Popular Post gstew Posted October 1, 2017 Popular Post Share Posted October 1, 2017 On 9/24/2017 at 5:44 PM, JohnSwenson said: Remember the leakage from an SMPS contains both high impedance and low impedance components, the ground shunt gets rid of the high impedance components. The low impedance components are still left. These components are usually still greater in amplitude than many linears, but some linears might be a bit more. BUT since what is left is the low impedance components, they CAN easily be blocked. You can do the same thing with linears, which gets rid of THEIR high impedance components, just leaving the low impedance components, which are REALLY low. I'm putting together another post in the next few days that shows how this works with network systems (I'll post that in the thread on the Sonore forum). John S. John, I don't understand the context & meaning of 'high impedance and low impedance components' of the SMPS noise. I've never heard of noise having an impedance. Can you provide some background and definition on that? TIA! Greg in Mississippi mansr and rickca 2 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 1, 2017 Share Posted October 1, 2017 27 minutes ago, sandyk said: Greg John will correct me on this if I am wrong, but HF noise from an SMPS would be coupled into the secondary side ( the other side of the HF transformer) via the obligatory low value, and high voltage, capacitors that are connected between primary and secondary to meet FCC RF/EMI standards. This would be a high impedance source. Incidentally, I have been looking at the high voltage leakage from my DTV STB as it connects into my TV and main system. It hovers around 100VAC from the STB chassis to mains earth, but when the STB is connected via HDMI to the TV it is reduced to Zero. I have disconnected the STB's Antenna and coax SPDIF leads to verify this. The interesting thing is that with nothing else plugged into the TV except the HDMI plug , and even when the TV's 2 wire mains plug is removed from the mains socket, the leakage STILL remains at Zero volts ! This shunting obviously is a good thing , but I am at a loss to see how this can happen. Alex Alex, Thanks for the response. Very curious if that's it or if something else is involved. I DO think I understand the need for very high impedance probes to measure it. AND also likely that the high impedance is part of why shunting it to ground works so well. Also curious about your DTV STB & TV puzzle. My 1st thought are some high-ohm (50k-1M) resistors across the HDMI inputs to ground in the TV and that this is sufficient to dissipate the high-impedance leakage in the resistors. Curious what John has to say on these questions/puzzles. Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 1, 2017 Share Posted October 1, 2017 21 minutes ago, sandyk said: Hi Greg I can't understand how this can happen where the TV is SMPS powered , and the 2 wire mains plug has been unplugged. There shouldn't be ANY connection to mains earth at all. Regards Alex Alex, I'm just musing that with at least some of the components of the leakage being super-high-impedance, a high-impedance resistance across it may not ground it, but in effect short it out. You could try 1M resistors across the signal pins to ground at the end of the HDMI cable... I can't imagine that would hurt anything, but could demonstrate a possible mechanism. Later! Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 4, 2017 Share Posted October 4, 2017 3 hours ago, ElviaCaprice said: After further listening, plugging in the ground plug and unplugging, my observations are that on my system this DIY grounding has no noticeable effect. I may be missing something... if so, please forgive me. I don't see in your post that you were grounding the negative side of SMPS outputs (like the Meanwell). As I understand it, the grounding trick is designed to sharply reduce the amount if 'high-impedance' noised transmitted from the SMPS into the powered unit. Is what I'm missing that you were using Meanwells to energize the LPS-1s? Or is something else? Pictures maybe? Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 4, 2017 Share Posted October 4, 2017 On 10/2/2017 at 2:19 PM, JohnSwenson said: Any electrical signal (useful or noise) has a "source impedance". You can think of this as a resistor the signal has to go through before it gets to where you are looking at it. It is an impedance because it can vary with frequency. As a concrete example, lets say your source has an impedance of 1k ohms, if you apply that to a 1k ohm resistor to ground, the signal level will be cut in half. That is actually how you usually measure output impedance, run the input through a known resistance to ground and measure the level across the resistor, the output impedance acts as a voltage divider with the known resistor, from the resulting voltage you can calculate the source impedance. For any given "noise source" there is at least one mechanism generating that noise. The mechanism will have a particular impedance associated with it. In many cases more than one mechanism is involved with generating noise, each of these mechanisms may have a different output impedance. There may also be different frequency response issues with the different mechanisms. This seems to be the case with SMPS, there seems to be at least two different mechanisms that cause the leakage and they seem to have very different impedances. I don't know what those mechanisms ARE, i have not spent time in figuring that out, I'm not really interested in building my SMPS so I don't really care what actually causes it, especially since it will take a LOT of work to find out and I would much rather spend my time working on other things. Leakage current causes issues in audio systems when it flows through a conductor, creating a voltage across that conductor which causes something to not behave the way you would like it. In some cases this is just directly creating noise on the shield of an interconnect and the receiving circuit sees this as noise. In a DAC this can show up as noise developed on a ground plane that can modulate an oscillator causing increased jitter on the clock. There are two ways you can attenuate the leakage noise, you can put a resistance in series with it, or you can shunt it. Lets cover both of the separately. The series resistor form works like this: you have the source impedance and you stick a resistor in series. Lets look at some possible values and outcomes. Say 100 ohm output impedance and you put 1 mega ohm in series, that is going to attenuate the noise drastically. But what if the source is 100 mega ohms, then that 1 mega ohm resistor is not going to do very much (a VERY slight attenuation). So for high impedance series resistor works good for lower impedance source, but not well for high impedance source. The shunt form works by shunting the source around your source around the "test point", frequently to ground. Say you have the 100 ohm source and you shunt with 1 mega ohm, nothing happens, but of the source is 100 mega ohms, that 100 mega ohm shunt will dramatically decrease the amplitude. I know the output of the SMPS is a combination of impedances by running a bunch of these shunt and series tests with different values and seeing what I get. The only way to get the results I saw is if the source consists of both high impedance and low impedance components at the same time. The upshot is that it takes BOTH methods to get rid of the noise, both a shunt and a series. John S. John, Many thanks for the explanation. Once you brought in the term 'source impedance', it all started to make sense! AND then you had to go and give more details! I now know a lot more about noise production mechanisms and how one goes about detecting them and eliminating them than I ever thought there was to know! VERY USEFUL!!! DANG, you give great explanations! Again, THANKS! Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 4, 2017 Share Posted October 4, 2017 1 minute ago, ElviaCaprice said: Greg, that is exactly what I'm doing. All my LPS-1's are powered by a meanwell. So I stuck the spade of the grounding wire to the outside barrel of the lead running from the meanwell to the LPS-1 input. At the input barrel the spade is attached to the outside of the barrel and plugged in to the input of the LPS-1, trapping the spade onto the barrel. Is this not correct? Thanks for the clarification. Yup, what you did should ground out that darned old 'high-impedance' SMPS noise. Sounds like it is all working as it should to me. While I now use only linear supplies to power my LPS-1s, I did get my 1st 2 with Meanwells. While I had some issues with buzzing in my setup (where the buzz was induced into a TVC in certain situations), I couldn't hear any difference in the performance of the gear I powered by those LPS-1s whether energized by the Meanwells or a couple of linear supplies. Greg in Mississippi ElviaCaprice 1 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
Popular Post gstew Posted October 29, 2017 Popular Post Share Posted October 29, 2017 John & Alex, Thanks for the great explanations & education on how AC power supply leakage works, how & why it was characterized, why that characterization missed the 'high-impedance' type of leakage, and what you can do to diminish its effects on our audio systems. The wealth of information you've both shared here (and embodied in your products) has made a world of difference in the sound quality of many audio systems, mine VERY much included. Please help me make sure I have understood all of this... First, linear AC-DC power supplies, have only the low-impedance type of leakage, though at various levels based on their 'different LPS implementation'**. AND all SMPSs, by nature of how they function, have both of the low-impedance & VERY high-impedance types of leakage. Uptone Audio LPS-1's (and similar non-AC connected supplies, such as pure battery supplies) do not pass the low-impedance leakage through the power supply feed, whether that leakage originates from an LPS