Case grounding questions

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

Moderators: Kent, luketeaford, Joe.

Post Reply
Johnbomb
Common Wiggler
Posts: 152
Joined: Wed Apr 26, 2017 11:57 am
Location: Nashville

Case grounding questions

Post by Johnbomb » Sat May 01, 2021 9:44 pm

Hello everyone,

The aluminum Vector rails in my rack contact aluminum side mounting brackets. I've read a few threads on grounding, and based on Mr. Graham Hinton's input, I think I have at least a partial understanding of the proper way to set this up, as below- Mr. Hinton, if you stop by, I'd very much appreciate any advice.

Let's assume a short lead connects earth ground to 0V on the buss board (on the +12 -12 side), a 2nd short lead connects earth ground to the 0V of the power supply (also on the +12 -12 side), and a 3rd short lead connects earth ground to chassis (in this case, Vector rails + aluminum mounting brackets). Also, let's assume that the faceplates of several modules have a direct electrical connection to the Vector rails via mounting screws (and thus chassis). In this setting, I would expect a module to have 2 potential paths to ground: 1) through the ground wire(s) in the power cable and 2) from faceplate to chassis.

Is this a setup for noise? If so, what is the solution?

Thanks,
John

User avatar
MikeDB
Veteran Wiggler
Posts: 653
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Re: Case grounding questions

Post by MikeDB » Sun May 02, 2021 5:42 am

There's many different thoughts on grounding in large audio systems. One is everything to a single point, i.e. star wiring, or everything to everything where you have possible multiple ground paths. There is also the concept of having two grounds, one for electrical signals joined everything to everything and one for the faceplates where you have to use insulated connectors so as to avoid connecting the two. The two grounds are then joined at a single common point. I prefer the last one but the 3.5mm connectors used in Eurorack synths makes this difficult.

All these techniques are used in mixers which is my main field, but the larger the mixer the more impossible pure star wiring is used simply because the number of wires becomes impossible to handle.

Note that in your final example you'll also possibly have another path to ground for some modules, i.e. from the module through an external device (amplifier, some MIDI controllers, etc) to ground. Graham's articles on this are good so best to follow his advice first and then fix any problems if they arise.
Always interested in audio startups needing help and investment, just PM me.

User avatar
Altitude909
Super Deluxe Wiggler
Posts: 3721
Joined: Wed Aug 24, 2011 5:13 pm
Location: Meesheegan

Re: Case grounding questions

Post by Altitude909 » Sun May 02, 2021 9:22 am

Here are a couple articles I keep bookmarked that go over the issues in a fairly clear manner.

https://www.analogictips.com/faq-ground ... round-faq/
https://circuitcellar.com/resources/ee- ... und-loops/

Its a complex question and not something I would just blindly follow someones advice one. For instance: In the USA, when you connect your 0V to the safety ground you are also connecting it to AC Neutral since those two are connected at your fuse box.

Obviously, the safety aspect is paramount but it's not necessarily good for noise.

Johnbomb
Common Wiggler
Posts: 152
Joined: Wed Apr 26, 2017 11:57 am
Location: Nashville

Re: Case grounding questions

Post by Johnbomb » Sun May 02, 2021 3:31 pm

Thanks, guys- It's a tough one. Along the lines of safety: in the eurorack world, how does one accomplish safe ground from exposed metal of a module's faceplate if an AC to DC supply is in the case? For example, in a wooden case, should you have a direct connection from Vector rack rails to earth ground? Otherwise, you are relying on the 26 gauge wire in the module power cable to provide a path to ground in the event of a fault.

John

User avatar
MikeDB
Veteran Wiggler
Posts: 653
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Re: Case grounding questions

Post by MikeDB » Sun May 02, 2021 7:08 pm

Johnbomb wrote:
Sun May 02, 2021 3:31 pm
For example, in a wooden case, should you have a direct connection from Vector rack rails to earth ground?
Yes.
Always interested in audio startups needing help and investment, just PM me.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1275
Joined: Fri Jan 04, 2013 8:12 pm

Re: Case grounding questions

Post by Mungo » Mon May 03, 2021 7:25 am

Johnbomb wrote:
Sun May 02, 2021 3:31 pm
Thanks, guys- It's a tough one. Along the lines of safety: in the eurorack world, how does one accomplish safe ground from exposed metal of a module's faceplate if an AC to DC supply is in the case? For example, in a wooden case, should you have a direct connection from Vector rack rails to earth ground? Otherwise, you are relying on the 26 gauge wire in the module power cable to provide a path to ground in the event of a fault.
Depends on the specific standard you are trying to meet, but as you say it generally needs more than the ribbon cable to provide safety bonding.
Johnbomb wrote:
Sat May 01, 2021 9:44 pm
... In this setting, I would expect a module to have 2 potential paths to ground: 1) through the ground wire(s) in the power cable and 2) from faceplate to chassis.

