Tom, thank you for your explanations.
So if I understand correctly we can create a star of ground planes with the option of separate traces for heavy currents.
Another question, what's you take on where signals cross the boundary between ground planes? My thoughts are that these should be avoided if at all possible.
So if I understand correctly we can create a star of ground planes with the option of separate traces for heavy currents.
Another question, what's you take on where signals cross the boundary between ground planes? My thoughts are that these should be avoided if at all possible.
What your opinion on how this amp is grounded, and how the supply is connected.
http://www.diyaudio.com/forums/solid-state/76290-studio-350-a-3.html
http://www.diyaudio.com/forums/solid-state/76290-studio-350-a-3.html
This is why I gave up on ground planes. There's too much I don't know. Didn't feel I could implement them properly. Star grounding seemed more intuitive.
Many people of the theory background are trying to analyze currents and behaviors of this type of plane that is only and commonly used commercially in RF applications , The missing point is most of the times theory takes as a fact that the mains is clean and together with that ground is clean . As we all know this is not real ....
A few reasons not to
--A badly designed plane will introduce both capacitance and inductance, interfere with the bottom layer , and the sensitive parts like LTP VAS and so on ...let us not forget that a few picofarad in feedback or VAs will alter the performance of the all amp
--A plane will have a direct impact on the choices and stye of total grounding which is an issue that seems that theoretical people seem to not understand and / or refuse to see what we call common practice IE :
you will have two options with a given fact that small signal ground is ""clean "" but still connected to the main ground via resistor or other way .
Option 1 a mains grounded enclosure .
In this case electrical ground goes to enclosure , then to CT according to Andrew's opinion ,Via a ground lift switch in common practice, and this is also used as star ground starting point which eventually will feed the plane , then the small signal ground up to the input plug ( isolated )
With modern devices that featuring SMPS power supply with real or virtual ground most of these devices try to divert sub and filter products or garbage to ground virtual or not . In real ground these will travel through the main cables while in virtual ground ( of the smps ) will travel through signal ground .
In both of cases i don't want to have so close to my LTP such a noisy source On the contrary i would like to have my LTP in an electrically free or sealed environment from any possible noise
Option 2 a non mains grounded enclosure with two leads .
Will mean that ground of small signal will be clean but electrically connected to main ground and especially to plane and then to enclosure ( this is under question though ) ( or can be done with more than one ways ) So what will happen in real life is such a large area ( plane ) including enclosure if and how this might be connected to plane and signal ground will become a huge antenna eager and horny to receive anything you might imagine like AM FM CB GSM and intent to seek connection to a real ground in the system obviously with the RCA leads with catastrophic results .
You might get away from the first part with no connection of the ground to the enclosure which then will act as a magnetically sealed environment..but still introduce inside the plane things that might be carried in the mains ground
This is common commercial practice any machine you look in the market will be made like that and NO plane
Kind regards
Sakis
A few reasons not to
--A badly designed plane will introduce both capacitance and inductance, interfere with the bottom layer , and the sensitive parts like LTP VAS and so on ...let us not forget that a few picofarad in feedback or VAs will alter the performance of the all amp
--A plane will have a direct impact on the choices and stye of total grounding which is an issue that seems that theoretical people seem to not understand and / or refuse to see what we call common practice IE :
you will have two options with a given fact that small signal ground is ""clean "" but still connected to the main ground via resistor or other way .
Option 1 a mains grounded enclosure .
In this case electrical ground goes to enclosure , then to CT according to Andrew's opinion ,Via a ground lift switch in common practice, and this is also used as star ground starting point which eventually will feed the plane , then the small signal ground up to the input plug ( isolated )
With modern devices that featuring SMPS power supply with real or virtual ground most of these devices try to divert sub and filter products or garbage to ground virtual or not . In real ground these will travel through the main cables while in virtual ground ( of the smps ) will travel through signal ground .
In both of cases i don't want to have so close to my LTP such a noisy source On the contrary i would like to have my LTP in an electrically free or sealed environment from any possible noise
Option 2 a non mains grounded enclosure with two leads .
Will mean that ground of small signal will be clean but electrically connected to main ground and especially to plane and then to enclosure ( this is under question though ) ( or can be done with more than one ways ) So what will happen in real life is such a large area ( plane ) including enclosure if and how this might be connected to plane and signal ground will become a huge antenna eager and horny to receive anything you might imagine like AM FM CB GSM and intent to seek connection to a real ground in the system obviously with the RCA leads with catastrophic results .
You might get away from the first part with no connection of the ground to the enclosure which then will act as a magnetically sealed environment..but still introduce inside the plane things that might be carried in the mains ground
This is common commercial practice any machine you look in the market will be made like that and NO plane
Kind regards
Sakis
Last edited:
forgot to mention that a properly designed and implemented plane is far out of many diyers and often so called sophisticated designers .
Giesberts TITAN had a lot of problems with the plane which only some innovative people that actually solved the problem by braking the plane in parts .
But i think Giesbert is a bad example most of his designs suffered from a number of practical and stability issues and i have often expressed my intention to slap his face for making so many people suffer with his designs while ready to face the consequences of my act
Kind regards
Sakis
Giesberts TITAN had a lot of problems with the plane which only some innovative people that actually solved the problem by braking the plane in parts .
But i think Giesbert is a bad example most of his designs suffered from a number of practical and stability issues and i have often expressed my intention to slap his face for making so many people suffer with his designs while ready to face the consequences of my act
Kind regards
Sakis
With modern devices that featuring SMPS power supply with real or virtual ground most of these devices try to divert sub and filter products or garbage to ground virtual or not . In real ground these will travel through the main cables while in virtual ground ( of the smps ) will travel through signal ground .
To make this more clear i am referring to the peripherals that will be connected in the system
To make this more clear i am referring to the peripherals that will be connected in the system
The HF interference that is grounded/shunted to the chassis is not transferred to the mains via the mains cables.
There is simply too much impedance to send the HF by that route.
The HF interference is capacitively coupled from the chassis to "earth", that great big capacitor under your feet.
There is simply too much impedance to send the HF by that route.
The HF interference is capacitively coupled from the chassis to "earth", that great big capacitor under your feet.
Signals crossing from one ground plane to another need to be balanced ones. Otherwise, avoid.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.