or SMPS. Grounding the negative side of the output of SMPSs effectively eliminates the majority of the high-impedance leakage, preventing it from being passed through to the driven piece, whether to an Uptone LPS-1 or another piece of equipment, such as a network switch. AND grounding the negative side of the power input into one of the tested & confirmed switches (I won't list them here, that list may grow over time) ALONG with proper Ethernet cable connection (currently using alternating jacks, but that also may change over time) will prevent high-impedence leakage from being passed through the switch to downstream connected gear. Have I got it right so far? Next, a few questions. First, there are situations like my systems, where ALL upstream supplies (WiFi router, file server computer, switch, FMCs) and source supplies (player computer, interface cards, DAC, output stages) are linear supplies. AND in the case of the interface cards (RPi I2S isolator and reclocking cards) and DAC, those supplies are LPS-1s. BUT the amps I currently have in my systems are powered by SMPSs. In building / modifying / tweaking these amps several years back, I did find that having the negative output of the power supply grounded produced the best SQ, so they are all configured that way. IS there anything else I should be doing to reduce the effects of leakage currents in my setups? Is it worthwhile converting these amps to use linear supplies (where I can) or going to amps that already use linear supplies (where I cannot convert them)? Or can one not make a blanket statement here? Second, I assume that leakage currents are produced by the AC to DC conversion process in the supply, not anything afterwards. SO in gear that uses switching DC-DC converters (such as most computers, router, switches & FMCs out there), they are not producing any more leakage currents than ones that use linear regulators? (Though they MIGHT be producing more noise on the internal DC supply line, depending on the quality of the DC-DC converter and its implementation?) Third, grounding the negative output of a linear supply does nothing to diminish its low-impedance leakage currents, though it might be helpful in some situations such as when powering one of the recommended switches and using that grounding to block high-impedance leakage from upstream gear being passed to downstream gear, correct? Fourth, leakage from a linear supply and the sonic effects it has is separate and distinct from other qualities of the supply that impact SQ, such as low noise, broad bandwidth, symmetrical response, and wide-band low output impedance? (Which is why supplies such as your JS-2 & Paul Hynes well-regarded supplies still matter, correct?)* Finally, again, thanks for both the great information and great products! Greg in Mississippi *P.S. I wrote up one experience how some of those other quality factors of a linear supply can matter EVEN when energizing LPS-1's, due to how noise they can create can be transmitted via the AC power connection to other supplies in my post here: https://www.computeraudiophile.com/forums/topic/29313-discussions-of-alternate-quotenergizingquotcharging-ps-units-for-use-with-ultracap-lps-1-not-that-any-will-make-any-difference-to-output/?do=findComment&comment=691641 . Note that I heard these differences (and eliminated them) before I even hooked the supplies up to an LPS-1 in the system... they were just plugged into the same outlets as my other gear with nothing connected on their outputs. I just had a related experience, where by chance I received a discarded linear regulated AC-DC adapter that had been dropped and broken open. Looking at it prompted me to crack open up some of the 5V linear regulated adapters I'd purchased from Jameco and apply most of the same techniques I used to quiet my 4x LPS-1 energizing supplies, such as good Schottky diodes, John Swenson's transformer secondary ringing damping snubber, better DC filtering, and filtering on the AC input. What surprised me was the magnitude of the difference I heard replacing 2 stock Jameco 5v supplies (powering sending and receiving FMCs) in one of my systems with the modified ones. Of course, the changes I made should have reduced both the noise fed back to the AC line AND the noise fed to the powered equipment. Hmmm, given what I heard, I really need to try LPS-1's there! We really need a DIY forum on CA for things such as this. Modifying those adapters would not be an easy project for someone with no DIY experience, but a snap for someone who has done at least a little electronics DIY'ing. Alex, would you be ok if I started a separate thread on your sponsored forum to detail those mods? **P.P.S. John, can you point to or further explain what 'different LPS implementation' schemes produced different levels of leakage? MikeyFresh and Puma Cat 2 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted October 29, 2017 Share Posted October 29, 2017 3 hours ago, jabbr said: @JohnSwenson nice investigation Help a simple country boy out please: the