Questionable Output Transformer Ratio. Maybe.

[dtbradio]

Hi All!

I have an OPT pulled from a 1950's Baldwin tube organ. It was in a 6l6GC P-P circuit, but I was unable to identify the total speaker impedance due to rust, etc., and no schematic for that organ. Anyway, I ran a ratio test using a 1khz tone into first the primary and measuring resulting secondary voltage (turns ratio showed 19:1), then ran the generator into the secondary and measuring the primary voltage to double-check my first result. Same 19:1 measurement each time with multiple meters and a 'scope. 19:1 seems like a very low turns ratio for an OPT, especially for a 6l6 circuit since the 6l6 sheets I've read show about a 10K-ohm plate to plate impedance spec. I saw no indication that the windings in either the primary or secondary were shorted, but I suppose its possible to have shorted windings and still have a clean sin wave reproduced across the transformer. Anyone have any thoughts?

[martin manning]

19.1 squared is 365. An 8 ohm speaker would give only 2.9k, so I would suspect a 16 ohm speaker, which would give you a 5k8. That would look pretty good with 6L6 at 450V plate, 350V screen.

[LOUDthud]

Are you sure it is for 6L6"GC" and not 6L6 "G", "GB", or "nothing" ? What is the B+ Voltage ?

[dtbradio]

Are you sure it is for 6L6"GC" and not 6L6 "G", "GB", or "nothing" ? What is the B+ Voltage ?

Its a GB, not sure why I typed GC, lol! For a supply I'm using a 5U4 powered from a center-tapped 700V transformer (develops a bit more like 730-750v since the mains are a around 120 volts here), 235uf worth of cap filter with a 560K bleeder. With that setup and no other load, powering up from dead 0V on the cap and then measuring after being on for 120 seconds, I'm seeing about 480 volts developed across the bleeder. Once I have some quiescent tube load it will probably settle out to the 430-450 range or so.

All the data I'm finding for the 6L6GB (etc) seems to indicate that 10K plate-to-plate load is appropriate, but I see posts and circuits indicating anything from 3K on up as a plate-plate load (differing supply voltages, of course).

19.1 squared is 365. An 8 ohm speaker would give only 2.9k, so I would suspect a 16 ohm speaker, which would give you a 5k8. That would look pretty good with 6L6 at 450V plate, 350V screen.

Would it be safe to run 8 ohms if I use a higher cathode resistor? I know it will drop volume, but that's not a big deal for my purposes. I know the tone will change, not sure if that will be a good or bad thing until I actually hear it.

[Stevem]

Let me get this straight, you have for the first filter node after the 5U4 a 235 uf cap?

If that's the case then you will eat up rectification tubes really fast!

[martin manning]

Would it be safe to run 8 ohms if I use a higher cathode resistor?

With 6L6GB you have 22W max plate dissipation. That says you need higher primary impedance. The 5k8 that you get with a 16 ohm load would be a pretty good match. If you run with 8 ohms and have 2k9, biasing cooler would help some, but when you have signal present the plate dissipation will be too high.

[dtbradio]

Let me get this straight, you have for the first filter node after the 5U4 a 235 uf cap?

If that's the case then you will eat up rectification tubes really fast!

Thank you for your reply! I forgot to mention that the 5U4 dumps through a large choke first, then the cap. The choke is from the same Baldwin organ, and is labeled "512-30 K".What cap value would be recommended as the first node after the choke? There is no cap before the choke.

Would it be safe to run 8 ohms if I use a higher cathode resistor?

With 6L6GB you have 22W max plate dissipation. That says you need higher primary impedance. The 5k8 that you get with a 16 ohm load would be a pretty good match. If you run with 8 ohms and have 2k9, biasing cooler would help some, but when you have signal present the plate dissipation will be too high.

Thank you for your reply as well! Then I will run 16 ohms, whether speaker coil or resistive load. I always set my output jacks up to connect a dummy load of similar resistance to the speaker load as a matter of course for tube safety when no speaker is plugged in, as well as it being convenient for testing AC and DC levels without having to run into a speaker.

