|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
Several people have asked me about low cost isolation transformer, to isolate their digital gear from the rest of the system. Most transformers advertised as isolation types command a premium, and are not readily available.Now there is a solution:
You go to RS, and get two of their 25 volt 2A transformers, part number 273-1512 at5 $10 each. Measure them and mark which one has the higher output voltage (they may be the same, in which case, don't worry about it), Connect the AC power to the primary of the one with the higher voltage on the secondary, then connect the secondary of the other one to the secondary of the first, and then wire in an outlet acrss the primary of the second one to plug the CD player/DAC into.A huge honking capacitor (I suggest a large AC voltage safety margin here, even though it is only 25 VAC) can be placed across the secondaries, effectively killing any digital hash coming out of the player, or coming in from the line. Placing the cap on the secondaries avoids any leakage currents, etc. Several uF is not too much, say at a minimum of 63 VAC (it still needs to be AC rated), although if you use a big cap, also use a smaller one, say 0.1 uF or so, to assure that RF frequiencies get shorted, and will help filter things nicely.
The end result on a decent system can be astonishing, the combination of the two transformer cores, and the capacitive short at HF's far exceeds anythig that can safely be acheived with just a cap directly on the AC line.
The only down side is that these particular transformers will only pass about 50W or so, which covers many of the modern CDP's and DVD's, except for the higher end models that draw a lot of current.
Of course, you are not limited to the RS units, you can do this with any step down transformers, and still have isolation, and use the capacitor filtering trick.
Overall rating of the composite transformer aassembly would be the secondary voltage times the current rating to give the amount of wattage it can handle. I would NOT use anythig lower in voltage than around 25VAC for the secondary, as losses through both transformers would be higher, as would the possibility of core saturation, etc.
If you find a higher voltage secondary unit, size the voltage of the caps appropriately. Obviously, an X or Y rated across the full AC line cap would be fine with just about any step down voltage.
This trick with the cap should also be able to be used with a pair of one to one transformers, providing excellent filtering, with a much higher level of leakage current safety.
As always, when dealing with the AC line voltages, take all necessary precuations, use properr build technique and safety measures (breakers, fuses, etc.) and place the entire assembly in some kind of box.
Jon Risch
Follow Ups:
Jon,
For the last couple of weeks I was searching the Tweaker's Asylum archives to get ideas on how to improve AC supply to my audio system. I was down to deciding using an isolation transformer with balanced secondary output, until I read your latest tweak. What I found it should be interesting is what you have proposed to Paul Barker of using a large front end transformer, and several smaller ones on the output side to allow isolation of digital, phono, etc, from the rest of the system. I have a couple of questions regarding this configuration:
First of all, is the latter option better than using a balanced isolation transformer?
Since I can order rhe 1:1 isolation transformers, is there a benefit of using a balanced secondary output?
How much larger the power rating of the isolation transformers (both front end and output side) should be in relation to the total (and individual) power consumption of my system?
Since the primaries of the output side Txs will be in parallel, do I have to put the cap on the secondary of the input side Tx or use several (smaller?) caps on each primary of the output side Tx? What value (approximately) for each case?
Is there a reason for placing a mains filter in front of the first transformer?
Why should the surge or spike protection be placed between the trannies and not in front of the input Tx?
Maybe a lot of stuff here, but your message arose a lot of questions in my head.
Regards,
Evan
[ First of all, is the latter option better than using a balanced isolation transformer? ]I assume you mean several individual isolation transformers rather than one large balanced one? Since there are trade-offs involved, it will depend on your particular situation, and how much hash you line has, and your equipment puts out.
[ Since I can order rhe 1:1 isolation transformers, is there a benefit of using a balanced secondary output? ]
Most folks concede that balanced is better, all other things being equal.
[ How much larger the power rating of the isolation transformers (both front end and output side) should be in relation to the total (and individual) power consumption of my system? ]
With transformers like this, it is hard to say how much is overkill, and how little is too little. I would get as much capacity as you can afford for the large front end transformer (up to several kVA), and use at least 2 to 1 overkill on the current for the localized trannies on the output side. Other wise, for the front end unit, at elast 2 to 1 overkill for the total system load.
