back to back isolation transformers:

[maxlorenz]

Hi all:

I live in a country with 220VAC mains.
Recently, I received a BIG, "EI" type, 220VAC to 120 VAC, transformer, as a gift. I put it to use as an "isolation Tx" feeding a cheap, moded Sony DVDP used as a transport for my secondary sytem. This DVD has a SMPS and accepts 120VAC.
The improvement in sound was huge! (I hate this word:huge) I got excited because every aspect of musical reproduction improved.

I have no luck and could not get a true 1:1 isolation Tx afterwards, to feed the rest of my equipment. Then I thought to get another 220 to 120 Tx and put it in series with the former in reverse sens, this way:

mains AC->220VAC(primary)->120VAC(secondary)->120VAC(secondary)->220VAC(primary)->system or DIY filter.

It would work, wouldn't it?
Maybe I'll get more HF isolation.

What do you think?

Thanks.
Mauricio.

 

[davidsrsb]

Yes it will work, regulatioon won't be as good as a single transformer.

Remember that your amp will take a lot more current than a cd player and will need a suitably large transformer.

I have a 3KVA 1:1 240V isolation transformer in my lab and its big, heavy and hums loudly.

 

[maxlorenz]

Thanks!

Mine must be around 2KVA. Only the first one will hum
and I will try a "DC block" before it.
Appart that, I made a heavy wood box to enclose the Tx and it reduces drastically audible humm in a manner that it doesn't boder.
My amps are UCD180, more than 90% efficient. I think the Tx's should have no problems feeding them.

Regards
Mauricio

 

[maxlorenz]

Here's my report on my back to back transformers to imitate an iso-trans:

The second 220VAC to 120VAC I was kindly given was a little smaller than the previous monster. I tried them on my active system. First using my DIY simple filter to plug DVD and both stereo UCD180 amps, living a Behringer active crossover on his own, unfiltered. Then, isolating only the UCD amp that feeds the midwoofer and tweeter (without filter).

I both cases (in my system, with my AC) I heard noticeable improvements. Even my non-audiophile father heard a change at once.

What I hear is a rounder, softer and calmer music presentation, with better clarity and warmth on midrange, and more natural timbres. Also soundstage improved in width and performers positioning.

I don't find a lack of dynamics at all.

Now I have to continue my search of true isolating transformers


I hope this helps.
Mauricio

 

[testlab]

For the ultimate in power source stability you want something like this http://cgi.ebay.com/California-Inst...591567718QQcategoryZ50968QQrdZ1QQcmdZViewItem
. It converts the incoming power to DC, then makes AC with a linear amplifier. You are completely isolated from the power company and line disturbances.

 

[jacco vermeulen]

Mauricio,

can not comment on placing 2 transformers in series.
I am using 4 big toroidal 1:1 isolation transformers in combination with high current Shaffner filters, there certainly is quite a difference.

Overhere such totally enclosed 2 KVA donut transformers are used by farmers in the field, and sometimes offered for sale at $40 the piece. No buzzing, and like regular toroidal transformers they can do 4 KVA the piece briefly. Best audio investments i made.

 

[pmkap]

Max,

The following URLs might be of interest to you -
http://www.audiocircle.com/circles/viewtopic.php?t=18441
http://www.audiocircle.com/circles/viewtopic.php?t=18443
as it takes a very similar approach to your experiments.

Some quick comments -
1. Ideally, the output of your 2nd transformer should be a dual winding that would allow you to ground the center tap which would give you balanced (technical) power. The benefits of this is the cancellation of reactive leakage currents on grounded components -
http://www.equitech.com/articles/articles.html
2. Both transformers should have either a shield that can be grounded between the primary and secondary if the windings are overwound, or a split bobbin construction. This minimizes the capacitive coupling between primary and secondary, minimizing noise coupling.
3. While transformers are wonderful for eliminating common mode noise, they're not particularly effective against differential(transverse) noise. Adding capacitors 'accross the line' between the transformers, X caps, will convert much of the differential noise to common mode which will be addressed by the common mode capabilities of the following transformer.

Regards,
Paul (aka Occam on AudioCircle)

 

[maxlorenz]

A thousand thanks to everyone of you that contributed with such good advices!

Dear pmKap:
My own filter uses some varistors followed by one X1 cap and then a special shotgun ferrite wich I think eliminates both kind of noise to some degree:
VDR: RS components catalog nº 289 7301

Capacitor X1: 47nF; RS components cat nº 240 5334.

Double apperture ferrite: 10uH each;
RS components cat nº 212 0617. I use at least 5 of these.

Best wishes.
Mauricio

 

[gmphadte]

If u have two step down transformers, why not add up the secondary voltages and use for 120 and 240. Connect primaries to mains as usual.

Gajanan Phadte

 

[maxlorenz]

Hi Gajanan:

If u have two step down transformers, why not add up the secondary voltages and use for 120 and 240. Connect primaries to mains as usual.

Do you mean like this???

What must I do not to burn the house?
Can I attach also primaries like upper winding with upper winding and lower with lower?

Will this give "balanced power"?

Any help welcome

Thanks for your interest
Mauricio

 

[gmphadte]

Yes, attach primaries in one plug as if they are two mains loads. If u get subtraction, just swap primary connection of one transformer.

Do not worry as long as u do not connect the secondaries in parallel.

U are right and then use 120 for 120 devices and added 240volts for others. Only take care u do not exceed each transformer power ratings.

Gajanan Phadte

 

[maxlorenz]

Thanks Gajanan!

It's time to get my variac to use
So, I connect secondaries "O" to ground and to ground pin of the power connector?
This will be a monstruous box

I discovered a big "reserve" of 220 to 120 transformers, here at my Hospital's cellar

I hope "they" don't catch me

I'll report back...IF i can!!

Best wishes to all.
Mauricio

 

[gmphadte]

U dont connect any secondaries to any ground.
Please draw what u want to do.

Gajanan Phadte

 

[maxlorenz]

My present transformer "source" has 220 to 110VAC TX's

Will this be OK? :

Thanks for your help.
Mauricio

 

[pmkap]

Max,

As you've configured the transformers in the above picture, you have 'isolated' power, typical for hospital environments. Typically one would rebond the appropriate side (so that you maintain appropriate phasing) of the output to reestablish typical 'hot' and 'neutral' configuartion. Alternatively (and IMO prefferable) you could connect the center point of the series connected secondaries to ground to make balanced (technical) power. Please refer to the Equitech URL on the previous page for explanations as to the benefits.

Regards,
Paul

 

[maxlorenz]

Err...sorry Paul. I didn't have the time to read those papers previously, but...
that's what I intuitivelly thought when I asked:

Will this give "balanced power"?

So, I connect secondaries "O" to ground and to ground pin of the power connector?

Will this be "balanced power"?

An externally hosted image should be here but it was not working when we last tested it.

Thanks for your support!
Mauricio

 

[gmphadte]

Good article.

Now that u have balanced power configuration for 220V but not for 110v.
For this u will have to wire the primaries in series and the secondaries as u have done. U will need 2 more of those trafos.

With ground connected, this configuration is as safe!!! as the mains. Beware, it has nothing called isolation.

Gajanan Phadte

 

[pmkap]

Hi Max,

As Gajanan, and your schematic indicated, the above schematic will yield balanced (technical) 220v, 2 antiphase lines both of 110vac referenced to ground. Its an interesting execise to put a spst switch at that junction between the secondaries to ground, allowing you to switch the configuation between isolated and technical power. You'd be able to switch while powering a component(s) and judge its effect for yourself.

Assuming you use 2 1kva transformers, the above configuration will nominally provide you with 2kva of balanced power. How balanced this configuration actually is would depend on the matching of the specific transformers. Alteratively, if you configure this with primaries in series, it will indeed provide you with balanced 110vac, +55vac to ground and -55vac to ground. Please note that the 110vac balanced configuration would yeild a 1kva conditioner with those same transformers as the current capabilities of the windings are the limitation of the power. It will also provide slightly cleaner power as the transformers are operating at a lower point on the B-H curve.

With this configuration, I'd also suggest you consider using Y capacitors to ground at the output on each of the output legs. With unbalanced power, using Y caps to ground. (this does not preclude your use of the ferrites). With typical 'hot' - 'neutral' mains, the use of Y caps can be problematic, often tripping leakage fault devices and/or shunting 'crap' and contaminating the ground line. But realize that differential(transverse) noise becomes balanced differential noise at the outputs, and because the shunted noise is nominally antiphase, they will 'cancel'.

Regards,
Paul

 

[maxlorenz]

Thank you, gentlemen

I will get two more 2KVA trafo's from the same make and build a technical balanced power supply.
Maybe I'll try my Y caps again. They are 10nF. Not too big?

One of my present trafos' is bigger than 2KVA (maybe 2.5KVA), judging for size. Will this yield some trouble for balanced operation?
The work load won't be more than 500VA, I presume, in peacks.

Beware, it has nothing called isolation.

There, I didn' t catch you.

I'm enjoying this

Maybe I will end with a respectable PS.

Gratefully yours...
Mauricio

 

[pmkap]

Max,

Running somewhat different transformers in cascade (back to back) is not problematic as long as you account for the limitations of the smaller, but I wouldn't advise running different transformers in parallel, as no matter how close the specifications, the measurement conditions are inevitably different and you're likely to have large imbalances if configured in parallel, as shown in your shematics above.

With regards to Y caps.... In non balanced environments, because of concerns over ground contamination and leakage currents, I don't use Y caps, relying on X caps to convert much of the . differential noise to common mode and treat that with a following cmc or transformer. But with balanced power out of a transformer, because symmetrical caps to ground will shunt noise that is antiphase, and ideally cancel, I use rather large Y caps 100-470nf, and with those transformers which I have custom wound with bifilar secondaries, I'm experimenting with substantially larger values. But I have access to instrumentation that can measure ground leakage, Sencore PR57, PR570. As you work in a medical establishment, you might also have access to the 220v equivalent for measuring leakage. Unless you can specifically measure the ground leakage, I wouldn't advise using Y caps even on balanced outputs larger than 47-100nf.

Regards,
Paul