (C6a, b, and c) are now in parallel with the low impedance coil winding, whereas in seems like they should parallel the high impedance winding (as was the case in my "alternative" circuit).

No problem, it's a learning process. I've been there. The high impedance winding should be connected to the resonant tank caps. And the low impedance winding should be in the tube plate circuit. Best to rewire to make it this way, and leave enough wire so you can swap the connections on one of the windings, if it doesn't oscillate.

Yeah, just gotta be sure C5 (the feedback cap) remains tied to its junction with C6 when swapping things.

"Tube tester? I don't have to show you no stinkin' tube tester!"
 

At Murri's Electronics, the man says, "I put my tube tester in storage. Only needed to use it a couple of times a year. You should try A-Gem." A-Gem Supply is the place with the bad tube tester.

On reading Old Coot88's comments about very low voltage at pin 8--well, it does consistently read either 8 mV or -8 mV.

Questions:
From the data--voltages, resistance--is there any indication that either the tube or the coil is bad?
Please identify for me high impedance/low impedance windings and which is primary/secondary.

Fortunately I followed Phil's advice to not solder the coil connections until the thing actually works. In fact, I've gone one step beyond:

https://lh6.googleusercontent.com/-i...7%25252052.jpg

I'm inclined to buy another tube and coil, revert to Phil's original design, and rewire it as if starting from scratch. That will wait until I open the Atwater Kent 37 and find out what I need for it.

Until next time, thanks much, and as my old professor said, "When you hear hoofbeats, don't look for zebras."

- Winky

The voltage drop across the 10K filter resistor (R3)indicates the tube is conducting. R3 is dropping 8V, which works out to .0008 amp, which squares just about perfectly with the 8mv you were seeing across the 10 ohm coil. Offhand i would say the tube is good, but will wait and see if Robert concurs.
If the coil shows good solid continuity on the ohmeter with no intermittency it is probably good.
The primary would be the 10 ohm coil and carries B+ to the tube's plate (pin 3), and the secondary would be the 78 ohm coil which forms the tank circuit with C6.
According to Phil's schematic, the primary is shown as lugs 3 and 4, and the secondary is shown as lugs 1 and 2. This appears to be backwards from the way it's been getting hooked up. Maybe Phil could have an eyeball at it and see if a little switchamaroo in the labeling mighta snuk in.

In truth, i was trying to devise a circuit that'd conform to the coil lugs' labeling shown in the schematic, while keeping the high impedance winding in the tank circuit.