[ Since the primaries of the output side Txs will be in parallel, do I have to put the cap on the secondary of the input side Tx or use several (smaller?) caps on each primary of the output side Tx? What value (approximately) for each case? ]
Since the two locations you describe are electrically identical, you can use just one cap on the "secondary of the input side" tranny.
[ Is there a reason for placing a mains filter in front of the first transformer? ]
Yes, to get full benefit of the filtering and protection, and to minimize any leakage currents.
[ Why should the surge or spike protection be placed between the trannies and not in front of the input Tx? ]
I was addressing his specific situation, and intent. If there is no step down, then I would place any surge protection ahead of all the trannies, the filtering in between the two trannies was able to be large and lower voltage due to the step down (and back up) in the original posted design. It helps to remember that the original intent of the use of the two RS transformers was to have low cost, and gain the benefit of a lot of filtering without other downsides.
Jon Risch
Jon,
Thanks a lot for your advise.
As I understand, if I would like to take it to the extreme, I'd use a large input 1:1 isolation transformer and several balanced 1:1/2+1/2 output transformers, using at least a 2 to 1 overkill for all trannies.
Of course, I'm aware that the intent of the original posting was to give the idea of using two low cost step down transformers just for the digital source, but it was Paul Barker's message that prompted me to take it a step (or two) further.
Thanks again,
Evan
Is a toroidal transformer (with electrostatic shield) a "worse" isolation transformer than the more usual Hammond type (with ES shield)? I ask because I've never actually seen a commercial toroidal IT, even though they can be had.Would regular transformers with lower secondary voltages (high current) be better/worse/same as the 25V sec. ones?
Thanks.
Is a toroidal transformer (with electrostatic shield) a "worse" isolation transformer than the more usual Hammond type (with ES shield)? I ask because I've never actually seen a commercial toroidal IT, even though they can be had.It pretty much boils down to interwinding capacitance. And off the top of my head I'd imagine that you can achieve somewhat lower interwinding capacitance with a split bobbin transformer with an electrostatic shield versus a toroid with an electrostatic shield.
Would regular transformers with lower secondary voltages (high current) be better/worse/same as the 25V sec. ones?
The lower the secondary voltage, the greater the overall losses by the time you get back up to 120 volts.
se
(I still can't believe the oldish Hammond IT I told you about is only rated for 50VA - it's way bigger in every dimension than their current 250VA model [which I will get]. I would have thought that size counts...)
I still can't believe the oldish Hammond IT I told you about is only rated for 50VA - it's way bigger in every dimension than their current 250VA model [which I will get]. I would have thought that size counts...Mmmmm. Got me stumped. Perhaps better magnetics in the newer model? I've really no idea. Typically for a given input/output voltage, the higher VA rated transformer will be larger so it's not unreasonable to think that "size counts."
se
Hi Jon,
I am sorry but I can not follow this thread.
You place a large cap across the secundary and use two transformers? Maybe a little schematic will help.
You make a difference between garbage coming out of a digital component say DAC and garbage coming in from the mains. As I see it any filtering between the circuitry and the mains will work both ways.
Some high Sony equipment had a 1 to 2 µF cap across the secundary of the transformer with a low value (50-75 Ohm)resistor in series with it and a 0.1 µF cap across the resistor. I tried that circuit and it did not make any difference in sound.
The only thing I can say from experience is that the sound from my DAC is better using two chamber transformers (low interwinding capacitance), from ERA instead of using toroids, from Amplimo. It is not a world of difference but is clearly audible. I use one trannie for the digital section and one for the analog, each with its own mains filter and one filter for both as the mains entrance.
ELSO
I am sorry but I can not follow this thread.
You place a large cap across the secundary and use two transformers? Maybe a little schematic will help.Here you go:
The differences depend on whether you want to use the center tap and add common-mode caps.
You make a difference between garbage coming out of a digital component say DAC and garbage coming in from the mains. As I see it any filtering between the circuitry and the mains will work both ways.
