Hiwatt DR504 build

[Pete Farrington]

Interesting results, thanks!
I don’t think that the g1=-20V point on the A3 curves is being suggested as a suitable operating point. They’re just characteristic curves, don’t read more into them than that.
Note that for the typical operating conditions, there’s a g2 resistor, which will significantly reduce g2 current / dissipation.

Anode dissipation with signal looks to be ignoring power being delivered to the load and lost in the OT.

 

[David Copeland]

Anode dissipation with signal looks to be ignoring power being delivered to the load and lost in the OT

I'm not sure what your referring to by this. Please explain. The power out of my amp went down from 110 watts at clip to 92 watts at clip or -0.77 db with the bias adjustment. I can live with that.

 

[David Copeland]

Now that I've had my first cup of coffee: Yes the anode dissipation is based on the DC readings across the 1 ohm cathode resistor, the screen grid resistor, and the plate voltage so by definition they do not include any AC signal current.

 

[Pete Farrington]

Power that the amp delivers to the load is power that’s not being dissipated by the anodes or the OT.
You’re measuring the power that’s going in (to the valves) but not subtracting the power that’s going out (to the load).

So assuming the OT is 90% efficient, 92W to the load equates to 102W at the anodes, so that’s 25.5W per anode we can subtract from the anode dissipation result in the spreadsheet.

Note that for valid power measurements, an rms reading meter needs to be used.

 

[StingRay85]

The topic of this thread is a DR504, hence an appropriate OT will have a primary impedance of around 3k5.

The conditions above are intended for an OT with a primary impedance of 11k and anode and screen grid voltages somewhat different to those of a DR504, hence don't seem to be apposite?
Vba = 800V at Vi = 0Veff means that the HT at idle is 800V.


Notice that the OT primary impedance is 4k or 5k, and that the screen grid HT node voltage is 400v or 375V respectively.

All the currents there are average, ie over a full signal cycle, rather than instantaneous peak.

I just checked the conditions in my Sound City 50Plus (3k4 OT, white shield Mullards).
I used a variac to maintain a constant 240VAC mains input. EL34 each biased for 35mA idle cathode current, 5mA of which was attributable to the screen grid.
At idle, HT voltage = 466, Va = 465, Vg2 = 459
At max unclipped sinewave output (22Vrms @ 8ohms = 60W), HT = 424V, a = 420V, g2 = 386V, Ik = 154mA, Ig2 = 27mA (Pg2 = 11.3W)
Note that Ik limiting value of 150mA is exceeded, and likewise for g2 dissipation 8W limit.
At max squarewave output (28Vrms @ 8ohms = 98W), HT = 398V, g2 = 340V, Ik = 222mA , Ig2 = 40mA (I'm not going to risk my Fluke by probing an overdriven anode ) (Pg2 = 13.6W)
Note that Ik is far in excess of the limiting value, and likewise for g2 dissipation.

Happliy the Sound City has a tap on its HT winding, which allows for a lower HT voltage. As stock, this was unused. I've put it onto a switch on the back panel. With this engaged -

At idle, HT = 389V, a = 397V, g2 = 389V, Ik = 35mA, Ig2 = 5mA
At max sinewave (onset of clipping 19.3Vrms @ 8ohms = 47W), HT = 372V, a = 369V, g2 = 339V, Ik = 138mA, Ig2 = 22mA (Pg2 = 7.5W)
Note that at a lower HT, everything is within its limiting value.
At max squarewave output (25Vrms @ 8ohms = 78W), HT = 357V, g2 = 306V, Ik = 203mA, Ig2 = 35mA (pg2 = 10.7W)
Note that despite Ik and g2 limits being exceeded, the degree is lessened.

All voltages are DC unless otherwise noted.

Hence my suggestion that a slightly lower HT voltage than typically used would be beneficial. This is especially the case if current production valves are to be used, as these may not accommodate their limiting values being exceeded quite as well as those of vintage manufacture.
An idle HT of 420V may be a good choice, as per the Philips info.

Very interesting. Also the fact that the 50 plus has an unused lower HT tap. Those are the EV transformers right? Look a lot like the Patridge but without the varnish.

 

[Pete Farrington]

Very interesting. Also the fact that the 50 plus has an unused lower HT tap. Those are the EV transformers right? Look a lot like the Patridge but without the varnish.

I’ve had 2 SC 50+, an earlier one with yellow label partridges (PT has the lower voltage HT winding tap), the other with what I suppose are EV, it was later model with the phase splitter valve between the EL34s. The later one didn’t have that HT winding tap.
Both sets of transformers seemed to work and sound the same - brutal!

 

[StingRay85]

I’ve had 2 SC 50+, an earlier one with partridges (PT has the lower voltage HT winding tap), the other with what I suppose are EV, it was later model with the phase splitter valve between the EL34s. The later one didn’t have that HT tap.
Both sets of transformers seemed to work and sound the same - brutal!

Makes me wonder if other Patridge transformers have this similar feature. I think all my 120s have at least 470V anode voltage at idle