|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
85.147.82.48
In Reply to: RE: cMP - the open source high-end Memory Player posted by cics on December 30, 2007 at 05:42:01
Hi all,I have a question on: earthing arrangements and position of ‘Live’ and ‘Neutral’. May be knowlegdeable inmates know someting about this. Feedback and thoughts on this are very much appreciated.
I used to have a NAIM system (CDX, NAP250, NAC82, 3 HICAPS) and with this system one could hear differences in sound quality when the power plugs had different positions of ‘Live’ and ‘neutral’ on the pins. I found this to be very annoying because one would think that NAIM uses circuit designs that are professional enough not to make such sound quality differences (for me this was another good reason to change too professional recording gear)
But now I discovered that this effect is also there between my 3 PSU’s (the Antec PSU + linear PSU on P4 + linear PSU on picoPSU-150-XT) . :-(
It does matter how the ‘Neutral’ and ‘Live’ 230 Volt positions are for each PSU.Are there any other inmates with have experienced the same?
And what is theoretically the best way to determine what the optimal 230 Volts positions are between the 3 PSU for ‘Live’ and ‘neutral’?
* Now over too the other side of the PSU (12 Volt output)
There are also very small voltage differences between the (black Neutral 12 v lines) and the ground (earth).
- On the linear P24 PSU the differences between ‘Neutral’ and GRND is 3 mVolt.
- on the linear P4 PSU the differences between ‘Neutral’ and GRND is –20 mVolt.Those two black ‘Neutral’ 12 volt line both go the mobo and 'meet' somewhere on the MoBo. I think.
- What happens with this 23 mVolt difference than? Is this earthed on the MoBo ?
- Should I ground both black ‘Neutral’ 12 volt lines directly at both PSU’s?
- should I connect both black ‘Neutral’ 12 volt lines together directly at both PSU’s?
- should I do both: connect them together at the PSU's and aerth them there a the PSUAny suggestions?
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Edits: 12/03/09Follow Ups:
Mark,
First, I need to be honest and say that it's been many years since I've listened to gear with the AC sexed the wrong way... I check this as a matter of course when I'm building gear... and honestly, I'm not sure if there's a similar effect with SMPS's. The measurements I've done with them so far have been inconclusive, possibly a difference at the limits of my meter, but not definitive. I THINK they should have similar difference, but since the transformer is after the initial rectification, I'm not sure... and if you wanted to change it, you'd have to swap the inputs on the transformer, which could get challenging.
So sorry, can't answer the question on whether it makes an audible difference in our digital world.
But I'd guess it does... noise on the ground (and this is what we're minimizing with the correct hot/ground orientation) will impact connected analog gear... and I suspect will show up on the digital side as increased jitter.
And as for the grounding scheme, I wasn't convinced that the 'ground at the preamp' recommendation for all-analog setups was right and I'm not sure what is right for our analog/digital systems.
Gack!
I guess I need to go do some research now. So thanks for the link to the reference book... that'll be a good place to start.
Later!
Greg in Mississippi
P.S. Here's a good post by Charles Hansen of Ayre Acoustics where he relates tracking down an issue related to grounding... and not being able to measure what caused the difference. I personally believe grounding is still very much a black art, not because we don't understand the mechanism, but because the complexity of interactions are so high we can't effectively analyze it. So my take on it is to start by following the best practices and then experiment a bit to see if it can be made better... like connecting the case to wall-ground.
Hi Greg,
Again thankx for your response. Through our conversation I have now have on all PSU’s the Live and Neutral the correct way. And as a spin off, I now also have a better understanding of the difference between the 0V (neutral) and the ground. This was always a bit confusing to me, since many make no difference between 0Volt (neutral) and ground. Which resulted in now having also all the modules correctly grounded in my setup (by the book).
Mark
P.s.
The book ( http://www.ivorcatt.org/digital-hardware-design.htm ) was referred too on this forum by (if remember correctly) inmate Ryelands. Which I think was a very good thing to do, since the different set of rules between processing sine waves (analog audio) and shock waves (digital audio) are forgotten or mis-understood all too easy.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
First, in my experience, what you saw with the Naim gear is common to all transformer-based-linear-supply-powered components. Here's some background (and please forgive me for being somewhat basic here, not sure what info you already have).
For power transformers in linear supplies (not sure if this is the same for switching supplies), there is a non-symmetry to the windings that will induce a stray AC voltage in the transformer core and nearby metal parts. You can see this by hooking up a raw (unconnected to anything else) transformer primary connections to the AC line Hot and Neutral (do this safely with a fuse in line and the secondaries secured and insulated). Plug in the AC line and measure the AC voltage between the transformer core and the AC Ground and you will almost surely get a small (a few millivolts) to medium (a few volts) AC voltage reading. Swap the transformer primary connections around so that the line that went to Hot now goes to Neutral and vice-versa. Plug in the AC line again, measure between core and AC Ground again, and you will again get a small AC voltage reading, likely higher or lower than the first reading. When you wire your transformer primaries up, you want to use the orientation that gives you the lower AC voltage reading.
I first heard about this back in the 1980's. The tweak was that you took your piece of audio gear, disconnected all the inputs and outputs, then used a 'cheater' plug that only connected the Hot and Neutral, and measured between the component's ground and the AC ground. The 'right' way was again the one that gave you the lower AC reading. And it worked... it gave you a lower noise floor and a more detailed and natural sound if you got all the components AC plugs oriented the right way. Someone even sold a little tester to make this easy to do. (BTW, I've heard this called 'sexing' your transformers).
