I have two Thordarson 22S86 'Universal' 3 watt transformers, but I have misplaced the instruction sheet that had the connection information.
The transformers are T-22S86, and I have found a scanned copy of the Thordarson catalog (from 1947!) online, but unfortunately, it does not include the connection information.
Per the catalog, the transformer characteristics are:
http://www.vintageradio.com/history/Thordarson1947.pdf
Universal Single Plate or Push Pull Plates to Voice Coil
Pri.-14,000 C.T.-8,000 C.T.-3,500-2,000
There are three terminals for secondaries: Common / 3.2 ohms / 8 ohms.
There are five wires on the primary side. Sadly, the colors have faded over time, plus I am color-blind. However, it appears that the wire on the left as I look at the front of the transformer is black, which I would presume would be the plate (?!?). I will call that wire 1.
I think the one next to the black wire is yellow with some kind of stripe. I think the wire next to that is yellow (or green). The wire next to that is green or brown, and the last wire is red (I think). Hard to say, they are very faded with age anyway.
With my DMM, I measure the primaries like this:
Wire 1 - 2: 196.9 ohms
Wire 1 - 3: 391 ohms
Wire 1 - 4: 480 ohms
Wire 1 - 5: 736 ohms
If I measure from the last wire (call it wire 5) going the other direction, I come up with:
Wire 5 - 4: 256 ohms
Wire 5 - 3: 344 ohms
Wire 5 - 2: 540 ohms
Wire 5 - 1: 736 ohms
I am trying to get to 2500 ohms, for a 50C5 amp I am nearly done building.
Can anyone help me with a diagram, or failing that, to deduce what the connections should be to get to 2.5K for the 50C5 output tubes?
Thanks!
The transformers are T-22S86, and I have found a scanned copy of the Thordarson catalog (from 1947!) online, but unfortunately, it does not include the connection information.
Per the catalog, the transformer characteristics are:
http://www.vintageradio.com/history/Thordarson1947.pdf
Universal Single Plate or Push Pull Plates to Voice Coil
Pri.-14,000 C.T.-8,000 C.T.-3,500-2,000
There are three terminals for secondaries: Common / 3.2 ohms / 8 ohms.
There are five wires on the primary side. Sadly, the colors have faded over time, plus I am color-blind. However, it appears that the wire on the left as I look at the front of the transformer is black, which I would presume would be the plate (?!?). I will call that wire 1.
I think the one next to the black wire is yellow with some kind of stripe. I think the wire next to that is yellow (or green). The wire next to that is green or brown, and the last wire is red (I think). Hard to say, they are very faded with age anyway.
With my DMM, I measure the primaries like this:
Wire 1 - 2: 196.9 ohms
Wire 1 - 3: 391 ohms
Wire 1 - 4: 480 ohms
Wire 1 - 5: 736 ohms
If I measure from the last wire (call it wire 5) going the other direction, I come up with:
Wire 5 - 4: 256 ohms
Wire 5 - 3: 344 ohms
Wire 5 - 2: 540 ohms
Wire 5 - 1: 736 ohms
I am trying to get to 2500 ohms, for a 50C5 amp I am nearly done building.
Can anyone help me with a diagram, or failing that, to deduce what the connections should be to get to 2.5K for the 50C5 output tubes?
Thanks!
Wires 1 and 5 are apparently the extreme ends of your 14K primary winding. The center-tap must be wire 3. Wires 2 and 4 are your 8K end-to-end primary, again using wire 3 as CT. Either end of the 8K primary to CT yields 2K single-ended, which shifts upward to 2.5K if you attach 4 ohms to the 3.2 ohm nominal secondary. Does that sound right?
I think I agree with BinaryMike. Also, I am interested in seeing what you are building... 50C5 stereo amp, or push-pull monoblock? Sounds cool, either way.
cheers
Rob
cheers
Rob
I will happily supply photos and schematic soon, but yes, its a stereo amp.
The idea for this started with a pair of surplus 120 volt 200ma transformers I picked up for next to nothing, new old stock.
I intended to use one for the heaters, one for the B+.
Problem was, with house mains at 124 VAC, I was getting 156 volts unloaded from the secondaries.
I didn't want to use big resistors to drop voltage, too much heat under the chassis.
I finally decided to series-string 2 50C5 tubes with a 50DC4 halfwave rectifier and a 12AX7A, all Matsushita with the four-way seams (Mullard tooling?). Ended up running the heaters a little cool, but within 10%. Don't know how it will work out long term.
I was originally just going to use the 50DC4 as a ballast tube, which would at least put the heat on the top side of the chassis. But I wired up a solid state bridge rectifier, had to use big resistors and a choke to get B+ down, and the PT got hot. And the amp had a lot of hum.
So, even though my current draw is at absolute max for the 50DC4, I went for it. Put a silicon diode in series with the plate for safety.
Dumped the choke, now I am running 33uf-100uf-100uf with 100ohm 5 watt resistors between them. B+ is 115 on the plates, 120 on G2. I will fiddle with the resistors a bit.
Hum is under control, output is ok, but I need a bit more gain and a bit less distortion. Need to add NF and tweak the 12AX7 cathode resistor value a bit. Measuring 117 on the plates, 1.5 on the cathode, with 100k plate resistors and I think 3.3k cathode resistors bypassed with 100uf caps.