term “high impedance leakage current” is making me a little dizzy as I try to drink my coffee on this Sunday AM. Impedance isn’t a term normally used to qualify a current so perhaps we should rename this before it takes on a life of its own — you mean “high voltage/impedance” = current ... what is a better term? Also these currents are going to be way more important at higher frequencies. I think if a circuit were made available to precisely illustrate the excellent point you are making here (this is all about parasitic capacitances and inductances) this frequency point will be better illustrated. (Yes— high impedance probes are essential when working with RF) @Jabbr, Great minds question alike (except one of them hasn't finished his Dual Bank Floating Supercap Supply DIY board yet... hint, hint!). SO I asked pretty much the same question a few weeks ago: On 10/1/2017 at 10:57 AM, gstew said: John, I don't understand the context & meaning of 'high impedance and low impedance components' of the SMPS noise. I've never heard of noise having an impedance. Can you provide some background and definition on that? TIA! Greg in Mississippi AND John gave one of his textbook clear answers... even a DIY tech like myself understood: On 10/2/2017 at 2:19 PM, JohnSwenson said: Any electrical signal (useful or noise) has a "source impedance". You can think of this as a resistor the signal has to go through before it gets to where you are looking at it. It is an impedance because it can vary with frequency. As a concrete example, lets say your source has an impedance of 1k ohms, if you apply that to a 1k ohm resistor to ground, the signal level will be cut in half. That is actually how you usually measure output impedance, run the input through a known resistance to ground and measure the level across the resistor, the output impedance acts as a voltage divider with the known resistor, from the resulting voltage you can calculate the source impedance. For any given "noise source" there is at least one mechanism generating that noise. The mechanism will have a particular impedance associated with it. In many cases more than one mechanism is involved with generating noise, each of these mechanisms may have a different output impedance. There may also be different frequency response issues with the different mechanisms. This seems to be the case with SMPS, there seems to be at least two different mechanisms that cause the leakage and they seem to have very different impedances. I don't know what those mechanisms ARE, i have not spent time in figuring that out, I'm not really interested in building my SMPS so I don't really care what actually causes it, especially since it will take a LOT of work to find out and I would much rather spend my time working on other things. Leakage current causes issues in audio systems when it flows through a conductor, creating a voltage across that conductor which causes something to not behave the way you would like it. In some cases this is just directly creating noise on the shield of an interconnect and the receiving circuit sees this as noise. In a DAC this can show up as noise developed on a ground plane that can modulate an oscillator causing increased jitter on the clock. There are two ways you can attenuate the leakage noise, you can put a resistance in series with it, or you can shunt it. Lets cover both of the separately. The series resistor form works like this: you have the source impedance and you stick a resistor in series. Lets look at some possible values and outcomes. Say 100 ohm output impedance and you put 1 mega ohm in series, that is going to attenuate the noise drastically. But what if the source is 100 mega ohms, then that 1 mega ohm resistor is not going to do very much (a VERY slight attenuation). So for high impedance series resistor works good for lower impedance source, but not well for high impedance source. The shunt form works by shunting the source around your source around the "test point", frequently to ground. Say you have the 100 ohm source and you shunt with 1 mega ohm, nothing happens, but of the source is 100 mega ohms, that 100 mega ohm shunt will dramatically decrease the amplitude. I know the output of the SMPS is a combination of impedances by running a bunch of these shunt and series tests with different values and seeing what I get. The only way to get the results I saw is if the source consists of both high impedance and low impedance components at the same time. The upshot is that it takes BOTH methods to get rid of the noise, both a shunt and a series. John S. I hope that helps! Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted November 6, 2017 Share Posted November 6, 2017 26 minutes ago, pgbcincy said: @JohnSwenson Concerning Ethernet connection leakage topic: Ever have Baaske MED MI 1005 or EMO Systems EN-70HD or similar network isolation devices on your test bench? I'm wondering if those address some of the network leakage you describe. Although these are a more expensive proposition than the Network Switch solution. TIA. Paul Paul, Search is your friend! The short answer is yes he has and no they don't. Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
