it can do if you connect the wrong wires to the star. Take for instance a two wire input to the power amplifier. Break the signal return wire and take it on a long route to the star reference node. That introduces a big loop area into the original signal pair. It will be susceptible to interference.What about star-ground? Does this make things worse?
I keep referring Members to the important (for me) rule that ALL connections between modules MUST include the Flow and Return parts of the two wire connections.This is interesting, but can you illustrate where it's wrong?
When I say "you" I mean "anyone" who feels "ground" is labeled wrong?
Do you have an example circuit?
I've been building electronic boxes since 1987 and still find electronics very confusing.
I think this is due to improper labeling. You can picture concepts or you can't. I can't, mostly because I expect laws to apply. So, maybe if things were labeled properly, things would be more clear. I'm still waiting for that bell to go off in my head, but I don't think it will happen in my life-time.
V.
This two wire connection must not be broken. The circuit will not work if you break either the flow route or the return route.
The Star ground is a voltage reference node. One takes a reference from here to those few parts of the assembly that need a reference.
A typical example are the power amplifier input and output. Compare the PA to an Opamp. They both have two inputs and two outputs. But often the Opamp return half of the output is hidden. It can be the same for the PA.
Just because it is hidden does not mean it does not exist. You must connect both the signal outputs. But that still leaves the output without a reference to what "Zero Volts" is.
The Opamp and the PA need a voltage reference connection from IN to OUT. This voltage reference can and usually is made at the Main Audio Ground (MAG, I prefer not to call it star ground).
In a mono amplifier the MAG can be on the amp PCB. Read Leach Lo Tim he uses a short resistor connected trace on the PCB to achieve a low impedance reference connection between the IN and OUT. And then he connects a long high current (high impedance) route to the off board MAG.
It is worth reading and re-reading the Leach Lo Tim articles. Lots of very valuable information in the long design guide. I found it before I joined this Forum and still return to it occassionally to check some detail.
Last edited:
Electronics always, without exception, obeys the laws of physics and mathematics. It does not always obey what simple recipes say it should do, as it knows nothing about these recipes. If you expect laws to apply then you are on the right track; you just need to learn the laws and ignore the recipes.vdi_nenna said:You can picture concepts or you can't. I can't, mostly because I expect laws to apply.
If you expect laws to apply then you are on the right track; you just need to learn the laws and ignore the recipes.
a gem of wisdom right there....
Star ground is a voltage reference node so where does the belief that this point needs to be earthed come from and why do people keep attaching things like metal cases and connector shells to it?
The mains powered equipment we can build is ClassI. This requires that the chassis be permanently and mechanically fixed to the Protective Earth wire in your mains supply.
There is a secondary requirement:
All exposed conductive parts should be connected to the protected chassis.
This is the one that requires all the metal bits sticking through the enclosure (even if it's plastic/wood) to be safe to touch. The receptacles inside an XLR socket cannot be touched without using some implement.
Compare to the barrel of an RCA. You can easily touch the metal part, some even leave the metal part touchable when the socket has a plug connected.
The recptacle inside the RCA cannot be touched so does not need to be connected to the Chassis.
It's compliance with this rule that forced some manufacturers to adopt protected speaker sockets like the BFA type.
These exposed metal parts, including screw heads, need to be connected and the easy way is to make a connection from Main Audio Ground to Chassis using a wire/circuit that can pass Fault Current for as long as it takes for the mains fuse to rupture and the arc to extinguish.
That Fault current can be some kA (yes thousands of Amperes !).
And this kA of Fault Current is why you must never use a resistor to make that Safety connection.
I have said it before but I think it worth repeating:
When a ClassI product has an interconnect cable connected to it's in/out, the other end is now exposed and connected to a piece of mains powered equipment. Is that other end safe for an inquisitive baby crawling around and picking up anything that is worth sucking on?
A piece of ClassII equipment that is interconnected with a ClassI product relies on the interconnect cable to blow the mains fuse if the ClassII product develops a catastrophoic mains fault that cannot escape down it's non existant PE wire.
There is a secondary requirement:
All exposed conductive parts should be connected to the protected chassis.
This is the one that requires all the metal bits sticking through the enclosure (even if it's plastic/wood) to be safe to touch. The receptacles inside an XLR socket cannot be touched without using some implement.
Compare to the barrel of an RCA. You can easily touch the metal part, some even leave the metal part touchable when the socket has a plug connected.
The recptacle inside the RCA cannot be touched so does not need to be connected to the Chassis.
