C
CryingDragon
Is the best place indeed the chassis? See my amp is being built in an old car amplifier case (witch the entire thing is aluminum heatsink with a steele panel cover) and there is a round terminal pedistal (used to mount bridge) and i wana know if i can use this to be my star ground? Also i wanted to put my transformer in a diffrent box because i nave no room for it in the amp case and run 2 wires (my +- rails) to the amp case and to te amp,Cam i do this? And put all the grounds on a star point inside the case?
OK EDIT: Nevermind i'll use an XLR jack (3 pin) to connect my poer box to the amp case (carry the transformers + _ and CT to the amp case and rectify it inside.....yes...) ok so i'm still dumbfounded about the star ground tho.
OK EDIT: Nevermind i'll use an XLR jack (3 pin) to connect my poer box to the amp case (carry the transformers + _ and CT to the amp case and rectify it inside.....yes...) ok so i'm still dumbfounded about the star ground tho.
My habit has been to use a dual row terminal block with jumpers across the barriers. I bolt this to the floor of the chassis at some convenient point. All ground lines that are supposed to meet at the star point have ring connectors and are screwed down tight to the terminal block. One (and only one) heavy wire runs from the terminal block to the earthing point where you also find the connection to the ground wire from the three wire power cord.
I never had a hint of hum with this. By the way I also use completely separate paths to the star point for each of the following1- power rail bypassing, 2- input signal return/shield, 3- Output signal return, 4- any other 0V network. Multiply by two for stereo.
I never had a hint of hum with this. By the way I also use completely separate paths to the star point for each of the following1- power rail bypassing, 2- input signal return/shield, 3- Output signal return, 4- any other 0V network. Multiply by two for stereo.
Safety Earth and Signal Ground
At some point all the grounds should have the same potential: 0 V.
It is a matter of keeping currents away from each other. . .
Power supply ground charges up all the potential and delivers
all the low current for input areas, output areas, lights, cool accessories.
Input signal ground has its low currents.
Output signal ground does its own high current thing.
Safety ground it the end of the line; something goes wrong, all
faults go here, and down into the 6 foot rod near your electric
company meter.
Elliot Sound talks ad nauseum about the subject; but for good reason:
it is still misunderstood!
All equipment should be three wire. . . Hot, Neutral, and Ground; but
much of the equipment out there uses the Neutral as if it were ground.
Neutral is ground at the meter! The rest of the time it is the return
for the Hot. . .I'm talking about unbalanced AC mains. . . in the USA.
Then we hook the whole thing to a transformer and create a new
ground with the center tap. . .which at some point should be
tied to the earth ground, unless it is a two wire setup. . .
At least that is how I see it. . . nobody wants to get killed listening
to their music.
At some point all the grounds should have the same potential: 0 V.
It is a matter of keeping currents away from each other. . .
Power supply ground charges up all the potential and delivers
all the low current for input areas, output areas, lights, cool accessories.
Input signal ground has its low currents.
Output signal ground does its own high current thing.
Safety ground it the end of the line; something goes wrong, all
faults go here, and down into the 6 foot rod near your electric
company meter.
Elliot Sound talks ad nauseum about the subject; but for good reason:
it is still misunderstood!
All equipment should be three wire. . . Hot, Neutral, and Ground; but
much of the equipment out there uses the Neutral as if it were ground.
Neutral is ground at the meter! The rest of the time it is the return
for the Hot. . .I'm talking about unbalanced AC mains. . . in the USA.
Then we hook the whole thing to a transformer and create a new
ground with the center tap. . .which at some point should be
tied to the earth ground, unless it is a two wire setup. . .
At least that is how I see it. . . nobody wants to get killed listening
to their music.
talk about misunderstanding and then this!.
NO. the safety earth does one job.
The audio ground does a completely different job.
Safety Earth and Audio Ground should NOT be connected.
The misunderstanding is in part caused by commentators repeatedly calling everything "ground".