Popular Post gstew Posted November 8, 2017 Popular Post Share Posted November 8, 2017 3 hours ago, sandyk said: Perhaps the resistance value is low enough to completely remove the high impedance sourced residual noise as measured from a typical SMPS, but high enough to reduce the possibility of inadvertent earth loops causing very low level hum where you may need to use 2 different mains outlets ? I'm with Alex on this... reading John's explanation below on the 'high impedance' leakage, using simple Ohm's law, shunting a super-high source impedance like 100M with 0R or 15R will have immeaurably identical results in how well the source is attenuated. AND I've seen equipment with low-ohm ground lift resistors (such as the Hafler DH-200's I had as my 1st mid-Fi-ish gear back in the late 1980s) to minimize possible ground loops before. Even the legendary Enid Lumley recommended this in one of her then far-out articles. On 10/2/2017 at 2:19 PM, JohnSwenson said: Any electrical signal (useful or noise) has a "source impedance". You can think of this as a resistor the signal has to go through before it gets to where you are looking at it. It is an impedance because it can vary with frequency. As a concrete example, lets say your source has an impedance of 1k ohms, if you apply that to a 1k ohm resistor to ground, the signal level will be cut in half. That is actually how you usually measure output impedance, run the input through a known resistance to ground and measure the level across the resistor, the output impedance acts as a voltage divider with the known resistor, from the resulting voltage you can calculate the source impedance. For any given "noise source" there is at least one mechanism generating that noise. The mechanism will have a particular impedance associated with it. In many cases more than one mechanism is involved with generating noise, each of these mechanisms may have a different output impedance. There may also be different frequency response issues with the different mechanisms. This seems to be the case with SMPS, there seems to be at least two different mechanisms that cause the leakage and they seem to have very different impedances. I don't know what those mechanisms ARE, i have not spent time in figuring that out, I'm not really interested in building my SMPS so I don't really care what actually causes it, especially since it will take a LOT of work to find out and I would much rather spend my time working on other things. Leakage current causes issues in audio systems when it flows through a conductor, creating a voltage across that conductor which causes something to not behave the way you would like it. In some cases this is just directly creating noise on the shield of an interconnect and the receiving circuit sees this as noise. In a DAC this can show up as noise developed on a ground plane that can modulate an oscillator causing increased jitter on the clock. There are two ways you can attenuate the leakage noise, you can put a resistance in series with it, or you can shunt it. Lets cover both of the separately. The series resistor form works like this: you have the source impedance and you stick a resistor in series. Lets look at some possible values and outcomes. Say 100 ohm output impedance and you put 1 mega ohm in series, that is going to attenuate the noise drastically. But what if the source is 100 mega ohms, then that 1 mega ohm resistor is not going to do very much (a VERY slight attenuation). So for high impedance series resistor works good for lower impedance source, but not well for high impedance source. The shunt form works by shunting the source around your source around the "test point", frequently to ground. Say you have the 100 ohm source and you shunt with 1 mega ohm, nothing happens, but of the source is 100 mega ohms, that 100 mega ohm shunt will dramatically decrease the amplitude. I know the output of the SMPS is a combination of impedances by running a bunch of these shunt and series tests with different values and seeing what I get. The only way to get the results I saw is if the source consists of both high impedance and low impedance components at the same time. The upshot is that it takes BOTH methods to get rid of the noise, both a shunt and a series. John S. On 11/6/2017 at 12:52 PM, pgbcincy said: Well... I did search... but on this topic. I know John's a busy dude, dont want to waist his time. Thank you Greg!! Paul, You are very welcome. I initially did a search on all content just on Baaske, thinking I'd have to limit it to just John's posts. But no, the post was in the top few responses. Greg in Mississippi P.S. I've seen some questioning of whether it is better to ground the negative of an SMPS and possibly have the noise in the ground or leave it in the SMPS output. My speculations... 