Is this a setup for noise? If so, what is the solution?
One solution is to only used unbalanced connections between modules inputs and outputs, but eurorack standard is for unbalanced connections.

Another solution is to not have a case that needs to use the modules as part of the safety system. Use a power supply that contains all that separate to the case and module panels.

... or just keep adding more metal to try and reduce the resistance between any two points in the system.

Johnbomb
Common Wiggler
Posts: 152
Joined: Wed Apr 26, 2017 11:57 am
Location: Nashville

Re: Case grounding questions

Post by Johnbomb » Mon May 03, 2021 10:42 am

What do you think about taking steps such that the back of the module metal faceplate makes solid connection to the earth grounded vector rail? This would seem to address both safety and low resistance grounding to minimize noise.

User avatar
EATyourGUITAR
has no life
Posts: 6342
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Case grounding questions

Post by EATyourGUITAR » Mon May 03, 2021 10:53 am

Johnbomb wrote:
Mon May 03, 2021 10:42 am
What do you think about taking steps such that the back of the module metal faceplate makes solid connection to the earth grounded vector rail? This would seem to address both safety and low resistance grounding to minimize noise.
This already happens with metal jacks and metal panels without the vector rail being grounded inside the case. The path to ground is through the rails through the panels through the jacks back into the power distribution system which is already mains earth referenced. With plastic jacks or fiberglass panels, the jack sleeve is still grounded. There is no mains anywhere on the DC power supply unless the regulators fail closed input to output with floating ground and the rectifier fails closed with floating ground. All very unlikely to happen at the same time but if they did the ribbon cables would self destruct maybe in flames then it would become safe again.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1275
Joined: Fri Jan 04, 2013 8:12 pm

Re: Case grounding questions

Post by Mungo » Tue May 04, 2021 3:27 am

EATyourGUITAR wrote:
Mon May 03, 2021 10:53 am
Johnbomb wrote:
Mon May 03, 2021 10:42 am
What do you think about taking steps such that the back of the module metal faceplate makes solid connection to the earth grounded vector rail? This would seem to address both safety and low resistance grounding to minimize noise.
This already happens with metal jacks and metal panels without the vector rail being grounded inside the case. The path to ground is through the rails through the panels through the jacks back into the power distribution system which is already mains earth referenced. With plastic jacks or fiberglass panels, the jack sleeve is still grounded. There is no mains anywhere on the DC power supply unless the regulators fail closed input to output with floating ground and the rectifier fails closed with floating ground. All very unlikely to happen at the same time but if they did the ribbon cables would self destruct maybe in flames then it would become safe again.
This sort of misleading and inaccurate discussion is why people shouldnt be relying on random strangers on the internet for safety advice.

Every accessible metal surface needs to be protected from any possible single fault, not just the couple of faults you can think of. For people trying to make a safety earthed case the typical panels and rails do not make a reliable or low enough resistance path for protective earthing.

These are not suitable protections:
Not it'll be ok after some time
Not probably ok
Not relying on unspecified or untested characteristics

Safety needs to be extremely reliable and well thought through. 99% of people here should not be working with mains wiring, helping others to do so, or providing advice about how it works ok for them so go for it.

User avatar
EATyourGUITAR
has no life
Posts: 6342
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Case grounding questions

Post by EATyourGUITAR » Tue May 04, 2021 7:31 am

If you are correct then everyone with a wood case is wrong. Every manufacturer with a fiberglass panels is wrong. Every wood case without an earth connection to the rails is wrong. Every open back happy ending kit is wrong.

I think probably what you are thinking of is because there is mains inside the case then a wire could come lose or break through the jacket where it lands on something that the user touches that has no path to earth. Yeah I understand perfectly. I had to comply with all these codes at work when I built factory control panels and industrial power supplies for automation that could lift a car. Every size and power rating. Cabinets 20ft long. We only used metal cabinets. Everything had an OSHA emergency stop button. Everything had a light stack. Something's had redundant contactors just in case the primary contactor became welded shut.