That's correct.
Some high Sony equipment had a 1 to 2 µF cap across the secundary of the transformer with a low value (50-75 Ohm)resistor in series with it and a 0.1 µF cap across the resistor.
That may have been some sort of resonance damping network rather than filtering. And if it is a dampnig network, it ideally needs to be designed for a specific transformer (depending on its inductive and capacitive characteristics) so replicating the circuit and applying it to a different transformer may not produce ideal results and may in fact make things worse.
se
Jon, many years ago I did just what you are talking about. But I found that the two transformer, isolation transformer had very poor output regulation due to high leakage inductance. It would drop the line voltage to 105-110v on its output with a normal pre-amp pluged into it. This put the regulator into low voltage dropout and caused hum to be in the system. With a Very light load this system might work, but I caution to watch for low output voltage. Also, the RS transformers ARE split bobbin types with about 30pf pri to sec C. They are a very good power supply transformer because of this, as they pass littel common mode noise, due to the low C.
Also, the RS transformers ARE split bobbin types with about 30pf pri to sec C. They are a very good power supply transformer because of this, as they pass littel common mode noise, due to the low C.Thanks for the information. Then this configuration could provide some benefit for equipment using single bobbin laminated or unshielded toroid power transformers (loading issues notwithstanding).
Though seeing as how the MagneTeks are readily available from Mouser, it's easy to go with a single transformer solution. And the 100 VA MagneTek sells for only $10 more than a pair of the Radio Shack 50 VAs.
se
Hi,I have been using a similarly composed commercial conditioner from Holger Stein for a good while. I have also experiemnted with similar circuits myself.
My Findings:
1) Add plenty of ferrite beads to the Lines (both incomming, outgoing and between the transformers).
2) Within reason you cannot have too much capacitance, however, note that having much capacitance will cause current to flow and heat up the transformers even without load. A better output Waveform is achieved with quite a few uF on the final output then between the transformers.
3) If you want "balanced Power" (the use of just one secondary however gives in essence floating balanced) use two Transformers on the outout side, connect the primaries and secondaries in series - voila - balanced power for the price of three transformers. Also, having a much less utilised core (core saturation on current peaks) improves the output Waveform.
Later T
The only down side is that these particular transformers will only pass about 50W or so, which covers many of the modern CDP's and DVD's, except for the higher end models that draw a lot of current.Of course, you are not limited to the RS units, you can do this with any step down transformers, and still have isolation, and use the capacitor filtering trick.
The RS units appear to have just a single 120 volt primary. That means no center tap. Which also means that the safety ground will need to be connected to the AC line's normal safety ground. Which leaves you with the same intechassis leakage currents you had to begin with.
The main reason for using isolation transformers is to take advantage of the secondary's center tap for the safety ground in order to balance the safety ground (and subsequently the various equipment chassis it's connected to) between two 60 volt windings so that the interchassis voltages cancel and thus (ideally) eliminate interchassis leakage currents and the noise they cause.
This arrangement doesn't appear to offer any advantage in that regard. So what exactly is the advantage?
se
Steve,You seem to be confusing isolation and balanced power.
I am only talking about isolation transformers, not balanced power.
Many CD players and DVD player only have a 2 prong plug any way, so any objections to not using the combination of balanced power with the isolation transformers would be a moot point any way.
If one did find a transformer that had a center tapped secondary, then they could also provide balanced power to the component as well, a further potential improvement.
The neat thing about this circuit is the ability to use relatively large caps to shunt HF's without the penalty of high AC leakage currents involved.
Jon Risch
You seem to be confusing isolation and balanced power.Nope.
My point is, the component's already isolated from the AC line. That's what the power transformers in the component do. So how do these Radio Shack transformers provide any greater isolation? Their interwinding capacitance isn't necessarily going to be lower than what's being used in the components. And they certainly don't have any electrostatic shield between the primary and secondary.
So seeing as the component's already electrically isolated from the AC line, what, specifically, is the benefit here?
I am only talking about isolation transformers, not balanced power.
But the component's already isolated from the AC line.
Many CD players and DVD player only have a 2 prong plug any way, so any objections to not using the combination of balanced power with the isolation transformers would be a moot point any way.
The more mass-market players do, yes. But most others do not. And even of those that do, many DIYers are keen on adding three prong IEC connectors and turning them into grounded components.
If one did find a transformer that had a center tapped secondary, then they could also provide balanced power to the component as well, a further potential improvement.
Ok. I'm still wondering what specific significant advantage the pair of Radio Shack transformers offers. If you were recommending split bobbin or electrostatically shielded transformers, I could see that these would have some advantage when used with components that employed unshielded toroidal power transformers. But these Radio Shack transformers appear to be concentrically wound. I could be wrong but unless they are, I fail to see that they'd offer any significant advantage in terms of isolation.
The neat thing about this circuit is the ability to use relatively large caps to shunt HF's without the penalty of high AC leakage currents involved.
What AC leakage currents? You said most CD and DVD players are ungrounded. So how do you end up with any AC leakage currents due to RF filter caps when using ungrounded components?
By the way, Mouser sells MagneTek isolation transformers ranging from 15 to 250 VA and priced from $14.49 to $62.16. The 50 VA model sells for $19.40. These use non-concentric, split bobbin construction which keeps interwinding capacitance to a minimum and subsequently AC line isolation to a maximum.
se
Steve,The whole point of doing this is that the RS units are readily available, and cheap. I was not aware of the cheap Magnetek's, so many things to check out, so little time.
A side benefit is the ability to use the large caps at lower voltages.
Yes, if one was going to use just an isolation transformer, then it would be good to gain some additional isolation via a low C pri-sec coupling, and an electrostatic shield, rather than just add an additional transformer.
All I know, is that several people have tried this, and it works really well, cleaning up the digital components and improving the sound fio the overall system.
Jon Risch
The whole point of doing this is that the RS units are readily available, and cheap. I was not aware of the cheap Magnetek's, so many things to check out, so little time.Yes, I understand why you recommended the Radio Shack transformers. The reason I responded was that it wasn't clear in your original post just what the underlying point (i.e. benefit) was intended to be and how using these transformers achieved it. Not everyone who reads these forums is interested in just cookbook recipies without any desire to gain an understanding of what's actually going on.
A side benefit is the ability to use the large caps at lower voltages.
True. Though using large caps may have implications with regard to power factor. But that's not something I've given a lot of thought to.
Yes, if one was going to use just an isolation transformer, then it would be good to gain some additional isolation via a low C pri-sec coupling, and an electrostatic shield, rather than just add an additional transformer.
Yes. My point was that the transformers wouldn't necessarily offer much benefit unless they were superior in terms of interwinding capacitance to the power transformers used in the component it's feeding.
All I know, is that several people have tried this, and it works really well, cleaning up the digital components and improving the sound fio the overall system.
Which is all well and good. Again, I only responded because it wasn't clear just what the transformers were doing in terms of benefit.
se
There's only one way to find out; make the thing and check the results :)Cor
There's only one way to find out; make the thing and check the results :)But it's easier and cheaper to ask Jon. :-)
se
Sounds very good, I will certainly try this.
Question though;
When I do not want to use any capacitors, can I instead
put two RF chokes serial in the secondaries?Cor
Sure. There is little reason NOT to use the caps here though, the usual down sides are virtually eliminated, as they are no longer directly across the AC line either incoming or at the equipment end.
Jon Risch
Thanks for your answer.
I should like to use RF chokes simply because I have quit a few of them left.Regards,
Cor
Very Interesting John.Were you in the position I am, because I wind my own iron, would you take this to it's logical conclusion and build a pair of 1:1 isolation transformers of say 1kv for the entire audio circuit? In which case which value in total of caps? I have plenty of high voltage ac caps, so whatever you think is best I would aim for.
Do these caps replace the roll of the spike suppressors that clamp down on dc spikes? or would you add those also?