1. Just as above, the grounding of a very high impedance source of noise, based on Ohm's law, will work like grounding a battery or other voltage source. The source will try to generate a voltage across the resistance of the short (OR) to ground. Given the extremely high ratio of source impedance to shunt impedance, it won't generate much. AND any voltage generated will be dissipated as heat across the 0R shunt to ground (though VERY little heat, given the very small voltage). BUT with leakage being a common mode source of noise (same on both positive and negative outputs of the SMPS), since you are only shorting one side of the common mode output, you might still expect to see some of the noise on the output... which is what you saw in the measurements, though it is GREATLY diminished. 2. Even if the above is not correct, I'd MUCH rather have the noise shunted to ground and away from my equipment than in the power feed to a piece of equipment in my system where the noise makes its way into the circuit. 3. AND re-reading John's descriptions, it all seems much more complex than this, so what I wrote was likely not useful! P.P.S. When grounding gear for purposes OTHER than dealing with high-impedance leakage from SMPSs, I've found it useful to think of my system in zones between different methods of galvanic isolation... for example, my larger system has both a WiFi router (used as an access point) AND a Zotac micro computer (used as an LMS server for a Squeezelite player endpoint) before my switch. After the switch is an FMC to FMC link to a RaspberryPi Squeezelite player. An isolator board (Allo.com or Ian's) sites on top of the RPi, which feeds an Allo.com I2S Reclocker board and a custom-HotRod build of a Dial Audio RPi DAC. That feeds a S&B TVC, which then feeds my amps & speakers. By my count, I have 8 galvanically-isolated (at lower frequencies) zones.... 1. WiFi router 2. Zotac micro computer 3. Switch 4. Upstream FMC 5. Downstream FMC 6. RPi->isolator input 7. Isolator output->Kali->DAC->TVC input 8. TVC output->amps->speakers While they aren't isolated at the frequencies of the high-impedance SMPS leakage where John's recommended solutions work best, they ARE isolated at DC either via the eithernet transformers, wired-to-optical ethernet interfaces, I2S isolators, or TVC (transformer volume control). Honestly, I'm not sure how that is important yet, especially in the context of these discussions, but something to think about. MikeyFresh and Cornan 1 1 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted November 8, 2017 Share Posted November 8, 2017 29 minutes ago, austinpop said: Greg, Have you tried grounding each of these zones? That's the path I'm on. Rajiv, I'm trying some different combinations. So far I've gotten better results grounding the isolator output ->DAC->TVC input AND amp->speaker zones. I have the RPi->Isolator input zone grounded, but need to go back and listen with/without. I haven't grounded any of the other zones as yet. ALL of the power supplies on the gear through the DAC are linear supplies, so no high-impedance leakage worries there. AND there are currently 5 LPS-1s in the setup... on the switch (a ZyTEL recommended by my UK friend Dave Bruce, which so far bests one of John's originally recommended Cisco SG100D-08 switches and the Netgear GS605AV switch I originally used, though I need to re-test both powered by an LPS-1), 1 each on the DAC digital and analog supplies, 1 on the RPi isolator board, and 1 on the Kali I2S reclocker. Lots more experimentation to do here. BUT a significant upgrade I recently did was modifying the 5V Jameco Linear Regulated Adapters I use to power the FMCs (the upstream one directly, the downstream one via a TeckNet 5V USB charging battery that is being charged by the Jameco). I replaced the diodes with some good Schottky diodes, added John Swenson's transformer ringing snubber, increased the size of the raw DC and post regulator capacitors, and added an AC filter cap. AND also added a C-L filter (built in-between an AC plug and socket) that provided the same level of filtering I used in my DIY 4-LPS-1 energizing supplies (and BTW, all of the other computer & networking linear supplies in my setups). I was inspired to do this by John's comments that the Jameco's have very low low-impedance leakage here: On 9/14/2017 at 12:58 AM, JohnSwenson said: I have been doing a bunch of leakage tests on a bunch of power supplies (what I had on hand). I have tested both the Jameco and the iPower, the Jameco has one of the lowest leakage tested and the iPower had the highest, about a million to one difference, that might have had something to do with your observations! John S. I could of course try to make them even better with better regulators, but my main rationale was to make them quieter on my AC lines, not make them better on the output side. Both seem to be accomplished based on what I'm hearing. I'll do a blog post describing the mods at some point when I get some time. Later! Greg in Mississippi P.S. Some may question the wisdom of modifying a roughly $10 adapter. BUT for about 2x the price in