But I'm not about to tell people they need to stop building wood cases. I'm not going to tell people to boycott fiberglass panels. I have opened up toaster ovens sold at WalMart. The statistics of how you are likely to be killed by an appliance accident is more likely that it will be the Hamilton bay toaster oven sold at WalMart. There are screws missing from the front panel controls that have mains running through them. The control housing is plastic. The spade terminals are not booted. The cloth wire jacket is thin, not impregnated with PVC and it is frayed. The wire is thin aluminum 4 strand tin plated. The chassis is not mains earthed but it is a metal chassis. How many of these things do you think we're manufactured? Probably over a million per year. Customs never stopped this shipment from entering the country to protect American consumers.

The things that would need to happen to a properly build eurorack case includes smashing it with a metal sledgehammer, cutting it in half with a chainsaw, or lighting it on fire. If you don't plan on doing that under normal operation then I think fiberglass panels are fine because there is no mains there. It requires two or three simultaneous failures for mains to get to the parts exposed to the user.

Please give me an example of how a user gets killed by mains and what would need to fail simultaneously. I'm interested to learn more. We can also talk about how likely this is.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

Johnbomb
Common Wiggler
Posts: 152
Joined: Wed Apr 26, 2017 11:57 am
Location: Nashville

Re: Case grounding questions

Post by Johnbomb » Tue May 04, 2021 2:58 pm

Mungo wrote:
Tue May 04, 2021 3:27 am
Every accessible metal surface needs to be protected from any possible single fault, not just the couple of faults you can think of. For people trying to make a safety earthed case the typical panels and rails do not make a reliable or low enough resistance path for protective earthing.
Is there any way to safely install an AC/DC converter in a case? Would each module metal faceplate need its own large ground wire going to earth?

User avatar
EATyourGUITAR
has no life
Posts: 6342
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Case grounding questions

Post by EATyourGUITAR » Tue May 04, 2021 3:19 pm

Johnbomb wrote:
Tue May 04, 2021 2:58 pm
Mungo wrote:
Tue May 04, 2021 3:27 am
Every accessible metal surface needs to be protected from any possible single fault, not just the couple of faults you can think of. For people trying to make a safety earthed case the typical panels and rails do not make a reliable or low enough resistance path for protective earthing.
Is there any way to safely install an AC/DC converter in a case? Would each module metal faceplate need its own large ground wire going to earth?
ask around. take a survey. how many people have a #12 ground wire with the correct color code and lock washer bonded to every metal panel eurorack module in a complete system. probably the number of people that have a system like this is close to %0. statistically insignificant. so then the argument can be made that we should only use metal panel no plastic no fiberglass. I dare someone to start talking smack about all the manufacturers that do this. these are DC +/-12v appliances. obviously the complaint goes against what has become standard practice in the industry. it is a well founded complaint but it is not so black and white to say that everyone is doing it wrong. there is some grey area in between. if the argument is being made that a wire could come lose inside the case making the DC side hot with mains then the argument can also be made that the ground wire could fall off and make the entire case a death machine even if it was built correctly. the argument goes both ways. so in the end, follow best practices. use the correct size wires for safety mains earth. everything should be booted and securely mounted with enough of an air gap between mains and DC components. the wiring should be clean and not coiled or strained. technically the whole system would need to be tested by UL for $50,000 with all the modules installed but they would also destroy the system doing some destructive testing. I don't think anyone is actually in compliance with the law but they get around it because the law does not apply to an incomplete product, components, or kits.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
MikeDB
Veteran Wiggler
Posts: 653
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Re: Case grounding questions

Post by MikeDB » Tue May 04, 2021 5:01 pm

Johnbomb wrote:
Tue May 04, 2021 2:58 pm
Is there any way to safely install an AC/DC converter in a case? Would each module metal faceplate need its own large ground wire going to earth?
Probably not, though it would arguably be the best solution. In large console mixers, which have a similar though larger construction to modular synths, the top and bottom rails are usually bolted to a grounded sub-chassis which sits inside the wooden case. However this is purely for safety earthing of the front panels, which are always metal. The signal/supply ground is totally isolated from this and goes through almost every other wire in the wide ribbon cable(s) which run the whole length of the mixer, not just the two pins of a typical Eurorack synth supply. And then as EatGuitar says, we then have to put the prototype through expensive testing to check everything is truly safe.
Always interested in audio startups needing help and investment, just PM me.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1275
Joined: Fri Jan 04, 2013 8:12 pm

Re: Case grounding questions

Post by Mungo » Tue May 04, 2021 6:10 pm

EATyourGUITAR wrote:
Tue May 04, 2021 7:31 am
If you are correct then everyone with a wood case is wrong. Every manufacturer with a fiberglass panels is wrong. Every wood case without an earth connection to the rails is wrong. Every open back happy ending kit is wrong.
Since you're talking out loud without actually forming any cohesive story or argument to follow its hard to address this. You're not linking any of what I said to what you say.

