[pabbi1]

OK, so I ordered wrong - for the c2, I ordered Cera mite ceramic, Mouser 75-440LD22... will this work?

Guessing not, since there is no polarity?!?

Argh.

Now, I have some .0039 or .0047uf - will either of these work, and what are the ramifications?

Is the fuse optional? If not, how can I wire a panel mount fuse holder?

 

[mono]

Yes those will work. In fact most PSU noise filters do use something more similar to those, yet Tangent prefers film caps for many areas in which ceramics will do (sometimes their greater temperature sensitivity is an issue, but relatively, those will be fine, is what they're designed for).

A fuse is technically optional, but a good idea to use one. Do you "need" a panel mount fuse holder? You seem taking opposite extremes, wondering if you need one at all at one extreme then wanting to go to extra effort (contrasted with the typical Steps implementation on the PCB itself) at the other extreme.

If you simply don't have a fuse holder for the board but do have the fuse, what you could do is solder thick wire "studs" to each end of the fuse. How to wire a separate fuse holder is quite simple- take the AC hot wire coming in, and instead of running it straight from the cord (or IEC socket, however you're getting the power into the box) to the Steps board, wire that to one end of the fuse (Holder) and the other end to the steps board. It just goes inline, in series with that AC lead. Mainly, be sure the wires have a good mechanical connection in addition to solder or crimp (whatever applies), and perhaps even heatshrink over that... you don't want those coming loose fuse or not.

 

[pabbi1]

Thanks... Do the C2 or C1/C3 have polarity?

And, yes, my parts store was out of the clips and only had panel mount, so I grabbed one... well, off to Fry's, and hoping...

 

[BrokenEnglish]

Quote:

Originally Posted by pabbi1 Thanks... Do the C2 or C1/C3 have polarity?

no... neither film nor ceramic caps are polarized. i'm using 3,3nf y2 ceramics myself in a netfilter similar to the one in the STEPS.

 

[tangent]

Quote:

Mouser 75-440LD22... will this work?

Yes.

Quote:

Is the fuse optional?

If you have a metal case and value your life, no. If you can live without the latter, then yes. Ahah.

Quote:

how can I wire a panel mount fuse holder?

Make the fuse holder break the AC line (L) wire between the inlet and the board.

Quote:

yet Tangent prefers film caps for many areas in which ceramics will do

The STEPS is a high-end design. No compromises. If that's not your goal, use an Elpac. Much simpler.

 

[mono]

Tangent, are you hinting that a Steps with Ceramics tested noisier? What real benefit is there in films here?

 

[tangent]

Quote:

Originally Posted by mono Tangent, are you hinting that a Steps with Ceramics tested noisier?

Not at all. They are, however, the best parts possible for the job without getting ridiculous. That's the point of the STEPS.

Contrast the Wellborne power supplies, which I think go over the line, particularly in the way of boutique capacitors. I'm no opponent of boutiques, but in a linear power supply, the size of the cap bank directly affects the amount of ripple, so I can't see why one would tolerate a smaller cap bank just to get an exotic brand label in there. In a headphone amp, where even 100uF is probably technically more than sufficient, this argument doesn't fly, so I don't oppose exotics there.

Quote:

What real benefit is there in films here?

Are you implying that audiophiles care about measureable results?

 

[ceRon]

may i use LT1085CT instead of LM317T as the regulator to drive my PPA?

 

[n_maher]

Quote:

Originally Posted by tangent Are you implying that audiophiles care about measureable results?

I love it when Tangent gets salty

!

 

[steinchen]

yes, but take care of your rail capacitance
http://www.tangentsoft.net/elec/teps/pguide.html#IC1

 

[ceRon]

Quote:

Originally Posted by steinchen yes, but take care of your rail capacitance http://www.tangentsoft.net/elec/teps/pguide.html#IC1

does it mean if i use just 1 cap. with 2000uF on the PPA, i cant use the LDO type regulator?
how about if i use 9 X 470uF cap.?

 

[steinchen]

LDO type regulators are sensitive to excessive rail capacitance, you should keep the total capacitance below 3.000uF, 2.000uF is fine

for ESR reasons it is better to take several medium sized caps (e.g. 4 or 5 caps of 470uF or 680uF each) than 1 big cap of 2.200uF

 

[ceRon]

Quote:

Originally Posted by steinchen LDO type regulators are sensitive to excessive rail capacitance, you should keep the total capacitance below 3.000uF, 2.000uF is fine for ESR reasons it is better to take several medium sized caps (e.g. 4 or 5 caps of 470uF or 680uF each) than 1 big cap of 2.200uF

that means i cant use that type of regulator since 9 X 470uF > 2000uF?

 

[tangent]

Quote:

Originally Posted by steinchen LDO type regulators are sensitive to excessive rail capacitance...2.000uF is fine

2000uF is not fine with an LM108x family LDO, in my experience. The regulator will not power up reliably with that much capacitance on its output.

Oh, you could do something ugly like put a 1 ohm resistor inline with the regulator's output to allow it to work, but then remind me why you're using a regulated power supply in the first place?

 

[mono]

I've been pondering over this LDO issue myself.
I DO have one running stabily with, hmm, let me count up the caps- 7300 mfd as 2 x 1500 mfd, 1 x 3300, 1 x 1000, (these all post-regulator) plus a couple ceramics and a couple tants.

The above wasn't a goal though, I had an old power board and started looking at where I could stuff caps, changing it from it's original design to a basic high-capacitance linear and planning on later swapping in a LM317- but AFAICT it's completely stable with the LDO still in it (all other old parts removed). The circuit design is nothing noteworth, pretty much a reference LM1084 but with more capacitance both before and after the LM1084.

I'm NOT suggesting "go ahead and do it", not at all. Rather I wonder if there's some trick to getting it to work that I'm just not seeing. There is definitely only minor trace resistance between the (1084 Adj. set to 13.2V with 21V incoming (no-load)) and the output caps. After hooking up a suitably-sized transformer it was putting out around 2.5A with a dummy load, seemed fine with that load, lesser load, and unloaded (nothing but an LED).