Using 'Bridging Transformer' as PP line output

[lassoharp]

I have a 20K (2 10K windings) :600/150 transformer that I wish to use as an output transformer in a PP circuit using 2 halves of a 6SN7. It was brought to my attention that this transformer was intended for use as a line bridging transformer.  It's rating in that context is +15dbm max level.

My concern is whether the pri windings can handle the apparent dissipation due to the large DC winding resistances in this particular transformer - 4.84K per side

Each half of the SN7 is pulling ~11.0ma. CT B+ is 250V. The plates sit at 220V. So a 30V drop with 11ma gives me an equivalent resistance of ~2.7K per winding if I'm figuring this correctly. I'm assuming this dissipates about 1/3W in the winding at idle. So far it performs like any other dedicated PP output and the windings have not fried - but how to tell if they will take it over long periods?

Better to use something else?

 

[emrr]

We were talking about this question offline, and it's a subject of interest since you see bridging transformers sold as outputs on ebay all the time.  In the one SE parallel feed application I tried, a bridging performed just fine, with similar losses.  But PP with DC may be another matter. 

The Altec 15095 is used as much for bridging as it is output, but always in parallel feed as output. 

The winding resistance of the Langevin 5116-B is particularly high for an output, and it pulls a lot of current for a tiny tiny transformer.  It's the closest actual output I can think of.  The Fairchild 660 output is also very high Z, and I haven't looked to see what current, voltage drop, winding resistance, etc exist. 

 

[abbey road d enfer]

I wouldn't be worried about long term heat dissipation; look how small a 1/2W resistor is. It is designed to never exceed 50°C (122°F). Your transformer is certainly bigger and has more thermal exchange surface than two 1/2 W resistors.
I would be more concerned about transients and mishaps, in particular in the event a tube goes short. Generally, dedicated OT's are designed to withstand a temporary short. I'm not sure your xfmr can withstand 90mA/23W for any length of time.

 

[PRR]

This looks OK.

I'm not sure why you say "4.84K per side" then calculate 30V/11mA= 2.7K per tube. The latter looks like a fair test; where did 4.7K come from.

I had some 20K bridging transformers that didn't add up. Turned out they were a 10K winding plus two 4.7K series resistors. This gave a lovely light load on whatever was being bridged. Even at DC or 1MC (we didn't Hertz yet) they were sure to be >9.4K, and could be tapped on and off a 300 ohm bus with negligible effect. But they would suck as output iron because they wasted half the power you put in. They sipped, they wouldn't glug.

> +15dbm max level.

+15dBm where? On the bridging bus? Then that's only +3dBm on the low-Z side. Or actual 31.6 milliWatts in _10K_? Than that's still 31mW or +15dBm at the low-Z port. And the rating could be "soft": it may exceed 1% THD at +15 at 20Hz but be quite tolerable iron-sound in the 70Hz peak of popular music even many dB higher.

 

[lassoharp]

This looks OK.

I'm not sure why you say "4.84K per side" then calculate 30V/11mA= 2.7K per tube. The latter looks like a fair test; where did 4.7K come from.

I had some 20K bridging transformers that didn't add up. Turned out they were a 10K winding plus two 4.7K series resistors. This gave a lovely light load on whatever was being bridged. Even at DC or 1MC (we didn't Hertz yet) they were sure to be >9.4K, and could be tapped on and off a 300 ohm bus with negligible effect. But they would suck as output iron because they wasted half the power you put in. They sipped, they wouldn't glug.

> +15dbm max level.

+15dBm where? On the bridging bus? Then that's only +3dBm on the low-Z side. Or actual 31.6 milliWatts in _10K_? Than that's still 31mW or +15dBm at the low-Z port. And the rating could be "soft": it may exceed 1% THD at +15 at 20Hz but be quite tolerable iron-sound in the 70Hz peak of popular music even many dB higher.

The 4.7K was the meter measured DCR for each winding.  The 30V was the drop from the CT B+ to the plate. I used the measured cathode current and plate voltage drop to calculate what resistance that would be equivalent to - then assumed that value for the wattage calc.

I'm not sure how they were rating it.  Doug may have that info in the catalog. I need to run some measurements at lo and hi-z loads.

 

[emrr]

I had some 20K bridging transformers that didn't add up. Turned out they were a 10K winding plus two 4.7K series resistors. This gave a lovely light load on whatever was being bridged. Even at DC or 1MC (we didn't Hertz yet) they were sure to be >9.4K, and could be tapped on and off a 300 ohm bus with negligible effect. But they would suck as output iron because they wasted half the power you put in. They sipped, they wouldn't glug.

ADC, ADC/RCA, and UTC/RCA bridging types; most show the resistors on the schematic on the case.  I compared one to a similar Z output in a parallel feed circuit, and it worked fine, with nearly identical transmission loss. 

Virtually no one ever gives a rating with a reference to which winding they refer to.  There must be some standard; assume it to be the side with highest level?  Or the side presented to the outside world?  I think probably the latter. 

 

[CJ]

just put an hp osc on the 600 side,

put a vtvm or scope on the hi voltage side,

crank it up at 20 hz and see where it craps out.

if it saturates at 100  volts on the hi side, then it probably will distort,


but if you can swing 200 volts line to line, you might be in bidness.

400 volts means you can run dc, maybe, just a guess,

 

[lassoharp]

just put an hp osc on the 600 side,

put a vtvm or scope on the hi voltage side,

crank it up at 20 hz and see where it craps out.

if it saturates at 100  volts on the hi side, then it probably will distort,


but if you can swing 200 volts line to line, you might be in bidness.

400 volts means you can run dc, maybe, just a guess,

Good test tip - I can manage that with my limited setup.

I believe it may already be distorting down there - kind of in a good way but overall amp response with stock coupling cap values says too much bottom. About 470H total on the pri(hi side).

 

[lassoharp]

Transmission loss fairly noticeable.

Compared in-circuit with a dedicated OT of same impedance/step gain spec.

Bridging unit lost 5.2dB gain at 1K relative to other OT.

 

[emrr]

that's a lot......

 

[lassoharp]

Yup.  Probably will come down to a tonal decision here.

I still want to test the reverse - as a step up - and see what it does.