Discussion of AC mains isolation transformers

[gererick]

I can get 5 KVA transformer below for only $380 and it is not terribly far away.  It is a 91005-31, .0005 pfd like my Elgar, and it has been tested and is working according to the seller.  It could replace my 2.5 KVA Elgar and be run 240V in and balanced 120V out, I believe.

 

While the price is right, it might be stupid to even consider it for obvious reasons but would be interested in the perspective of others, assuming I test it with test equipment and then with a load that is not audio gear.  Other than the obvious risk of using it with expensive equipment and safety, is there a potential damage to the faraday shield, or is that likely intact?  I may be able to source a used one from Radwell.com that is not damaged for $1,100, but don't know the shipping yet.

 

CLEAN USED UNIT / TESTED WITH 120VAC / BROKEN SIDE AT CONDUIT ENTRY POINT MGE TOPAZ 91005-31 ULTRA-ISOLATOR 5KVA NOISE SUPPRESSION TRANSFORMER / 120-240V - Other Electrical Equipment & Supplies (bonanza.com)

 

[gererick]

23 hours ago, oneguy said:

@gererickthats a great price. I’d jump on it at that price. The end caps are aluminum and that transformer weighs 100lbs. It would take much of a fall to crack that especially with the conduit holes already drilled. From the looks of the pics the core looks straight so the faraday shield should be intact. 

 

If they tested it and it tested fine I think it’s worth the gamble at $380. 

The technician who works for the seller can hook up 124.4V in and 63.6V out balanced, but says there is no 480V connection to connect a 240 incoming voltage to get it down to 60V via 4:1.  I thought this 91005-31 accepts 480V in but perhaps it doesn't.

 

Maybe it is still worth getting.  But I wonder what models accept 480V in, that seems to be a limiting factor. 

[gererick]

53 minutes ago, Talley4217 said:

That unit will not work with a 240v IN and a "balanced" 120v output.  You need to find a unit that is part number "911xx" which allows  480v input design.  "91xxx" will not.

If your interested in my Eaton Power Surpress100 7.5kva it does allow the 480v input so you can wire 240v into the 480v input and receive "balanced" 120v output if that is what you are wanting.  You would end up with 3.75kva capability.  My unit is only a year of use made in 2019 but not installed until late 2020.

Talley, thank you.  I would be interested but want to retain .0005 pF.  Thanks for the info!

[gererick]

1 hour ago, One and a half said:

My guess on the large hole. The transformer was connected, and fell off a bench/stand. The cable gland or conduit was strong enough still to hold the transformer, and the connections, but it was too much for the casting to take. The rest of the transformer is OK, so it didn't drop to the ground.

Thanks, I bought it earlier today and it will likely ship next week.  Shipping from Geneva NY to Larchmont NY, so shipping wasn't too bad at $140 or so given it weighs 100 pounds.  I wasn't sure about the other models that accept 480V in because it wasn't clear if they were .0005 pfd or if I had to get a very heavy unit in order to get .0005 pfd combined with accepting 480V in.  91105-31 is .001 pfd, and Surplus Sales is selling one for $1,525 plus shipping from Nebraska, which was also not ideal versus spending $380 plus shipping.

 

The tech is going to cut some short wires to show me what wires color wires go where, which is nice of him.  He was a maintenance tech for Int'l Paper (I think) and seems knowledgeable.  He also said it was not humming when hooked up with a live input but without a load.

 

You showed a pic earlier of a cap across an outlet (white and black) with a shunt to ground to reduce transverse noise on an isolation transformer.  Was that on a unit wired with a balanced output?  In other words, to reduce transverse noise, can I do the same thing with mine if I don't have an on-board outlet and if my output is balanced?  I will be hard wiring to a Tripp Lite strip, same as now.

 

 

  

[gererick]

On 1/15/2022 at 11:54 PM, Talley4217 said:

OK Read into it some more.  