Finally, is there any mileage in either using C cores for slight gap, say at least on the first transformer in order to operate well even during DC transients, as the gap prevents saturation?
Yours in anticipation
Paul Barker
In your case, with a lot of poptions available, I would use a large front end transformer, and several smaller ones on the output side to allow isolation of digital, phono, etc, from the rest of the system.
Since the caps are isolated from the final output, they can be quite large, and there are no leakage current issues. Be sure to bypass larger caps with smaller less inductive ones.If surge or spike protection is desired, then I would place themin between the trannies also. Remember to use 150 VAC rated devices with a nominal 120VAC line and 1 to 1 trannies.
As for dealing with possible DC component coming in, yes, this could be a help, and once it is gone, then it doesn't matter for the second transformer.
Jon Risch
Thanks John, I sort of arrived at that thinking too since I posted. I actually have a great big door stop 240v to 110v oil filled tranny that may well be C core, many were of that genre (UK based) from weight and size and what I know of transformer design it's about 5kva, I also have a builders isolation 240v to 110v 3KVA transformer, that I could do the inital back to back pairing with somewhere near the board's supply. Your capacitor trick within. Then I too was thinking a little double shuffle sized appropriately at each stage. As a valve man, I have to wind a transformer for each input, so may aswell add another 1:1 with caps between for the ultimate supply.On another tack, I harbour ideas to create a 3 phase supply using a rotary inverter and build all my amps to 3 phase. I'd need to order 3 phase EI lams but that's no problem. Looking at Terman, for us valvies 3 phase has some major advantages, namely, minimised ripple in the first place, and an actual voltage gain with choke input (1.1 times rather than 0.9 times with conventional single phase). Disadvantage other than the obvious nothing available off the shelf (until I go into production) is 6 valve rectifiers per amplifier!
Thanks again for the assistance. I'd be interested in your views on 3 phase.
Regards
Paul Barker
Thomas Mayer of Munich is running 3 phase power supplies on all his valve gear and knows a lot about it. You are correct about the advantages, they are legion.Post to the Joenet with your question, he's known to frequent sidewalk cafes in Munich and the Joenet, among other disreputable places.
-j
OK JI knew there had to be a lot to be gained.
More life here than SET isn't there J? I may stick around.
Paul Barker
- but the stench is difficult to bear.re: 3 phase - there are not many people doing it (I only know of TM) because the iron is even bigger and more expensive than standard tube audio gear. You are well situated since you can wind your own, and if you have a 3 phase source, you might havce some fun with it.
-j
Well I could wind my own, but I'd need to get the shells in, and there'd be 3 bobbins to wind each time, this may remain a pipe dream, but in persuit of excellence it seems plain to me from Terman's explanation of 3 phase power, that if as I believe in a valve amp power supply minimal inductance AND minimal capacitance yields more musical results then starting with lower ripple in the first place, a lot lower, makes sense.Thanks for the address also, I'll talk to the man when I have more time to explore this, it really is just in the grey matter at the moment stored against the unlikelyhood of enough time to do anything about it. But thanks. Paul Barker
Hi Paul,I know I'm not Jon, but from my experience, nothing is better than individual isolation. Using, say, four 250 watts, will prevent interstage noise, something that a big isolator can't. Really usefull with phono and digital devices. Did you read the review of the Furman Sound IT-1220 review on the enjoythemusic.com ? Not excatly the same device, but it offers the idea of individual isolation.
As for caps, I tried both .47 and .22 and it works flawless both AC high and low voltage and DC when sub'ing trans for chokes.
And as we talk of isolation with two trans, you can push the limit and put a regenerator between the two, as the Walker Motor Control. I'm in the process of DIY for that type of device.
Regards,
Daniel Trudeau
Using this or a regular isolation transformer, do you have any suggestions for some DIY cordage i.e. between outlet and IT, and between IT and digital gear? For instance, say a Volex 14 gauge shielded cord from AC source to IT, and IT output to an outlet box/strip. Then regular cords to gear. Or should the better PC go from AC source to IT? Thanks.
This post is made possible by the generous support of people like you and our sponsors: