Use of isolation transformer


For about a week now I've been using a 160 watt isolation transformer over my NuForce DAC-9 D/A converter(power consumption: 9 watts). I was recommend this by an audio-friend who said that my HTPC and "analogue switching amp," a NuForce Stereo 8.5V3, could possibly "contaminate" the power to the DAC-9 (via the return-wires, I suppose?) being that they're all coupled through the same power distributor(i.e., I don't use a power distributor per se - the bare wires are connected directly; leadwire to leadwire, return- to return, and earth to earth). Actually, about a few months ago I started out by placing an isolation transformer over my HTPC, to good sonic effect, and then proceeded with an extra one over my DAC-9 to possibly "shield" it from whatever noise may be emitted backwards through the power cord from the power amp - or even from the power wall-outlet. There was nothing conclusive or overly presumptive about the use of isolation transformers as proposed by my friend(who has incorporated the same tweak in his, in some respects, roughly similar stereo setup) in regards to the hypothesized effect or its theoretical (dis-)advantages; it was simply a suggestive "what if?"

Now, the use of an isolation transformer over the DAC-9 has a definitive sonic effect, and one that I would also call an overall improvement. My immediate, and remaining impressions are that a more finely resolved/differentiated top end, a more full sounding and clearly outlined midrange(more organic, even), and a better integrated and "coherently distributed" low end. In fact, coming about these three frequency spectrums in a rather disparate fashion, necessary it may be, feels a bit awkward in that the overall impression is that of a rather homogenous presentation.

However, another friend of mine who came over some days ago, remarked, upon listening to my setup, that while he conceded the mids had become more full and present sounding since his last listen(i.e., prior to the iso. transformer over the DAC), and better integrated with the low end as well, he found that the reverberative nature of the top end had become somewhat constricted - and to be honest, I had thougt about this as well. Moreover, he now found the width of the soundstage had been narrowed, instead hearing a more center-focused soundstage. This, also, I could confirm via my own impressions, though contrary to him I find it to be a more distinct quality, and even disagreed on the narrowing of the soundstage width.

Being that my friend had made verbal the observations on the top end's reverberative less lively nature, and that I found this to be in tandem with my own impressions, I thought it possibly illuminating to remove the isolation transformer from the DAC powerline and see how its omission would affect the reproduction in this area. In short, after a few seconds of listening with the transformer removed it was clear to me that the sound had not only changed, but also (subjectively) deteriorated; the overall presentation seemed to have fallen apart as if a connective (invisible) glue was missing, the center-fill and fullness of the mids lacked sorely, the ease of flow likewise, and the top end appeared less resolved and now exhibiting a white-ish or pale "color" patina to it; a more withdrawn, less spacious and anonymous sound - simply not as involving. That is also to say: the top end energy or reverberative nature seemed in no discernable way to improve or correct the beforementioned "shortcoming," if indeed that's what it really is.

Having now re-inserted the isolation transformer over my DAC the sonic presentation has yet again fallen well into place, though periodically the round-ish nature of the highs, extremely well differentiated and resolved they truly are, continue to come off a tad uninvolving with some music as if the slightest "edge" is missing. Everything in me tells me the general imprinting of the isolation transformer is that of leading to a definate sonic improvement save for this small issue, making me believe that another issue in my audio chain could have been addressed this way. It could also be that over time one has become more or less accustomed to a slight excess of energy in the highs, or simply a specific sonic nature here, that the more finely resolved and organic presentation leads me to believe sometimes that something is missing. Or, that the fullness of the mids and (soundstage-)center presence has somehow reverted attention from the highs to the lower frequencies. Anyhow, it's an interesting and overall satisfying developement of the sound, but also one that may have exposed improvements or changes could be made elsewhere - if indeed over more time I should conclude that more top end reverberative energy is needed.

I would appreciate the input of others who're using isolation transformers in front of their stereo setups as well - be it either (though preferably) with smaller and separately placed isolation transformers, or large singular ones used over the outlet group feeding the entire setup - possibly even on a related note to what I'm writing above both with regard the general nature of the perceived sonic changes isolation transformers lead to, as well as more specifically, where noticed, the reverberative or overall nature of the highs.
128x128phusis
Thanks to all for the replies. I would seem a balanced isolation transformer, feeding all components, is a viable solution both with regard to size and noise level - disregarding cost, not that I believe it'll be an excessive figure. I'll have this in mind, and consider such a solution in the near future.

Yesterday, however, I made a (seemingly) minor correction with the ISO transformer over the DAC-9 that addresses my quibbles on the slight softness and rather center-focused presention; earlier when I started out with an ISO transformer over my HTPC, the initial setup led to an overall sonic character(i.e., pronounced softness and center-focus) quite simliar to that heard when the second ISO trsf entered before my DAC a little over a week ago. The friend of mine who'd been the "architect" of this initiative had in the meantime found out (first and foremost through hearing) that the wiring of the ISO trsf, and how they were hooked up with the lead- and return-wires to both DAC and outlet(secondary and primary side respectively), had significant impact on the sound. Being that our ISO trsf's were alike(same brand and wattage size), he relied on and found a way to (visually) localize the wires on the ISO trsf that were to be connected to lead and return(or 'hot and cold') on both sides, and once I had implemented this change over the HTPC, for a change it was, the sound fell into place: the soundstage opened up providing a more evenly distributed presentation, and a natural "edge" had found its way back into the sound as well. All was good.