Romex is literally garbage for audio delivery due to it's PVC insulation being a poor dielectric you build up capacitance within the cable and thus create a natural resistance which will create noise issues.   You need good polyethylene cabling and you also need an external ground ran with the cable and do not use the any internal cabling as a ground.   Your hot and neutral cables in romex are a pure parallel run with the ground in the middle perfect for contaminating the ground with noise induced by the hot and neutral.  You want the hot and neutral cable to twist in its run (any TC rated cable) and then have the external ground run along this twist straight (not twisted with).   Twisted cable prevents noise buildup and the external ground would not pick up any noise thus giving a clean system.

Now I cannot comment beyond the power delivery for your audio system as my expertise is in electrical signaling.   I work in petrochemical and we deal with facility wide noise issues on some very sensitive equipment.  Fieldbus devices are the worst.   

The 6 gauge wire which Jim Weil of Sound Application sells is triplex construction, which I believe he said means that the white, black, and ground wires are all twisted around one another.  I am not sure what the insulation is yet, or if I want to pursue triplex in any event given your suggestion.  From your post, "You need good polyethylene cabling and you also need an external ground ran with the cable and do not use the any internal cabling as a ground."    "You want the hot and neutral cable to twist in its run (any TC rated cable) and then have the external ground run along this twist straight (not twisted with).   Twisted cable prevents noise buildup and the external ground would not pick up any noise thus giving a clean system."

 

I am interested in following your suggestion but google searches aren't quite getting me to an adequate level of understanding to be able to execute on it.  One question, if the ground is external to the twisted cable and run straight in your suggestion, is the ground merely not twisted with the other two and run straight but otherwise in immediate vicinity of the twisted hot and neutral inside of a hose or conduit, or is it also external to a hose or conduit that the twisted hot and neutral cables are in?

 

For example, right now I have star quad cable from VH Audio feeding my transformer from the outlet through a hose, and ground wire outside the hose.

 

I would consider doing the same thing from the panel to my IT, but the VH Audio star quad is 12 AWG x 4 which results in 9 AWG, not 6 AWG.  It is cryo'd, but I could buy 6 AWG and get it cryo'd while following your suggestion, though your suggestion wouldn't be star quad of course.  Its insulation is FEP (flourinated ethylene propylene).

[gererick]

3 hours ago, One and a half said:

I wonder what the starting inrush current is on a 5kVA transformer at 240V. If there are breakers in the main board (dual for 240V), they do need to clear the inrush current. Somewhere back in this thread, the breaker should be a slow blow parlance for fuses, or Curve C to IEC rules or NEMA equivalent. Perhaps your vendor can advise. For 1-2kVA types, I never worried about it, but when paralleling transformers from the one breaker, I ran into trouble when the power went off, then on again. The combined inrush tripped the breaker.

 

In this case, I used a 20 Ohm resistor in the primary, and shorted it out with a contactor coil fitted on the output. Can also use incandescent lamps or flood lights. The resistor reduced the current to a smaller level, the contactor once applied with at least 80% on the secondary, bypasses the resistor. Can show a drawing. 

 

Where does the 480V requirement come from?

Thanks!  Regarding your question of where does the 480V requirement come from, I had interpreted Talley's January 15 post which said "One thing that article doesn't mention is using a 240V split phase input with a 120V output.  Having 240V input ends up cancelling out alot of noise and is the preferred method for wiring these units.   120V in and 120V out will allow many transverse noise to pass through without filtration."    as meaning it having 240V come into the transformer is better than having 120V come into the transformer from the standpoint of lower noise.  I will have "120V in and 120V out".  Though the out will be balanced and though the transformer will presumably be wired as if 240V is coming in (not sure of this last bit, but in order to get a balanced out), it will "allow transverse noise to pass through without filtration", unless I misinterpreted Talley's post.

 

Because I will have 120V coming in, I will probably be using a 20A breaker.  I am using a 20A breaker now with 120V coming in to a 2.5 KVA transformer with no issue, but presumably I am not paralleling transformers now?  

[gererick]

10 hours ago, oneguy said:

I think you meant to tag me in your post.
 

That units wasn’t wired for balance because it was a -32 which can’t be wired that way. You don’t need to have a receptacle on the transformer add a capacitor.

 

With that 5Kva you’ll need to strip the ends of the capacitor wire and add it to the terminals underneath the screws that are clamping in your output wires. You can leave the capacitor inside the gray endcap  You can use a zip tie and mount to secure it in place if you want.

