New Felix Power Line Conditioner “FO-Felix”

WireNut · « **on:** 10 Apr 2018, 07:54 pm »

There’s a new updated Felix power line conditioner pcb in town call the “FO-Felix”.
It’s designed by a different person from the original Felix featured here on AC a few years back.

Since the original Felix was an AC project I thought this would be the best place to get opinions from builders of the original vs this new pcb.

The new updated pcb will now except 15-16 amp coils. I’m considering getting in the group buy at DIYaudio as I already have four 15 amp coils setting here ideal.

The designer states the new version does not have capacitors after the choke. Also, the caps before the choke are not varied in size as in the original Felix.
He’s using four 0.47uf caps on this pcb rather then 6 caps of varying sizes. Bypass resistors for the chokes are also added to the pcb as well as a convenient fuse holder.

Any opinions on this new version vs the original are appreciated.

More info and GB link below.
http://www.diyaudio.com/forums/group-buys/321223-gb-fo-felix-emi-filter-ac-mains-120-230v-folsom.html#post5399670

Attached is a photo of the FO-Felix pcb.

wushuliu · « **Reply #1 on:** 10 Apr 2018, 08:02 pm »

Folsom is an AC member so hopefully he will chime in.

Also I'll jump in before Anand and ask if there are any measurements...

And Folsom: Is there a more in-depth reason why your layout is more effective? I thought smaller cap values were important for different frequencies? Same with choke size. Can these be stacked one after the other like the original for improved performance?

Folsom · « **Reply #2 on:** 10 Apr 2018, 10:02 pm »

The first thing about why the Fo-Felix is more effective is the resistors that bypass the coils. Independent people have confirmed that they are a sonic improvement, on top of the objective reason which is to stop electrical resonances with the CMC, from the diversity on the AC lines.

As stated in the other thread, the problem with using multiple capacitor values is they can start talking to each other, and when they resonate with each other it's hard to predict just how far those effects may be audible; but they could go from nearly benign to fatigue. The negative consequence does not out weigh the (small) benefits of varying sizes.

They can be arranged any way you want. But if you mean ran in series, it is possible. But I'm not condoning that because it isn't an ideal circuit. It creates the same circumstance of having capacitors following the CMC, which does not appear to be optimal in results.

The reasons why I've done the different changes is from years of development working with CMCs. When you really start digging you find out very quickly that trying to squeeze the numbers for the highest/ideal attenuation is pointless when you ruin the sound at the same time. It looks like you're making a compromise, but when you hear what a filter sounds like without simultaneously butchering the complex impedance, you quickly know that it's way better. You're really getting the benefit of my R&D, that has been on going for some time now.

I'm trying to think of a good analogy... The goals are to create a stable improvement that doesn't have negative consequences, as opposed to stacking improvements on one side of a teeter-totter while the problems shoot up on the other. It's important to understand that the AC mains are very complex electrically, so what you think you know may not be true. The reality of the original Felix is that it was merely OK. Some of the results may have been negative from an engineering perspective, but were significant enough sonically that people enjoyed the difference in sound.

I'll post a graph or such in a bit.

poseidonsvoice · « **Reply #3 on:** 10 Apr 2018, 10:18 pm »

Quote from: wushuliu on 10 Apr 2018, 08:02 pm

Folsom is an AC member so hopefully he will chime in.

Also I'll jump in before Anand and ask if there are any measurements...

And Folsom: Is there a more in-depth reason why your layout is more effective? I thought smaller cap values were important for different frequencies? Same with choke size. Can these be stacked one after the other like the original for improved performance?

Love it!

Best,
Anand.

Folsom · « **Reply #4 on:** 11 Apr 2018, 02:25 am »

Here is the difference in settling times.

And a comparison between resonances.

WC · « **Reply #5 on:** 11 Apr 2018, 03:18 am »

So how many amps can they handle?

So could you put one or more of these in line before a power strip with all the audio gear plugged in? Just trying to figure out how people use these.

Folsom · « **Reply #6 on:** 11 Apr 2018, 04:00 am »

The amperes depends on the choke used (PCB pours & parts exceed or match capability of available ampere from breaker). I'm recommending for amplifiers to use one 4x the ampere rating of your amplifiers ampere rating. In fact I think I may increase the sizes available after some review.

The plan is to include the comoco no matter what, which can do 3/5/10A, and then the other holes are for much larger like 17A or 20A.

These can be used inline but with the shape and connecting terminals on them can easily be put into a power distribution box. With fuses you can make sure you never overload the wrong CMC by plugging something into the wrong socket.

I'm looking at possibly larger CMCs to go on them, because for amplifiers 4x the size is a good way to make very sure you never get any saturation.

WireNut · « **Reply #7 on:** 11 Apr 2018, 04:21 am »

Are you saying that if my amp is rated at 10 amps (10 amp fuse) that I would want to use a 40 amp coil?

Folsom · « **Reply #8 on:** 11 Apr 2018, 04:24 am »

Quote from: WireNut on 11 Apr 2018, 04:21 am

Are you saying that if my amp is rated at 10 amps that I would want to use a 40 amp coil?

That depends on a few factors. Is it 10a ClassA?

WireNut · « **Reply #9 on:** 11 Apr 2018, 04:28 am »

No, AB. I'm using 2 Adcom GFA 555's early 1986-87 mark 1's. Fuse is 10 amp.

BTW. I currently have four Coilcraft CMT4's. Two 9-amp coils and two 15 amp coils.

I wonder if anyone on AC has experience with the CMT4's compared to the Bournes and smaller chokes by Coilcraft.

Attached are a few pics of my CMT4's. They are much larger then the coils used in the original Felix.
Due to the larger size the pin spacing is 1.500 by 0.900 and may not fit your board.

However, It would be interesting to see how many folks would be interested in trying the CMT4's with your new board.

Maybe if there's enough interest in the 4's it would be worth modifying your pcb if needed.

Folsom · « **Reply #10 on:** 11 Apr 2018, 05:45 am »

Quote from: WireNut on 11 Apr 2018, 04:28 am

No, AB. I'm using 2 Adcom GFA 555's early 1986-87 mark 1's. Fuse is 10 amp.

ha, no. It's more like a 20A CMC. The fuse might be 10A, but the amplifier doesn't use that.

Folsom · « **Reply #11 on:** 11 Apr 2018, 05:47 am »

The advantage of using the CMT4 is that it won't saturate quiet as easily. But at 15A is is still somewhat limited. I'm looking into Wurth right now because they go even higher by using a nanocrystaline core that doesn't have to be so big.

wushuliu · « **Reply #12 on:** 12 Apr 2018, 05:52 pm »

Folsom, did you try cap values higher than 2uf and if so what were the results?

Folsom · « **Reply #13 on:** 12 Apr 2018, 07:08 pm »

There are benefits but it gets more complicated. As you increase the capacitance you have to start compensating with increased size, different kinds of resistors, limitations due to heat, etc. It quickly devolves out of DIY, and into something most forum owners/mods don't want to see because the necessity of self deployed safety aspects rapidly encroach. Where as this iteration is at a heightened safety compared to the original, and will remain so even if you stuffed a box full of them.

The last time I helped someone build a larger power conditioner, well, it basically was erased from DIYaudio. It was safe, but the forum itself wasn't having it... and yet people are still building 400/600v tube amps on there...

Anyways, I hope that suffices as an answer. Basically due to safety and complicity, this is a great amount of capacitance. (in CFC no less)

wushuliu · « **Reply #14 on:** 12 Apr 2018, 09:40 pm »

Got it. Thanks.

I have a bajillion Bourns 8118 (9.3A 7.3mH) lying around from the original Felix. Can these be used instead of the 8120?

Folsom · « **Reply #15 on:** 12 Apr 2018, 10:13 pm »

You could probably force the pins with very little effort. But I am going to say I do not condone it as it circumvents Folsom's built in safety for ampere concerns, so at your own risk knowing there is no fuse, and it does not match a 15A socket. So as long as you are not trying to use too much current, it would work.

The original Felix was often inline (of the power cord), so it was cord specific to equipment that would not exceed the rating.

Syrah · « **Reply #16 on:** 13 Apr 2018, 05:13 pm »

It would be great if we could source good boxes etc. as well. I found that to be the hardest and most time consuming part of building my Felixs. I ended up having cord in / cord out with a steel box in between, like a snake that just swallowed a box. I think it would be safer to have built them with IEC in socket out.

I have 4* Dodd 120w tube monoblocks and I use one for each amp.

Folsom · « **Reply #17 on:** 13 Apr 2018, 08:59 pm »

Boxes with cut outs or? Polycase has nice boxes for in line or distribution for a few. They offer services for cut outs but are not so practical for orders under around 10.

Syrah · « **Reply #18 on:** 14 Apr 2018, 06:12 pm »

Maybe if someone makes a power bar that is hackable and has enough room inside.

wushuliu · « **Reply #19 on:** 14 Apr 2018, 06:57 pm »

Quote from: Syrah on 14 Apr 2018, 06:12 pm

Maybe if someone makes a power bar that is hackable and has enough room inside.

There should be plenty of options on ebay:

https://goo.gl/17vait
https://goo.gl/a3X93A