Felix meets the Squeezebox

GBB · « **Reply #60 on:** 18 Feb 2008, 01:11 am »

Quote from: BradJudy on 17 Feb 2008, 11:59 pm

Point to point is more compact, cheaper and pretty easy for this project.

You forgot sounds better aa .

---Gary

mikef · « **Reply #61 on:** 18 Feb 2008, 01:12 am »

In answer to the earlier question by Tan, my Felix runs cool to the touch. I also agree with Gary that a circuit board really isn't necessary, and the breadboard I used wasn't needed either, but made the end result look a little neater.

I'm looking forward to how this little thing will sound once the caps get some break-in time on them.

Tan - I also like your all in one box with the wall wart inside. Maybe I'll re-do mine to copy that.

Mike Fox

tanchiro58 · « **Reply #62 on:** 18 Feb 2008, 02:17 am »

Quote

Tan - I also like your all in one box with the wall wart inside. Maybe I'll re-do mine to copy that.

Mike,

You need to have a little bigger box than my aluminum box in the picture before you attempt to put Felix (as Gary's picture) and a stock switching PS in (DO NOT try to squeeze them in)

Tan

Folsom · « **Reply #63 on:** 18 Feb 2008, 02:56 am »

PCB board does allow for easier, safer, mounting however. I do not plan on using one, but for some folks it might be preferable.

mikef · « **Reply #64 on:** 29 Mar 2008, 08:20 pm »

As you may recall, I built Gary's implementation of the Felix for my SB ps, and am really enjoying it, prefering it over the linear power supplies I had before.

I'm thinking about trying to build a Felix after the Squeezebox switching ps, on the DC side, to complement the one I built on the AC side, based on previous discussion of this idea. Has anyone implemented this, and if so, what were the values of the CMC and capacitors you used? And what were the listening results?

Also, what do you guys think about parts for this project -
Should I use the Coilcraft P3717-A CMC or 30 mh 2.3A JWMiller 8109 CMC?

For caps Paul (Occam) suggested in an earlier thread using a Jantzen 400vdc CrossCap .47uf paralleled with a .1uf on both sides of the CMC. Any other suggestions?

Thanks in advance for any help.

Mike Fox

Folsom · « **Reply #65 on:** 29 Mar 2008, 08:44 pm »

I am still running a (nice) linear power supply with a filter (can check my topic about it). It is not a Felix but...

TO ANYONE CONSIDERING SOME TYPE OF FILTERING IT IS WORTH IT IN A WAY THAT USUALLY WOULD COST HUNDREDS OF DOLLARS TO GAIN THE DETAIL YOU MAY POTENTIALLY GET FROM SOME FILTERING!

The subtle jump, is really huge, in the difference in sound. As far as small noticeable things, stuff that is detail that most of the time is not trying to be approached for realistic function of increasing, even on commercial expensive stuff, I have had no bigger change than filtering. Now sonically that is an entirely different story as all my capacitors in signal are Blackgates with the exception of two Sonicaps, and nice wires, big differences, but in different ways. Also filtering has smooth out most of the sound, the Squeezebox DAC shines a lot more as far as being smooth. I am much more inclined to just sit and listen to an entire album as if it where vinyl due to the ease of listening. I did before but it was more of a technological experience from enjoying aspects of terms like imaging, dynamics, etc, not so much just pleasing to the ear until now. Granted I did listen to music for sweet beats, lyrics, riffs, etc, too, now it is just soothing.

I have not tried the switcher just because my linear PS is so nice I have trouble believing that the massive gains from my linear PS before filtering would somehow be lost and recreated much stronger with a switcher. I have looked in the switcher and frankly it does not have the beans to match a very nice linear PS, from what I can tell. Now a few modifications to a switcher might make it worth your wild. My switcher already has a noise suppression cap in it for instance. However some fast switching diodes and new capacitors, especially a large bank for the output would be appealing. However while I am not sure, someone else can confirm or disconfirm, that the single little diodes probably could not feed some large banks of capacitors.

My linear has four large surface mount transistors (to heatsink), multiple diodes, etc.... (adjustable to 15v) so the huge capacitor bank is backed up well.

kyrill · « **Reply #66 on:** 30 Mar 2008, 12:05 pm »

i was pondering on Gary's observation that a basic but good linear pws sounds ( with or without Felix) less than the original; cheap SW PWS with the Felix in the AC line..
He assumed logically that SW PWS negative effects were "blocked" in a way by the Felix to enter the other audio components.

Suppose you do the Felix in the 5V line. It will "block" all the unwanted interference from the Ac line PLUS the unwanted effects from the SW PWS for the SB3 but will not stop those SW effects for the rest of the audio components

If now the SW PWS still sound better than the linear PWS, another explanation must be found than Gary's original assumption. If not, the assumption is proven to be fact, no?

mikef · « **Reply #67 on:** 30 Mar 2008, 01:08 pm »

Quote from: kyrill on 30 Mar 2008, 12:05 pm

i was pondering on Gary's observation that a basic but good linear pws sounds ( with or without Felix) less than the original; cheap SW PWS with the Felix in the AC line..
He assumed logically that SW PWS negative effects were "blocked" in a way by the Felix to enter the other audio components.

