Bricolo, an isolation transformer isolates primary and secondary windings by placing isolation shields between them, reducing interwinding capacitance. That capacitance, in a non-shielded transformer, is sufficiently high to transfer high frequency noise between the windings. The best isolation transformers currently available reduce capacitively coupled noise, through shielding, 150dB. MGE UPS manufactures a variety of such transformers under the brand name Topaz, typically available on the used market at good prices. Elgar once manufactured similar transformers, but recently pulled the (single-phase) line.
Re: toroid noise rejection
I'm not sure I got it.
What's the advantage in balancing the AC? I don't see how it would cancel noise.
And AC is already some kind of balanced signal: it's not (directly) referenced to ground, and a transformer is a 2 wire thing. So it balanced or unbalanced, it doesn't see the difference
serengetiplains said:
I'm not sure I got it.
What's the advantage in balancing the AC? I don't see how it would cancel noise.
And AC is already some kind of balanced signal: it's not (directly) referenced to ground, and a transformer is a 2 wire thing. So it balanced or unbalanced, it doesn't see the difference
serengetiplains - good post! good point! Balanced power, now you're talking. Something that just dawned on me that may be a great source for headbutting drag down arguments is that some if not all countries in Europe have 240v (???) which would be just the 2 legs of 120v (???) (i.e. balanced ac??) In the states we just have 120v single sided non-balanced power. I'd think the two would probably have different cures for maybe the same ailment sometimes? I can definitely see apples being compared to oranges here.
serengetiplains - oops, i just reread my previous post about interwinding capacitance and isoxfmrs. I messed up, so sorry. I guess I got ahead of myself and was describing the split-bobbin type. Yes, the isos are wound with an electrostatic shield that reduce the capacitance. My boner,..can't believe I did that. I need more coffee...will be back. Bricolo, also it's good to ground the static shield if it's not built in. At least all were done that way I've seen.
Re: Re: toroid noise rejection
A balanced transformer has a center-tapped split secondary. The center tap becomes the ground reference. The two legs measure + and - 60VAC from this reference, ie, are opposite phase from each other, therefore balanced. The Equi=Tech website explains this well.
As to the question of noise reduction via balanced AC, that topic seems very complex. I'll explain a little of my understanding. In a typical power supply circuit, AC current flows in current peaks, charging the input capacitors only where the input voltage exceeds capacitor voltage. These current peaks represent a form of power factor loss, and themselves create voltages across any resistance (any piece of wire will do). The current peaks, for their part, create reverse-going voltages through inductance. Imagine a wire through which current flows. That current will create a magnetic field around the current carrying wire that will collapse to the degree the current reduces. The collapsing magnetic field in turn induces current/voltage opposing the current/voltage that created the magnetic current, shooting this noise back into the AC line. This noise sometimes goes by the name current harmonics, reactive current, etc. That's one form of noise reduction.
AC drawn from 2 x 120VAC lines (= 240VAC) stepped down should be cleaner than AC drawn from 1 x 120VAC. In North America, the neutral line carries current from both 120VAC legs, so has double current harmonics, among other things. Ridding your AC supply of these harmonics is a step in the right direction. The step-down option also, as you said, offers noise reduction in the manner of balanced AC, but only to the degree the secondary windings are precisely balanced. Any winding or coupling difference in the two halves of the secondary winding will reduce noise rejection. A precisely balanced 240VAC-120VAC step down isolation transformer seems to me the best supply transformer a person can obtain.
VA requirements depend on overall current requirements of your system. As a good rule of thumb, transformers should operate cool. Heat in a transformer indicates the transformer is working closer to capacity. Transformer distortion increases the closer a transformer is operated to capacity. Which is to say a larger than smaller transformer is generally preferred.
Mind you, a balanced transformer that reduces noise probably has much higher overall operating headroom than a transformer that does not: the former will of necessity operate with a higher power factor and thus with greater apparent power than the latter.