Use the correct names and much of the misunderstanding could be eliminated.
Ground is 0V potential
Upon entering a metal box the Earth Ground should be attached to
the interior with a bolt. . . Now if anything 'bad' happens inside
the box with the circuit or lighten strikes the box, the current has a
safe dedicated path to Earth Ground. The box becomes an extension
of the good old Earth. . .the 0V reference.
Again, at some point the transformer shield should be bolted to the
metal box to insure that if something goes wrong that it is shunted
to the earth. It also makes it electrically 0V (earth ground).
I completely agree that the various 'grounds' perform different jobs,
but they should all be 0V potential independent of the currents they
may carry; the currents shouldn't infringe upon each other. Keeping
lower current grounds upstream from higher current power supply
grounds is important. . .but they should not all be floating.
I think the problem arises when a two prong AC piece of equipment
and a three prong piece of equipment are connected together.
The two prong equipment "creates" a ground point that this not
referenced to 0V (earth ground). When connected to the three
prong equipment that does reference it's 0V point to earth ground
you can have movement through the three prong equipment since
the two prong equipment 0V reference might really be a millivolt
higher or lower than true 0V.
I agree that I've used the word 'ground' to liberally. There are
differences in function between Floating Ground, Signal Ground,
Power Supply Ground, Earth Ground. But ultimately they need
to be 0V reference to each other.
Great Article:
http://www.aikenamps.com/StarGround.html
http://www.rane.com/note151.html
Upon entering a metal box the Earth Ground should be attached to
the interior with a bolt. . . Now if anything 'bad' happens inside
the box with the circuit or lighten strikes the box, the current has a
safe dedicated path to Earth Ground. The box becomes an extension
of the good old Earth. . .the 0V reference.
Again, at some point the transformer shield should be bolted to the
metal box to insure that if something goes wrong that it is shunted
to the earth. It also makes it electrically 0V (earth ground).
I completely agree that the various 'grounds' perform different jobs,
but they should all be 0V potential independent of the currents they
may carry; the currents shouldn't infringe upon each other. Keeping
lower current grounds upstream from higher current power supply
grounds is important. . .but they should not all be floating.
I think the problem arises when a two prong AC piece of equipment
and a three prong piece of equipment are connected together.
The two prong equipment "creates" a ground point that this not
referenced to 0V (earth ground). When connected to the three
prong equipment that does reference it's 0V point to earth ground
you can have movement through the three prong equipment since
the two prong equipment 0V reference might really be a millivolt
higher or lower than true 0V.
I agree that I've used the word 'ground' to liberally. There are
differences in function between Floating Ground, Signal Ground,
Power Supply Ground, Earth Ground. But ultimately they need
to be 0V reference to each other.
Great Article:
http://www.aikenamps.com/StarGround.html
http://www.rane.com/note151.html
Rane Notes
http://www.rane.com/note151.html
Chassis Ground vs. Signal Ground
Let us examine the distinction between chassis and signal ground in audio devices. Chassis ground is generally considered any conductor which is connected to a unit's metal box or chassis. The term chassis ground may have come about since units with 3-conductor line cords connect the chassis to earth ground when plugged in to a properly wired AC outlet. In units with a 2-conductor line cord (consumer equipment), the chassis does not connect to earth ground, though the chassis is normally connected to the signal ground in the box in both unbalanced/consumer and in balanced/pro equipment.
Signal ground is the internal conductor used as the 0 V reference potential for the internal electronics and is sometimes further split into digital and analog ground sections. Further signal ground splits are also possible, though it is important to remember that all "divisions" of signal ground connect together in one place. This is usually called a star grounding scheme.