Its very hard to make a hodge podge of random parts make a safe system if each part needs to provide some part of the safety. Note that when I say most panels are able to mount solidly enough to provide safety bonding I'm not taking the extremes of non-conductive panels but "normal" metal ones. Its possible to build a wood case, nowhere do I say its wrong.
EATyourGUITAR wrote:
Tue May 04, 2021 7:31 am
Customs never stopped this shipment from entering the country to protect American consumers.
We don't care about your life story or that you can find examples of unsafe equipment (a quick browse of the low cost case option on etsy etc easily comes up with more relevant examples). The American centric view of electrical safety is already an outlier (110V power is minority worldwide) where the lower voltages are akin to someone throwing a few nails at you, unlikely to kill and probably just annoying, vs the rest of. the world where 240V mains is more like throwing a brick at someone, still not likely to kill but much higher risk.

But the worst of it is you keep going back and repeating the misleading and more dangerous points:
EATyourGUITAR wrote:
Tue May 04, 2021 7:31 am
Every open back happy ending kit is wrong.
No, the happy ending kit keeps all the electrical safety neatly contained within the power brick so the user doesn't ever have to know anything about it and can't mess it up. Those sorts of setups are the safest option and should be made clear.
Last edited by Mungo on Tue May 04, 2021 6:17 pm, edited 3 times in total.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1275
Joined: Fri Jan 04, 2013 8:12 pm

Re: Case grounding questions

Post by Mungo » Tue May 04, 2021 6:15 pm

Johnbomb wrote:
Tue May 04, 2021 2:58 pm
Mungo wrote:
Tue May 04, 2021 3:27 am
Every accessible metal surface needs to be protected from any possible single fault, not just the couple of faults you can think of. For people trying to make a safety earthed case the typical panels and rails do not make a reliable or low enough resistance path for protective earthing.
Is there any way to safely install an AC/DC converter in a case? Would each module metal faceplate need its own large ground wire going to earth?
There are several ways but its not easily done DIY. Just because you can't rely on the faceplates for protective bonding doesn't mean a safe case can't be built. But its very difficult and beyond any simple description on a forum (especially since any rational attempt to discuss it get buried in the noise as above). The easiest way to get to a safe case is to start with a power supply that contains all the electrical safety inside its enclosure, then the case doesn't need to worry about it.

User avatar
EATyourGUITAR
has no life
Posts: 6342
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Case grounding questions

Post by EATyourGUITAR » Tue May 04, 2021 6:26 pm

Mungo wrote:
Tue May 04, 2021 6:10 pm
But the worst of it is you keep going back and repeating the misleading and more dangerous points:
EATyourGUITAR wrote:
Tue May 04, 2021 7:31 am
Every open back happy ending kit is wrong.
No, the happy ending kit keeps all the electrical safety neatly contained within the power brick so the user doesn't ever have to know anything about it and can't mess it up. Those sorts of setups are the safest option and should be made clear.
this is exactly my point. even if you have mains inside the wooden case. how is wood any different compared to plastic. if the wood case is fully enclosed than the only difference is that the isolated laptop power supply passed hipot testing but it doesn't pass hipot testing where IEC mains entered the brick. hipot testing is done at 6000v when I do it. not 110vac or 240vac. the isolated laptop supply sends DC to other DC devices. there is no mains AC on the DC side. if you are concerned that the mains to DC supply will fail in such a way that the mains will be on the DC where it is then exposed to the user then you could say the same about an isolated supply that fails in such a way that the mains is on the DC output of the brick. these are all pretty close although not entirely identical. I am not advocating for mains AC to be exposed to the user. what I am saying is that how is mains going to get to the user in a wooden case. you completely ignored the question because you can't answer it or you don't want to.

my problem with your tone is that you do the same thing hinton did. you have your opinion which is completely fine and that is your right. but to go onto a forum and start spouting that other people are spreading misinformation and they shouldn't even speak because they are not qualified. dude. give me a break. that is the only reason why I post my qualifications here. if everyone was going to die from such statistically rare circumstances that you are talking about then there would be a million people dead already from modular synthesizers. what is the actual failure mode that you are beat around the bush. put your money where your mouth is and show us the failure mode that you claim is a big problem.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
SkyWriter
Veteran Wiggler
Posts: 667
Joined: Fri May 16, 2014 6:18 pm
Location: Absolute Elsewhere
Contact:

Re: Case grounding questions

Post by SkyWriter » Tue May 04, 2021 6:41 pm

For the vast majority; it's a hobby. It's not even a dangerous hobby like racing.

