• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Breve Tufvassons PFS 250 protective isolation transformer used against audio ground loop issue

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Breve Tufvassons PFS 250 protective isolation transformer test – any good for audio?

1. Introduction


There is a lot of misunderstanding on use of isolation transformers for audio. It should be noted that such device purpose is to get rid of possible or existing ground loop issues in audio, especially in case of single ended cable connections between class I instruments. For this reason, from many types of 1:1 isolation transformers, only class II isolation transformers make sense for audio.

1a. A brief explanation of the differences between transformer classes I, II and 2

Class I transformers
are designed and constructed so that protection against electrical shock hazards relies not only on basic insulation but which includes an additional safety feature to provide earthing, (grounding), of accessible conductive parts in the fixed wiring of the installation. In the event basic insulation should fail, the conductive parts cannot become live as they are grounded or earth protected.

Class 2 is a classification referring to the NEC – National Electric Code. To avoid potential cable overheating due to excessive currents and electric shock, the output of the power supply is limited to 60VDC or 100VA, (100W when used with an AC-DC power supply).

Class II transformers are designed and constructed to prevent electrical shock hazards by relying on double insulation or reinforced insulation, (terms common to many IEC and EN standards, including IEC 60664, IEC 61140 and IEC 61558-1)

Double Insulation is defined as insulation comprising basic insulation plus supplementary insulation.

Reinforced Insulation is defined as a single insulation system applied to hazardous live parts, which provides a degree of protection against electric shock equivalent to double Insulation.

To protect against electric shock, a Class II transformer does not rely on basic insulation only but includes additional safety precautions such as double insulation or reinforced insulation.

Class II transformers have no provision for grounding, protective earthing, or reliance upon installation conditions.

A Class II power supply rating label will show this symbol:
classIIsymb.png

For audio, to avoid ground loops, it only makes sense to use class II isolation transformers, without continuous protective PE circuit, with plug and double contact sockets, 2-wire power cord. It is the only way how to avoid ground loop via mains power cables, in case we are connecting two class I audio components.

2. Breve Tufvassons PFS 250 isolation transformer

Encased protective transformer suited predominantly for those applications that require protection against electric shock and high IP grade. PFS is a portable version, fitted with a handle, power cord, and an output socket with a plug. The transformer is made with class II insulation and protection grade IP44, thermal class of insulation B (130°C). Manufactured in compliance with EN61558-2-4 and EN61558-2-6. It is a 1:1 230/230V transformer rated at 250VA. It looks like this:

PFS250.JPG



I have measured 120 pF stray capacitance between primary and secondary and maximum of 8.6 uA (microamperes) leakage current from secondary to mains earth. This makes its use absolutely safe and in case of a single-pole touch of its live output against earth there is no danger of electric shock for the user. In fact it makes the connected appliance safer.

3. Audio test

The test circuit was connected as per the following schematics:

amp_isoltrafo_test.png



From PC notebook the DAC (Topping D10s) was connected by a 50 cm of single-ended link cable to the stepped attenuator and from the attenuator by 1.5 m of single-ended link cable to a power amplifier laded with 8 ohm resistor load. The output voltage from the load was connected to the balanced input (+IN, -IN, no ground) of the ADC USB soundcard connected to the PC. The PC was alternately supplied from:

a) directly from 230V mains which made the PC a class I component
b) via the PFS 250 isolation transformer to 230V mains which made the PC a class II component

So there was a ground loop from mains PE through the PC, USB, DAC, stepped attenuator, link cables and power amp in case (a). This ground loop was interrupted by PFS 250 in case (b).

Measurements

Noise voltage
at amplifier output – class I (a) red, class II (isol. tr) green

classI_vs_isol_noise.png

We can see at least 10 dB lower noise floor with PFS 250 and more importantly the mid and high frequency audible spurious components have disappeared.


Distortion at amplifier output – class I (a) red, class II (isol. tr) green

classI_vs_isol_THD.png


Same result, distortion is considerably lower with the PFS 250

4. Conclusion

The class II safety isolation transformer may make a huge improvement if the user has troubles with ground loop hum, buzz and whistles in his system. It is a safe way to fix the issue.
 
Last edited:
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
THD and THD+N vs. power plots at low level, without and with the isolation transformer. We can see considerable improvement of distortion with the isolation transformer, which results in both objectively and subjectively cleaner sound and better low level signal resolution. And the buzz and whistles resulting from the ground loop are gone.

classI-classI_thdpower.png


classII-classI_thdpower.png
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
This is the recorded noise from the power amplifier output, for both configurations, with and without the isolation transformer, in 2 flac files. Full scale, 0dBFS of the data file refers to 10Vrms amplifier output voltage.
 

Attachments

  • classI_II_noise.zip
    1.1 MB · Views: 68

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
What happens if you would use the TrippLite instead (which leaves safety ground connected) ?
Immitate by feeding through the isolation transformer and making a connection between safety ground from mains and that of mains ?
The Tripplite also connects N to safety ground. What changes when doing this ?
I have seen the D10 mentioned often in conjunction of ground loops. Perhaps internal routing of ground(planes) or some small resistance is added.
Other SE DACs may not do this in the same level.
Perhaps draw in the 230V input from the amp. Maybe leakage current is responsible instead of safety ground.
Same mains outlets used for power amp/isolation transformer input ?

You state the notebook is class-I but I assume it has an external SMPS which is class-I and feeds DC to the notebook.
Mains leakage (through Y cap) could be the root cause of the ground loop.
Can this be solved in the same amount of improvement with an USB isolator or DI box ?
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
What happens if you would use the TrippLite instead (which leaves safety ground connected) ?
Immitate by feeding through the isolation transformer and making a connection between safety ground from mains and that of mains ?
The Tripplite also connects N to safety ground. What changes when doing this ?
I have seen the D10 mentioned often in conjunction of ground loops. Perhaps internal routing of ground(planes) or some small resistance is added.
Other SE DACs may not do this in the same level.
Perhaps draw in the 230V input from the amp. Maybe leakage current is responsible instead of safety ground.
Same mains outlets used for power amp/isolation transformer input ?

You state the notebook is class-I but I assume it has an external SMPS which is class-I and feeds DC to the notebook.
Mains leakage (through Y cap) could be the root cause of the ground loop.
Can this be solved in the same amount of improvement with an USB isolator or DI box ?
To answer your questions:

1. A mere continuous safety ground closed across the isolation transformer makes the same effect as case (a), in other words direct class I without isolation transformer

2. N connected to safety ground makes not much difference. In fact my home mains network is an old-style L-PEN 2-wire system.

3. D10s used is no more prone to the ground loop effect than another DACs with single-ended RCA output. Of course the DACs with balnced output and the amp with the balanced input are usually a cure, if properly wired inside.

4. Safety ground is responsible, not the amp leakage current. The input RCA connector of the amp is directly connected to the chassis and the chassis is grounded (connected to mains PE via the mains cord and plug).

5. Same mains outlet is used for both PC and the amp in THIS test.

6. The USB Olimex isolator makes the same cure as the PFS 250 isolation transformer, with same measured results.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
1: So connecting the laptop to a 2 prong mains socket (possible in Europe) without connecting safety ground solves the problem.
This would be a much cheaper solution in that case than using an isolation transformer.
The TrippLite Amir tested thus does not solve this particular issue nor will sound quality improve.
In your case audible gremlins are removed. When these gremlins were not be there to begin with, then there would be no improvement.

Most people using PC or laptop connected to stereo systems may not have ground loop issues. For them the TrippLite nor the PFS20 will offer any audible improvements.

When safety ground is the issue (for the laptop power supply) then the Tripplite tested by Amir cannot solve the issue.
@Dan Clark reported he solved ground loops using these isolators. It would thus seem that not all ground loop issues (weird noises that should not be there) are not always caused by safety ground.

6: As expected, as long as the loop path of unwanted signal is broken somewhere all is good.
 
