Meanwell PSU wiring question

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: luketeaford, Joe., lisa, Kent

User avatar
kurodama
Veteran Wiggler
Posts: 744
Joined: Wed Feb 13, 2013 2:50 am
Location: Italy

Post by kurodama » Fri Nov 27, 2015 2:16 am

Let me revive this thread to say thank you! This is the first time I find a thread about mains wiring where there's some actual, precise and understandable info on what to do and what not to do.

But let me ask one more question, since a lot of people seem to be disagreeing on this: what's the main benefit of wiring all ground connections to protective earth? Is it for personal safety or just to remove hum and noise?

I think this is relevant also to judge if a commercial case one is about to buy is really worth its money when it comes to the PSU.

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3279
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Post by Graham Hinton » Wed Dec 09, 2015 5:37 pm

kurodama wrote: But let me ask one more question, since a lot of people seem to be disagreeing on this: what's the main benefit of wiring all ground connections to protective earth? Is it for personal safety or just to remove hum and noise?
Both, although the latter component is very small in comparison. Consider how small the signal is from a ribbon microphone where you are deliberately putting a single turn coil in a strong magnetic field. Hum is not something that exists and then has to be got rid of, it is something that should not be occurring and the usual cause is power and microphone amplifiers feeding back through 0V wiring.

The third reason when using unbalanced equipment is that the mains Earth wiring, even if there is no Earthed mains outlet, forms part of the signal return paths. You tend to get a hum when this is lacking rather than the opposite where people believe that creating an "earth loop" is bad.

If you connect two pieces of equipment each with their own PSUs using an unbalanced patch lead the signal return current will split between the cable screen and the Earth wires. These are similar resistances, several hundred milliohms each, so effectively there are two resistors in parallel. The noise you get between different equipment 0Vs is not something that appears magically because the cable exists, rather you create it by patching signals and the currents develop a small voltage across the cable resistances. V=IR.

The way to reduce the voltage differences is to reduce the cable resistance. One technique is to run a Parallel Earth Conductor (PEC), an unfortunate terminology because it confuses its purpose. Sometimes you find a green/yellow striped wire in STP multicore cables, this is a lower resistance path that is wired in parallel with the cable screen resistances.

To get a 20dB reduction 90% of the currents have to flow in the parallel conductor which means that its resistance has to be a tenth of the normal paths and to get a 40dB reduction 99% has to flow in a hundredth of the resistance, a few milliohms. Here is a method I tried recently using a 30x12 aluminium bar to join all the TEs of various unbalanced synthesizers spread over 3m. The 2.5m long bar measured 0.56 milliohms end to end and the connecting cables were up to 20 milliohms each.

Image

Before you could not patch between one case and another without hearing wobbly CVs, after it did what you imagined it should do in the first place. N.B. all the PE wiring remains intact and no patch cables have disconnected screens.

User avatar
kurodama
Veteran Wiggler
Posts: 744
Joined: Wed Feb 13, 2013 2:50 am
Location: Italy

Post by kurodama » Fri Jan 08, 2016 9:07 am

I kind of missed this update in the thread earlier (sometimes it doesn't show up)

Thanks again for taking the time to reply in depth to my questions.

Still there is one thing I'm not sure I fully understand (but it might be just a linguistic thing, due to me not being perfectly fluent in English), what do you mean by this sentence:
Both, although the latter component is very small in comparison. Consider how small the signal is from a ribbon microphone where you are deliberately putting a single turn coil in a strong magnetic field. Hum is not something that exists and then has to be got rid of, it is something that should not be occurring and the usual cause is power and microphone amplifiers feeding back through 0V wiring.
Do you mean that the above "grounding technique" is more important to prevent hum than as a way to protect your health if something goes wrong inside the case? Or what is very small in comparison, and in comparision to what?

Thanks again! Despite this thread basically having convinced me that I'd better leave it to the experts to wire up cases, it's knowledge that is also very previous when having to choose which case/psu to buy and makes one less prone to buy some overpriced but ultimately worthless junk. it also shows how much knowledge, time and work has to go into the design of a PSU, to get a good product.

solaris
Super Deluxe Wiggler
Posts: 2125
Joined: Mon Sep 13, 2010 6:49 pm
Location: ̋͋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̋̏̏ ̏ ˳̊̊̊̊̊ ˳̥̥̥̥̥̊̊̊̊̊̊̊˳̥˯Ξ̿̿̿̿̿̿̿̿̿̿▂ ̋̋̋̋̋̋̋̋̋̋

Post by solaris » Fri Jan 08, 2016 9:12 am

kurodama wrote:I kind of missed this update in the thread earlier (sometimes it doesn't show up)

Thanks again for taking the time to reply in depth to my questions.