You both got good points.

If you're producing product; by all means protect yourself from litigation.

But, this is amateur hour, as long as folks don't start patching 120vac into their VCA's, it's going to be OK.

Oh, and please don't synthesis standing in a puddle you performance folks.
-----
Yes, I spent my career on fault tolerance computing - so I GET it. This, ain't that.
Prologue 16 - an adventure worth having!

User avatar
MikeDB
Veteran Wiggler
Posts: 653
Joined: Fri Feb 08, 2019 8:28 am
Location: UK

Re: Case grounding questions

Post by MikeDB » Tue May 04, 2021 6:47 pm

Whilst I haven't heard of anyone killed using a modular synth, I would like to point out that we still seem to lose one or two musicians a year to electrocution. This often gets blamed on 'faulty mic wiring' which is usually a nonsense as the mic is usually the ground route for a live voltage already getting to the musician but unable to get to ground through their shoes or whatever. And whilst mains voltage doesn't usually kill (as Keith Richards, Greg Lake, George Harrison and countless others have proven in the past), occasionally it does. And this doesn't just happen on stage, Keith Relf died in his home studio.

So spending the effort making sure all your equipment is as safe as can be is well worth the time and effort involved.
Always interested in audio startups needing help and investment, just PM me.

User avatar
Mungo
Super Deluxe Wiggler
Posts: 1275
Joined: Fri Jan 04, 2013 8:12 pm

Re: Case grounding questions

Post by Mungo » Tue May 04, 2021 6:54 pm

EATyourGUITAR wrote:
Tue May 04, 2021 6:26 pm
Mungo wrote:
Tue May 04, 2021 6:10 pm
But the worst of it is you keep going back and repeating the misleading and more dangerous points:
EATyourGUITAR wrote:
Tue May 04, 2021 7:31 am
Every open back happy ending kit is wrong.
No, the happy ending kit keeps all the electrical safety neatly contained within the power brick so the user doesn't ever have to know anything about it and can't mess it up. Those sorts of setups are the safest option and should be made clear.
this is exactly my point. even if you have mains inside the wooden case. how is wood any different compared to plastic. if the wood case is fully enclosed than the only difference is that the isolated laptop power supply passed hipot testing but it doesn't pass hipot testing where IEC mains entered the brick. hipot testing is done at 6000v when I do it. not 110vac or 240vac. the isolated laptop supply sends DC to other DC devices. there is no mains AC on the DC side. if you are concerned that the mains to DC supply will fail in such a way that the mains will be on the DC where it is then exposed to the user then you could say the same about an isolated supply that fails in such a way that the mains is on the DC output of the brick. these are all pretty close although not entirely identical. I am not advocating for mains AC to be exposed to the user. what I am saying is that how is mains going to get to the user in a wooden case. you completely ignored the question because you can't answer it or you don't want to.
EATyourGUITAR wrote:
Tue May 04, 2021 6:26 pm
what is the actual failure mode that you are beat around the bush. put your money where your mouth is and show us the failure mode that you claim is a big problem.
I'm not going to play whack a mole trying to discuss every single tiny detail of electrical safety with you or anyone else on a forum. Buy a copy of your relevant standard and stick to it.

You can't separate the basic concepts of a case that requires safety bonding (class I) vs a case that does not require any such complex and specific details to achieve the same level of safety (class III). So there is no discussion here but just your misleading noise that prevents people from understanding whats actually going on.

User avatar
EATyourGUITAR
has no life
Posts: 6342
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Case grounding questions

Post by EATyourGUITAR » Tue May 04, 2021 10:26 pm

this is the DIY section. not the manufacturer section. not the ISO section. not the underwriters laboratory section. if someone wants to build a DIY wood case with mains inside then maybe you should help them do it instead of attacking me. you have not contributed anything to this thread. only your own noise and attacks. do you want to teach or do you want to start fights? MikeDB is a professional and a nice guy. he shared his knowledge. he pointed out that there were some interesting and correct points that I made. but he did not launch any personal attacks. take a lesson on how to share your opinion without insulting others. maybe you can find an ISO IEEE IEC NEMA ANSI whitepaper on how to do social interaction.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
htor
Veteran Wiggler
Posts: 538
Joined: Sun Dec 01, 2019 6:20 am

Re: Case grounding questions

Post by htor » Wed May 05, 2021 7:12 am

grounding threads — where people go to vent

Post Reply

Return to “Music Tech DIY”