It is easy to confuse chassis ground and signal ground since they are usually connected together -- either directly or through one of several passive schemes. Some of these schemes are shown in Figure 3. The key to keeping an audio device immune from external noise sources is knowing where and how to connect signal ground to the chassis.
http://www.rane.com/note151.html
Chassis Ground vs. Signal Ground
Let us examine the distinction between chassis and signal ground in audio devices. Chassis ground is generally considered any conductor which is connected to a unit's metal box or chassis. The term chassis ground may have come about since units with 3-conductor line cords connect the chassis to earth ground when plugged in to a properly wired AC outlet. In units with a 2-conductor line cord (consumer equipment), the chassis does not connect to earth ground, though the chassis is normally connected to the signal ground in the box in both unbalanced/consumer and in balanced/pro equipment.
Signal ground is the internal conductor used as the 0 V reference potential for the internal electronics and is sometimes further split into digital and analog ground sections. Further signal ground splits are also possible, though it is important to remember that all "divisions" of signal ground connect together in one place. This is usually called a star grounding scheme.
It is easy to confuse chassis ground and signal ground since they are usually connected together -- either directly or through one of several passive schemes. Some of these schemes are shown in Figure 3. The key to keeping an audio device immune from external noise sources is knowing where and how to connect signal ground to the chassis.
Hi,
The audio ground does not need to be tied to safety earth for ANY audio reason.
The ONLY reason for attaching the disconnecting network between Safety Earth and Audio Ground is to ensure the safety of operator(s) in the event of a mains fault developing in the unit that could cause some exposed conductive part to become connected to live.
The chance of lightning hitting the unit WHILE the operator is touching it is VERY low risk and would not be designed for in normal domestic use.
In any case the conductor sizes involved would not survive a lightning strike.
Loss or damge to equipment due to a lightning strike would be covered by insurance if the owner had bothered to obtain cover.
Rane's fig 3 shows the disconnecting network options, but I disagree with the use of a switch in domestic equipment when untrained operators could misuse the device. Similarly as explained in many posts I completely disagree with a direct connection between the Audio Ground and Safety Earth. The other components are sensible but I have added Pass' Power Thermistor recommendation to my listing.
No they don't.
The audio ground does not need to be tied to safety earth for ANY audio reason.
The ONLY reason for attaching the disconnecting network between Safety Earth and Audio Ground is to ensure the safety of operator(s) in the event of a mains fault developing in the unit that could cause some exposed conductive part to become connected to live.
The chance of lightning hitting the unit WHILE the operator is touching it is VERY low risk and would not be designed for in normal domestic use.
In any case the conductor sizes involved would not survive a lightning strike.
Loss or damge to equipment due to a lightning strike would be covered by insurance if the owner had bothered to obtain cover.
Rane's fig 3 shows the disconnecting network options, but I disagree with the use of a switch in domestic equipment when untrained operators could misuse the device. Similarly as explained in many posts I completely disagree with a direct connection between the Audio Ground and Safety Earth. The other components are sensible but I have added Pass' Power Thermistor recommendation to my listing.
Earth Ground to Signal Ground Connections
Here is an example of something simalar that you mentioned.
http://sound.westhost.com/project04.htm
Putting some protection between Earth and Signal but still allowing
faults to reach a real earth connection.
I do agree that signal and earth don't have to touch each other when
using a floating ground as most consumer electronics use.
But it seems very safe to connect the two properly observing upstream
and down stream currents to avoid humm. All the amplifiers that I have
ever looked at have the power supply CT and the Common output
signal (- speaker output) and input signal (-) connected at some point.
The power supply is going to swing the voltage positive then negative
in reference to the CT of the transformer (Power Supply) and it would
be an introduction of noise if the input signal had a different signal
ground reference voltage.
Maybe we are splitting hairs here. . . In the end, we want good
clean audio. . .some equipment is designed to use floating
grounds, some is designed to reference signal ground to earth
ground. Both work. I would think the more pieces of equipment
that you interconnect, the more likely you would want all the signal
grounds referenced to the same 0V point. . .long interconnects
could set up voltage potential in the signal ground.
Most home equipment probably doesn't benefit from such a practice.
I looks like we have both seen our share of different grounding
designs.
Okay, lightening was an extreme example.. . I was kind of looking
at an antenna. . .grounding would be important. . .sure the
equipment would be fried. . .but it all is earthed. . . you are less
likely to become the earthing point.
Here is an example of something simalar that you mentioned.
http://sound.westhost.com/project04.htm
Putting some protection between Earth and Signal but still allowing
faults to reach a real earth connection.
I do agree that signal and earth don't have to touch each other when
using a floating ground as most consumer electronics use.
But it seems very safe to connect the two properly observing upstream
and down stream currents to avoid humm. All the amplifiers that I have
ever looked at have the power supply CT and the Common output
signal (- speaker output) and input signal (-) connected at some point.
The power supply is going to swing the voltage positive then negative
in reference to the CT of the transformer (Power Supply) and it would
be an introduction of noise if the input signal had a different signal
ground reference voltage.
Maybe we are splitting hairs here. . . In the end, we want good
clean audio. . .some equipment is designed to use floating
grounds, some is designed to reference signal ground to earth
ground. Both work. I would think the more pieces of equipment
that you interconnect, the more likely you would want all the signal
grounds referenced to the same 0V point. . .long interconnects
could set up voltage potential in the signal ground.
Most home equipment probably doesn't benefit from such a practice.
I looks like we have both seen our share of different grounding
designs.
Okay, lightening was an extreme example.. . I was kind of looking
at an antenna. . .grounding would be important. . .sure the
equipment would be fried. . .but it all is earthed. . . you are less
likely to become the earthing point.
Re: Earth Ground to Signal Ground Connections
No more is required.
Allow yourself to open the door to the possibility that there is a better way.
Yes, I am on 240Vac but why should that make any difference. A hum loop is a hum loop no matter which end of the earth it's at.
Definatelygni said:
Agreed
Not necessarily safer. The disconnecting network should achieve sufficient conduction to blow the mains fuse in event of a mains fault going to earth.
No more is required.
I think the audio ground (the reference) in the way you describe is the only way this can and must operate.
No. Both don't work. An Audio Ground directly connected to Safety Earth introduces the probability of a hum loop that WILL happen if a similarly mis-connected source is plugged into any part of the source to speaker chain.
If each unit uses it's own audio ground and this audio ground has no direct connection to safety earth then the linking together does not introduce a long interconnect problem
But some are wired incorrectly and to compound it by also wiring the power amp incorrectly cannot be good practice. With the proliferation of double insulated audio units, the mis-wiring of Safety Earth to Audio Ground will become less common, but we still have all the old three wire mains connected sources to contend with, so the domestic situation can benefit from correctly wiring the power amps within our control.
Yes, I did it wrongly until recently, because I didn't know any better.
Allow yourself to open the door to the possibility that there is a better way.
Yes, I am on 240Vac but why should that make any difference. A hum loop is a hum loop no matter which end of the earth it's at.
Both Rod Elliot and Doug Self describe about the same grounding scheme. I found the diagrams confusing at first but eventually made sense of them. Part of the problem is that they are half way between a schematic and a physical wiring diagram -- something that is probably unavoidable.
Self's practice of using the input signal shield to tie the star point to the chassis/earth ground seams weird and dubious at first glance but I think I understand the point. However, I think connecting the shield to the chassis ground at near end may be equally effective and often mechanically easier if you happen to be using an insulated RCA connector. Conversely, if you are using non-insulated RCA connectors you will have tied the input singnal by default.
As far as my limited experience goes I never had a hum problem arrising from a DIY project that uses this method. Any problems have alway come from commercial equipment.
Self's practice of using the input signal shield to tie the star point to the chassis/earth ground seams weird and dubious at first glance but I think I understand the point. However, I think connecting the shield to the chassis ground at near end may be equally effective and often mechanically easier if you happen to be using an insulated RCA connector. Conversely, if you are using non-insulated RCA connectors you will have tied the input singnal by default.
As far as my limited experience goes I never had a hum problem arrising from a DIY project that uses this method. Any problems have alway come from commercial equipment.
Grounding. . .oh Grounding. . . why!
I specifically asked about the 220/240 VAC situation since here
in the United States all 220/240 VAC plugs are either three or even
four prong. . . I wasn't sure if any plugs in your part of the word
are 2 prong 220/240 VAC. . .
I guess we are not going to agree on this point:
Some equipment operates on a floating ground (the signal and power
supply ground do not connect to the earth ground) usually with
a two prong to AC mains.
Some equipment operates on a ground referenced to earth (the
signal and power supply ground connect to earth ground) usually
with a three prong to AC mains. Often seen on equipment with
a 'ground lift' switch. I agree that the switch is a bad idea. . .
it should be on or off but not either. . .
At on point I said:
All the amplifiers that I have ever looked at have the power supply CT and the Common output signal (- speaker output) and input signal (-) connected at some point.
You replied:
I think the audio ground (the reference) in the way you describe is the only way this can and must operate.
Thus the final point that we disagree on is that I have seen and
like that some units also tie the Center Tap to the transformer
case and the Earth Ground without passive elements. Where as
you think that it is never tied together or only with passive
elements. . .See the next statement.
We do agree on your point that some sort of passive form of network
should be used if a connection does exist between earth ground
and the signal/power supply ground; such as the following:
Pass' Power Thermistor
Resistor
Capacitor
Diodes
Variations of said passive devices.
I think audio engineers (and electrical engineers also) have come
up with different solutions to the problems of grounding. Different
situations require different circuits to counter the problems.
What a great conversation this has been. . .I hope we can contribute
further in other threads. . .may our threads cross again.
But let me know about the two prong/three prong 220/240 VAC
question; it just seems 'cool' that many countries routinely use
such high voltage mains. . .we only use them for dryers and electric
stoves (30 Amps for the dryer circuit and 50 Amps for the electric
stove circuit). The rest of our wall sockets are three prongs but
the neutral and ground plug are tied together at the electric
meter. Some are 20 amp and the rest are 15 amp plugs.
Thank you Andrew T
Chris Browne
Northern California (1 hour north of San Francisco)
I specifically asked about the 220/240 VAC situation since here
in the United States all 220/240 VAC plugs are either three or even
four prong. . . I wasn't sure if any plugs in your part of the word
are 2 prong 220/240 VAC. . .
I guess we are not going to agree on this point:
Some equipment operates on a floating ground (the signal and power
supply ground do not connect to the earth ground) usually with
a two prong to AC mains.
Some equipment operates on a ground referenced to earth (the
signal and power supply ground connect to earth ground) usually
with a three prong to AC mains. Often seen on equipment with
a 'ground lift' switch. I agree that the switch is a bad idea. . .
it should be on or off but not either. . .
At on point I said:
All the amplifiers that I have ever looked at have the power supply CT and the Common output signal (- speaker output) and input signal (-) connected at some point.
You replied:
I think the audio ground (the reference) in the way you describe is the only way this can and must operate.
Thus the final point that we disagree on is that I have seen and
like that some units also tie the Center Tap to the transformer
case and the Earth Ground without passive elements. Where as
you think that it is never tied together or only with passive
elements. . .See the next statement.
We do agree on your point that some sort of passive form of network
should be used if a connection does exist between earth ground
and the signal/power supply ground; such as the following:
Pass' Power Thermistor
Resistor
Capacitor
Diodes
Variations of said passive devices.
I think audio engineers (and electrical engineers also) have come
up with different solutions to the problems of grounding. Different
situations require different circuits to counter the problems.
What a great conversation this has been. . .I hope we can contribute
further in other threads. . .may our threads cross again.
But let me know about the two prong/three prong 220/240 VAC
question; it just seems 'cool' that many countries routinely use
such high voltage mains. . .we only use them for dryers and electric
stoves (30 Amps for the dryer circuit and 50 Amps for the electric
stove circuit). The rest of our wall sockets are three prongs but
the neutral and ground plug are tied together at the electric
meter. Some are 20 amp and the rest are 15 amp plugs.
Thank you Andrew T
Chris Browne
Northern California (1 hour north of San Francisco)
This one's for sam9
This one is for you sam9:
Thank you for waiting. . . Andrew T is a smart cookie. . . My lesson
is just because you have seen a circuit doesn't mean it will work
in every instance. . .someone designed that circuit to work a specific
way. . . don't mess with it unless you have one of those degrees
after you name: Jon Doe, PhD Electrical Science!
I had to google Doug Self. . . I think in a fully balanced audio
system. . . the chassis is earthed. . . thus the transformer case
will be earthed. . .this creates a big shield. . .if pin 1 of the balanced
cable(the shield) is connected to the chassis then it is an extension
of the 'shield' that is connected to earth.
It is going to take me some time to read his article. . . it is quite
extensive.
Maybe I will learn something new. . .probably a bunch of things.
I'll check back soon. . . after I've learned a bunch.
This one is for you sam9:
Thank you for waiting. . . Andrew T is a smart cookie. . . My lesson
is just because you have seen a circuit doesn't mean it will work
in every instance. . .someone designed that circuit to work a specific
way. . . don't mess with it unless you have one of those degrees
after you name: Jon Doe, PhD Electrical Science!
I had to google Doug Self. . . I think in a fully balanced audio
system. . . the chassis is earthed. . . thus the transformer case
will be earthed. . .this creates a big shield. . .if pin 1 of the balanced
cable(the shield) is connected to the chassis then it is an extension
of the 'shield' that is connected to earth.
It is going to take me some time to read his article. . . it is quite
extensive.
Maybe I will learn something new. . .probably a bunch of things.
I'll check back soon. . . after I've learned a bunch.
I allways make a thick ground wire of around 3mm copper in midline of (tube, pre, power) amp isolated from chassis. All Gnd signals wil be mounted on that wire, it will form a "star" then. The cabinet is connected by a 1K resistor on Gnd star. The HF rubbisch from outside on cabinet has to pass a 1K res. before its on (signal)Gnd. All connectors have to be isolated from chassis then, and a Gnd connection of an eventual mains socket must not be connected to prevent ground loops with other equipment.
As far as I've learned so far, this always works for me...
1. Common point of power supply caps are tied together with good sized copper plate.
2. Bolt through this copper plate holds varoius different audio ground returns in order of noisiest to cleanest.
3. What is cleanest and noisiest can be debated to some extent.
4. Safety earth must always be connected to chassis close to where it enters (legal requirement round here)
5. Safety earth not required for double insulated equipment.
5. Safety earth /chassis is to be connected to audio ground bolt as mentioned in point 2 via suitable network (reverse parallel diodes / resistor / cap)
For gni - I believe the UK has the same system of electrical distribution as us in that there are 3 phases spaced 120deg apart and a neutral grounted at the point of supply with 230vac from any one of the phases to neutral and 400vac between phases. Most domestic installations are single phase hence 230vac.
1. Common point of power supply caps are tied together with good sized copper plate.
2. Bolt through this copper plate holds varoius different audio ground returns in order of noisiest to cleanest.
3. What is cleanest and noisiest can be debated to some extent.
4. Safety earth must always be connected to chassis close to where it enters (legal requirement round here)
5. Safety earth not required for double insulated equipment.
5. Safety earth /chassis is to be connected to audio ground bolt as mentioned in point 2 via suitable network (reverse parallel diodes / resistor / cap)
For gni - I believe the UK has the same system of electrical distribution as us in that there are 3 phases spaced 120deg apart and a neutral grounted at the point of supply with 230vac from any one of the phases to neutral and 400vac between phases. Most domestic installations are single phase hence 230vac.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.