OT ratio too low?

[DaveZ]

I have a tranny testbed, basically everything in the power section save the phase inverter. I'm testing a PT/OT bought off ebay (salvaged).

PT looked good with just recto & filtercaps and power tubes drawing heater current. But B+ drops 100V when the OT was put into the circuit. Example: 440V B+ with no power tubes; with 6L6s & OT, dropped to 265V, and the 6L6s were dissipating 20W each. This with 275 ohm cathode resistor. Roughly basing the thing on the tweed pro (390V & a 250ohm rK).

6V6s also dissipated 20W each (briefly, I hit standby fast). That with the same 275 ohm rK & plates only at 300V! With a plate of only 300 a 275 rK should work.

Screens are an appropriate 20V below. I tried a stiffer recto (5ar4) and the fuse on the B+ (250mA fastblo) blew.

So I suspect the OT, which measures a voltage ratio of 372, impedence ratio is 19.3.

Compare to a standard 8000/8 OT for 6V6 imped ratio would be 31.6. Or a 4000/8 for a 6L6 22.3. This tranny is more like a 3000/8. I'm guessing this is not enough resistance and thus the tubes are idling way to high, even though the cathode resistor and plate voltages should be working OK. Am I right?

Would you expect there an internal short in the OT or likely just built for a different output tube?

[Amp Kat]

IMO I don't think it will cause idle currents to be any difference however it could have effect on the tone. What does the transformer ohm out on the primary side from center tap to eack leg and across the winding ? Have you tried RG's debugging page at Geofex ? That will give you a good idea if it's good or bad. The only worry may be the wattage rating of the tranny and from the measurements looks like maybe a 6L6 so for 6V6's at those voltages should be fine.

[MWJB]

So with .5VAC applied to the secondaries, you are getting 9.7VAC, end to end, accross the primaries? Are you getting identical voltages from the centre tap to each end of the OT primaries? If not, this would suggest a short in the primary, you could also check DCR from CT to each end for a "rule of thumb" guide (a few volts difference in dc plate voltage, is not necessarily a cause for concern).

A 3K primary is a bit on the low side for 6V6s (I'd use a 16ohm speaker before trying them again). 6L6s/EL34 should tolerate it OK (subject to heater current & B+ current available from the PT).

What happens to the heater voltage when the 6L6s are installed? Assuming that the OT primaries are in good shape, try a separate 470ohm cathode resistor for each power tube (how's the cathode bypass cap by the way, should be 50v or better & some space between in & resistor, if this cap shorts you're in trouble) and check out voltages (inc. heater voltage) with one tube, then with 2. What happens?

A tweed Pro might hit 500vdc+ with no load/tubes? Not that you would typically verify voltages with no load on the PT. In the absence of any other details, this may suggest that your PT may not be designed for typical 6L6 voltages & currents?

[DaveZ]

test results

Hi gents, thanks for the hints. Ohms on the OT are OK (I have some notes on OT testing, may have been from geofex). My DMM may not be accurate enough to tell if they are mismatched.

MWJB I tested the OT with 8.5V AC into the primaries and got .44 out. (I have a little AC tranny) but I did not test both legs, will do that tonight.

heater voltages are a bit high unloaded due to 120V AC. They measure 7.12. With the 6V6 and a 5U4G recto (3 amps of heaters there) it drops to 6.97. I'll measure heaters tonight w/ 6L6.

great hint doing 1 tube at a time...will try that.

I agree 6L6s may be asking too much of this PT. But the OT may well be a mismatch for 6V6...the voltages look good for 6V6 so I may also try a 330 ohm rK just to see if that gets the 6V6s under control. I don't have a 16-ohm speaker or speaker sim, only 4 & 8.

Might have to spring for a tweed dlx OT & use 6V6.

FWIW tried an old 5AR4 recto and blew my 240mA fastblo every time.

[MWJB]

Don't overlook the cathode bypass cap & cathode voltage, something's drawing lots of current, this cap would be the next thing I'd check after eliminating tubes & OT.

[DaveZ]

Don't overlook the cathode bypass cap & cathode voltage, something's drawing lots of current, this cap would be the next thing I'd check after eliminating tubes & OT.

I'm going to cut that cap out as voltages are twice what they should be over it. The resistor is fine, I doublechecked it.

[Enzo]

OTs run on impedance, not resistance. At idle, that is not pushing signal, the impedance is irrelevant. The DC resistance of the winding will have some small effect, that's all.

You didn't mention what these trannys were out of. Perhaps the PT was never intended to run a power amp? Or for that matter maybe the PT has the problem. You can load the PT secondary with a resistance, value set for current desired. See what sag there is.

You can calculate the tube current either by shunt or by the voltage dropped across the cathode resistor. Current is current. Either the PT an provide it or not. Impedance ratios is not the issue here.

[DaveZ]

OTs run on impedance, not resistance. At idle, that is not pushing signal, the impedance is irrelevant. The DC resistance of the winding will have some small effect, that's all.

You didn't mention what these trannys were out of. Perhaps the PT was never intended to run a power amp? Or for that matter maybe the PT has the problem. You can load the PT secondary with a resistance, value set for current desired. See what sag there is.

You can calculate the tube current either by shunt or by the voltage dropped across the cathode resistor. Current is current. Either the PT an provide it or not. Impedance ratios is not the issue here.

