[samplesj]

Jeremy, you have a scope?     What transformers will you be using?

A 1a fast blow fuse has sufficed for the Modwright pre and the TACT pre, a 3a switch should be fine for testing, I'd presume...  I've 3 or 4 spare DPDT 25a toggles if you'd like to borrow one for a while, actually you can have one if ya promise to report back on your scope results!!!

No, I don't have a scope, but I can get access to one.  When my father-in-law's isn't loaned out and is working I could get him to run a few measurements.

Thats why I said I couldn't really do the scope stuff real time.

[samplesj]

Converging on the 'ideal' values and configurations for the various components is harder that one thinks in that aural memory really not as good as we think. What JoshK and I did was build 2 complete Felicias that we could rapidly switch (through actual switches) between to evaluate different caps, transformers, etc... in various roles feeding the same component.

Oh, I misunderstood where you were going with the caps and switch.  I thought you meant to wire the hot and neutral through the center tabs and just wire in a cap at the top and one at the bottom.

[Occam]

Jeremy,

And as you're not switching between two pairs of different transformers, your way makes a lot more sense. You could also make some male quick connects that would hold 'always in circuit' caps such as the bypasses and/or some of the bulk capacitance and use the dpdt to switch in parallel different caps into the circuit to hone in on the optimal.

Regards,
Paul

[samplesj]

Hmm, after seeing AudioFerret's measured results out of the much bigger MPJA transformers I'm begining to wonder if I shouldn't just use those.  

It looks like they started sagging between 25-30 if I'm not misunderstanding something about the power factor (resistive load vs cap load).

Those others were only rated at around 59VA so if they sag around the same amount it won't be the 20 I need, but those MPJA look to be right at the level I need with a little safety margin.

[ackcheng]

Dear Occam

I do have a balanced transformer (100VA) set up already for everything. Do you think I will get additional benefit by added the circuit described on this thread?

Thanks!

[Occam]

ackcheng,

That would depend on the specifics of the balancing transformer you have. You weren't specific. Is it a single transformer, is there flanking circuitry, and if so, what? If its the equivalent of a BPT transforer, I'd say its more than good, as is. Is the transformer a split bobbin or a toroid? Is it shielded between primary and secondary?
Cascadeing transformers, within current constraints, is generally beneficial. It provides an opportunity for more common mode rejection via the extra transformer. It also provides a place to put capacitors in shunt accoss the lines connecting the transformers, which is very effective in dealing with differential noise, whithout concern about leakage to ground.
If you've access to very low cost surplus isolation transformers like Powervar or Oneac, you can feed your balancing transformer with a 3 amp version which should be more than adequate for a 100va transformer, which I assume, you're using for source components.  Then you'd have to experiment with capacitance accross the conections between caps. There are no hard and fast rules, and much would depend on the specifics of your particular transformer.

[ackcheng]

Thanks Occam,

I think I have made a typo. I should have been a 1000VA transformer. It is a toroid transformer with 0 - 220v primary and 2 sets of 0 - 110v secondary. Yes, there is a shield between the primary and secondary too.

I follow the circuit shown in this page

http://www.diyzone.net/phorum/read.php?f=5&i=224418&t=224130&v=f

I use this balance transformer to feed my CDP DAC and preamp. Do you think there will be any benefit to put another balanced transformer say, a 200VA 1:1 transformer in the CDP?

If you look at both the CDP and my 1000VA transformer on the whole, it just have the configuration you posted on the 1st page. But the transformers will be connected via power cable. The other difference is that both of my transformers will be 1:1

Thanks for your comment in advance!

[YoungDave]

Hello, Occam & all,

I was wondering, why not insert a single 115 in/60-0-60 out transformer into the individual units?  The unit's transformer could be the second half of Felicia.

That way, you could size the transformer appropriately for each preamp, etc.  You would still have the balanced power, the filtration, etc.  You would just be using an already-present 2nd transformer - that is, 2 transformers rather than the 3 transformers (Fel1, Fel2, and audio unit transformer) inherent in the Felicia system discussed heretofore.

You might achieve a reduction in cost that way, and you would certainly avoid the additional shelf footprint of separate Felicias if you have room for a second suitably-sized transformer in your audio unit.

Of course, sourcing the right transformers would be necessary...

The circuit I have in mind would look something like this:

[/img]
C1 (not shown) could be an "x" capacitor; C2 & C3 would be selected as per previous threads, and everything to the right of the dashed vertical line is the original transformer, filter, and rectifier in the audio unit.

Does this look feasible to you?

[Occam]

YoungDave,

Absolutely spot on! The idea behind the Felicia project was the use of surplus transformers, and with their recent unavailability, Felicia's topology makes less and less sense. When the $10 Signal A41 split bobbin transformers were available, it made sense to uses these dual 14v secondary transformers, arse to arse, to kludge a balanced power conditioner. This had the following benefits -
1. The pair of transformers was $20
2. Runnings them at 120v into 240v windings operate them at a very low flux level, albeit limiting the power available. Operating them in this manner is certainly inefficient, but as they were inexpensive, we could afford to literally throw massive amounts of iron at the problem.
3. The split bobbin construction (unlike conventional toroidal transformers) had minimal primary to secondary capacitive leakage which tremendously aided the power conditioning capabilities
4. The cascade topology allowed massive filtering (18uf) between the transformers, facilitating differential (transverse) noise filtering, which is quite a bit harder to do otherwise without putting substantial leakage and noise on the mains ground.

Felicia also has rather severe constraints -
1. Heavy and bulky, and ideally demanding a separate pair of transformers for each source component fed.
2. Sourcing problems for the transformers. Tammie and Billy at BGMicro really do want to fill those back orders for 100s of Signal A41 transformers, and will, if their source ever gets their act together.
3. Output constraints. A dual A41 based Felicia can power a component drawing 50watts before its output voltage drops below 110vac when fed from my 124vac mains. The $18ea alternative from Marlin P Jones can only provide about 35watts, despite its stated larger va rating, 240 vs 175. The smaller transformers, that Gordy has so graciously tested, are even more limitied.

Given this, the alternative of a more conventional balancing power conditioner is appealing. Now the trick is to do this with a single balancing transformer while maintaining the power conditioning performance.  BPT (and no doubt, others) have done this quite well. But it requires additional circuitry above and beyond just the transformer. This will obviously require capacitors, and iff'n you want excellent performance, inductors, most likely in the form of common mode chokes, CMCs. You never know what type of transformer you're feeding in that component. If it has a conventional toroidal transformer, that simply isn't going to offer the contribution of the second Felicia transformer. If the component has a switching supply, not only do you give up whatever the common mode noise rejection of that PS's transformer, but you've usually a net noise contributor.

I'm working on it.... I've had some very high performance 150watt balancing transformers custom made. They don't have the voltage sag problems of Felicia, and can actually power 125+watts in components. But the circuitry needed to provide conditioning equal to Felicia and isolation between components, is not done to my satisfaction.  Hopefully, we'll be able to build a smaller, lighter Felicia II to power 3-4 stout source components, with equal performance for less cost and grief that 3 individual Felicias.
Much of that circuitry is also applicable to providing non balanced power conditioning. So a Felix is possible that would basically be a DIY Audience Adept.

But the process is nowhere near the point I can offer anything concrete. And as flawed as Felicia is (the conditioner, NOT her namesake), within her constraints, she is IMO a wonderful addition to most systems.

[andyr]

Hi Paul,

I'm confused as to why you seem to be giving up on Felicia?  Surely, just because the appropriate surplus (ie. cheap!    ) transformers are no longer available, the concept of having a pair of trannies with some caps in between to provide smoothed, balanced power shouldn't be discarded.

Shirley, it just means those of us who haven't got the cheap trannies will have to buy expensive trannies for their Felicias!  

Regards,

Andy

[samplesj]

Ok, I'm officially building a couple now.  I went with the MPJA transformers instead of the smaller ones just to be safe.  

PE was out of the .1 Jantzen's so I just added those to the order for the Auricap .01s.

I'm going to go ahead and hardwire the input power cable, but I had a good outlet extra and I'm using a wall wart right now anyway so it would have been more interesting to go the hardwired output route.

It looked like everyone was just using rubber bands or tape to keep hardwired input lines from pulling back out.

I've got TONS of mdf scrap lying around so I'll be using that for the case.  I figure in and out on the back, the front, top, and bottom blank, and then a grid of venting holes on the sides.

[Occam]

Hey Andy,

Its not a question of giving up, rather one of building on what has already been done. Frankly, literally throwing a tremendous amount of iron at the problem is a straightforward, successful method of improving the power for your audio system. I've indicated that with my line voltage a single Felicia (with 2 of those Signal A41 transformers) can reliably source 50 watts (at 110vac when fed my typical 124ac line voltage). So its not unreasonable to ask how one can power 2 components with this while providing isolation between components. Isolation pretty much demands separate inductors for each component. Which leads to the consideration of how much power filtering can be provided from a combination of capacitive and non-transformer inductive components? Which leads you to the Audience Adept, PSAudio UPC/UO.....
When Felicia was first developed we had a shootout at a NYRAve meeting. At that meeting the Adept was clearly better than Felicia. Susequently, Larry and I spent a substantial amount of time tweaking Felicia, and repeatedly comparing her to a friends Adept. Now, we both prefer Felicia to that Adept when powering source components. But the Adept supports 12 individually isolated and filtered outlets, and is not current constrained, able to power amps, plasma TVs, and other power hungry applications. Similarly, Felicia is 'up there' with the tweaked BPTs, but again, constrained in what she can power.
I've certainly not given up on balancing as one facet in power filtering/ noise reduction. But IMO, noise rejection, and to various degrees, isolation between components are similarly important. And if our intent is to power more than 2 source components, other techniques should be applied and incorporated. Nor have I abandoned the non-norte Americanos. In addition to those 150va custom balancing transformers, I've also gotten a pair of 'World Power' 175va 115:230/115:230 (Triad vps230-760 usually $35usd, Signal A41-175-230, Hammond 185G230) transformers which will provide the same functionality for 220-230vac users.

Presently, Felicia is the best powerconditioner I know how to build. If you actually want to pay retail, the 'world power' 175va, with dual 18v secondaries would work a treat, with lower lossses, and slightly better filtering. These are made by Signal (A41-175-36), MCI (449), Triad(VPS36-4800 $35 from Allied or Mouser), Stancor, Hammond (185G36), etc..... all the same. But at $35ea, its getting quite spendy.

[studley]

If you want balanced power, you need dual windings (or a center tap) on the output side.

Under the right set of design conditions, you could get away with just a single input on the primary side; i.e., if you had a transformer designed to take 120V input and split it into two 60V outputs on the secondary side. Such transformers are not all that common, but they do exist, and some of them were designed with balanced power applications in mind.

A much more common type of transformer is one that has dua ...

I'm a johnny-cum-lately to this thread so apologies for this question.  As far as I can tell some of the commercially produced balanced power supplies use just a single toroidal transformer.  Do you guys believe that the 2 transformer recipe has advantages over the single transformer, or was it just driven by (cheap)  transformer availability?

[studley]

OK before everyone shouts at me I've just read further back and answered my own question.  Sorry!!

More interestingly perhaps, I'm in the UK and I'm determined to make a Felicia.  Any other UK lurkers out there who might want to join the search for some suitable, locally sourced, transformers?

[YoungDave]

Really, Studley,  I can't imagine anyone here shouting at you.  I'm just stumbling around in the dark, and nobody has shouted at me yet...

[studley]

Thanks YoungDave.

Just finished work so I'm off to the pub . . .Cheers!

[Occam]

Studley,

Pages 16-19, 24 & 25 of this thread are have discussions about building for 220-240vac, with some Circle members in GB, Builder Brad and RonR.
Those exchanges mention and link to specific transformers.

Along the lines of a single transformer based conditioner, you do have some nice transformers available to you -
http://uk.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=7740174&N=401
http://www.farnell.com/datasheets/53687.pdf
and Farnell has them from 60va to 500va at very reasonable prices. They'd provide you the balanced (split phase) 230vac directly, though you'd need ancilliary circuitry to provide additional filtering and isolation between components.
There is also -
http://uk.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=3304644&N=401
which would allow some cusomization of output voltage.
I've not used either, nor has anything progressed on the single transformer based conditioner that I'd have anything of value to offer at this point.

RS also carries Block manufactured balancing transformers. Simply search on RS for 'block safety isolating transformers' and the ones marked 'seconday 2x115' are the ones. They carry the same TIM series encapsulated as Farnell, at a higher price, but in addition they carry the STEU EI frame transformers which allow +-15vac adjustment to the 230/400VAC primary.

Regards,
Paul

[studley]

Occam

Thanks for that.  Having now read at least some of the earlier posts I think I'll go for the 2 transformer recipe as it seems the better bet.

And whilst we're talking can I just say what a top bloke ( I'd hazard that roughly crosses the pond as "righteous dude"?) you are for your contributions to these threads.

[Occam]

Hey Studley,

Dunno about the 'top bloke' part, but I do recall overhearing some Geordies refer to me as a 'right wanker'.....

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 Occam cheap SOB special.

The first transformer has a single primary of 230v, while the second has split primarys 120/120vac (grounding the center tap of the series connected primary will provide the needed 240vac balanced output), while both have dual 20v secondarys. This will move the output voltage upwards, compensating for some of the losses.

But I've not used any of these transformes, so I'd suggest sending PMs to RonR and Builder Brad and get their advice and feedback.

Regards,
Paul

[studley]

Occam
 Don't fret, Geordies think everyone is a right wanker.  Me, I'm a Yorkshireman; we know what's what.

I'll check out the Farnell gear.  I've already sent a pm to Brad.  (Hope he's not a Geordie!)

I'll be interested to see how the development of a Felicia 2 goes; I have a pair of Aleph 2 monoblocks that drink juice like there is no tomorrow - 600 watts each!