Search on eBay for "Topaz isolation transformer" (Audiogon also). You should find something.
Bricolo said:
A balanced transformer has a center-tapped split secondary. The center tap becomes the ground reference. The two legs measure + and - 60VAC from this reference, ie, are opposite phase from each other, therefore balanced. The Equi=Tech website explains this well.
As to the question of noise reduction via balanced AC, that topic seems very complex. I'll explain a little of my understanding. In a typical power supply circuit, AC current flows in current peaks, charging the input capacitors only where the input voltage exceeds capacitor voltage. These current peaks represent a form of power factor loss, and themselves create voltages across any resistance (any piece of wire will do). The current peaks, for their part, create reverse-going voltages through inductance. Imagine a wire through which current flows. That current will create a magnetic field around the current carrying wire that will collapse to the degree the current reduces. The collapsing magnetic field in turn induces current/voltage opposing the current/voltage that created the magnetic current, shooting this noise back into the AC line. This noise sometimes goes by the name current harmonics, reactive current, etc. That's one form of noise reduction.
RetroAudio said:
AC drawn from 2 x 120VAC lines (= 240VAC) stepped down should be cleaner than AC drawn from 1 x 120VAC. In North America, the neutral line carries current from both 120VAC legs, so has double current harmonics, among other things. Ridding your AC supply of these harmonics is a step in the right direction. The step-down option also, as you said, offers noise reduction in the manner of balanced AC, but only to the degree the secondary windings are precisely balanced. Any winding or coupling difference in the two halves of the secondary winding will reduce noise rejection. A precisely balanced 240VAC-120VAC step down isolation transformer seems to me the best supply transformer a person can obtain.
Bricolo said:
VA requirements depend on overall current requirements of your system. As a good rule of thumb, transformers should operate cool. Heat in a transformer indicates the transformer is working closer to capacity. Transformer distortion increases the closer a transformer is operated to capacity. Which is to say a larger than smaller transformer is generally preferred.
Mind you, a balanced transformer that reduces noise probably has much higher overall operating headroom than a transformer that does not: the former will of necessity operate with a higher power factor and thus with greater apparent power than the latter.
Search on eBay for "Topaz isolation transformer" (Audiogon also). You should find something.
Bricolo - it sounds as though you already have your mind made up about trying an iso xfmr. While it's so easy to get caught up with all the theory spouted off around here and get taken up with what people say as gospel, you might want to back up here and think about why the original xfrmr is doing what it's doing. If it is faulty and humming because of it, an isolation xfrmr may not do what you want. Think about it,...it's hum frequency is probably that of the line, in which case it's humming just because you applied power to it. An iso xfrmr would apply power just the same. If it is faulty, is there any way to swap your own gainclone xfrmr for another known to be good? From a friend or someone? I'm of the opinion that maybe the trouble needs to be taken care of directly, not covered up. - Just a thought from someone that still needs more coffee..
I thought about this too.
I can't get another transformer for my amp, I can borrow an isolation transformer from a friend.
when I'll test with the isolation transformer, there are 2 possibilities:
-the amp's transformer becomes quiet -> DC on the line (that causes the core to saturate)
-the buzz is still here, and since it's not due to DC, my trafo is faulty
I can't get another transformer for my amp, I can borrow an isolation transformer from a friend.
when I'll test with the isolation transformer, there are 2 possibilities:
-the amp's transformer becomes quiet -> DC on the line (that causes the core to saturate)
-the buzz is still here, and since it's not due to DC, my trafo is faulty
Hi,
Bricolo,
Let's suppose it is DC on the line for a minute and you won't be able to keep your friends' isoxfmr. What can you do?
1/ You could buy a new one in which case I'd choose one with an EI core or better, NOT a toroidal one, with an electrostatic shield that you connect to Earth.
Chosing one with a higher current rating will have a better chance to run cooler if you're lucky. Most xfmrs heat up due to copper losses in the wires.
2/ You could build a DC blocking circuit as explained on Thorsten Loesch's website and this one:
DA BLOCK.
Watch the diode and cap polarities, follow all the necessary safety precautions and it should work.
3/ if you think of investing in an isoxfmr, think 5KVA, not 500VA if you plan to run the entire system from a single unit, otherwise use smaller rated ones where it matters most, i.e. digital naughties.
4/ Whatever you build that uses a powerxfmr nowadays will likely benefit from a well built high quality xformer. Most toroidal powerxformers could double as decent OPTs which should tell you enough to know they'll likely let a truckload of RF junk riding on the mains through as well. This is why I almost always avoid toroidal powerxfmrs like the plague.
5/ A cheap powerxfmr is a cheap powerxfmr so don't expect stellar results from it.
I haven't read the entire thread so if I repeated someones elses' suggestions, my apologies.
Cheers,
Bricolo,
Let's suppose it is DC on the line for a minute and you won't be able to keep your friends' isoxfmr. What can you do?
1/ You could buy a new one in which case I'd choose one with an EI core or better, NOT a toroidal one, with an electrostatic shield that you connect to Earth.
Chosing one with a higher current rating will have a better chance to run cooler if you're lucky. Most xfmrs heat up due to copper losses in the wires.
2/ You could build a DC blocking circuit as explained on Thorsten Loesch's website and this one:
DA BLOCK.
Watch the diode and cap polarities, follow all the necessary safety precautions and it should work.
3/ if you think of investing in an isoxfmr, think 5KVA, not 500VA if you plan to run the entire system from a single unit, otherwise use smaller rated ones where it matters most, i.e. digital naughties.
4/ Whatever you build that uses a powerxfmr nowadays will likely benefit from a well built high quality xformer. Most toroidal powerxformers could double as decent OPTs which should tell you enough to know they'll likely let a truckload of RF junk riding on the mains through as well. This is why I almost always avoid toroidal powerxfmrs like the plague.
5/ A cheap powerxfmr is a cheap powerxfmr so don't expect stellar results from it.
I haven't read the entire thread so if I repeated someones elses' suggestions, my apologies.
Cheers,
Topaz 5KVA on eBay
Bricolo, check it out:
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=4661&item=3819629512&rd=1
I have a Topaz 5KVA and am very happy with it. You can run this transformer in step-down quasi-balanced mode 240VAC to 120VAC balanced. I say "quasi-balanced" because I'm uncertain just how balanced the secondary windings are.
Bricolo, check it out:
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=4661&item=3819629512&rd=1
I have a Topaz 5KVA and am very happy with it. You can run this transformer in step-down quasi-balanced mode 240VAC to 120VAC balanced. I say "quasi-balanced" because I'm uncertain just how balanced the secondary windings are.
eBay II
A smaller Xentek 1KVA model is also for sale on eBay. Looks like a repackaged Topaz, but in any event gives -146dB noise reduction.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=71391&item=3820308409&rd=1
A smaller Xentek 1KVA model is also for sale on eBay. Looks like a repackaged Topaz, but in any event gives -146dB noise reduction.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=71391&item=3820308409&rd=1
Bricolo said:
http://www.toroid.com/index.html
http://www.equitech.com/products/products.html
While I prefer my Topaz Ultra-Isolation transformer
(.005 pf interwinding capacitance), toroids with
carefully balanced windings and Faraday shields
do make for compact isolation transformers.
That's the trick: it's got to be a quality component
designed as an isolation transformer for best
isolation results. Given a toroid's inherant wideband
capability, magnetic coupling can defeat electrostatic
shielding so carefully balanced windings are necessary
for best common mode rejection.
And I can still get some dimmer buzz even through
my Topaz, at some settings on the dimmer. I need
a dedicated 220 V line, but I doubt my landlord's
going to be sympathetic to major wiring changes.
I'm also using heavy-duty Isobars, but that's because
I originally lived in lightning-prone Georgia.
http://www.tripplite.com
--Damon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.