Felicia Balancing PowerConditioner Constructor's Thread

samplesj · « **Reply #320 on:** 11 Jan 2006, 07:11 pm »

Ok, my dual Felicia is alive now.

My son came down with Mono last week and was in the hospital for extra fluids for a day so I didn't get a lot of extra time to work on it this weekend. I got the box finished up Monday night and was up till midnight wiring all the transformers last night. I know the fuses wander from a straight line, but I was in a hurry so I didn't get too concerned about lining up the rear holes (can't see them from the seating

). I also didn't think about 3/4" mdf being too deep for panel mounted fuse holders so I had to drill from the backside with a bigger bit to let the lock screw catch.

4 MPJA transformers feeding an extra Hubble hospital grade outlet I had lying around with a hardwired volex cord. 1amp fast blow fuses in all 5 holders. I'm using the big Jentzen and .47 Jentzen, but I also have .1 and .01 Auricaps

Right now I've got it plugged into Running Springs Audio Duke, but when I get time this weekend I'm going to compare it to right out of the wall. I just got the Duke so I'm going to borrow the HT systems BPT 3.5sig to do a head to head so I may also compare to the Felicia.

Pictures are linked instead of inline so they don't slow down page loading.
Inside from front
Inside from rear
Fuse holders, cord, and outlet on rear
Side vents

I've got a few questions for other builders.

1: Is my venting enough? Its a larger box so I didn't think I needed swiss cheese, but I don't want to catch it on fire either.

2: Is it stable right away or does it take a while to "charge" up the caps? When I first plugged it in the Benchmark was good right away, but the Squeezebox was really flaky. Now, it just so happens that I had a Windows update "fixing" the server box at the same time so I went ahead and booted it just as a safety measure. After lunch when I checked on it, it seemed to be fine. Was that just a server fluke or do the caps take a while to stabilize after getting voltage?

3: What kind of field EMI/RFI have people expeienced from these? How close have people been able to put them to other gear? I know without a metal case I've hurt myself in that aspect, but the MDF and paint were free. I'm not seeing any problems myself, but it never hurts to ask around.

I like to again thank Occam and all the other people involved in getting this design just right.

JoshK · « **Reply #321 on:** 11 Jan 2006, 07:52 pm »

you know a cheap as mud solution to the power cord strain relief I often use is the clamps used in electrical service boxes sold at HD. They aren't anything special to look at but work a charm and can be used and reused.

Gordy · « **Reply #322 on:** 12 Jan 2006, 03:12 am »

Congratulations SamplesJ! I'm anxious to hear your thoughts on the BPT / Felicia comparison!

As for vents, the a41 transformers I used have always remained stone cold and no venting is needed.

I also was running 6 of the a41's in a wooden box / no internal shielding within 6 inches of my FM tuner with absolutely no effect on the tuner's reception or sound... all IC's were located pretty far away though...

Occam · « **Reply #323 on:** 12 Jan 2006, 03:52 am »

Samples,

Very nice indeed!

My experiences are similar to Gordy's, My 3 x Felicia, 6 transformers in total is housed in a sealed wooden box, and I've not had any heat issues.
Unlike toroids, EI core transformers do have a goodly amount of EM field (not RF), so I'd keep the box at least 2' away from components.

I'd suggest you twist the 'paired' leads together, input, output, and the wire to and from the secondaries, minimizing potential interference pickup.

Dan Banquer also judyifiably 'smacked me upside the head' for leaving the uninsulated leads of the Jantzen caps uninsulated. He is right.

While ideally the consequence of a short between leads is a fuse blowing, the laws of unintended consequences seems to have grave hostility towards me.

I'm looking forward to your comparisons to those 'big dogs' (Gulp)

Regards,

Paul

JoshK · « **Reply #324 on:** 16 Jan 2006, 05:48 am »

Paul,

Which way is the magnetic field around the Signal tx's? I got to thinking about my layout in my case and was wondering if it was decent or not. I never took E&M and don't think I've ever read which way it is. Does the right hand rule apply to E&M the way it does with torque?

randytsuch · « **Reply #325 on:** 16 Jan 2006, 05:53 am »

I'm not Paul, but I can tell you the right hand rule does apply to EM fields, so if you know the direction of the current, then you can use the right hand rule to figure out the direction of the mag field..

Randy

studley · « **Reply #326 on:** 16 Jan 2006, 02:46 pm »

Quote from: Occam

Hey Studley,

A low cost way of building a unit that should be able to power a 60watt source component would be to use for the input transformer a -
http://uk.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=696821&N=401
and for the output -
Part # 835-175
http://www.jprelec.co.uk/
at 'advanced search' simply seach on the part #
These total less than 22GBP, so vaugely qualify as an Occa ...

Occam
re the output transformer, JPR have a minimum order value of £30 so I'm looking for an alternative from Farnell. Would this be suitable do you think? (This question will probably reveal to you how little knowledge I have of this subject!):-

https://secure3.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=696821&N=401

Thanks

Ian

Christopher Witmer · « **Reply #327 on:** 16 Jan 2006, 08:16 pm »

By design, the link provided goes dead quickly. The trannie in question is 696821 on page 3 of the following datasheet:
http://www.farnell.com/datasheets/67538.pdf

studley · « **Reply #328 on:** 16 Jan 2006, 09:12 pm »

Doh! just realised that the alternative output transformer I asked about is the same one that Occam recommended as the input transformer, which obviously means that its not optimal as the output, otherwise Occam would not have recommended a different one.

Logic in action. Shame I didn't engage brain before typing in the first place.

samplesj · « **Reply #329 on:** 17 Jan 2006, 02:09 am »

Well I finally had a chance to listen to the Felicia as part of a critical comparison in my 2 channel system. I hauled down my BPT 3.5 sig and I had a RSA Duke down there. I went ahead and put my system config in my sig.

Felicia is using a hardwired Volex cord. The RSA Duke is using their supplied Cardas power cable. The BPT is using the supplied Litz power cable.

I just bought the Duke used about the same time I finished the Felicia so I've been running them in together. Since I had a BPT in the theater room I wanted a chance to see if a "nice" balancing conditioner made a difference compared to the Felicia.

The biggest thing I found is that the Felicia and the BPT do NOT play well together. Plugging the Felicia into the BPT yielded a very congested sound. It was almost a muffled sound, but it still had a bit of an edge to the highs.

While I really like my BPT in the HT system for what it does for the video compared to the other gear I've had in there, it isn't the optimal for audio to my tastes. My 2 channel system is a bit on the analytical side so I was happy to have the extra PRAT of the RSA conditioner. The RSA also brought out much more detail and had much more powerful bass. Maybe it was just the recent impressions of lugging the thing downstairs, but the BPT seemed to weigh "heavy" on the music. Ironically the BPT caused much more transformer noise out of my amp than the RSA device.

Now while my primary purpose was to compare the RSA to the BPT I also wanted to compare the BPT to Felicia since they are both balanced power conditioners. Even though the Felicia is a DIY design for much less money than the BPT it comes very close (inside its limits). Like I said before the BPT feeding Felicia sounded bad to me. Using the BPT for the amp only and the Felicia for the Benchmark and SB3 yielded very good sound. It was very dynamic and I could easily be happy with it. However with everything plugged into the BPT (on seperate duplexes of course) it was just a little bit more extended on the top with a little more sparkle in the highs at the cost of a tiny bit of PRAT. This was a very very close comparison and on a couple of the pieces I actually preferred Felicia over the BPT.

Once I switched back to the RSA + Felicia I think I got the best of both worlds. The nice extended highs were back, but it was still very dynamic and PRATful. I did try the RSA feeding only the amp with Felicia out of the wall, but it just didn't have the sparkle in the highs. I also like the idea of keeping the system in the RSA gear because occassionally I have a voltage drop and the RSA gear passively smooths that out (no drop outs since I've put it in the system :->) I can't really compare the RSA Duke to the Felicia because the Duke only has two outlets.

The Felicia was definately worth it just for the piece of mind having the two devices totally isolated from each other via transformer. The TrippLite ISOBar I was using to isolate the two devices (nowhere near as effectivly) was almost as much as the transformers. Even with the added costs of the caps I think Felicia is a smoking deal. It was so so close to the BPT and potentially better depending on the other system factors and taste. Of course I don't expect the Felicia design to put Chris out of work because it has a very limited application due to the 20-30watt max drive. The BPT is a big brute and can easily power my 2 channel system or even whole HT system. I actually had a 200x7 amp and the rest of the theater running from just the BPT at one point. So while the BPT has certain advantages there is something to be said for the dual transformers and extra filtering caps.

As a way to make my Felicia better I wonder if the missing sparkle in my Felicia straight out of the wall means I need to tweak the filtering caps. Maybe some noise is sneaking through it that both the BPT and RSA stop. I also have to admit that I've not yet gone back and twisted my wires to help reject the last bit of noise so maybe it is as simple as that.

EDIT:
I should point out that just using the Felicia alone was a major improvement over the wall and is well worth the time and expense. Any "failings" I heard were minor issues (except the BPT feeding the Felicia) and probably wouldn't have been noticed without the other device right there to compare to. There is also a 10x to 20x cost difference between the Felicia and the other gear so a minor nit when compared is actually pretty nice praise.

Occam · « **Reply #330 on:** 17 Jan 2006, 07:19 pm »

Josh & Randy,

Indeed, EM fields follow the 'right hand rule' but we've to consider that massive hunk of permeable iron diverting the magnetic flux. Assuming the question is in regards to optimizing the orientation of the transformers with respect to each other, ctviggen made an excellent suggestion- use that second transformer as a transducer, i.e. power one of the primaries with ac loading the secondary with 30-60ohm resistor accoss so that that transformer is passing 30-15 watts (actually half that if the primaries are connected in series rather than parallel, but you get the jist...) Vary the orientation of a second transformer in actual practical positions with the primary connected in series to a high impedance set of headphones. When the 60hz in the phones is minimized, you're also minimized the coupling between the transformers. I think?????

Studley,

While the primary of the input transformer can be a single 230 or 240v winding, the primary of the output transformer must be a split 115:115 or 120:120volt winding so that when we connect them in series, we can ground the center tap and produce 230-240v split phase balanced AC.
I'd suggest sourcing the input and output transformers from different vendors because using a 230v input primary and a 240v center tapped output primary (with the secondary volages equal) would have raised the output voltage of your Felicia by about 4% increasing the range of output wattage prior to the output voltage falling below some arbitrary point.
Jprelec.com is the only UK source I know of that carries transformers with 120:120 v primary windings, with the appropriate secondary volages. I can't help you with their mininum order requirements, but you might get them dirctly from clairtronic.com , their manufacturer.

[EDIT - CORRECTION OF]CORRECTION:Rapidelectronics actually carries the whole series 230, 115:115, 120:120 of primaries from Clairtronics and the highest secondary voltage matching amongst those primaries is the dual 20v secondary, which I'd wire in series.
http://www.rapidelectronics.co.uk/rkmain.asp?PAGEID=80010&CTL_CAT_CODE=30523&STK_PROD_CODE=M30995&XPAGENO=1
http://www.rapidelectronics.co.uk/rkmain.asp?PAGEID=80010&CTL_CAT_CODE=&STK_PROD_CODE=M72692&XPAGENO=1
So for the output transformer I'd suggest the 88-0452 and for input either the 88-3460 or 88-3511. While the 88-0452 is a bit more expensive than the same 120:120v pri trans from JPRelec, and the 88-3460 a bit more than the 230v pri trans from Farnell, you can order both transformers from the same vendor and they're both from the same manufacturer. Dunno if Rapid has a minimum order though....

Rapid also has a polyprop film/foil 1000v cap PN 10-2526, that is inexpensive and might? do well for bypass duties.

