Duncan's PSU, 12AX7/6V6/EL34 plate current?

[Normster]

I've been fooling around with Duncan's Power Supply Designer and I'm not too sure about the values for plate (and screen) load for each tube type. For example, if a 12AX7 tube chart shows 1.2mA, do I use 2.4mA to represent both triodes as a load? I've seen many existing models that show 1.5mA for a single 12AX7 so it doesn't make sense to me.

Also, when entering values for pentodes in a push/pull circuit, do I enter the current draw for both tubes combined or is it some other value based on PP operation? (i.e., 70mA for a 6V6 or 140mA for the pair.) Same question for the screen current.

The reason I ask is that I'm trying to design a PS using a 5AR4 rectifier that can be used with both 6V6 and EL34 output tubes. The simulation shows that I exceed the limits of the 5AR4 with EL34s but I know the circuit should handle it based on Marshall and Fender circuits I've seen in the past.

Sorry to be such a newbie, but this is the first time I've ever tried designing around a tube rectifier and I'd like to verify the operation in Duncan's simulator before I commit it to a circuit board.

Thanks,
Norm

[skyboltone]

You need a tube manual. Your 12AX7s are probably not goint to use that much per side. I would say total. Subtract the plate voltage from the voltage on the other side of the plate resistor (say 75 volts?) and devide that voltage by the resistance of the plate resistor. That's your current for that triode. Do that for all stages other than the output. I'm not certain that's the best way for the PI but it seems to give a reasonable current.

For the output tubes enter a stepped current tap separated by ten seconds or so. The first entry is the 0 signal current off the tube sheet. The second entry is the full signal current at full power.

Set up the time base of the simulation for 1000 miliseconds after 2 seconds, for the 0 signal load, and 10 seconds for the full signal load. Things will fall together better now. You can see that you will probably be over the top of your transformer specs at full signal but remember unless you are building a siren or an organ amp you won't get to full power often.

good luck
Dan

[Normster]

Thanks for the excellent explanation, Dan! It makes perfect sense now. The transformer is a pretty hefty piece of iron from Moose. (300mA) It's the 5AR4 that I'm having difficulty with. I'll plug in the numbers using your method and see what happens. I want to build the amp primarily as a 6V6 but thought it might be nice to know if I could pop EL34s in.

Thanks again,
Norm

[skyboltone]

How are your noodlings going Normster?

I've got this 375-0-375 Hammond organ transformer I got off of Ebay and I'm trying to model a supply using Kt66 finals. It's rated at 250MA but I figure that's got to be conservative because it weighs 8.5 lbs. Everything else will be Rev A except for running the KT66s ultra linear. Hey, Dumble says his Distortion is all in the pre amp. It's worth a try. You can run KT66s up to 550 on the plates and they are one of the few tubes that will run ultra linear.

Anyway, I've modeled the Rev A power supply every which way from Sunday trying to ascertain the loads of each stage. Change one node and all the others change as well eh? Clearly the way I told you isn't working. You may be able to tell tube current that way but there are obviously losses in the plate resistors and cathode resistors as well. Not to mention the PI which doesn't follow any easy rule I can figure.

Anyway, he he, what I need is somebody to lift the B+ leads at each Power supply node and insert a milliamp meter between. Let's find out what the zero signal loads are for clean and OD and PI stages. The output stages are fairly simple to model and once signal is added everything starts flopping all over the place anyway.

Anybody out there with an an ODS and a keen interest in helping others along in their design efforts want to experiment a little??

Thanks in advance

Dan