But what you were really detecting was the transformer orientation. And if you had a component that had more than one transformer (like my linear-powered computer supplies), then you had a problem if the manufacturer didn't know to do this and had them hooked up so that one transformer was oriented correctly when the other was not.
And as far as your Naim gear went, this is likely why it sounded different depending on the Neutral and Hot orientation... and this is still a factor with professional gear (assume that means balanced), but it may be less critical there... or maybe the professional gear you use just has all the transformers oriented the right way and the Naim gear didn't.
It's a lot easier when you are building components and power supplies from scratch. Then you just check each transformer before you wire it's AC connections permanently and make sure to do it correctly, so that when your AC plug is in the normal way (Hot-Neutral-Ground), each transformer has the lowest AC reading.
Now this gives you the right Hot-Neutral orientation. Grounds are something else. First, I have much less knowledge and experience here on what is right and how to implement it. What seems to be best with grounds is what's called a 'star' ground, where all the ground connections come back to a single point. This can be challenging in an audio system because you'll often have multiple components with ground connections on their AC cords. When you connect them to each other with your interconnects and speaker wires and plug them into the AC outlet, then there are multiple paths to ground which will often cause ground loops, where current is flowing through the grounds in ways that increase the noise level of the system. So one way to deal with this is to use those darned 'cheater' plugs on all components in your system except one, typically the amplifier, but sometimes the preamp. And if you have a bi-amped or tri-amped system, it becomes complicated and challenging. And of course, who knows how each component is grounded internally. It is complicated... and says a lot for the benefits of a single-manufacturer setup that has spent a lot of time optimizing grounding and AC line filtering, like Ayre's.
So now back to our cMP setups.
First, you want to do the transformer orientation thing for each of your linear supplies. Then if you have multiple power cords (I have three in my cMP^2 setup... one for the P24 & P4 power supplies, one for the separate supplies for the Juli@ & DAC, and one for the linear 'dirty' supplies), you want to have only one have the ground connection active... the other ones should have only Hot and Neutral connected. And the ground should go to the case... and somewhere there should be one connection from the circuit ground to the case. Of course, most computer cases and motherboards have multiple connections from the circuit ground to the case... it'd be worthwhile to insulate all but one and see if that makes a measurable or sonic difference.
In my cMP^2 setup, the P24/P4 linear supply power cord has the ground connected. In addition, since my transformers and heatsinks are on wooden platforms, I have a pigtail lead connected to a mounting screw on each that I take back to a single circuit/case ground point on the motherboard.
But of course, I don't have my setup 'right' because in addition to the cMP^2 being grounded, my amps are too. Need to fix that... in a setup like this, I'd likely have the computer be the main ground path.
My $50 (lots more than 2 cents here!)
Greg in Mississippi
Hi Gstew,
After I checked ‘the sexual orientation’(muwahaha) of the linear PSU’s in my setup, as you discrebed, I was still puzzled about the grounding. Would the common rule that says: in a system with more than one module there should only be one (1) grounding point (preferably at the source module), also apply to digital systems?
After reading chapter 11 and 12 of ‘Digital Hardware Design’ I think the same grounding rules apply. I write ‘think ‘, because I have no professional education in electronics. Only some basic stuff. So I don’t know if I understand everything in chapter 11 and 12 real good.
http://www.ivorcatt.org/digihwdesignp73.htm
In a digital system with more than one module, the 0 Volt also should be grounded preferably at one point. And the grounds of the modules should be inter-connected.
Mostly this is already done by the main leads. But since in a cMP setup it is advised that the ATX PSU should be powered from a different AC circuit (thus also different grnd), the cMP case and the linear PSU grounds should be connected together. At least: that’s what I think needs to be done after reading chapter 11 and 12.
If I understand your post corretly, this is indeed what you do with your linear PSU’s and the pig-tail wire.
And may be that’s why TheoB noticed an improvement in sound quality (SQ) when he grounded his cMP case.
But, when I took an extra wire and grounded the cMP-case (and thus the ATX-psu) with this wire to the grounds of the 2 linear PSU’s, there wasn’t a SQ improvement in my setup. But I will leave the wire that’s connecting the cMP-case to the grounds of the linear PSU in place.
since it’s the suggested method to do so (if I understand correctly)
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Hi Mark just a slight clarification...when I grounded my mobo I found a slight increase in transparency.
Hi Gstew,
Thank you for your response and such generous feedback. Bit and peaces of your post I knew already but you nicely ‘painted the big picture’. Your response waa helpful for me but I think your response is also much appreciated by other inmates who also power the P4, P24 and sphericals with separate (linear) power supplies.
I posted the question because:
1. there’s always a big chance of being tricked by ‘the placebo effect’:
2. and I also don’t know if these phenomena from the analog audio world also apply to the digital audio world. I know all ‘digital’ audio is analog, but I also know that a different set of rules applies too: processing analog sinus-waves and processing (square) block pulses (‘digital’ signals).
But did you hear (and other inmates who read this) difference in sound quality when powering the linear PSU’s with ‘Live’ and ‘Neutral’ in different positions?
If not, than most likely the placebo effect is fooling me.
fully AOB optimized cMP2 PC -> Lynx AES16 -> XLR AES/EBU -> Lavry Black DA10 -> XLR Mogami Gold -> Klein & Hummel O300
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: