PT-10 specs

[JeffYoung]

I've been trying to model my S.E.X. PSU in PSUD2, but I'm missing some numbers on the PT-10 transformer.


I've gotten a reasonable approximation of the free-load voltages assuming a 5A filament secondary with 10% regulation, and a 1A B+ secondary with 10% regulation.


Are those close, or are the real specs available somewhere?


Thanks,
Jeff.

[Paul Joppa]

It's easy to measure. If you pull the tubes, you can measure the voltage across each winding. Unpowered, you can measure the resistance of each winding. I could give you those numbers based on the design calculations, but they will be pretty much the same as what you measure. Try it! It's fun!!

It won't be precise, because PSUD does not deal with multiple secondary windings, and it's even more complicated when multiple secondaries also have rectifiers. But the error is fairly small, and I don't have any better tool to suggest.

[JeffYoung]

OK, I put the off-load voltage (6.9V) and primary (33R3) and secondary (0R2) resistances into PSUD2's Source Impedance Calculator, and it gave me an estimated impedance of 0R233.


I then used the Off-load Voltage Calculator and edited the regulation until it gave me the right off-load voltage, and then edited the current until it gave me the same impedance as estimated above.  Results: 10% regulation, 2.7A.


I ran it again for the B+ secondary (180V, 33R3, 40R0) to get 12.5% regulation and 330mA.


These figures certainly appear to be in the right ball-park.  Are they a reasonable match to specs?


Thanks,
Jeff.

[Paul Birkeland]

Are they a reasonable match to specs?

You can put the rest of the power supply into PSUD to see how these specifications match the DC voltages you're measuring.

Since the SEX splits the power supply, you can model it as though there was only one power supply that's a voltage doubler CRC with 340 Ohms as the R, 100uF as the second C, and a current draw of about 60mA.

[JeffYoung]

Yeah, that's exactly how I have the PSU mocked up.  Trouble is I need to adjust the transformer impedance up to 90R and the output down to 225mA to get that to produce 400V.


Things tend to get spec'd in round numbers, though, and if I put in 250mA for the transformer then I get a working voltage of 407V.  Given that my no-load voltage was also a smidgen high, and neither Paul has yet told me that I'm out in left field, I'm going to assume that's accurate enough.


Thanks for your help Pauls,
Jeff.

[Paul Birkeland]

I'm not sure how you're attempting to factor in the 6.3V winding into the analysis of the HV winding performance.

For the HV winding, I have measured the 160V winding to be about 168V unloaded.  I also estimate the 160V winding to be about 0.3A AC (not DC!).  This corresponds to about 5% regulation, and is enough to populate the fields in PSUD II for the power transformer.  I did not find that filling out the source resistance fields was necessary to provide an adequate simulation, YMMV.

I get 407V at the output of the HV power supply in PSUD with a voltage doubler, RC filter with 340 Ohms of series resistance, and a 65mA load.

The manual says that there should be 400V here, so we are 1% off, not too bad.  With the heater circuit taxing the power transformer a bit, we would expect that voltage to drop a little.

[JeffYoung]

I'm not sure how you're attempting to factor in the 6.3V winding into the analysis of the HV winding performance....

I'm not; I'm just mocking up two different PSUs in PSUD2.  But your point that this will make things an approximation at best is well taken.


Cheers,
Jeff.

[Paul Birkeland]

Oh, the low voltage supply won't model well in PSUD.

[JeffYoung]

Murphy's Law, no doubt.  I actually had an easier time getting what I expected with the filament supply:

[Paul Birkeland]

I'm surprised those results came out, especially with 1N4007's!  I can assure you that if you build that supply, the diodes would go snap, crackle, and pop!

[JeffYoung]

Good catch!


I built a reasonably accurate model of an MBR745 and added it to PSUD2's rectifier database*.  It doesn't model quite as well, but it is what it is. 


Cheers,
Jeff.


* If anyone else is interested, you can download the current rectifier database here: http://www.duncanamps.com/psud2/rect_list.php