Discussion of AC mains isolation transformers

[Speedskater]

On 12/9/2017 at 11:48 PM, One and a half said:

The reality of what can be found is a compromise of what technically 'will do'. Agreed otherwise in all except for the last sentence.

 

I would any day trade 22.6pA of 60Hz current across the primary to secondary winding for a 0.0005pf Topaz, than 7.2uA for a nominal inter-winding capacitance of 160pf of a standard toroid. This is stray current to earth/ground at 120V, 60Hz.

The only place that any of this AC power line stuff matters is at the power amplifier's speaker terminals!  It's the amps output signal to noise ratio and only that ratio that matters.  We are not doing medical procedures on humans.

[sukyung]

1 시간 전, sukyung said :

 

 

[One and a half]

3 hours ago, Speedskater said:

The only place that any of this AC power line stuff matters is at the power amplifier's speaker terminals!  It's the amps output signal to noise ratio and only that ratio that matters.  We are not doing medical procedures on humans.

There's a video on Youtube explaining the leakage paths to ground, will see if I can find it again. Every little bit helps ya know!

[Ralf11]

4 hours ago, Speedskater said:

The only place that any of this AC power line stuff matters is at the power amplifier's speaker terminals!  It's the amps output signal to noise ratio and only that ratio that matters.  We are not doing medical procedures on humans.

 

what about noise injection into a DAC ?

 

or do you mean that it matters only if some distortion is present at the speaker terminals?

[Speedskater]

1 hour ago, One and a half said:

There's a video on Youtube explaining the leakage paths to ground, will see if I can find it again. Every little bit helps ya know!

Only if it helps at the amplifier's speaker terminals.

 

The big leakage paths to chassis/Safety Ground/Protective Earth are all your components power transformers.

[Speedskater]

Once we get to large isolation transformers wired as Separately Derived Systems, we are well down the path of diminishing returns. Worrying about picofarads is yet father down that path.

[SoundSparks]

The Sum of all Watts, VA of audio equipment connected in my case is about 600. Should I go with 1000VA isolation unit?

 

Would Plitron Medical Power Isolation Units serve the purpose, please? Couldn't find any info on the inter-winding capacitance of the isolation transformers employed... I understand it should be no more than 0.005pF.

 

Ditto regarding Eaton Power-Suppress 100 Ultra-Isolator Noise Suppressors, please.

 

In case neither Plitron not Eaton works, what Topaz model should I be looking for, please?

[Ralf11]

Ssparks, I posted a summary up above somewhere - dunno if it got moved to the top or not

[SoundSparks]

Thank you Ralf11! Will try to find it...

[SoundSparks]

Excellent summary, Ralf! Thank you!


Any experience working with Second Generation Balanced Power System from Goertz? Cannot find any detailed specs, unfortunately... Is it fit for the purpose?
 

 

[Ralf11]

Thx - I did it to try and understand things for myself and then thought it might help others.  I am still looking for that mythical beast, the $75 Topaz...

[Speedskater]

@SoundSparks, a 100VA isolation transformer is as small as you should go. But Plitron makes excellent transformers so I'm sure that their VA rating are very conservative. I won't worry about the inter-winding capacitance of any well made isolation transformer, you're not doing medical surgery.

[Ralf11]

doesn't  the inter-winding capacitance determine the amount of nose leakage??

[asdf1000]

11 minutes ago, Ralf11 said:

doesn't  the inter-winding capacitance determine the amount of nose leakage??

 

Yep check out this post by the Guru. It's more about noise and surge suppression, than leakage current isolation.

 

 

[tapatrick]

I installed a new Topaz 2.5Kva iso trans (0.0005pf) yesterday to replace a 1Kva (0.001pf) and it added a significant improvement over the 1Kva. So the extra lowered interwinding capacitance does seem to make a difference (or maybe the added headroom). Blacker background, cleaner all round sound, more details revealed, Natural voices and better, tighter bass. It does hum a bit even though I added a DC blocker too. I will probably put it in a box next.

[tapatrick]

1 minute ago, tapatrick said:

 

[esmit]

4 minutes ago, tapatrick said:

I installed a new Topaz 2.5Kva iso trans (0.0005pf) yesterday to replace a 1Kva (0.001pf) and it added a significant improvement over the 1Kva. So the extra lowered interwinding capacitance does seem to make a difference (or maybe the added headroom). Blacker background, cleaner all round sound, more details revealed, Natural voices and better, tighter bass. It does hum a bit even though I added a DC blocker too. I will probably put it in a box next.

I use mine at 230v. It does hum a lot, even with ATL DC blocker before it. How about yours?

[One and a half]

20 minutes ago, tapatrick said:

I installed a new Topaz 2.5Kva iso trans (0.0005pf) yesterday to replace a 1Kva (0.001pf) and it added a significant improvement over the 1Kva. So the extra lowered interwinding capacitance does seem to make a difference (or maybe the added headroom). Blacker background, cleaner all round sound, more details revealed, Natural voices and better, tighter bass. It does hum a bit even though I added a DC blocker too. I will probably put it in a box next.

It could be a mechanical resonance that's causing the hum on the Topaz. I have the same model, it doesn't make a noise.

 

Note the four bolts that hold the Topaz together, they are 8mm quite large at either side of the core. Place the transformer over a poor insulator, like timber, this will keep the heat in the transformer. Put a heater on the output, full 2400W power and let the heat build up in the transformer over an hour or two. Switch off the load, and immediately, tighten the four bolts in a cross hatch manner, 17Nm torque should be OK. The bolts are a UNC type, possibly using a 1/2 in socket to tighten, from memory. Let the Topaz cool, and try on the audio system again.

 

The blacker background is the reduction of leakage currents across the windings. Other users have reported that wiring the output in balanced mode reduces the noise (and heat). 

 

Connections for balanced mode:

X1 - Hot 1

 

Join X2 to X3. Wire an earth from the small 3/16 in stud to this point and all shields (the smaller wires in the black sleeves)

 

X4 - Hot 2

 

 

[SoundSparks]

1 hour ago, tapatrick said:

I installed a new Topaz 2.5Kva iso trans (0.0005pf) yesterday to replace a 1Kva (0.001pf) and it added a significant improvement over the 1Kva. So the extra lowered interwinding capacitance does seem to make a difference (or maybe the added headroom). Blacker background, cleaner all round sound, more details revealed, Natural voices and better, tighter bass. It does hum a bit even though I added a DC blocker too. I will probably put it in a box next.

 

Why is the new unit better? Is it because of the lower capacitance (2 times lower) or is it due to the higher power (2.5 times higher)? What is the major contributing factor here?

 

Should the distance from the gear to the iso trans be taken into consideration? 2' vs. 10', e.g.?

 

Is 1000VA iso trans powerful enough for 600VA (total) of connected gear? I.e. is (Total VA)/0.6 = (iso trans power rating) an accurate enough approximation?

 

On 2017-12-20 at 9:58 AM, SoundSparks said:


Any experience working with Second Generation Balanced Power System from Goertz? Cannot find any detailed specs, unfortunately... Is it fit for the purpose?

 

Goertz does not publish any detailed specs, unfortunately...

 

Please chime in!

[Speedskater]

17 hours ago, Ralf11 said:

doesn't  the inter-winding capacitance determine the amount of nose leakage??

I doubt that there is much difference in a real world test situation. If I were ambitious (but I'm not right now) I would do a test with a power transformer similar to what you might find in a hi-fi component connected to the secondary of the isolation transformer. The to the secondary of that component transformer I would connect a suitable load resistor (so that it dissipates say 25 to 100 Watts).

Looking at an unloaded circuit you will see all kinds of crap that disappears when a load is connected.

[Speedskater]

Most power transformers are rated for continuous resistive loads. That is based on core temperature and hi-fi components like pre-amps, players and processors are similar to resistive loads.  But power amps don't draw continuous max power (class A amps excepted) so they don't heat the transformer near as much.  However on the peaks, they draw lots & lots of current which can saturate the transformer core. 

[lmitche]

15 minutes ago, SoundSparks said:

 

Why is the new unit better? Is it because of the lower capacitance (2 times lower) or is it due to the higher power (2.5 times higher)? What is the major contributing factor here?

 

Should the distance from the gear to the iso trans be taken into consideration? 2' vs. 10', e.g.?

 

Is 1000VA iso trans powerful enough for 600VA (total) of connected gear? I.e. is (Total VA)/0.6 = (iso trans power rating) an accurate enough approximation?

 

 

Goertz does not publish any detailed specs, unfortunately...

 

Please chime in!

A couple years ago I visited the Bridgeport magnetics factory, met the owner and found a stash of 7 Alphacore balanced power transformers in audiophile cases.  I sold 6 here on CA and kept one for myself.  Everyone seems to be pleased with their purchase. I know I wouldn't part with mine.

 

While at the factory I was told about the new transformers referenced above. They are likely to sound the same.  Plug a tripplite power strip into the transformer outlet and you are good to go.

[Speedskater]

1 hour ago, One and a half said:

The blacker background is the reduction of leakage currents across the windings. Other users have reported that wiring the output in balanced mode reduces the noise (and heat). 

Those reports are seldom well documented. The gains from a balanced transformer over an isolation transformer are rather small.

Why?  Because the transformer in a hi-fi component is asymmetrical with respect to the power line and the chassis.  Good circuit designers know this and connect the correct transformer terminal to the Neutral.

[tapatrick]

2 hours ago, esmit said:

I use mine at 230v. It does hum a lot, even with ATL DC blocker before it. How about yours?

Yes does hum a fair bit, more than the 1Kva, which is a expected but its only noticeable when the room is completely silent.

[tapatrick]

2 hours ago, One and a half said:

It could be a mechanical resonance that's causing the hum on the Topaz. I have the same model, it doesn't make a noise.

 

Note the four bolts that hold the Topaz together, they are 8mm quite large at either side of the core. Place the transformer over a poor insulator, like timber, this will keep the heat in the transformer. Put a heater on the output, full 2400W power and let the heat build up in the transformer over an hour or two. Switch off the load, and immediately, tighten the four bolts in a cross hatch manner, 17Nm torque should be OK. The bolts are a UNC type, possibly using a 1/2 in socket to tighten, from memory. Let the Topaz cool, and try on the audio system again.

 

The blacker background is the reduction of leakage currents across the windings. Other users have reported that wiring the output in balanced mode reduces the noise (and heat). 

 

Connections for balanced mode:

X1 - Hot 1

 

Join X2 to X3. Wire an earth from the small 3/16 in stud to this point and all shields (the smaller wires in the black sleeves)

 

X4 - Hot 2

 

 

Thanks @One and a half - I will give that a try to tighten the bolts and am thinking to try balanced output too... sounds so good right now.