Power Transformers and chokes

[Nikon1975]

Dear All,

I got the the OPT makes a big difference, and it's quite clear why.

I also got that capacitors has to be chosen according o their function, and I found explanations for that here and there.

But what about power transformers and chokes ?

What are the unsaid characteristics that distinguish a good transformer or chokes from a bad one ?

I hear in my power transformer ot in the choke, from time to time, a noise as a rattle snake, I don't know if it's normal, but I don't think so.

Additionally, I unregulated PSUs have many variations, but basically are all based on RLC filters, so what is behind the choice of the L value, from what I understand the bigger the better, as far as you keep the oscillation frequency of the RLC low enough. Is there something I am missing ?

simply, if I want to filter the B+ of a driver, that has few mA, and for the same price I can have a 160H 10mA or a 30H 30mA, I would go for the first one.


Thanks,

Davide

 

[analog_sa]

so what is behind the choice of the L value, from what I understand the bigger the better

Or is it the opposite? It may be entirely coincidental but i seem to only like low resistance, low henry chokes and slighly underdamped power supplies.

What makes a good sounding power transformer seems almost impossible to define but they vary greatly in subjective quality. Same goes for chokes. I probably have 20-30 different chokes and the only ones i like are made by Lundahl. Some of the classic output transformers i've tried are really great sounding but strangely, none of the PS chokes.

 

[Glowbug]

My advice would be to download Power Supply Designer 2, and start playing around with configs 🙂

http://www.duncanamps.com/psud2/index.html

You'll find that there are very few absolutes, it's all about tradeoffs. Component choices have both positives and negatives that have to be weighed...I'm doing this right now, and it seems like every time I ask a question, it raises more questions than answers 😉

I hear in my power transformer ot in the choke, from time to time, a noise as a rattle snake, I don't know if it's normal, but I don't think so.

Sounds like lamination rattle?

 

[Kashmire]

It may be entirely coincidental but i seem to only like low resistance, low henry chokes

I find DCR beneficial in several of my power supplies, because the resistance creates a CRC along with the CLC "pi filter".

Any comments about pros/cons on operating a choke's core at nearly full saturation vs. oversizing the core? I have a magnetics handbook. I'll do some reading.

For my big solid-state power supplies (Pass amplifiers), I use air-core inductors, which do not saturate (no maximum rated current, other than voltage drop due to DCR).

 

[Glowbug]

I find DCR beneficial in several of my power supplies, because the resistance creates a CRC along with the CLC "pi filter".

Any comments about pros/cons on operating a choke's core at nearly full saturation vs. oversizing the core? I have a magnetics handbook. I'll do some reading.

For my big solid-state power supplies (Pass amplifiers), I use air-core inductors, which do not saturate (no maximum rated current, other than voltage drop due to DCR).

How much resistance have you found to be useful?

 

[Nikon1975]

I use a lot psu2, but my question was maybe simpler: let's say you have done the maths, now you go to your favourite iron dealer and you find three chokes, same values, different brands, different price. I am wondering, is there any reason why i shouldn t buy the cheapest?

Then, let's say you need a choke of 10H 180mA and for the same price you can get either 10H 300mA or 15H 200mA. Any reason why not to pick the 15H?

Thanks,
Davide

 

[zigzagflux]

The filter is an R, L, C circuit; as a result, not only does it perform attenuation of the 120 Hz ripple, but it also has a resonance and damping associated with it. So while the 15H choke might be beneficial in reducing ripple even further, you still want to consider resonance and transient behavior of the PSU. These issues are dealt with fairly well in Morgan Jones Valve Amplifiers, and in my experience following his suggestions makes for a more stable supply (one that doesn't ring and bounce around).

While there are advantages to a low DCR choke, modeling it in PSUD will show the tendency for ringing when the supply is hit with a step current change. Whether or not this affects the sound quality of the amp is open for discussion, but there is benefit from a theoretical standpoint to having some series R in the supply. This is further affected by SET vs. PP topology, as the power supply draw characteristics are different.

 

[Nikon1975]

I have ordered Morgan Jones book on Amazon last week, it should be on his way. I look forward to reading it.

I understand the resonance problem, and I always check the behavior with a 10 mA step, as I should not have anything bigger than that in a SE class A amplifier.
I don't want a PSU oscillating for a while after a transient.

But what about the choice of chokes and transformers with the same declared characteristic. Is a 10H 200mA choke that costs 40 $ the same (at least electrically) compared with a 150 $ one ?

Thanks,

Davide

 

[analog_sa]

Is a 10H 200mA choke that costs 40 $ the same (at least electrically) compared with a 150 $ one ?

Not necessarily. The chokes have obviously different parasite capacitance which will determine whether they continue being chokes at high frequencies. Perfect winding techniques will minimise this. Adding a separate RF choke seems to work as well although i've never tried it.

Are the chokes you are comparing with the same dc resistance?

 

[Nikon1975]

Yes, the chokes I am speaking about have more or less the same DCR, let's say around 80-90 ohm for a 10 H.
I am looking at Tango, Hashimoto and Noguchi.

Best Regards,

Davide

 

[torrence]

A perfect power supply would be a constant voltage source, which has zero output impedance (same voltage for any load). Chokes with large DCR add to the output impedance of a power supply, which might be OK in a class A amp where the load is nearly constant, but is very problematic for class AB.

Generally speaking, physically larger chokes are better, both because the inductance is higher, but also because the higher current rating is related to a lower DCR. Of course these become very large, heavy, and expensive.

As many others have said, PSUD is a very good way to test the advantages and disadvantages of different configurations. There is no best solution for all problems, just different optimizations.

 

[analog_sa]

let's say around 80-90 ohm for a 10 H.

This seems really excessive. The LL1638 i use everywhere are about 32ohms for the same inductance.

 

[Nikon1975]

I checked the Lundahl, they are very low in DCR, but depends on the current rating. My choke is 10H rated 300mA and has 81 ohm. The Lundahl 200mA has 60, I guess you have the 100mA. It makes sense. Actually I wanted to try Lundahl for the possibility to use the common mode rejection configuration. Has someone ever tried ?

D.