All on the breadboard, but I ran out of weekend.
Hope you found that interesting!
Thanks for your help!
Last edited:
Heck yeah! I have been thinking about doing a set of single ended 50C5 or 50L6 amps for casual listening for years, but never seemed to get around to it. I have quite a few of those NOS tube around from old radio restorations. Looking forward to seeing your creation when it gets to that point.
cheers
Rob
I also have several sets of tubes of that type, including 50L6, which can take more B+ than the 50C5 can, so I was tempted to let the B+ rise and simply use the 50L6, or even a pair of 50C6G, which I have previously played with and which look cool as heck in ST-14 bottles. But I really wanted to try to put these 50C5 tubes into service somehow, so I've stuck with it for now. I'm also giving some thought to replacing the 50C5's with 35C5's, which leave me with extra voltage on the heater string; I could add a small power resistor and a 6 volt pilot lamp, I suppose. Dunno yet. Still playing.
But whichever route I go, I am going to try to get these old OPTs working. They are tiny, not for high-fidelity clearly, but they're matched in terms of power output to what I'm building. The OPTs I am currently using to test are much too large (12 watts) for this design, and I don't want to commit them to the project permanently.
I'll post in a separate thread when I get a schematic and photos together.
A set of four 50C5s would drop cleanly into my old Parsimon push-pull amp design. The OPTs are still available at MCM for $4.99.
I have a small stack of those OPTs, but I didn't think about doing a P-P design. Thanks!
The DPC-120-200 yields just 120VAC on the secondary, so your heater power concept needs revision.
Why is the silicon diode present?
If we assume a half-wave rectifier 'form-factor' of 2.3 (per Stancor) and 120mA total DC load current, then you're overloading T1 by 38%.
Why is the silicon diode present?
If we assume a half-wave rectifier 'form-factor' of 2.3 (per Stancor) and 120mA total DC load current, then you're overloading T1 by 38%.
Well, it's supposed to be 120 on the secondary, but even loaded, they are putting out much more. I guess because my household voltage is so high?
The diode was to protect the PT iF the overloaded tube rectifier shorts. I read guitar guys do it. I didn't realize I was overloading by so much though.
I've got a big 120vct 850ma transformer I can sub in for the two little ones. Would that work?
Thanks!
You're probably conflating rectified DC voltage with secondary AC voltage. The DC value is higher because the filter cap charges to the peak value of the AC waveform. If you're thinking of using the bigger xfmr with your 162V heater string operating on DC, then the 50DC4 would be overloaded. I suppose I would ditch the 50DC4, leaving just the 1N4007, and tie the remaining 112V heater string to the 120VAC secondary with some series resistance to sop up the remaining 8V. 56R/2W should be about right. Use a wirewound resistor, to survive the cold heater surge.
Good to know. Thanks!
Have you considered a negative feedback loop to reduce distortion, or will the gain suffer too much for the level of input you are going to drive it with?
Last edited:
Ok, yes I was, and I see now. Thanks for the info! So I'll ditch the 50DC4 as you suggested. I can also add a 6v pilot light to the string.
I was debating back and forth between that and maybe just eliminate the bypass cap on the 50C5 cathode. Either way, I'm going to lose some gain to tame the beast I fear.
You could add the second half of the 12AX7 for more gain, and then use "global" feedback I think. I like your signal circuit as is, and it might sound just the way you want it...but just thinking.
Well, if I use both halves of the 12AX7, then I have nothing for the other channel, you know? Sometimes it seems a tube amp is nothing but a collection of compromises! Take from here, give to there. Cut distortion, lower gain. Add gain, distortion rises.
I'm also thinking that BinaryMike was absolutely correct about the fact that I'm running the heaters for all these tubes overspec for that transformer; I need to drop the 50DC4 and simply use a dropping resistor to get my heater voltage dialed in. Then since I'll be using a solid state diode for rectification, my B+ will be higher and will require more dropping resistors or a choke in the power supply to get the B+ under control. Fun, fun, fun!
I will redraw the schematic and hopefully this weekend I'll have the time to put the changes in and see how it sounds afterwards.
I cleaned up the schematic per BinaryMike's suggestions. I added a 6V pilot light to the heater string, which I believe only leaves me with <2 V to drop, so I calculated a 10 Ohm resistor to do that, uprated to 2 watts and wire-wound as suggested.
PSUDII suggests that with a 680 Ohm and a 100 Ohm resistor in the power section, I should end up with 125 VDC on B2 with .5 volt ripple, 114 VDC on B+ with .1 volt ripple, which I am pretty sure I can live with. I assumed 110ma current draw for the two 50C5 and one 12AX7 tube. Might be a bit more than that.
As always, I appreciate any comments or suggestions!
PSUDII suggests that with a 680 Ohm and a 100 Ohm resistor in the power section, I should end up with 125 VDC on B2 with .5 volt ripple, 114 VDC on B+ with .1 volt ripple, which I am pretty sure I can live with. I assumed 110ma current draw for the two 50C5 and one 12AX7 tube. Might be a bit more than that.
As always, I appreciate any comments or suggestions!
Attachments
Indeed they are. Took me a few months to work the bugs out of my first 6V6 guitar amp, and I was already plagiarizing part of the schematic from Fender.
- Status
- Not open for further replies.