Suppose you do the Felix in the 5V line. It will "block" all the unwanted interference from the Ac line PLUS the unwanted effects from the SW PWS for the SB3 but will not stop those SW effects for the rest of the audio components

If now the SW PWS still sound better than the linear PWS, another explanation must be found than Gary's original assumption. If not, the assumption is proven to be fact, no?

If we can assume that a Felix on either side of the switching power supply blocks unwanted interference from getting into the AC line and from getting into the Squeezebox, unless the SW PS is radiating RFI, how else can the interference get into the other components?

Mike Fox

kyrill · « **Reply #68 on:** 30 Mar 2008, 01:11 pm »

as i understand it the Felix cannot block SW noise entering the other audio components when placed after/behind the SW PWS, so when placed in the 5V line

mikef · « **Reply #69 on:** 30 Mar 2008, 01:33 pm »

Oh, I was thinking about 2 Felixes used at once, before and after the SW PS to block both sides.

Using only 1 on the DC side might be an interesting experiment. For me, all of my components are plugged into a PS Audio Quintet, so I think that is already a big help on the AC side.

If anyone has read my earlier post (5 posts before this one) I'm still looking for advice on components to use for a Felix on the DC side....

Mike Fox

kyrill · « **Reply #70 on:** 30 Mar 2008, 01:50 pm »

Hi Mike
on the DC side you dont change the Felix, But because it is low voltages DC all danger is considerably lessened. So you can experiment with (much) better sounding film caps ( auricap, Mundorf or the Platinum Gen 1l) or even expensive teflons

tanchiro58 · « **Reply #71 on:** 30 Mar 2008, 01:52 pm »

Mike,

I am doing Felix before and after switching PS (5VC first then 9VDC for Duet) in a few days. Let see what will happen to the sound.

Tan

mikef · « **Reply #72 on:** 30 Mar 2008, 03:19 pm »

Hi Tan,

Looking forward to your results. Let us know what you end up building on the DC side.

Mike Fox

kyrill · « **Reply #73 on:** 8 May 2008, 08:43 am »

Quote from: tanchiro58 on 30 Mar 2008, 01:52 pm

Mike,

I am doing Felix before and after switching PS (5VC first then 9VDC for Duet) in a few days. Let see what will happen to the sound.

Tan

Hi Tan

any news in how it sounds?
kyrill

art · « **Reply #74 on:** 8 May 2008, 05:02 pm »

Has anyone measured the effects of 0.6 A DC through one of those CMCs???

No.............didn't think so.

Pat

kyrill · « **Reply #75 on:** 8 May 2008, 05:52 pm »

hi Pat

Where does this DC comes from?

BradJudy · « **Reply #76 on:** 8 May 2008, 06:35 pm »

Quote from: kyrill on 8 May 2008, 05:52 pm

Where does this DC comes from?

I think he's referencing using a Felix style filter between a DC power source and the device, as people are discussing in this thread.

Occam · « **Reply #77 on:** 8 May 2008, 07:04 pm »

Quote from: art on 8 May 2008, 05:02 pm

Has anyone measured the effects of 0.6 A DC through one of those CMCs???

No.............didn't think so.

Pat

Hey Pat,

Good to see you haven't given up you curmudgeonly ways.

Think again.... and realize its not being used as differential mode choke, but a cmc.
Then again, iff'n you forget to hook up the return coil (or hook it up backwards), you magnetize it, and you've got to demagnetize it with a variac, sorta like us old folks did with bulk tape erasers

.

Nice to see you posting,
Paul

kyrill · « **Reply #78 on:** 8 May 2008, 10:24 pm »

Hi P&P
can you elaborate on this? What are you suggesting?
what should we do if we use it in a DC line? hooking backwards?

Occam · « **Reply #79 on:** 9 May 2008, 11:38 pm »

kyrill,

A cmc is simply 2 equal coils wound on the same permeable core. Its configured such that magnetic flux generated by a current flow imposed on one coil is canceled by the flux imposed on the second (return) coil (save for its leakage inductance). When fed by a single ended or balanced signal, AC or DC, the fluxes cancel,(assuming you've wired it correctly, save for that previously mentioned leakage inductance. When presented with a common mode signal, those fluxes add presenting a large inductance. But note, that inductance provides a frequency dependant impedance which is exactly what we want; that impedance increases with frequency, attenuating hf noise.

That leakage inductance, to whatever extent present, provides a frequency dependent attenuation of differential/transverse mode noise, increasing in impedance with increasing freqency. Along with 'across the line', X caps, it forms a 2nd order low pass filter.

Regards,
Paul