good parts & a little time (I did 4 of them on Sunday in about 5 hours), they make nice and easy to site units. The modification is not DIY beginner-level because of the effort and care needed to open the cases, remove the existing parts and position the now slightly larger parts so that you can re-close the cases, but someone who's done a little DIY and solder reasonably well could do them. AND the sonic improvement over the stock units was significant, even though they were only powering FMCs. P.P.S. I do plan some switch & FMC mods starting with replacing the switching DC-DC converters with good linear regulators. I'll try replacing the regs in the Jameco adapters when I do that work. I'd been considering this for awhile, but Fob69's posts here have revived my interest: 13 hours ago, fob69 said: Hi I just did it. First, i took them away 6 monthes ago and replace them with an etalon isolator with a huge benefit, and i did add too a long ethernet distance between my switch and my etalon isolator (15m cat6 grounded only on switch side). 15 days ago i did change my internet box on my ethernet network and plug new equipments. I had to admit that even with the etalon isolator and the long ethernet wire, my new network config had downgrade image and timbres. It confirg to my the huge SQ impact of the noise coming from the network equipments (even with galvanic isolations). I was about to buy linear supply for my new internet box, fiber modem, NAS ... But before i did try to replug the MC110Cs btween my etalon filter and switch, on switch side. I did see a good improvement especialy on the image but there was still some network impact on trebles as is was 6 monthes ago when i removed the MFCs (due to the 348kz signal of the PWM component). Then, considering again the PCB, i found a FAST, CHEAP way to bypass the PWM schitching supply in the MFC : i did unsold the "+" leg of the female jack plug of the pcb, an link it with to basic 2X 1A diods in serial to the PV3V3 hole close to the green electro condo. Supplied by a standard 5V linear supply, the to diods downgrade the voltage to around 3,5 V just before the regulator, and the switching PWM component is short cutted. This is the best 1$ upgrade ever. All my network noise is now invisible from my player with the fiber bridge, and i bypass the switching power that is the main problem with the MFC. The image ans definition is incredible. I will post a photo this evening. B. austinpop 1 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted November 13, 2017 Share Posted November 13, 2017 1 hour ago, lmitche said: Thanks John, I am curious, what is the theory of operation for the end to end wire? Does it make the cable a poor antenna? Adding the tinned copper sleeving and wire to the Lush USB cable this afternoon had a noticably positive impact. I'll do the same to my ethernet and usb 3.1 cables later in the week. Larry, See here: Greg in Mississippi Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
Popular Post gstew Posted January 13, 2018 Popular Post Share Posted January 13, 2018 16 hours ago, Cornan said: Since this is a SMPS thread I hope it is OK to post this question. Have someone by any chance tried the Silent Switcher? https://diyaudiostore.com/collections/kits/products/linear-audio-silent-switcher I am kind of interested to know if this could be a good option for the GI output powered by a 5v powerbank (or Gophert csp-3205II floating SMPS 6.7v via DC Y-split with LT3045s on the other leg)? This is a floating SMPS according to the maker and looks pretty interesting to these eyes. Cornan, I bought a couple of these from the initial KickStarter. I've used one to provide the +-15v to a Dial Audio Raspberry Pi DAC (one of my 3 favorites of that genre): https://dial-audio.jimdo.com/ My thoughts... meh. It works and is not bad, at least powered by an inexpensive 5v cell-phone / R-Pi SMPS adapter. It was noticeably better powered by a good linear power supply and even better powered by an LPS-1 (energized by a linear supply). But a pair of small shunt regulators (like simplified Salas) powered by an above-average linear power supply (oversized R-Core transformer, selected soft-recovery diodes, +-47000uF Jensen 4-Pole caps) provided more music... more dynamic, stronger & better defined bass, better definition top to bottom, sweeter mids-highs. Mine will get relegated to bench supplies and test setups. I also tried the 5V output powering an Allo.com Kali I2S reclocker under the Dial Audio DAC...also a meh, compared to a Acko previous-generation AKR75 (ADM715x-based) regulator powered either by the linear PS or LPS-1. Nothing obviously wrong with it, but IMHO, I didn't get any magic. YMMV. Greg in Mississippi P.S. BTW, an SMPS is a switch-mode power supply... something intended to be fed AC and produce a DC output. The Silent Switcher uses switching DC-DC converters, devices intended to be fed DC and produce a different DC voltage. Entirely different. They can be used with each other and some SMPS's include switching DC-DC