 

To answer your question about the capacitor and balanced power, yes it’s fine. All mine have capacitors on them. 

 

Thanks. So the cap wires would go in the clamps that will clamp the black and white wires, and the black and white wires would go in those same clamps, but on top (and not underneath) the capacitor wires, correct?

 

How would the shunt to ground be connected - the same way as the capacitor wires but instead between the clamps for the white and green wires?  

[gererick]

On 1/17/2022 at 3:32 PM, oneguy said:

With the topaz T1 units it’s visible but I can’t speak to others. The T100R did not have one installed and not all T1’s have it installed. From

what I have seen it seems more prevalent in the receptacle version vice the terminal version but both have come from the factory with them. Capacitor sizing for this task is talked about here:

https://www.surplussales.com/Transformers/Pdfs/tp-91407-11.pdf

 

The topaz units come with an Aerovox brand capacitor. Here’s a pic of one installed from a T1 that I opened up:

So if the formula is C max = 265 * VA  / output voltage squared, the IT is 5 KVA, and output voltage is 120V balanced, the the max cap value is [265 * 5,000] / [120 * 120], correct?

 

That math equals 92.

 

Can you tell me what the 92 is?  Is that farads etc.?

 

Also, since the IT will output in balanced configuration should I use one-half the 5,000 VA, or should I use the entire 5,000 VA in the formula?

[gererick]

3 hours ago, One and a half said:

An effective way to filter Transverse (aka differential) and common mode noise is to use a standard EMC filter on the input, and why not on the output.

This pdf from Murata discusses what techniques are used in suppressing both noise parameters. Experiments with filters on the output of a Topaz cleaned up the noise a lot further. Note the EMC filter reflects the noise back to the source. The Topaz already shreds common mode noise, the EMC filter makes sure the noise is contained. At 50/60Hz frequencies the filter has no effect.

 

Most EMC filters have differential mode caps fitted from hot to neutral with a common ground centre point. In a system where the neutral is grounded, one half of these caps is hunted out. For a balanced network, the EMC filter then can actively use both caps They are typically about 1-10uF, sine there are few of them in the complete filter.

 

26to30.pdf 58.28 kB · 1 download

 

In the context of an isolation transformer,

 

(1)  Output end.

 

The drawing seems to show a capacitor between the white and black wires' clamps on the (balanced) output end of the isolation transformer, which we've discussed.  In my case the value of the cap would be 50 uF or less.

 

(2)  Input end.

 

In addition, the same on the input end of the transformer, which in my case would a black wire with 120V, a white with 0V.  No shunt to ground in the diagram.

 

Is that correct so far?

 

What value for the cap on the input?  

 

(3)  Output end (in front of the cap in item (1))

 

If correct so far, then the line bypass capacitor seems to involve, when going from the white to black or vice versa on the output end and before or in front of item (1),

 

From black wire (or white wire) (each of which are 60V) -->

(x) a capacitor (of what value?) -->

(y) then a connection to something (not sure what the midpoint connection is?) -->

(z) another capacitor (of what value?)  -->

To black wire (or white wire)

 

Are the cap values in 1-10 uF as referenced in your post?

 

In the context of an isolation transformer, where would all of (3) fit, given space constraints?  Presumably it would have to be in the output end of the transformer?

 

 Thanks!

[gererick]

5 minutes ago, Talley4217 said:

I'm willing to bet a regular topaz IT with the .0005pf is plenty enough for 99% of the users out there.  Many won't even have a system capable of resolving the differences.   After all this is just one component of many that helps create good accurate music.  The room and speaker type is by far the largest part.  

I suppose what I'm saying is don't over think it.   I would suggest a power quality analysis to determine if you even need to go the extra step of adding filters or capacitors.  You may not even need this.   

Things to remember, you can add all the components you want to clean the signal at the transformer but if your cable design is poor (i.e. romex) then your just going to add problems in the cable which will pass to your equipment. 

Contributors to this thread and others have changed their power source from a .0005 pfd IT to the much more expensive, more transverse-noise-focused conditioner and they have been thrilled.  I don't wish to do that at this stage because of the substantial cost.  My system is capable of resolving differences big and small.  Agree, not trying to / don't wish to overthink.  I am very interested in feeding the IT with something other than Romex.  Still interested in learning best approach with respect to that.