It's compliance with this rule that forced some manufacturers to adopt protected speaker sockets like the BFA type.
These exposed metal parts, including screw heads, need to be connected and the easy way is to make a connection from Main Audio Ground to Chassis using a wire/circuit that can pass Fault Current for as long as it takes for the mains fuse to rupture and the arc to extinguish.
That Fault current can be some kA (yes thousands of Amperes !).
And this kA of Fault Current is why you must never use a resistor to make that Safety connection.
I have said it before but I think it worth repeating:
When a ClassI product has an interconnect cable connected to it's in/out, the other end is now exposed and connected to a piece of mains powered equipment. Is that other end safe for an inquisitive baby crawling around and picking up anything that is worth sucking on?
A piece of ClassII equipment that is interconnected with a ClassI product relies on the interconnect cable to blow the mains fuse if the ClassII product develops a catastrophoic mains fault that cannot escape down it's non existant PE wire.
Last edited:
I actually manually disconnected the shell of my RCA's from the chassis. I've never had any major electrical discharge happen and when I work on the devices I shut them off. I encountered some interference when the shell was attached to the chassis.
Last edited by a moderator:
The RCA outer must be connected to something. Is that something grounded?
'It has never happened, so it won't happen' is the way NASA used to approach safety until Feynman exposed it.
'It has never happened, so it won't happen' is the way NASA used to approach safety until Feynman exposed it.
The RCA outer must be connected to something. Is that something grounded?
'It has never happened, so it won't happen' is the way NASA used to approach safety until Feynman exposed it.
The RCA's shell is the Neutral which is connected to the "black" wire on the connector. The center pin is Live, the shell is neutral. and the neutral shouldn't be connected to ground, in my opinion.
i'm not saying that safety ground/earth practices be tossed out or ignored.
but why is the 0 volt node of the device's power supply attached to this point?(chassis in this case)
if by code we can't or are not allowed to isolate connectors from safety earth isn't that just a recipe for noise hum problems?
but why is the 0 volt node of the device's power supply attached to this point?(chassis in this case)
if by code we can't or are not allowed to isolate connectors from safety earth isn't that just a recipe for noise hum problems?
i'm not saying that safety ground/earth practices be tossed out or ignored.
but why is the 0 volt node of the device's power supply attached to this point?(chassis in this case)
if by code we can't or are not allowed to isolate connectors from safety earth isn't that just a recipe for noise hum problems?
That's exactly what I think. That's perhaps why we have a signal ground wire after all. But I guess from now on that wire is mored aesthetically pleasing as our signal ground bleeds every possible way, now that ChassGround is the new SigGround.
Last edited:
BGW amps where some of the first amps i encountered where a 2 terminal jumper was provided so the user could isolate circuit from safety ground.
this simple provision went a long way in eliminating hum loop and noise problems.
this simple provision went a long way in eliminating hum loop and noise problems.
Is this ambiguous?....................
There is a secondary requirement:
All exposed conductive parts should be connected to the protected chassis.
........................
I thought I had made it clear and unbderstanable !
Depending on what is being disconnected, this could be illegal.BGW amps where some of the first amps i encountered where a 2 terminal jumper was provided so the user could isolate circuit from safety ground.
this simple provision went a long way in eliminating hum loop and noise problems.
I have a pro crossover. Inside it has a ground lift option. One manually moves the spade connector across to the other connection.
Both options keep the exposed parts connected to the enclosure. It is legal.
as a manufacturer that distributed there products to several countries i don't think BGW broke any laws.
do tell AndrewT what x-over are we talking about....i'm curious to see if they made made the PIN 1 connection direct to the case.
do tell AndrewT what x-over are we talking about....i'm curious to see if they made made the PIN 1 connection direct to the case.
back to the original question where oh, where is it written that the 0 volt node of a supply must be earthed?
from what i've seen this is why a good many homebrew devices (and alot of "professionally manufactured" products) turn into hum generators in use. somebody thought it was a good idea to ground/earth the 0 volt node of the supply.
from what i've seen this is why a good many homebrew devices (and alot of "professionally manufactured" products) turn into hum generators in use. somebody thought it was a good idea to ground/earth the 0 volt node of the supply.
Last edited:
that's a shameI can look but I can't show you anything.
perhaps if you care to tell me what "pro" crossover you have i can possibly provide an image.... i've serviced and installed so many .
- Status
- Not open for further replies.
- Home
- Amplifiers
- Power Supplies
- when is ground not ground?