converters to produce final output voltages, but they are not interchangeable things. MikeyFresh and Cornan 2 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted January 13, 2018 Share Posted January 13, 2018 8 hours ago, Cornan said: Great and very useful reply for me. Thanks a million Greg! ? I think I will pass using it powering my GI output and go for LT3045 as planned earlier, but powered by Gophert csp-3205II. I might order it anyway when the updated boards (under construction) have been released. It is still a quite useful board for two voltage sources that might come in handy some day. Thanks for the Dual Audio link as well. Even if I am not looking for a DAC at the moment this one surely goes directly into my collection of interesting purchases (that works with Silent Switcher supplying a Pi3 as well). BTW. I have got my two 3A Stammheim's single ps LT3045 boards the other day. I will have some time all alone later today and have the chance to try one of the single ps out between Gophert csp-3205II (floating SMPS) and Brooklyn DAC. Will be fun to see how that turns out! ? Cornan, Very welcome. Very curious on how Michael's regs work for you. Do remember they take some time to break-in... my experience is that circuits with ceramic caps take about 3 weeks of constant running to finally settle. I have one of Michael's dual-boards powering my SDTrans384 SD Card player. Its been running about 2 1/2 weeks now and starting to sound good... at 1/2 week I put the previous supply (a dual 4||LT3042 board from DIYAudio poster OPC) back in and slightly preferred that. Now I suspect it'd be the other way around (though not by a huge difference... both the LT3042 and LT3045 are great regulators. BUT the in-line ballast resistance of the LT3045 setup is 1/2 of that on the LT3042 setup, so a little less reduction in the 'goodness' of the regulators when paralleled). I power both of these with 2 LPS-1s with each powering 1 side of the dual board. I only have Michael's dual boards... he didn't have the 6||LT3045 boards when I ordered them. BUT for low-level digital electronics, the 2 LPS-1s into a dual paralleled LT304x regulator board is the best I've tried. One would need current-matched LPS-1s to do that into the 6||LT3045. Pix of my build attached. I used different caps than Michael's BOM... I wanted to use the same ones as on OPC's BOM for his boards as I knew they worked well in the setups where I've used them. I also added a couple at the output to make setups more like each other. AND I used fixed resistors instead of the adjustable pots... again to make the two different implementations more like each other so I can get a better feel for the basic differences between 2x4||LT3042 and the 2x3||LT305 boards. Later! Greg in Mississippi P.S. I added a couple of pics of OPC's 2x4||LT3042 boards for comparison. Then the other boards in the build picture are alternative reg boards from OPC for powering ES9028/9038/9038 DACs' AVCC rails. Working on implementations of a couple of those. Gotta love building SMD boards! Cornan 1 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
gstew Posted January 14, 2018 Share Posted January 14, 2018 On 1/12/2018 at 6:30 PM, gstew said: Cornan, I bought a couple of these from the initial KickStarter. I've used one to provide the +-15v to a Dial Audio Raspberry Pi DAC (one of my 3 favorites of that genre): https://dial-audio.jimdo.com/ <SNIP> Cornan, I'd forgotten I'd posted this about the Silent Switchers awhile back... it doesn't add much, but just shows I have sweetened on them over time. Later! Greg in Mississippi On 9/3/2017 at 12:25 PM, gstew said: hfxrzw, Curious what you're powering it with. A bit off-topic... I've got a couple here and have used one to power the +-15V side of a Dial Audio R-Pi DAC. It was ok with the recommended phone charger, a bit better with an iFi iPower 5V, and a bit more better powered from an LPS-1 set to 7V. BUT using a +-pair of some small shunt regulators (sorta like simplified Salas) powered by a hefty AC-connected linear supply was definitely a step up. I haven't tried the 3.3V-5V-6.3V output as yet, I'll have to do that and see how it compares to my other options for things like powering an RPi or an RPi DAC. Jesus, IF you'd like to try one, I can lend you one for awhile. Greg in Mississippi P.S. Ted, that lend offer goes for you too, if you are interested. BUT I'm not opening it up to the world at large. sorry! Cornan 1 Everything Matters! 2 systems... Well-Tempered Refs->ET-2.5->DIY or Lounge LCR MkII phono stages Standalone digital Sony HAP Z1-ES or SDTrans384/Soekris DAM DAC Networked digital Zotac PI320-W2 LMS Server -> EtherRegen -> USBBridge Sig -> Katana / Ian GB / Soerkis / Buffalo-IIIPro DACs Passive S&B TX102 TVC or ladder attenuators -> BHK-250 -> Eminent Tech LFT-VIII / IV / VI ALL gear modified / DIY'd; cables MIT; all supplies DIY’d or LPS-1.2s w/HUGE Ultracaps; Audio gear on DIY AC filters + PS Aud P15s; misc gear on separate AC w/filters Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now