[gererick]

3 hours ago, JohnSwenson said:

Star Quad would be even better, but since you are using 4 wires you could probably go with 10AWG wires. With Star Quad the magnetic field is focused into the center of the bundle which really limits induction to the ground wire.

 

John S.

Chris Venhaus / VH Audio sells 12 gauge star quad, resulting in 9 AWG, that has been deep cryo’d.  Presumably this means it is easier to get directly into a 20A breaker and into the transformer without a pigtail and without interruption of a straight run of wire.  Not sure yet.  But it doesn’t result in 6 AWG.

 

Can one DIY 10 AWG into star quad without production equipment?

[gererick]

40 minutes ago, Talley4217 said:

 

You can make all sorts of DIY anything.  You shouldn't be concerned over pigtails.  Pigtailing offers better connection than any prong/socket plug/receptacle makeup can provide.  

Also didn't you say your only 20' from your panel?  Something like that?...  Your need for 6awg is probably pretty low.  You would be better served having an isolated ground system (properly bonded to main ground) and again the method of how the wiring is ran and type of wire is more important.  Small tricks makes wonders.  It's power to a device, your just trying to maintain a cleaner signal so your equipment has to work less manipulating it into audio signals.
 

Answer this for me:

Distance from panel to IT:
Distance from IT to Outlets:
Sustained load:

I like to keep voltage drop to less than .25v at sustained load and no more than .5v at peak.  Depending on your equipment and its use I can halfway determine peak. Would be about 2x anyway than sustained.  My system is for theater equipment.  I have larger need on amplifier power than most especially running dual subs.  This will only get worse when I move to quad 18's with 8kw of amplifiers just for subs to achieve reference level dynamics.   This is why I built my system the way I did.  Do once cry once.

Hi Talley, thanks. 

 

After going up the wall from the panel and back down to the floor, it would be about 35' plus the footage from twisting the black and white wires.   I could instead run the conduit at panel height on the garage wall into the adjoining room, but it would be more in harm's way.

 

I may go directly from the panel to the IT, and only be able to shut off the IT at the breaker.  That is what I am planning at the moment.  Haven't decided yet (interested in input, may put a male and female plug near the IT), but there won't be an outlet. 

 

Sustained load probably about 1,100 watts as each monoblock draws ~425 watts at idle.  That is not including two Revel Performa B15 subs, which at present are connected via preamp but aren't powered by the IT.

 

 

 

 

[gererick]

Below is a pic of how I understand, via seller’s tech, the incoming 120V line should be connected.  If incorrect, please LMK.

 

Assuming this is correct, the tech said that X1 + X3 on the output would be one 60V output, and X2 + X4 the other output (need to remove the jumper that is between X2 and X3).  He wasn’t certain whether either set needed to be connected to the ground clamp on the chassis (I would think not, but unsure).  I assume the ground wire going out to my power strip will connect to that ground clamp.

[gererick]

7 minutes ago, oneguy said:

With that style terminal block I would put ring terminals on the capacitor leads and screw them into X1 and X4. You can flip one ring terminal at each location so that they can be flat against each other she. You screw them in. You can leave the capacitor inside the end cap and mount as you like. 
 

Always grab a multi meter and test after any rewiring to verify you are get what you expect and the wiring is safe. 

Before applying power check with the ohmmeter function:

H1 to ground = open circuit

H4 to ground = open circuit

H1 to H4 = low ohms

X1 to ground = low ohms

X4 to ground = low ohms

X1 to X4 = low ohms

Ground to case = low ohms

 

After applying power check with the voltmeter function:

H1 to H4 = 120V

X1 to ground = 60V*

X4 to ground = 60V*

X1 to X4 = 120V*

*The isolation transformer will slightly bump up the voltage approximately 1-3V from input so expect slight higher than half the inputs on X1 to ground and X4 to ground along with X1 to X4 being higher than H1 to H4.  

Thanks.  Also wonder what the best (low-impedance) termination would be to 6 AWG wire going into the input side, when I am ready to do that.  I assume some sort of copper lug, preferably with some heat shrink.

 

Won't be doing that for awhile, just planning ahead.

[gererick]

19 hours ago, oneguy said:

With that style terminal block I would put ring terminals on the capacitor leads and screw them into X1 and X4. You can flip one ring terminal at each location so that they can be flat against each other she. You screw them in. You can leave the capacitor inside the end cap and mount as you like. 
 

Always grab a multi meter and test after any rewiring to verify you are get what you expect and the wiring is safe. 

Before applying power check with the ohmmeter function:

H1 to ground = open circuit

H4 to ground = open circuit

H1 to H4 = low ohms

X1 to ground = low ohms

X4 to ground = low ohms

X1 to X4 = low ohms

Ground to case = low ohms

 

After applying power check with the voltmeter function:

H1 to H4 = 120V

X1 to ground = 60V*

X4 to ground = 60V*

X1 to X4 = 120V*

*The isolation transformer will slightly bump up the voltage approximately 1-3V from input so expect slight higher than half the inputs on X1 to ground and X4 to ground along with X1 to X4 being higher than H1 to H4.  

What about a capacitor across the white and black terminals on the input side as well?  Plus a short wire from the white to the ground?  A bit more transverse noise filtering, with no downside?

[gererick]

I've been looking into an audio grade capacitor as an across the line filter, and heard back from a parts retailer the following:  "For AC Line voltage, X or Y rated caps are typically used, because they have a safety rating in case of catastrophic failure. While people DO use metalized caps, there are instances where catastrophic failure could result in electric shock or fire hazard. Therefore, for liability reasons, I suggest using X or Y rated caps…" 

 

It was also suggested to me, separately from someone else who is an EE, that a cap that is good up to 380 VAC (not DC) would be a good level to shoot for in my situation, to avoid failure and for better safety.

 

Seems perhaps a bit more to it than just the uF value?  Yet people DO use metalized caps?

[gererick]

On 2/11/2022 at 12:36 PM, oneguy said:

With that style terminal block I would put ring terminals on the capacitor leads and screw them into X1 and X4. You can flip one ring terminal at each location so that they can be flat against each other she. You screw them in. You can leave the capacitor inside the end cap and mount as you like. 
 

Always grab a multi meter and test after any rewiring to verify you are get what you expect and the wiring is safe. 

Before applying power check with the ohmmeter function:

H1 to ground = open circuit

H4 to ground = open circuit

H1 to H4 = low ohms

X1 to ground = low ohms

X4 to ground = low ohms

X1 to X4 = low ohms

Ground to case = low ohms

 

After applying power check with the voltmeter function:

H1 to H4 = 120V

X1 to ground = 60V*

X4 to ground = 60V*

X1 to X4 = 120V*

*The isolation transformer will slightly bump up the voltage approximately 1-3V from input so expect slight higher than half the inputs on X1 to ground and X4 to ground along with X1 to X4 being higher than H1 to H4.  

Measurements:

 

The measurements before applying power were fine.

 

 I bought a 12 AWG extension cord, cut off a couple feet on the female end, made the connections of the male end of that cord into the IT’s input, plugged it in and got the following measurements.

 

118.0V     H1 - H4

60.2V      x4 to ground

60.2V or 60.3V x1 to ground 

121.1V       x1 to x4

There was a little hum but not bad considering it was on a concrete floor.  My existing Elgar transformer sits on Townshend Pods (rubber-coated springs) and is quiet.

 

i connected the cut-off female end of the extension cord to the output, plugged a fan into it and the fan worked fine.  I shut the fan off and unplugged the IT.  I would like to try some other household items, but am not sure what to try given it is balanced power.  Maybe I’ll plug an old Yamaha receiver into it, connect old speakers and let it play for awhile.

 

I am wondering about putting it inside a metal enclosure, maybe one from Hammond Manufacturing in NY, so I have more room for capacitor(s) across the black and white wires outside of the end caps, and more room for thick wires going into and out of the unit.  Or putting metal boxes on the end caps with most of the boxes’ bottoms cut out except for screw hole locations.

[gererick]

On 2/13/2022 at 2:29 PM, oneguy said:

Both of my 5’s are very quiet. I had a 2kva -32 that had a really bad buzz so I suppose some of it is luck of the draw. What I can guarantee though is yours would be much louder not in balanced configuration.

 

You can put it inside a metal enclosure and it shouldn’t be an issue. I did that with a 2.4kva and a 1kva. Also, as previously shown, putting the capacitor inside the power strip case works too. 
 

I circled the capacitor in the 2.4kva build:

 

 
And here in the 1kva build:

 

If you go with a case pick one that is vented. 

 

After putting my transformer on a softer surface, it makes almost no noise with my ear up against it.

 

Regarding crossing the black and white wires on the output with a possibly better and bigger capacitors from Mundorf, I heard back via email from Mundorf.  The English is a little German, but it may be useful to some folks.  My email to Mundorf began with the following:  "I am interested in buying quality capacitors, such as Mundorf caps, to go between the black and white wires on the output of an isolation transformer from Topaz.  It is a 91005-31, and it is 5.0 KVA.  The original cap that comes with some versions of the Topaz (including a 2.4 KVA model) for reducing transverse mode noise is the Aerovox 7 uF 250 VAC cap.  The value can go up to 50 uF safely, according to Topaz documentation."

 

 

"Thanks for your feedback at all and interest in our range.  It is a very specific use to bypass the transformer but I think we have some benefits ones for your project.

 

We have some cooperation’s you have nearly same layout for transformer & power supply circuit.  My first idea is the TCAP47-600 but at least 100VAC … not reach the minimum requirement.

 

The supreme-evo caps like SEO-47uF or SESGO-47uF are great performers J (especially in cross-over settings)

 

… but from my point of view the supreme classic winding technology (biggest metalized surface with lowest possible inductivity) could have the main benefit you are looking for.  I would start with 1x SUP8-10uF and after 20x-30x hours I would bypass a MCAP400-22uF for 1:1 compares."

[gererick]

I cut part of the back out of an 8” x 6” junction box, drilled holes for screws, and fastened the box to the output end of the transformer.  I bought two terminal lugs for the transformer, and with my size transformer I can go up to 50 uF in capacitors across the + and - terminals to reduce transverse noise.  On the lugs at the terminals, I paired up a 22 uF 400V Mundorf MCap MKP Classic and a 10 uF 600V MCap Supreme.  I connected two short wires between the + / - terminals and a pair of split bolts that are connected to my power strip.  Between the two split bolts, I connected a .01 uF Duelund Silver Bypass capacitor, a .1 uF 1200 V Jantzen Superior ZCap , a 1.0 uF MCap, and a 2.2 uF 800V ZCap.  These voltages are DC, but I made sure they worked at their respective AC voltages from a safety perspective.  (I tested them for over a week at 24/7 on a Yamaha receiver tuned to an FM station, with no issues but connected at the terminals instead of some at split bolts).  The isolation transformer sits on six Pods from Townshend Audio for vibration control and reduced hum (only four are needed but I don’t have the right spec’s unit on hand).  I plugged it in yesterday and it seems to be working fine with 62+ volts AC at each split bolt.  After measuring, I taped up the split bolts with 2” wide Scotch.Super 33+.  I did not listen yet.  The idea behind the different values of caps is to address a variety of frequency ranges.  I will probably pop open a top and maybe a bottom hole in the junction box to allow heat to escape.  

 

 

[gererick]

I had a small unbalanced isolation transformer plugged into my power strip when the power was unbalanced.  Two digital devices are plugged into that IT.    If my power strip is now getting balanced power, can I still plug an unbalanced isolation transformer into it?

[gererick]

1 hour ago, Speedskater said:

I hope that value was a typo.

The maximum line to ground value is 0.01 uF.

Typical line-to-line values are 0.1  to 2.0 uF.

Remember the AC power line capacitors need to be 'X' & 'Y' types.

 

Regarding "Typical line-to-line values are 0.1 to 2.0 uF", please see One and a half post Jan. 15 with .pdf attachment, and on Jan 17 with photo, my post on page 75 / Feb. 5 asking about the calculation of max value, and Talley4217's reply on Feb. 5 indicating 50uF is the max based on the formula provided by Topaz. 