Still there is one thing I'm not sure I fully understand (but it might be just a linguistic thing, due to me not being perfectly fluent in English), what do you mean by this sentence:
Both, although the latter component is very small in comparison. Consider how small the signal is from a ribbon microphone where you are deliberately putting a single turn coil in a strong magnetic field. Hum is not something that exists and then has to be got rid of, it is something that should not be occurring and the usual cause is power and microphone amplifiers feeding back through 0V wiring.
Do you mean that the above "grounding technique" is more important to prevent hum than as a way to protect your health if something goes wrong inside the case? Or what is very small in comparison, and in comparision to what?.
you mentioned "personal safety" followed by "hum&noise", so
"the latter" is "hum&noise"
▂̵̶̵̥̤̳̤̇̈̄̅̈□ □ □ □ □ □ □ □ □ t r a n s f e k t i o n ▔▔▔▔╲▂▂▁▂▂̵̶̵̥̤̳̤̇̈̄̅̈▂▁▁
▂̵̶̵̥̤̳̤̇̈̄̅̈□ □ ▂̵̶̵̥̤̳̤̇̈̄̅̈▂̵̶̵̥̤̳̤̇̈̄̅̈▂̵̶̵̥̤̳̤̇̈̄̅̈□ ▂̵̶̵̥̤̳̤̇̈̄̅̈▔╲▂̥̥̊̊̊̊▂ ‬::.:::.h++p://z.x-xx---x.info □ □ □ □ □ □ □ □ □‬

User avatar
kurodama
Veteran Wiggler
Posts: 744
Joined: Wed Feb 13, 2013 2:50 am
Location: Italy

Post by kurodama » Fri Jan 08, 2016 1:32 pm

you mentioned "personal safety" followed by "hum&noise", so
"the latter" is "hum&noise"
Ok thanks! Actually I think that it's the second part that confuses me: "Consider how small the signal is from a ribbon microphone …" is it relating to the previous sentence about the hum being "small"?
But I guess the two sentences are not really related and just went on explaning the hum&noise problems.

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3279
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Post by Graham Hinton » Fri Jan 08, 2016 2:33 pm

kurodama wrote: Actually I think that it's the second part that confuses me: "Consider how small the signal is from a ribbon microphone …" is it relating to the previous sentence about the hum being "small"?
But I guess the two sentences are not really related and just went on explaning the hum&noise problems.
Yes. The safety aspects have to be in place, but are rarely invoked. At least not with properly built equipment, some of the DIY systems I have seen have been potentially lethal. Same with a lot of gigged systems that get heavy use.
Safety grounding has to be there to protect you IF a fault develops.

Whenever there is a hum and noise problem people wave their arms in the air and cry out "maybe I've got a ground loop" usually coupled with a statement that they know nothing about electricity. Loops exist everywhere, you make one every time you insert a patch cable and they are only usually a problem if followed by a lot of gain. A ribbon microphone amplifier will give you an idea of the amount gain needed on a single turn loop in a strong magnetic field when you are deliberately intending to pick up a signal.

Most hum problems are caused by unbalanced connections and too high resistance in the 0V paths. "Too high" being less than one ohm, you have to start thinking in terms of milliohms and less and cannot assume that any length of wire is zero resistance.

User avatar
kurodama
Veteran Wiggler
Posts: 744
Joined: Wed Feb 13, 2013 2:50 am
Location: Italy

Post by kurodama » Fri Jan 08, 2016 5:16 pm

Thanks a lot for the additional explanation! This cleared all my doubts!

User avatar
Randy
Super Deluxe Wiggler
Posts: 1860
Joined: Sat Sep 22, 2012 9:50 pm
Location: Mississauga, Ontario, Canada

Post by Randy » Tue May 10, 2016 9:45 am

Hi, sorry to dredge this up again, just one thing I'm not sure about. Seems easy enough to feed the grounds from the mains ground, I just have a single ground source from the mains, and then star that out to the ground on the power supply (Meanwell again) and the 0V on the busboards.

First, is that correct, and second, what do I do with the COM output on the power supply? The reason I ask this is that if I leave the COM output unconnected to the busboards, I don't get any power to the busboards.

Thanks

Randy

User avatar
roglok
Super Deluxe Wiggler
Posts: 2441
Joined: Fri Dec 30, 2011 3:49 pm
Location: germany

Post by roglok » Tue May 10, 2016 10:04 am

Randy wrote:Hi, sorry to dredge this up again, just one thing I'm not sure about. Seems easy enough to feed the grounds from the mains ground, I just have a single ground source from the mains, and then star that out to the ground on the power supply (Meanwell again) and the 0V on the busboards.

First, is that correct, and second, what do I do with the COM output on the power supply? The reason I ask this is that if I leave the COM output unconnected to the busboards, I don't get any power to the busboards.

Thanks

Randy
Have you actually read the thread? The first page is all about what the COM terminal is and how it should be connected ideally.

User avatar
Randy
Super Deluxe Wiggler
Posts: 1860
Joined: Sat Sep 22, 2012 9:50 pm
Location: Mississauga, Ontario, Canada

Post by Randy » Tue May 10, 2016 10:08 am

Whoa, sorry, missed one sentence it seems:

"Do not put a wire between Gnd and Com on the terminal strip, instead connect the PSU to your distribution board and the mains Earth to your distribution board and join the Com there"

Will certainly try to be more careful next time.

Randy

User avatar
Randy
Super Deluxe Wiggler
Posts: 1860
Joined: Sat Sep 22, 2012 9:50 pm
Location: Mississauga, Ontario, Canada

Post by Randy » Wed May 11, 2016 1:42 pm

And, just to shake the tree a bit more, I did the recommended mod to my smaller case and there was no effect, nothing, nada. It made no difference to anything.

So, enough talk about theory, does anyone have any actual, real world, physical proof that any of this stuff has any effect on the final sound?

Randy

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Wed May 11, 2016 7:15 pm

SEE MY LAST POST
Last edited by peteone on Thu May 12, 2016 6:15 pm, edited 5 times in total.

User avatar
Randy
Super Deluxe Wiggler
Posts: 1860
Joined: Sat Sep 22, 2012 9:50 pm
Location: Mississauga, Ontario, Canada

Post by Randy » Wed May 11, 2016 9:33 pm

No, I don't think that is what was suggested. The ground should go from the mains socket to the ground on the power supply and to the ground on the busboard. So, that's two separate wires going from a single point on or near the mains socket. And then, you also take a wire from the COM on the power supply to the same point as the ground wires (0V) on the busboards.

Read the one sentence I missed in my post a bit earlier.

If I'm wrong, hopefully someone will step in, but that is how I interpreted the instructions. This whole thing reminds me of some of the I.T. Enterprise Architects I've worked with.

Randy

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Thu May 12, 2016 12:22 am

SEE MY LAST POST
Last edited by peteone on Thu May 12, 2016 7:03 pm, edited 1 time in total.

User avatar
roglok
Super Deluxe Wiggler
Posts: 2441
Joined: Fri Dec 30, 2011 3:49 pm
Location: germany

Post by roglok » Thu May 12, 2016 2:12 am

Randy wrote:And, just to shake the tree a bit more, I did the recommended mod to my smaller case and there was no effect, nothing, nada. It made no difference to anything.

So, enough talk about theory, does anyone have any actual, real world, physical proof that any of this stuff has any effect on the final sound?

Randy
what "mod" are you referring to and what do you mean with "any of this stuff"?

the reasons for the recommended wiring scheme are given in the respective posts in this thread, above all is security. nobody stated that you should expect the "final sound" to change by wiring your PSU correctly. However, if you experience instabilities or crosstalk in your signals, especially under heavy load, a proper grounding scheme with low resistance wiring will definitely help. Yes, there is actual, real world, physical proof for this.

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3279
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Post by Graham Hinton » Thu May 12, 2016 5:31 am

peteone wrote:Would this be a proper way to wire this system?
The connection from the inlet Earth to the PSU Com is wrong, that should go to the distribution 0V (often incorrectly called "Gnd"). The devil is in the detail, you have to think how current flows and not just in terms of being connected together. The order is significant. The wire gauge used is important and it is best to use accepted standard colours to prevent confusion.

Note that your mains inlet in your picture has a link between the Live input and the fuseholder and that it is single fused. The inlet should be covered with a rubber insulating boot so that nothing can touch this point. If something touched it, e.g. a patch lead dangling in the case, this point is before the fuse so unless you are in the UK and have fitted a 5A fuse in the IEC mains lead the next fuse is probably your main trips and that is will be something like 30A. Not that a 2A fast fuse in the inlet would protect you from electrocution.
Also countries that use Schuko plugs can reverse Live and Neutral so there might not be a fuse in the Live side at all.
It is best not to solder to mains inlets, but used shrouded crimp terminals. The crimp tool for red/blue/yellow terminals is one of the few that are very cheap so there is no excuse not to get one.

A PSU as pictured is really intended for use in a digital system that has some analogue circuitry, like a computer or some non audio device. The clue is just the one COM terminal. This means that the 5V return currents flow in the same wire as the analogue returns and will interact. A PSU intended for analogue use will have a separate +ve and -ve terminal for each rail and allow you more flexibility in wiring them. It would also have a lower ripple and noise figure.

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Thu May 12, 2016 1:27 pm

SEE MY LAST POST
Last edited by peteone on Thu May 12, 2016 6:14 pm, edited 1 time in total.

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3279
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Post by Graham Hinton » Thu May 12, 2016 3:18 pm

peteone wrote:ok here is the fixed version

HOPEFULLY
You have dropped two wires: The Neutral to the PSU and the one we are talking about, a connection from the mains Earth directly to the distribution 0V. Without that the PSU is "floating" without a reference.
The Neutral was correct before, the fused Live and Neutral should go through the DP switch so that the mains is isolated when off.

As you are in the US the mains wires should be Live = black, Neutral = white, Earth = green. In other countries Live = brown, Neutral = blue, Earth = green/yellow striped. Anybody with electrical training looking at that would immediately recognise them.

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Thu May 12, 2016 5:06 pm

EDITED - SEE NEXT POST
Last edited by peteone on Thu May 12, 2016 6:14 pm, edited 1 time in total.

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Thu May 12, 2016 5:18 pm

edit
Last edited by peteone on Tue Jun 07, 2016 7:16 pm, edited 1 time in total.

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3279
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Post by Graham Hinton » Fri May 13, 2016 6:13 am

peteone wrote:
Here is Final, Tested with multi-meter outputs +12 -12 +5
You still haven't referenced 0V to mains Earth. This is all floating, including your DMM.

Connect one of the so-called "Gnd" terminals on the distribution directly to mains Earth, connect the other two to the PSU COM with two wires with ring terminals stacked on the screw terminal.

User avatar
Joe.
Wired for sound
Posts: 4994
Joined: Sun Jan 30, 2011 10:58 pm

Post by Joe. » Fri May 13, 2016 6:48 am

It's not 'tested' if you haven't used a dummy load.

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Fri May 13, 2016 10:24 am

Graham Hinton wrote:
peteone wrote:
Here is Final, Tested with multi-meter outputs +12 -12 +5
You still haven't referenced 0V to mains Earth. This is all floating, including your DMM.

Connect one of the so-called "Gnd" terminals on the distribution directly to mains Earth, connect the other two to the PSU COM with two wires with ring terminals stacked on the screw terminal.
let me put it in a simpler English here for idiots like me

you want me to make a connection between Tree and Com on the PSU itself?
Last edited by peteone on Fri May 13, 2016 12:27 pm, edited 1 time in total.

User avatar
peteone
Wiggling with Experience
Posts: 467
Joined: Fri Feb 07, 2014 12:06 am
Location: Los Angeles, CA

Post by peteone » Fri May 13, 2016 12:03 pm

edit
Last edited by peteone on Tue Jun 07, 2016 7:16 pm, edited 1 time in total.

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1508
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Fri May 13, 2016 12:29 pm

As graham wrote further up the thread:
Almost. Don't use the PSU terminal as a junction. The mains Earth should go to a big post where it enters your system, if you have a mains inlet on a chassis it should be right next to the inlet. From there wires should go out separately to each place that needs to be earthed.
That was in response to a different diagram from someone else but applies now. To you, he wrote
Connect one of the so-called "Gnd" terminals on the distribution directly to mains Earth
where it is only the word "directly" that means your diagram does not fit Graham's spec.

Post Reply

Return to “Music Tech DIY”