-
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!
Mains Leakage Currents
- Thread starter Panelhead
- Start date
Those companies who do well -- chinese or otherwise -- have proper engineering skills. They are not chasing placebo effects that they themselves don't understand as is the case with you and John. Please use common sense and not get into arguments here where science and objective data rules. You will always be at the losing end of it.Exaggerations, misdirections, personal attacks, whatever. I’ll leave you to preside over your minions and to continue to promote cheap Chinese DACs as the pinnacle of audio excellence.
But yes, thank you for resigning from the discussion.
Exaggerations, misdirections, personal attacks, whatever. I’ll leave you to preside over your minions and to continue to promote cheap Chinese DACs as the pinnacle of audio excellence.
No, just ignoring misguided rhetoric and getting to the root-cause of the issue.
- Thread Starter
- #24
Even though this discussion broke down. I gathered a few ideas on how to power my interface.
1. As far as connecting the wall outlet ground directly to the output of the switcher supply it may be happening in my interface. I doubt it. My computer is totally isolated by the Corning Thunderbolt connection. The amplifier does not. So I suspect the leakage is high level from an iPower.
2. After a few hours of swapping and listening to different tracks the iPower is a better choice than the Jay’s LPS. The Jay’s may not be up to delivering the 1.25 amps pulled by the interface. Sonically it does not deliver the resolution of the 49.00 iPower.
3. My connecting of iPower output back to wall ground does not seem like a good resolution to eliminate leakage. That is tied to neutral in the breaker box and will reintroduce power line hash directly into the dc feed for the dac. Kind of exactly what a upgraded power supply should eliminate.
1. As far as connecting the wall outlet ground directly to the output of the switcher supply it may be happening in my interface. I doubt it. My computer is totally isolated by the Corning Thunderbolt connection. The amplifier does not. So I suspect the leakage is high level from an iPower.
2. After a few hours of swapping and listening to different tracks the iPower is a better choice than the Jay’s LPS. The Jay’s may not be up to delivering the 1.25 amps pulled by the interface. Sonically it does not deliver the resolution of the 49.00 iPower.
3. My connecting of iPower output back to wall ground does not seem like a good resolution to eliminate leakage. That is tied to neutral in the breaker box and will reintroduce power line hash directly into the dc feed for the dac. Kind of exactly what a upgraded power supply should eliminate.
AnalogSteph
Major Contributor
Us folks in non-polarized mains plug territory hold a potential advantage here - on the kind of supply with just one Y cap going from secondary ground to mains neutral, we can just turn the plug around. Doesn't help if it's two caps connected to both phase and neutral, of course. (This design originally stems from IEC Class I SMPS, i.e. those with a 3-prong power plug including protective earth, where mains leakage currents are promptly disposed of in PE. You use 2 caps from phase and neutral to PE because, again, non-polarized power plugs. Using the same kind of design but more or less just disconnecting PE to obtain a Class II supply is nothing short of moronic, but that's exactly what happened - there are many supplies like this out there. What should have been done is using the PE connection for the mains filter, and then running a Y cap from secondary-side ground to PE. I think Thinkpad supplies do it roughly like that, but with an R||C combo between secondary-side ground and PE. Granted, there still is the issue of countries where ungrounded outlets remain common - more than you think.)
One way to shut up these two-cap-to-secondary-ground mains filters is using "technical" AC, which doesn't have a phase (with all the voltage) and neutral (with no voltage), but rather splits mains voltage between the two conductors evenly, with zero common-mode voltage. (Some countries actually use this as their standard mains. Wasn't it Norway or something?) Ideally you want an isolation transformer with a center-tapped secondary for this, but a conventional isolation transformer with good CMRR should also go a long way (maybe with a pair of equal-value resistors of suitable power and voltage handling to PE in the box for some extra persuasion). Unfortunately isolation transformers tend to be somewhat uncommon, a bit costly and generally far bigger than required for a wall wart SMPS or two.
One way to shut up these two-cap-to-secondary-ground mains filters is using "technical" AC, which doesn't have a phase (with all the voltage) and neutral (with no voltage), but rather splits mains voltage between the two conductors evenly, with zero common-mode voltage. (Some countries actually use this as their standard mains. Wasn't it Norway or something?) Ideally you want an isolation transformer with a center-tapped secondary for this, but a conventional isolation transformer with good CMRR should also go a long way (maybe with a pair of equal-value resistors of suitable power and voltage handling to PE in the box for some extra persuasion). Unfortunately isolation transformers tend to be somewhat uncommon, a bit costly and generally far bigger than required for a wall wart SMPS or two.
- Thread Starter
- #27
Guess I will need to get the Noise Trapper down and see if it has any effect. Did not show much back when new. Just was a fancy power strip for line level sources.
Not sure if these leakage currents are audible after entering the interface. There are filter caps, multiple regs and dc to dc converters.
Not sure if these leakage currents are audible after entering the interface. There are filter caps, multiple regs and dc to dc converters.
AnalogSteph
Major Contributor
What sort of interface are we talking about anyway?
Common-mode currents should not (in theory) be impressing USB and its differential signalling much, but I mean, the same ought to be applying to HDMI, too, and there have been plenty of cases of ground loops impacting signal integrity on certain devices (like newer Denon AVRs). In fact, I also remember one case where USB devices on a computer would briefly check out whenever the fridge or something was turning on. (I think it was connected to an audio system reasonably far away via a USB sound device, at least there was some sort of ground loop involved.) My bets in each case are on the equivalent of a Pin 1 Problem, i.e. a grounding issue that lets shield currents affect signal integrity.
Common-mode currents should not (in theory) be impressing USB and its differential signalling much, but I mean, the same ought to be applying to HDMI, too, and there have been plenty of cases of ground loops impacting signal integrity on certain devices (like newer Denon AVRs). In fact, I also remember one case where USB devices on a computer would briefly check out whenever the fridge or something was turning on. (I think it was connected to an audio system reasonably far away via a USB sound device, at least there was some sort of ground loop involved.) My bets in each case are on the equivalent of a Pin 1 Problem, i.e. a grounding issue that lets shield currents affect signal integrity.
Last edited:
- Thread Starter
- #29
My setup eliminates these issues. The Corning Optical Thunderbolt cable totally isolates the computer from the interface (dac). There is no conductor in the cable, just the fiber.
The interface is a Focusrire Clarett 4Pre if that is the question. The TB version.
Hooked up the Noise Trapper last night. Checked and the system showed no hum pickup from inserting it in the rack. I remember needing to move it due to hum pickup with phono stage when used before. That was late 80’s or early 90’s.
It is labeled as a 500 watt unit. Big, ugly, isolation transformer.
The interface is a Focusrire Clarett 4Pre if that is the question. The TB version.
Hooked up the Noise Trapper last night. Checked and the system showed no hum pickup from inserting it in the rack. I remember needing to move it due to hum pickup with phono stage when used before. That was late 80’s or early 90’s.
It is labeled as a 500 watt unit. Big, ugly, isolation transformer.
Being rude and obnoxious, in fact almost gloriously obnoxious will inspire offence in folks. Some have more grace than others when mitigating those feeling in their posts.Exaggerations, misdirections, personal attacks, whatever. I’ll leave you to preside over your minions and to continue to promote cheap Chinese DACs as the pinnacle of audio excellence.
You made a product that lead to some getting worse performance and at best obtaining no benefit, argued tooth and nail about it and were proved to be wrong. You then come back after you fixed it ( so now its influence is likely is just benign other than the effect on folks wallets) making out like your the champion of the world , belittling people left right and centre being fantastically brash and obnoxious.
No grace or decency whatsoever while maintaining a conviction that you are not just technically superior but morally too it seems, wow .. how presidentially à la mode.
- Thread Starter
- #31
Well I learned a few things from this thread. Mainly it has been that it is good to try items and listen to their impact on the system.
Hooked up a Noise Trapper isolation transformer in the system. Only plugged in an iFi iPower 12 volt, 1.8 amp low noise switching supply. Used an Eupen medical grade shielded cable from wall to transformer. Expected nothing positive from usage 25 years ago.
Well I was wrong. Made a nice improvement. The hardness I have been hearing when the recording got loud has disappeared. I thought it was loudspeaker related.
Also think note decays are a little cleaner and extend slightly longer. I wondered if this was due to stacking the power amp directly above the Noise Trapper in the rack. The field from the transformer had created issues before and required moving it away from sources.
The input transformers in the amplifier have shielded cans and the case of the amp is lined with mu-metal. Cannot hear any hum with ears in the mouth of the midrange or bass horns. I think the 3/4” sheet of black granite under the amp helps here too. There is a fair amount of magnetite in black granite. Magnets stick to black better than grey or green granite, and white granite is almost non-magnetic.
Hooked up a Noise Trapper isolation transformer in the system. Only plugged in an iFi iPower 12 volt, 1.8 amp low noise switching supply. Used an Eupen medical grade shielded cable from wall to transformer. Expected nothing positive from usage 25 years ago.
Well I was wrong. Made a nice improvement. The hardness I have been hearing when the recording got loud has disappeared. I thought it was loudspeaker related.
Also think note decays are a little cleaner and extend slightly longer. I wondered if this was due to stacking the power amp directly above the Noise Trapper in the rack. The field from the transformer had created issues before and required moving it away from sources.
The input transformers in the amplifier have shielded cans and the case of the amp is lined with mu-metal. Cannot hear any hum with ears in the mouth of the midrange or bass horns. I think the 3/4” sheet of black granite under the amp helps here too. There is a fair amount of magnetite in black granite. Magnets stick to black better than grey or green granite, and white granite is almost non-magnetic.
- Thread Starter
- #33
You may be correct. The Noise Trapper at best did nothing when I used it before. Has been sitting in attics for 25 years. But it was always feeding linear supplies. My first source with switching supply was the Brilliant supply in a Linn Linto. That was a few years after the NT was relegated to attic duty.
Stupidly old thread, but it came up on google and I had to reply because of how wrong this part is...
They were probably confused by your request as...
You appear to be using their GST series power supplies (which capacitively couple the negative output to ground) and not their GSM series (which short the negative output to ground).
Here's the datasheet for the GSM40A series: https://www.meanwell.com/webapp/product/search.aspx?prod=GSM40A
This is apparent on the second page in the withstand voltage rating section; "O/P-FG:SHORT" means that that the output is shorted to ground. A failure to read and/or understand a products datasheet (and thus use/specify the wrong part) is not a failure by the manufacturer.
They were probably confused by your request as...
You appear to be using their GST series power supplies (which capacitively couple the negative output to ground) and not their GSM series (which short the negative output to ground).
Here's the datasheet for the GSM40A series: https://www.meanwell.com/webapp/product/search.aspx?prod=GSM40A
This is apparent on the second page in the withstand voltage rating section; "O/P-FG:SHORT" means that that the output is shorted to ground. A failure to read and/or understand a products datasheet (and thus use/specify the wrong part) is not a failure by the manufacturer.
Similar threads