[dtbradio]

I'm adding a 10-ohm, 20-watt resistor to the output of the choke before the capacitor to drop charging current a bit. That should make the 5U4 less unhappy.

[TUBEDUDE]

So you have a choke fed supply? With the B+ you mentioned I find that unlikely. Are you sure there's no cap before the choke after the rectifier?

[dtbradio]

So you have a choke fed supply? With the B+ you mentioned I find that unlikely. Are you sure there's no cap before the choke after the rectifier?

100% sure, as I built the thing from scratch. I do plan on putting a low-val cap in front of the choke, something in the range of 2 - 10 uf or so. I did a bit more research on 5U4 supply circuits, and that seems to be a common recommended range for pre-choke filtering. My problem is finding appropriate caps in the junk boxes, lol! I'm trying to avoid spending money on the project as much as I can, making use of bits I already have.

[TUBEDUDE]

A smaller cap, 100-200nF should be enough to keep the magnetic field from collapse. Larger values run the risk of it becoming a capacitive input filter (if I'm not mistaken).

[dtbradio]

A smaller cap, 100-200nF should be enough to keep the magnetic field from collapse. Larger values run the risk of it becoming a capacitive input filter (if I'm not mistaken).

Are we talking about the same cap, IE between the 5U4 and the choke? Connected to ground from that point, obviously, lol! Anyway, 200nf (.2uf) doesn't seem like it would be much of a reservoir for the 6L6 plates to use, unless I'm misunderstanding some of the schematics I see. For example, the Fender 5E5 (http://www.thevintagesound.com/ffg/sche ... _schem.gif) shows 16uf at the 5U4 output before the choke. Granted, the B+ is lower, but still, I'd think a few uF at least would be warranted at the 5U4 output.

[sluckey]

You keep calling this a choke input filter. A choke input does not have any cap between the rectifier and the choke. And the 6L6 plates would be getting their voltage ***AFTER*** the choke.

The 5E5 schematic you linked to doesn't even have a choke! I think there's some confusion that could probably be cleared up if you post a schematic of what you actually have. Maybe some hi-rez pics of your project.

[wpaulvogel]

With choke input filter arrangement, you’ll get about 350 volts rectified DC.

[dtbradio]

You keep calling this a choke input filter. A choke input does not have any cap between the rectifier and the choke. And the 6L6 plates would be getting their voltage ***AFTER*** the choke.

The 5E5 schematic you linked to doesn't even have a choke! I think there's some confusion that could probably be cleared up if you post a schematic of what you actually have. Maybe some hi-rez pics of your project.

Sorry for the confusion, I posted the wrong example schematic. The Fender Bandmaster Reverb (TL5005D), shown at http://www.thevintagesound.com/ffg/sche ... _schem.gif, is what I meant to post. It shows that the 5U4's output has the equivalent of 35uf to ground in parallel with 440k-ohm's-worth of bleeder resistors, before the choke. The plate is fed BEFORE the choke in this schematic, screens after the choke. Rectified/filtered voltage specified in this schematic is 440V before the choke and 435 after the choke, pretty close to what I expect to see in my circuit once the tubes are loading the power supply. My supply has the filter caps after the choke, not before as in the Bandmaster, so at this point, I'm thinking that the rectifier in my supply will be fine.

As to details of my build, that's not nailed down yet other than that I'll be using a PP 6L6GB power amp circuit with cathode biasing, and with plate supply taken after the choke. The pre-amp section is still not set yet, but leaning toward a Soldano SLO30-style minus the clean channel and effects loop.

With choke input filter arrangement, you’ll get about 350 volts rectified DC.

Already getting 480 filtered DC after the choke with a 560K-ohm load. Tube loading will drop that of course, but not down to 350. As per the Bandmaster mentioned above, I expect closer to 430 or so with loading.