Now, the ISO trsf I had used over my DAC to begin with, for I replaced it yesterday, was of the same brand and size, though the wiring(two yellows) on the secondary side was placed physically different and in such a way to make it hard(in effect, impossible) to differentiate them and determine which one needed to see the lead- or returnwire(also, the particular ISO trsf seemed to be of a different batch judged by the color of its label). I therefore wondered whether the wiring on the secondary side here had been swapped in regards to producing the desired sonic outcome, even though I had loosely assumed they were connected correctly. While it would have seemed most obvious to simply switch over the two yellow wires to see how that turned out, I instead opted for another similar ISO trsf I had been given(with the others), being that its wiring(i.e., their physical placement) equalled that of the one used over the HTPC exactly, indicating it might have been from the same batch as well(also judging by the label color). And sure enough, after replacing one ISO trsf with the other over the DAC-9 and getting the wiring right, the sonics fell into place as described above - and with everything sounding somewhat more full, clearly chiselled out, see-through, and organic/natural than before the ISO trsf entered the stage, so to speak.

06-02-12: Jea48
Phusis,

A test you may want to preform on your current ISO transformer is to check the AC phasing to see if it matches the AC wall outlet your other equipment is plugged into.

If the output of the ISO xfmr is in phase with the wall outlet there should be little or no difference of potential, voltage, between the hot contacts of the two receptacles.

If the two are out of phase the voltage will be 240V nominal.

Jim

Jim -

Perhaps you're addressing what I'm getting at above? But thanks, I'll try this out. I must stress that the ISO transformers I'm using only sports four wires (per trsf) in all, two on each side.
I must stress that the ISO transformers I'm using only sports four wires (per trsf) in all, two on each side.
06-04-12: Phusis

Phusis,

Yes I assumed that to be the case.......

That is why I suggested you check to make sure the 120V AC grounded output of the ISO xfmr was in phase with the 120V branch circuit that feeds the rest of your IC connected audio equipment.

Basically it sounds like that is what you have done through listening tests by switching the secondary output leads of the xfmr. (Changing which one of the output winding leads that is made the grounded neutral conductor.)

My test would have told you if the xfmr secondary hook up to the 120V receptacle phasing matched that of the 120V branch circuit wall receptacle.

Just insert one test probe of the volt meter in the "hot" small slot contact of the receptacle that is connected to the output of the ISO xfmr and the other test probe in the "hot" small slot contact of the wall receptacle.

If the 2 hots are in phase the voltage should read near zero volts.
If the 2 hots are out of phase the reading will be 240V nominal.
.
I've got a few. 300W, 2 and 5 KVA. The latter 2 step-downs fed 240V, wired for balanced and $100 each off Ebay (not included shipping). All I want them to do is reduce noise and if they changed the sound any more than that, they wouldn't be there.
Jea48(Jim) -

Thanks again for your reply. I'll have a friend over with a voltage meter this coming weekend, and we'll then make the test and see what it reveals.

I'm guessing my sonic evaluations, being that they're consistent in both cases, will translate into at least the same phasing configuration over both ISO trsf's in relation to the wall outlet phasing, and I would further assume the current configuration being in-phase - though that's certainly only a big assumption. Is there even a general consensus on differentiating the in-phase and out-of-phase sound when ISO trsf's are in use? I seem to have made no similarly discernable sonic observations in regards to power phasing when no ISO trsf's have been in use, I must add, but from the day I started making my own power cords(prior to the ISO's) I've seen to it that all of my equipment is hooked up in-phase, so the in-phase sound, in whichever way it's discernable here, has been my reference for some time now.

06-06-12: Ngjockey

... All I want them to do is reduce noise and if they changed the sound any more than that, they wouldn't be there.

Indeed, though I'd be thinking how noise in its different incarnations (and sources) really affects sound and what its omission would be perceived as, for how do I even begin to fathom or decide in advance what is a by-product of other than the more overtly perceived noise? It's system dependend, for sure, and my general rule of thumb is that of estimating on a gut-level whether restrictions enters the sound, things go largely unchanged, or something good comes of the initiative. An economical principle dictates that where anything other than the latter is the case, it won't be implemented; certainly less is sometimes more, and I don't thrive on chasing noise reduction or use of excessive filtering per se. All of my cables, both power and IC's(and in a sense speaker cables as well), are unshielded and assembled in a minimalist way, and are either braided(IC's and the balanced, digital AES/EBU cable) or "lightly" twisted, which is sort of a build-in feature that acts as a kind of a shield anyway - though without the use of anything other than the wires(and their insulators) themselves, and of course the electrical characteristics that comes of their different configurations. Shielded cables have oftentimes turned me sonically off, so to speak, and therefore I don't shield them. The use of ISO trsf's in my setup, configured properly, very obviously seems to do good, and therefore they stay.