FELIX questions for my setup

anthony a. · « **Reply #60 on:** 10 Apr 2009, 04:04 am »

Quote from: JoshK on 7 Apr 2009, 01:46 pm

P.S. I just realized yours doesn't have fuses. I wouldn't test this as is. Also, the ground is only partially covered. This is no problem if this were incased with strain reliefs. But as is, the wire could be moved by force and cause a short. This is somewhat dangerous in its current state.

thanks for the info josh. i debated for a while in regards to fuses or not and ultimately decided against. i may end up putting them in for safety. i also realize what you are saying about the ground not fully covered. the pic shows the "prototype" felix and thus has exposed wiring and no strain reliefs. im just going to test this out on my amps and see if it improves. just wondering if it is safe to connect this filter directly to the wall using a power cable and plugging in a light bulb to test if the circuit is functioning. will this be okay and a simple enough test to ensure the circuit is fine? and i guess i would leave it plugged in for a few minutes? please let me know if this is a viable test. thanks.

kyrill · « **Reply #61 on:** 10 Apr 2009, 07:44 am »

that would be a good test for the soldering and the coils

JoshK · « **Reply #62 on:** 10 Apr 2009, 05:34 pm »

That is an ok test to do. Paul and I did the same thing back when we built our first Felicia. The problem though without fuses is if you managed to wire up the cmc wrong, as both Paul and I managed to do independently, you'll trip the breaker but maybe not before taking out a couple parts. If you put in a small fuse, you'll likely not ruin the parts, just the fuse. Its your call.

kyrill · « **Reply #63 on:** 10 Apr 2009, 05:45 pm »

with the board it is not possible to wire the cmc wrong, isn't it?

JoshK · « **Reply #64 on:** 10 Apr 2009, 06:04 pm »

Quote from: kyrill on 10 Apr 2009, 05:45 pm

with the board it is not possible to wire the cmc wrong, isn't it?

Your probably right....I do it all p2p, so its different for me.

anthony a. · « **Reply #65 on:** 11 Apr 2009, 02:47 am »

okay great, im gonna try it out that way and see what happens. btw, a standard light bulb socket only has a hot and neutral, no ground. is it okay to bypass the ground from the bulb?

JoshK · « **Reply #66 on:** 11 Apr 2009, 03:52 am »

yeah, its not needed for the purposes of testing.

anthony a. · « **Reply #67 on:** 11 Apr 2009, 05:05 am »

great, thanks. so it turns out that i tried it out on my amps and using the miller 8118 choke, i find it does limit dynamics to a small degree. i do however love the black background it provides, so i will try out the 8120 and hopefully regain all the dynamics. i am hoping it achieves what i want, as the felix so far sounds very promising. one thing i did notice however was that there was an ever so slight buzzing emitting from the felix. this was only audible if you put your ear within a foot or so, but still audible. another thing i noticed is that my amps now hum much louder than usual. when i unplugged the felix it was definitely less. i guess this is why people use dc blockers? what can i do to reduce this hum i hear when the felix is connected? thanks.

BradJudy · « **Reply #68 on:** 11 Apr 2009, 01:15 pm »

You CANNOT use a larger JW Miller CMC on the PCB. The traces are not substantial enough to handle more current and I intentionally did not include pads that fit the larger CMC's (so you can't do it on the PCB anyway). If you want to try an 8120 point-to-point, go ahead, but be safe.

anthony a. · « **Reply #69 on:** 13 Apr 2009, 05:25 am »

i know, i will be doing point to point for the 8120 millers. just as an update, i tried the miller 8118 circuit on all my sources before they are plugged into the balanced power transformer. as mentioned earlier, i found the felix (on the amps) to offer a beautiful black background but at the minor expense of compressed dymanics. this held true on the sources as well, but i think i will keep one attahced to the preamp. althought it compresses just a smidge here (much less than everywhere else), it does offer a more solid and layered presentation which i like. so my next experiment is to try the coilcraft circuits on the sources and see how they compare to the millers. in the meantime, i just got a spare ac/dc power adaptor for my pc which i plan on installing 1 or 2 felixes inside. the adaptor is rated as follows: ac input - 1.5A................ dc output - 3.42A. so i have built 2 felixes, one for the input (coilcraft 3717) and one for the output (coilcraft 4007). below is a picture of the internals. so here are my questions:

i want to confirm which is the input and output. (the left plugs into the pc with the barrel type connector, and the right plug attaches to a power cord that goes directly into the ac receptacle). so input would be the right side (3717 felix attached here with cable plugged into ac outlet) and output would be left side (4007 felix there attached with barrel plug into pc). is this correct?

daz_bike · « **Reply #70 on:** 13 Apr 2009, 05:51 am »

Not good to hear that it impacts that much on dynamics .... esp. considering the dynamics are already heavily compressed with new CD's! This suggests my held belief that conditioning 'robs' the music to some degree. Maybe it's a nice change initially but in the long run not so good?

Occam · « **Reply #71 on:** 14 Apr 2009, 04:14 am »

Anthony,

Button up the switching supply in its (plastic?) enclosure. You can get at everything you need to via the C7? inlet and dc output cord. Simply build you ac conditioning filter into whatever cord you're using like the photo a couple of pages back of Mjosef's filter/cord. Same for the dc filter, use something like that RadioShack box, put an inlet that accepts the powersupplies output cord's plug. and run cord from the output of the dc filter and terminate with that same plug.

The coils you chose should work quite well. The dc filter should have no input caps, and you don't have to use line rated x2 caps. Anything up to about 10-30uf on the output side should be fine. Make sure the cap's voltage is at least 25% over the switcher's output voltage, and if you use electrolytics, be sure to use the proper polarity.

It looks like your AC inlet is a C7 or C8 IEC, rated at 2.5 amps max. Given your choice of AC Felix coil and its rating of 3amps, you really do have a valid excuse for not fusing it iff you build it into a cord whose output plug is a C7 or C8.

Looking forward to your take on the 8120 based Felix. What exactly are your components?

FWIW

daz_bike · « **Reply #72 on:** 14 Apr 2009, 04:56 am »

Occam,

Thanks for your reply and taking the time to deduce my post. You are right that I have Nuforce Ref 9SE V2 amps and I find it interesting your recommendation. I will follow that through and use the recommended CMC. My source is a Modwright Transporter and the figure I quoted was of 250 Watts was from Dan Wright.

What size do you recommend for this component (again 230V land)?

I will bite my lip until I have built a Felix for these components and then I can hear for myself> I'll report back to the group. I love 'humble pie' btw. Regards

Occam · « **Reply #73 on:** 14 Apr 2009, 05:25 am »

daz_bike,

Sorry, I not my usual easy going self.

I'm a bit stressed with trying to finish my taxes....

-Paul

anthony a. · « **Reply #74 on:** 15 Apr 2009, 04:23 am »

Quote from: Occam on 14 Apr 2009, 04:14 am

The dc filter should have no input caps, and you don't have to use line rated x2 caps. Anything up to about 10-30uf on the output side should be fine. Make sure the cap's voltage is at least 25% over the switcher's output voltage, and if you use electrolytics, be sure to use the proper polarity.

paul, are you saying i should NOT use the standard felix board with caps for the output side, or that it is not necessary? reason i ask is because i have already soldered everything to each board and if it should NOT be done that i will desolder... but if it still would be fine then i'll keep them intact.

i am going to take your advice and just wire everything up using some inlets and power plugs. i tried making short runs of wire, but it began to be messy so i think i'll just stick with your idea. i may house them all in a single box for a cleaner look.

p.s. my amps are synthesis the theater monoblocks from italy. they are rated at 70watts per channel and use a pair of 12au7 input tubes and quad el34 drivers per amp. i am happy to report that the 8120 millers sound awesome and have no dynamic limits that i can detect. i will definitely keep them on the amps.

Occam · « **Reply #75 on:** 16 Apr 2009, 01:37 am »

Anthony,

The input AC side, and output DC side filters seem fine, and should work quite well. BradJudy's boards are an excellent way to implement those filters. Its only when you use inductors that don't fit on the boards do you need an alternative.
The advantage of filtering on the DC side is that typically, there is a higher current on the dc than on the ac line, and its higher current draw tends to minimize unwanted peaks in a filter's response.
Because of the higher current, I suggested a higher capacitance, and because its on the output side, the input caps of the dc filter aren't required to minimize noise coming from the fed component, as the AC input filter addresses that. I suggested different caps as the caps on the dc filter because they are not subjected to high voltage spikes like x2 caps on the ac side filter.

Its good to know the 8120 based filter is working well on your amps, and not suffering from compressed dynamics. I'm curious how a 8120 based filter works on your preamp. Although I assume your amp(s) draw more power than your preamp, this doesn't guarantee that your pre will work as you desire with the filter. It all depends on the specific implementation of the pre's power supply.

Regards,
Paul

anthony a. · « **Reply #76 on:** 16 Apr 2009, 03:17 pm »

thank you paul for the advice. i appreciate all the time you and others here have given to help me out on this project.

anthony a. · « **Reply #77 on:** 16 Apr 2009, 07:20 pm »

in response to your question about the 8120 on the pre, i actually did try it out against the 8118 choke and found them to be too close to call. my preamp is an accustic arts mk3 solid state unit that draws only a few watts so perhaps that would explain why the 8118 and 8120 sounded the same. my next experiment is trying out the coilcraft circuit on the preamp and see how that sounds. same goes for the pc power supply. i'll keep you all posted with my progress.

anthony a. · « **Reply #78 on:** 21 Apr 2009, 03:47 pm »

1.) since the power cable is a 2 prong non-polarized cable (with no ground), does it matter which way i wire this into the felix input (hot and neutral)? i will simply be cutting off the 2 prong connector and soldering the cable directly to the felix. on the other end of the felix, i would like to put an iec socket, which i realize will have no ground intact, but atleast i can use a better power cable if i need to.

2.) the other end of the ac/dc adaptor has the barrel type connector which shows which is marked + and - on the circuit board. again, if i cut this cable and solder it directly to the second felix (dc filter), which wire do i attach to the hot and which to the neutral?

thanks.

mjosef · « **Reply #79 on:** 21 Apr 2009, 11:05 pm »

That plug looks like it polarized, meaning one blade looks wider(wider one being the neutral) than the other, so you should maintain that polarity. And of course it usually essential on the DC side. THe Felix doesn't have a set polarity until you assign it one.