Last edited:

xaviescacs

Major Contributor
Forum Donor
Joined
Mar 23, 2021
Messages
1,494
Likes
1,971
Location
La Garriga, Barcelona
Sorry to bother.
1: So connecting the laptop to a 2 prong mains socket (possible in Europe) without connecting safety ground solves the problem.
This would be a much cheaper solution in that case than using an isolation transformer
Is this why some amplifiers like Marantz or Denon don't have grounded mains (two pin only)?
In your case audible gremlins are improved. When these gremlins would not be there to begin with, then there would be no improvement.
IMHO, a thread title like "Breve Tufvassons PFS 250 protective isolation transformer test against ground loops" or something of the sort, will represent better what is being presented.

In any case, thanks for you time an effort @pma.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,785
Likes
1,519
Is this why some amplifiers like Marantz or Denon don't have grounded mains (two pin only)?
Simply put, Yes. if there is no ground pin there is no ground loop.
(there still is over the parasitic capacitance... but way higher impedance)

@pma
Thanks for making this demonstration.
Can you please also demonstrate how an isolation transformer can lower leakage (mains to signal ground capcitive coupling) in an class 2 device.

In your case audible gremlins are removed. When these gremlins should not be there to begin with, then there would be no improvement.
Some amount of gremlins are always present. or at least in most real world systems.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,785
Likes
1,519
But how isn't that against regulations with an equipment that has a mains input? :rolleyes:
Class 2 isolation.

Strange phrasing of an question actually.
It is not against regulations "because regulation says so"
There is no globally unified regulation on his anyways. but in most of the world class 2 is absolutely fine.
 
Last edited:
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
1: So connecting the laptop to a 2 prong mains socket (possible in Europe) without connecting safety ground solves the problem.

Yes, almost. It is possible with some but not all notebooks. However not equivalent, the notebook SMPS has much higher stray capacitance and the result is measurably worse than with the isolation transformer. But it is impossible to use a 2 prong socket for the desktop PC where the issue is absolutely same and is again cured by the isolation transformer.

I am concentrated on reduction of noise and interfering voltages in the complete audio system, which is in line with my professional work on instruments and systems for measurements in highly hostile EM environment like VHV testing labs and high current short-circuit testing of high power circuit breakers. The issues, though on very different absolute number level, are very similar. We share the same physics :).

6: As expected, as long as the loop path of unwanted signal is broken somewhere all is good.

Yes, but the isolation transformer makes a better cure than either the Olimex isolator or usual class II instrument. Olimex has been fine for audio range however for higher BW there is a lot of mess in the spectrum, probably due to on-board supply leakage. Compared to "ususal" class II component like CD player etc. the isolation transformer has almost always lower capacitance between primary and secondary so the capacitive loop current is lower.
 
Last edited:
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Class 2 isolation.

Strange phrasing of an question actually.
It is not against regulations "because regulation says so"
There is no globally unified regulation on his anyways. but in most of the world class 2 is absolutely fine.
Yes but the difference between class 2 and class II must be taken into account (see post #1) because it may be confusing to many readers who might assume it is the same thing.

 

xaviescacs

Major Contributor
Forum Donor
Joined
Mar 23, 2021
Messages
1,494
Likes
1,971
Location
La Garriga, Barcelona
Class 2 isolation.

Strange phrasing of an question actually.
It is not against regulations "because regulation says so"
There is no globally unified regulation on his anyways. but in most of the world class 2 is absolutely fine.
Yeah, I get it, I was asking why regulations allow this. In some cases ground is a must and in some cases not. I guess there are other safety mechanisms to avoid one getting electrocuted if there is a current leak into the chassis, so one of this Marantz amplifier is equally safe than another with ground connection.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,785
Likes
1,519
Yes but the difference between class 2 and class II must be taken into account (see post #1) because it may be confusing to many readers who might assume it is the same thing.
And so was i.
Actually i was referring to class ll devices

I was asking why regulations allow this
Why would they not?

In some cases ground is a must and in some cases not.
But "ground" don’t need to be connected to PE
airplanes seam to work fine without earth ground.

I guess there are other safety mechanisms to avoid one getting electrocuted if there is a current leak into the chassis,
first thing would be to make sure this can’t happen.
Generally the testing to make make sure this can’t happen. is expensive tahts why many "high end" aka low volume hifi devices don't do it and just ground all exposed parts to PE. (protective earth)
But an RCD helps not getting electrocuted even with class II devices without ground.


so one of this Marantz amplifier is equally safe than another with ground connection.
Most consumer devices do this. it is "save" and an RCD gives extra protection.
 

xaviescacs

Major Contributor
Forum Donor
Joined
Mar 23, 2021
Messages
1,494
Likes
1,971
Location
La Garriga, Barcelona
IMHO a device with a metal chassis asks to be grounded, regardless of other safety levels as RCD. So, rewording my original question, and trying to focus a bit more to avoid talking about planes, are those companies sacrificing a bit of safety just to avoid ground loops? Or there are other reasons not to ground those amplifiers?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Yes, almost. It is possible with some but not all notebooks. However not equivalent, the notebook SMPS has much higher stray capacitance and the result is measurably worse than with the isolation transformer. But it is impossible to use a 2 prong socket for the desktop PC where the issue is absolutely same and is again cured by the isolation transformer.

I am concentrated on reduction of noise and interfering voltages in the complete audio system, which is in line with my professional work on instruments and systems for measurements in highly hostile EM environment like VHV testing labs and high current short-circuit testing of high power circuit breakers. The issues, though on very different absolute number level, are very similar. We share the same physics :).



Yes, but the isolation transformer makes a better cure than either the Olimex isolator or usual class II instrument. Olimex has been fine for audio range however for higher BW there is a lot of mess in the spectrum, probably due to on-board supply leakage. Compared to "ususal" class II component like CD player etc. the isolation transformer has almost always lower capacitance between primary and secondary so the capacitive loop current is lower.

I think we both know where the real problem lies though and using the PFS is not the real solution here ;)
Yes, you showed that a real isolation transformer can cure a problem in a system.
You say the D10 is not the issue so I expect you have used other DACs that have the same issue.
Knowing your technical prowess I guess you know where I am going.
Time to remove the root cause and as only 1 remains you know what I am hinting at.
Why not address the real issue and free your transformer ?
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,785
Likes
1,519
IMHO a device with a metal chassis asks to be grounded,
Maybe bey your standards. but not necessarily regulations.
are those companies sacrificing a bit of safety just to avoid ground loops?
No.
Or maybe they do but you sure can’t tell from this.

Of cause there are unsafe class 2 and ll decides! many cheap phone chargers...
but there are also unsafe class l devices
 
Last edited:

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,636
Likes
3,593
Location
Sweden, Västerås
We used these in my line of work before scopemeters , to use with oscilloscopes that had chassis ground to measure for example across thyristor gates in large drives DC drives . Do not touch the chassi only the plastic knobs on the scope :facepalm: By today's standards this was not very safe behavior :rolleyes: I even recognize the brand .
Nice measurements explaining what they do .

Double insulated stuff is supposedly built to a standard and should be safe .

And in 2022 you have a GFCI in your house to protect you .
 
  • Like
Reactions: pma

DDF

Addicted to Fun and Learning
Joined
Dec 31, 2018
Messages
617
Likes
1,355
@pma Pavel, thanks for sharing these test results. It's good to get testing based on sound electrical engineering.
Excellent quality aftermarket classII laptop power supplies can be had for $35, accomplishing the same as the isolation transformer. It completely eliminated any background noise in my system. The best audio upgrade I ever made:
 
  • Like
Reactions: pma
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
@pma Pavel, thanks for sharing these test results. It's good to get testing based on sound electrical engineering.
Excellent quality aftermarket classII laptop power supplies can be had for $35, accomplishing the same as the isolation transformer. It completely eliminated any background noise in my system. The best audio upgrade I ever made:
Thank you Dave, I agree. As you know, I use the PFS 250 class II isolation transformer for multiple instrumentation purposes so it was a temptation to make some basic audio measurements as well. BTW, the cost was about $90, so not especially high.
 
  • Like
Reactions: DDF
Top Bottom