Input chokes

[sonicman]

Hi all. I'm new to electronics and trying to learn, so please bear with me if this is a silly question, but it was recommended to me to consider using choke inputs for a power supply I'm building. I was looking at Figure 4 seen at this link http://www.tnt-audio.com/clinica/ssps2_e.html and was wondering if anyone has tried to build this and incorporate chokes, and what did you find out.
Thank you.

[pacificblue]

For an amplifier those chokes are interesting, if your mains are very dirty. That can be, because you live near an industrial zone or because in your vicinity many dimmers, smps and similar devices contaminate the net, etc.
Those chokes are not too expensive, so buy and try its effect, if you suspect dirty mains. Important is to buy the right one. Filters have to be rated exactly for the current the amplifier draws. In this case bigger is not better, because that leads to a mismatch of the filter corner frequency. Don't buy a filter for more than the maximum current your amplifier draws. If you can find out, how much your amplifier draws typically, not peak, then go for a filter that matches the typical average current.

[pacificblue]

http://myweb.tiscali.co.uk/nuukspot/...nditioner.html

[tubesguy]

I confess that I'm confused, but I don't know which chokes are referred to as "those chokes" in the above posts. Dirty mains are a problem usually addressed by common-mode chokes, on the primary side of the transformer. They are inexpensive, and you can certainly try to incorporate them.

A "choke-input" power supply, on the other hand, has an inductor as the first element on the secondary side of the transformer. Whole different kettle of fish, and one that requires a total redesign of the supply, compared to a capacitor-input supply. Much lower voltage output, but there are advantages to be had. Cost is not one of them.

Check the Hammond website. It has a page that shows the various combinations of rectification schemes and cap vs choke inputs, and it will get you started. - Pat

[ostripper]

By "chokes" do we mean those ferrite cores
on the mains with 5-20 turns
of wire?? I thought they were for RFI rejection.
OS

[sonicman]

Originally posted by tubesguy
I confess that I'm confused, but I don't know which chokes are referred to as "those chokes" in the above posts. Dirty mains are a problem usually addressed by common-mode chokes, on the primary side of the transformer. They are inexpensive, and you can certainly try to incorporate them.

A "choke-input" power supply, on the other hand, has an inductor as the first element on the secondary side of the transformer. Whole different kettle of fish, and one that requires a total redesign of the supply, compared to a capacitor-input supply. Much lower voltage output, but there are advantages to be had. Cost is not one of them.

Check the Hammond website. It has a page that shows the various combinations of rectification schemes and cap vs choke inputs, and it will get you started. - Pat

The diagram I referenced does not use the chokes I was considering. I am asking about chokes as an inductor on the secondary side. The idea that was proposed was to place something like a 10H choke after the rectifier and cut back on the capacitance. Does this make sense? You mention the first element on the secondary, but many of the schematics I'm seeing have them "after" the bridge. Thanks.

[pacificblue]

My answer was about the common-mode chokes used for RF filtering.

The choke on the secondary is a different thing. A "standard" unregulated supply uses one or several pairs of capacitors to create a 6 dB ripple filter. With a resistor or coil in series before the capacitors you turn that into a 12 dB ripple filter. Add another pair of capacitors in front of the resistor or inductor and you get an 18 dB ripple filter.

As usual it is a question of finding one's personal best compromise. Higher order filters have a better filtering effect vs. higher component coount (= price) and effort. Some people, who have tried them, are very happy with the effect, others feel that they slow down the amplifier.

It certainly depends on the PSRR of your amplifier, if a higher order power supply filter is worth the effort. Voltage drop should be low, if it is well designed, therefore a choke is preferable to a resistor.

If you have a picture of a 10 H choke, do post. I am surely not the only one, who wants to see that.

[sonicman]

Here's a link to the Plitron chokes I was looking at. Looks like 10H and 20H in size.

[pacificblue]

There is no link in your post, but I searched for them and found them. The size of a 1300 VA transformer, but look at the current rating: 0,15 A. That will lead to incredible 0,18 W of output power with an 8 Ohm speaker.

If you want to use chokes for your power supply, the math is the same as for speaker crossovers. The goal is to create a low-pass filter that filters 50/60 Hz out. The load resistance is your supply voltage divided by the output current, which will be similar to the speaker impedance. You will end up with something in the mH range, and the choke should be rated for the peak output current.

[sonicman]

Hmmm... I see what you mean about the curent rating. So, am I on some wild goose chase then? My friend pointed me to http://www.sdinfo.com/volume_4_2/nelpass.html that discusses putting "pi" filters (large inductor w/ caps) to reduce ripple noise. This seems to be a CLC filter no?