PSUD II help

[Bengali]

Hi,

After simulation is complete, how do you display ripple voltage?

Right now, I just use 5sec with 0 delay. if I zoom it, I don't see anything. Do you use the delay?

Attached is psu circuit for reference. I(L1) shows perpedicular line, what does this mean?

What else should I be looking for other output voltage?

This is for an Aikido Line Stage, 6CG7.

thanks.

[zigzagflux]

For starters, you might want to go to just a 1 second simulation, for better resolution on the screen.

To see your ripple, first of all, click the V(C4) checkbox. The output voltage will then be displayed. If ripple is high enough, you can see it right in front of you.

Otherwise, drag to zoom on the graph, at the area under concern. Your ripple looks VERY low on this circuit. The buttons in the upper right of the graph help to zoom in/out/extents.

To get an actual number easily, put in a delay of 2 second, and simulate for 50 msec. You can then either look at the graph directly, or see the Diff column for a peak to peak number. I get 1.7mV peak to peak with your circuit.

[Bengali]

Thanks!

The key was to 'drag to zoom on the graph, at the area under concern'.

[Bengali]

Should I be concerned about the peak for V(L1) ?

Any suggestions to improve this circuit or it is okay?

Ripple seems to be in the uV.

What is a rough # to use for the transformer resistance?

I will use the Hammond 269JX and don't have one yet to measure.

Thanks.

[zigzagflux]

I would not be concerned about the peak for L1. This is in large part a product of PSUD having a constant current source drawing amps while the supply voltage is coming up. In reality, your tube circuit will not behave in this manner.

As far as improving the circuit, that's subject to personal preference. In general, your circuit should work; an Aikido does not draw a lot of current, and the consumption is fairly constant. I would think you might have a lot of parts involved in an attempt to keep ripple as low as possible. Keep in mind one of the benefits of the Aikido is its high PSRR. A little bit of ripple won't kill you, IOW.

My personal preference is to keep the number of stages low; you have four. I would not go over three, but again, that's personal preference.

If I were to suggest anything, it would be in the first stages. You are running a high cap input, at 200 uF, and the choke doesn't enter in until the later stages. This somewhat defeats the purpose of the choke; you already have low ripple on C2. In conjunction with this, a high C input results in high peak currents from your transformer; this will result in additional heat, audible transformer buzz (especially with the Hammond 200 series), and unwanted harmonics. You would help yourself by lowering the input capacitance (significantly), going with CLCRC. I typically size C1 for less than 1uF.

As you mentioned, you don't have the transformer yet; so maybe buying a different ratio to compensate, for the benefit of smoother current draw, would be suggested.

We can address the transformer impedance a little later, maybe provide some feedback if changing the design is something you are willing to move forward on. As well, others' opinions to either refute or support my thoughts might give you some confidence.

Cheers.

[Bengali]

Thanks for the input. You are right that I may have too many parts and the Aikido does not necessarily need such a low ripple.
I will try the CLCRC model and play with that.

There is a an option in the PSUD to turn on/off Soft Start. When is this option used?

I was also considering use of the 5AR4 for Full or Bridge. Do you know if the 5AR4 has the characteristics of the soft start?

[Bengali]

For C1, why such a low capacitance, 1uF or below?


I tried the following:

T1= 300V 100ohms
C1= 1uF
L1= 15H
C2= 40uF
R1= 5ohms
C3= 40uF
Load 35mA

I get a ripple of 258mV.
What's an acceptable ripple volt range for the Aikido?

How does the performance improve by adding a shunt current regulated circuit?

Thanks.

[zigzagflux]

For C1, why such a low capacitance, 1uF or below?

Good question. The idea is to come very near a choke input supply, which is the most favorable with respect to harmonic spectrum, peak currents, and transformer efficiency. By using small increases in the value of C1, you can easily tweak the DC output voltage with a pretty wide choice of transformer.

PSUD really is nice to see the effects this capacitor has on the transformer current. For preamp duty, I've had some very nice looking currents, smooth with no peaking, with the addition of a little R immediately after the rectifiers. Purely a personal choice. You will also see a much more sinusoidal inductor current when you keep the input capacitance low.

I get a ripple of 258mV.
What's an acceptable ripple volt range for the Aikido?

I'm not an Aikido user, but you could search around here or go to Tubelab's web site for more info on ripple.

How does the performance improve by adding a shunt current regulated circuit?

Do you mean by adding the current source in PSUD, or by adding a series CCS feeding into a regulator tube?

[Bengali]

quote:
Do you mean by adding the current source in PSUD, or by adding a series CCS feeding into a regulator tube?

A series constant current regulator(tube or s.s.) after the CLCRC,
nothing to do with PSUD.

For power supply caps, what's the preference, high quality electrolytic, paper in oil or propelyne, or this just changes how it will sound. or depends on the cap values.


thanks.

[Sheldon]

Originally posted by Bengali
For C1, why such a low capacitance, 1uF or below?


I tried the following:

T1= 300V 100ohms
C1= 1uF
L1= 15H
C2= 40uF
R1= 5ohms
C3= 40uF
Load 35mA

I get a ripple of 258mV.
What's an acceptable ripple volt range for the Aikido?

How does the performance improve by adding a shunt current regulated circuit?

Thanks.

You can reduce the ripple quite significantly by increasing R1.

Look up the power supply schema for some of the versions of
Aikido reported here. Model the supply and see how it compares with yours.

A series or shunt regulator would dramatically lower the ripple, but as stated, you won't need that for good noise performance with the Aikido. My or may not sound better to you, though.

Sheldon