Ian,

I'm quite pleased with your evaluation and that Felicia is even in the same ballpark as the BPT and Running Springs.

I've never used the transformers from MPJA and wish you'd been able to construct yours with the cheaper (but possibly better performing surplus ones from BGMicro). I simply don't know if twisting the wires would have helped much, but do suggest you give it a try. Now, if we can find some brave soul who would implement Felicia with a real bespoke powercord (costing substantially more than Felicia herself) whe could see if (as Emril says)we could bring her up a notch.

Feliciatations to all,

Paul

studley · « **Reply #331 on:** 17 Jan 2006, 11:52 pm »

thanks again Paul.

JoshK · « **Reply #332 on:** 5 Feb 2006, 08:13 pm »

Thought I'd add a couple pictures of how I ended up constructing the Felicia.

Here is the back of my multi-Felicia:

Insides from the side:

Insides from the top:

Its just about to be fired up to test some voltages. Just need to hook up the non-filtered outlets. I still plan to add a voltage meter and an ammeter to the front panel that will show the line voltage and current draw from my system. However, I need to mill the front panel first and I will probably have to take out one of the Felicias because I have a feeling it will be too cramped as configured.

I used a 15amp, dual pole breaker instead of fuses, much like the Audience Adept Response does. I hard wired in my power cord, made from VenHause flavor 4 star quad wire and a Furutech 20amp plug I had lying around. The strain relief is a $.40 basement breaker box special, but it does the job quite nicely.

One of the Felicias has an ERO (siemens) 0.1uf 1200V cap in position #7 instead of the Auricap since I was one short, so I'll have to make a comparison.

Like I said, I will most likely pull one of the Felicia's out and build it into its own enclosure for use probably with my HT system instead. I just have a feeling they are too close together and there will be too much EMI interaction.

JoshK · « **Reply #333 on:** 6 Feb 2006, 01:17 am »

Just plugged it in to test voltages....everything cool....AND....no hum at all! No audible mechanical hum even with my ear right down next to them. I had always gotten a hum up really close.

Occam · « **Reply #334 on:** 6 Feb 2006, 03:02 am »

Dang Josh, thats lovely. How do you make those cutouts for the outlets, etc...?

JoshK · « **Reply #335 on:** 6 Feb 2006, 04:37 am »

painfully....the manual way with a nibbler. I first drill some holes with a step bit and then nibbled it to what I wanted. Did it a few years back when I orginally had the OneACs in there. It took a really long time and my hand way aching for a week.

04dgmsilv · « **Reply #336 on:** 18 Feb 2006, 01:38 am »

To add a few more pictures to this thread.... This is a combo Felicia with a 5V linear PS for my SB3 which is off being modded by RWA, it'll be back soon....

These are the MPJA xfmrs, with just the load of the linear power supply, I have a very faint audible hum (ear right next to the xfmr) in the orientation shown here. I also had it set up with the output xfmr laid down like the input and it had no noise, so I'll fool around with that when I get the SB3 back....

Thanks to all for the great idea and technical info!

mgalusha · « **Reply #337 on:** 18 Feb 2006, 03:10 am »

Very nice and looks great, considerably better than my Felicia/SB combo.

Srinivas · « **Reply #338 on:** 22 Feb 2006, 04:08 pm »

I have a quick question - does this provide a similar functionality as a Monster 3500 MKII or different ?
The MKII is on sale for $99 at onecall. It provides surge suppresion too and has some other features like delayed switching etc. I was thinking of making a Felicia but am also considering buying it and just wanted you guys opinion. You can also power your receiver through it.

Thanks

Gordy · « **Reply #339 on:** 22 Feb 2006, 04:39 pm »

Hello Srinivas,

Welcome to AC! Here's a nice option to the Monster and with any luck it'll mount inside your receiver for a invisible solution! http://www.diycable.com/main/product_info.php?cPath=21&products_id=536&Cid=9067a10eba948294f2ed682195a7d699

I'll pm you a link to a monster cable thread...