• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

PSU Design question

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I am using PSU Designer II to model a PSU. My PSU will feed both channels of an amp.

Should I model the load as a resistance with value equal to the paralleled output resistance of the last stage of each channel? Or should I use only the load (OPT 1ry) resistance of each channel in parallel?

There is also the possibility of modelling a constant current load, but I thought that was not realistic as under dynamic conditions the curent varies widely, while the plate resistance does not vary that much.


Hi Radames,

The constant current load option is the most useful because it's the current figures we normally have to hand. The simulator can only deal with a static load anyway.
If you are designing for class B or AB you can make 2 simulations, for no signal and full signal. You should know the current for each case.


I apologize for the lack of information.

SE 211 amplifier with cathode bias, 70mA quiescent curent at 980V or so. Lundahl 11.5KOhm OPT.

I am looking for a full-wave bridge rectifier RCLCLC PSU and wanted to determine the right mains transformer and components. I am evaluating the cost of different components and the PSU Designer seemed a useful tool to plug different things in and out.


I am looking for a full-wave bridge rectifier RCLCLC PSU and wanted to determine the right mains transformer and components.....

Well, i think psu designer is not so good for determining your power transformer....
A big deal of the calculations inside psu designer is done with parameters you have to Measure on your transformer (like off-load voltage...dc resistance prim etc...)
You CAN use it to get an indication of what you need but you still have to do a little guessing...

Or am i incorrect???? and IS there a way to determine your transformer with PSUD?????
Guessing transformer DC resistance


have a look at some other transformer websites for similar transformers, most likely some will give a resistance. Put in that figure, then put in half and double the figure. You'll find it doesn't make a big difference.

If you want my guess, try between 150 and 300 ohms as a starting figure...

You've mentioned before that when operating in class A, the current will keep constant. When signal comes into the tube, isn't it the operating point swing on the loadline and the current will change? Should we consider the current swing when designing the PSU for a class A amp?
Hi Family,

This is not the first time that I've fallen into this exact trap by my imprecise use of language.
What I should have said is "The average current stays constant". This is due to the presence of the smoothing capacitor.
You are quite right to question my statement, and without such a capacitor, what you say would be true.

My apologies.
Family and Dharen,

Yes, I got the constant curent approximation. That's why it is important to design the PSU so the last cap (and the resistance it sees), can deliver constantly to the peaks of the music at the frequency of those peaks.

There is a nice discussion about the value of the last cap and the total resistance of the PSU, but I am not finding the site now. I'll try to fetch it and post it later.


PSU last RC


I cannot find the article I was referring to. But basically, the idea is that every time you get a peak of the music that requires energy beyond the operating point, the last capacitor of the power supply will provide that energy.

To do that, the cap needs to have the energy stored. Therefore, the capacitor charge time needs to be such that alows full re-charge after one peak, before the next peak arives.

If we consider bass peaks in, let's say dance or rock music, that's a frequency of around 150 - 300 bpm (beats per minute). The capacitor needs around 5*R*C to fully recover after complete drain, so I tried to design my PSU so that 5*R*C is way below 300/minute or 200msec. In fact, try to keep it as low as possible. To do that, I try to choose the chokes with lowest possible DC resistance, and the mains transformer such as not much R is needed to drop the voltage to the required B+. Due to voltage issues I am limited to a 50uF capacitor, but I think that's OK.

All this design constrains are, of course, superimposed to the ripple reduction necessary by my stages, taking into account their PSRR.

I think this makes some sense, but I may be wrong. If there is some obvious reasoning flaw with this, please let me know.



You are well on the way to designing a good power supply:)
The energy stored in the final cap will be the major source of current for the output stage during conductive music cycles. But remember, if you've got enough capacity to reduce the rectified DC ripple sufficiently, you've also got enough energy stored for music cycles.

Please excuse my term"music cycles" I hope you know what I mean:xeye:

You are quite right to aim at the lowest DC resistance in the power transformer and choke, but we are often forced to use what is available, and pretty much all that's available can be made to work. In fact with class A you can substitute the choke(s) with fixed resistor(s), as long as you can get high enough values of C. The resistor(s) may dissapate quite a lot of heat, so consider where you'll mount it/them.
I am sure you are on the right track!

About a month or two ago I had exactly the same difficulties you have but after playing with PSU2 for a while I basically worked out all of my questions. PSU2 is superb!

If you don't have the parameters of your transformer you can simply click on the RMS V botton and it will work out an approximation for you that is reasonably accurate (I guess +-20%). Without entering the parameters things can get dramatically wrong. I fell into that trap and it took me a long time to realise it.

In fact, you won't buy a transformer first then model it with PSU2. You use PSU2 to determine what transformer you need. That is my logic anyway.

The higher DCR of a choke may not be a bad thing because you may need it to further reduce the ripples. Of course, nothing is right or wrong and things must be taken under the contexts. Higher DCR gives better damping and reduces ripples but at the same time burning out (wasting) power. The whole PSU must be design as a unit in whole and components chosen to work together.

My preference for my valve preamp is to have an overrated transformer with higher voltage, tube or solidate rectifier, choke input possibly with a snubber (cap) of a small value, then a good quality cap (long life, high ripple current and low impedance), then appropriate RC filters to get the DC completely flat, if no regulator is in use.


I have some stupid questions. You mentioned the equation 5*R*C:

1) The C you mean is the capacitence of the LAST cap or the sum
of all the caps in the PSU?
2) How to dertermin the R? The sum of the resistance of the trans,
the choke and the ESR of the caps?

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.