 

I am at 35.31 uF versus the 50 uF max indicated on 2/5.  I don't know if the 22 uF is worthwhile, though. 

[gererick]

1 hour ago, Speedskater said:

I hope that value was a typo.

The maximum line to ground value is 0.01 uF.

Typical line-to-line values are 0.1  to 2.0 uF.

Remember the AC power line capacitors need to be 'X' & 'Y' types.

Regarding 'X' and 'Y' types, I instead went with rated measurements provided by the manufacturer so that I could use better quality caps.  

 

Email to mfgr:  I am interested in buying quality capacitors, such as Mundorf caps, to go between the black and white wires on the output of an isolation transformer from Topaz.  It is a 91005-31, and it is 5.0 KVA.  The original cap that comes with some versions of the Topaz (including a 2.4 KVA model) for reducing transverse mode noise is the Aerovox 7 uF 250 VAC cap, which is a lesser cap from the standpoint of sound quality.  The value can go up to 50 uF safely, according to Topaz documentation.

 

The isolation transformer will be powering my audio equipment.  I have a smaller Topaz in place at the moment, but want to operate in balanced mode so am going larger.  It is wired for a balanced output.  120V on the black wire at the input, and 0V on the white wire at the input.  60V on the black wire at the output, 60V on the white wire at the output.  The cap would go between these two wires to filter transverse or differential mode noise.

 

My equipment draws about 1,100 watts when all equipment is on and music is playing.  The caps should not total more than 50 uF.  I could use one, or several, caps to get to the 50 uF.  You have Supreme EVO oil caps that are rated for 550VAC from 6.8 uF up to 15 uF, some from 18 uF to 27 uF at 480 uF.  Your 33 uF and 47 uF caps in this product are rated at 410VAC for dielectric.  Your Supreme EVO Silver Gold has a similar number of caps with these VAC ratings for dielectric.  Would these caps be suitable for what I am considering?  If helpful, below is the relevant documentation."

 

Reply:  "Thanks for your feedback at all and interest in our range.  It is a very specific use to bypass the transformer but I think we have some benefits ones for your project.  We have some cooperation’s you have nearly same layout for transformer & power supply circuit.

 

My first idea is the TCAP47-600 but at least 100VAC … not reach the minimum requirement.  The supreme-evo caps like SEO-47uF or SESGO-47uF are great performers J (especially in cross-over settings)

 

… but from my point of view the supreme classic winding technology (biggest metalized surface with lowest possible inductivity) could have the main benefit you are looking for.      I would start with 1x SUP8-10uF and after 20x-30x hours I would bypass a MCAP400-22uF for 1:1 compares."

 

 

[gererick]

1 hour ago, Speedskater said:

Using only 'X' and 'Y' type capacitors is a safety requirement!

 

I think that you are getting into dangerous territory.

Is it a fair question to ask whether the capacitors in my amps, which draw 400 watts each, should be X or Y rated as a safety requirement also?

[gererick]

7 hours ago, Superdad said:

Well if you are talking about capacitors placed directly across AC mains (line-to-neutral X-types, or line-to-ground Y-types) those are indeed very specific applications where the cap must be capable of withstanding very high peak impulses (thousands of volts) and be designed to fail either as a short (X-types) or fail open (Y-types). 

Don't substitute, don't mix them up, and don't electrocute yourself! 

On Jan 17, One and a half showed a photo of a 7 uF Aerovox brand cap with a part # beginning with W43A250, as outfitted from the Topaz factory, wired from line to neutral at the outlet (mine is line to line, hot to hot and not to neutral due to it being wired as balanced output).  The high peak impulses you referenced are potentially an issue, but apparently Topaz felt the transformer did a fair enough job suppressing them as part of its inherent capabilities in that regard (I don’t think the Aerovox was an X type, but may be wrong).  Or perhaps X type caps were not as prevalent when Topaz was shipping units outfitted in that fashion.  X type caps would be a safer and proper approach, and the higher quality caps I’ve used may not make an improvement in sound quality. 

[gererick]

Would it be okay to plug an unbalanced topaz isolation transformer into a power strip that is fed by a balanced one?