OK thanks for the clarification on the earlier post (even a dead short...)

The PT I think is OK, it is delivering plenty of current, the 6V6s are idling at 20W each, 6L6s a bit more than that. PT is not even getting warm & other voltages are fine. I think I have a problem with the cathode bias. Going to work that angle tonight.

[Enzo]

One question to answer is this: Does the OT drop your B+ WITHOUT power tubes? After all, with power tubes the transformer is simply acting as a current path for the tubes.

If your tubes are idling at 20 watts - a perfectly reasonable level for 6L6s - and your B+ is sagging 175 volts (!!!), then I would be real suspicious of the PT.

Cathode or any other bias sets the current through the tubes. Getting 20 watts with cathode bias won't treat the B+ any different from getting 20 watts with fixed bias.

If the PT is puttting out pplenty of current, why is the B+ dropping from 440 to 265 when you draw that current? Is the AC at the PT dropping? or is it just on the DC side of the recto tube that it drops so much?

That 250ma fuse probably should be a slow blow, the charging cap current took it out, I'd bet.

[DaveZ]

cathode bias results 6V6

I'm figuring this PT might be best with a pair of 6V6.

Unloaded, B+ is 350V. This is dead on for a tweed deluxe circuit, which usually uses a ~270 ohm bias resistor. Voltage across the bias resistor is usually 25V in a tweed deluxe.

Loaded (e.g., standby switch on), B+ drops to 325. So far so good. But voltage drop across the bias (275 ohm) resistor is 52 V. That's 189mA across that resistor...With the plate to cathode voltage of 258V, that is about 24W per tube, double the plate dissipation for a 6V6. This is really making me scratch my head.

All the voltages are about right, the tubes are known good, WTH is going on here? If I double the rK, I still get close to 20W per tube. Are the screen volts too low - they are about 30V below the plates in all cases.

6V6 with 275 ohm bias:

B+ 324V Plate 313V cathode 52V 24.3W/tube


6V6 with 605 ohm bias

B+ 342V Plate 332V cathode 105V 19.5W/tube

[Enzo]

Ah. 350 down to 325 is more like it. Where did that 440v in the first post come from?

Have you measured current through EACH tube? Use shunt method or a bias probe. See if one side is out of whack.

Are the grids on pins 5 sitting at zero DC? A positive voltage there leaking through a coupling cap would up the current.

Since you built this - right? - did you remember to include grid resistors to ground from the pins 5? Unterminated grids can float to whatever voltage they like. For that matter, even the 10 meg of your meter will terminate them to some extent.

And also, scope the output to make sure there is no high freq oscillation cranking through the stuff. What load is on this tranny?

[pdf64]

Also worth double checking that there's no dcV getting on to the power tube grids, ie from phase splitter leaky coupling cap - Peter

[DaveZ]

Ah. 350 down to 325 is more like it. Where did that 440v in the first post come from?

Have you measured current through EACH tube? Use shunt method or a bias probe. See if one side is out of whack.

Are the grids on pins 5 sitting at zero DC? A positive voltage there leaking through a coupling cap would up the current.

Since you built this - right? - did you remember to include grid resistors to ground from the pins 5? Unterminated grids can float to whatever voltage they like. For that matter, even the 10 meg of your meter will terminate them to some extent.

And also, scope the output to make sure there is no high freq oscillation cranking through the stuff. What load is on this tranny?

HI Enzo, thakns, that is I think the missing piece, I did forget to ground the grids - no connection. I was assuming they'd be at 0V. I'll fix that up tonight & report back. Thank you very much for the help, this is the first powerup of the testbed hardware....it is not surprising I forgot something.

[DaveZ]

ALL FIXED. Just needed grid resistance, d'oh!

All is working fine. Sample measurements w/ 275-ohm rK

Recto Tube B+ Plate Screen P-K Cathode Watt/tube
5AR4 6L6WGB 395 389 383 355 30.8 19.9
5U4GB 1 6L6WGB 376 366 362 337 28.9 17.7
5Y3 1 6L6WGB 365 362 355 330 28.6 17.2
5Y3 1 6L6GC 370 356 353 330 27.6 16.6
5U4GB 2 6L6WGB 346 344 336 318 26.8 15.5
5Y3 2 6L6WGB 341 330 333 307 26.3 14.7

the two 5U4GBs are a small bottle (1) and a big bottle (2). The first 5Y3 is a Sovtek fake - you can see it is stiffer than one of the 5U4GB. The other is a NOS.

The PT voltage is a bit high for 6V6s, I'd have to have a very large rK, probably 400 ohm so I will stick with my original plan & build with the 5U4G & 6L6s.

thanks all for the help!
--Dave Z

[MWJB]

"The PT voltage is a bit high for 6V6s, I'd have to have a very large rK, probably 400 ohm so I will stick with my original plan & build with the 5U4G & 6L6s." Not necessarily. Depending on the effect of the 6L6 current draw pulling down B+, it's only when you fit the 5AR4 that you start exceeding 6V6 voltage with a 250/270ohm rK. At 395v on the 6V6 plate, a 330ohm Rk would probably be OK (B+ may jump 15v or so, but cathode voltage will rise also, marginally bringing down plate dissipation).