ESP's double shorted bridge rectifier is copied from the posts on this Forum that predated his work/article.
The single shorted bridge rectifier inserts two Vf drops into the link.
The double shorted bridge rectifier insets a parallel pair of single Vf drops into the link.
I use the double shorted version, because it has a lower Vdrop AND has a doubled Fault Current capability.
It's the kA Fault Current that MUST be taken back to Protective Earth (PE).
Last edited:
As a general guideline, don't skimp on the ground lifter rectifier power.
I use a 25A continuous rated rectifier or a 35A chassis mount type. In both cases the peak (half wave) rectified current is hundreds of amps, and for a few cycles high tens of amps - so enough to blow a fuse or trip the MCB.
I think either works well - but I agree the double shorted version does have the added advantage of the higher peak carrying current capability.
I use a 25A continuous rated rectifier or a 35A chassis mount type. In both cases the peak (half wave) rectified current is hundreds of amps, and for a few cycles high tens of amps - so enough to blow a fuse or trip the MCB.
I think either works well - but I agree the double shorted version does have the added advantage of the higher peak carrying current capability.
http://hifisonix.com/wordpress/wp-c.../How-to-wire-up-a-Power-Amplifier_Updated.pdf
is that 15R decoupler on the power amps wired correctly?
is that 15R decoupler on the power amps wired correctly?
I have, big non isolated switched mode supply I built with a three phase thyristor bridge and 10mH 40A choke as a prereg, fed with 400V delta (Entertainment lighting application).
Turns out the board had a bit of badly thought out clearance when the igniter pulse (~60kV @ 100KHz) was triggered which started the arc, and DC arcs are reluctant to go out....
It took all three 40A fuses in the supply and made one hell of a mess inside the case, surprisingly the thyristor bridge survived intact and is still in use in the rebuild.
Turns out the RCD as not sensitive to pulsating DC faults.....
Note, non isolated supplies are outside the scope of this forum, this was a very specific application and appropriate precautions were taken (Especially as both sides of the output are in contact with the lamp/cavity DI water cooling loop).
Turns out the board had a bit of badly thought out clearance when the igniter pulse (~60kV @ 100KHz) was triggered which started the arc, and DC arcs are reluctant to go out....
It took all three 40A fuses in the supply and made one hell of a mess inside the case, surprisingly the thyristor bridge survived intact and is still in use in the rebuild.
Turns out the RCD as not sensitive to pulsating DC faults.....
Note, non isolated supplies are outside the scope of this forum, this was a very specific application and appropriate precautions were taken (Especially as both sides of the output are in contact with the lamp/cavity DI water cooling loop).
Thank you so much for posting the updated document, and answering my question so clearly!
Geary
Thank you for your explanation
Andrew:
Thank you for your reply, and let me say that i appreciate all the straightforward answers.
Geary
Andrew:
Thank you for your reply, and let me say that i appreciate all the straightforward answers.
Geary
Here's the explanation:-
A problem in many systems, even if the chassis' are earthed via the mains plug, is that ground loops are formed between equipment through the interconnecting cable returns. Since the resistance (and at HF, inductance) of the interconnect grounds is of the same order as the safety ground connections between the equipment (few tens of mOhm at DC), loop currents are quite high and will simply flow where the resistance is lowest - often along the interconnect return connection = noise. The decouple resistor forces any ground loop currents to flow through the chassis earths and not the signal interconnect.
This solution should not be used to plaster over bad internal wiring practices.
How to test for the case described above: if your power amp is absolutely silent when powered up but with nothing plugged in, but then becomes noisy (i.e hum) when you plug your source in (powered up but muted), it's likely you have an earth loop formed through the interconnect.
Also note: check that your mains live-neutral is correctly phased on your transformer. Further, I always try to plug the amp and the source into the same outlet - I use a strip plug. This ensures the safety ground loop is small and also helps.
As a general rule, the signal return (aka signal ground) interconnect should never be used to provide a safety ground between equipment. Use a proper safety ground from your chassis to the mains plug.
My system is very quiet - all power amps, two preamps and Oppo and Pioneer SACD players.
Last edited:
I think the problem is worse than what Bonsai is describing. A pre amp and a power amp, each with their own mains earth connection. This has created three ground connections between the two units (two interconnects and the mains ground). All these connections are low impedance and so there is nothing to stop the return audio current from splitting into three with some returning through the other channels interconnect and some returning through the mains cable.
We don't only have ground currents flowing through the interconnects, we also have audio current flowing through the mains cables!!!!
We don't only have ground currents flowing through the interconnects, we also have audio current flowing through the mains cables!!!!
BUT why connect ground signal to the main ground if your transformer are break and have a resistance (different from zero ,i think is a real case) than interconnect main power to signal ground and on the amp ground connection is break .
When you tray to disconnect an RCA its becomes so dangerous for life
When you tray to disconnect an RCA its becomes so dangerous for life
I don't think that the ground lift with a bridge rectifier between circuit ground and metal housing is a good idea.
Think what may happen if there is a second fault (e.g.bridged main fuse or production fault of the not-for-safety-build bridge rectifier)...
If the bridge rectifier is open-circuit: The 230 V goes over a non-isolated RCA connector to the connected CD-Player..
A lot of amps flow down the RCA cable and produce an interesting smell...
A solid wire connection between signal ground and metal housing is the way to go in a Class-I scheme. This is a long proven and accepted system.
A solid wire connection works even without a main fuse as the PE resistance is so low that the voltage on chassis stays below 50 Volts.
Deviation from this scheme need a lot of testing and expertise.
It is mutch easier to build a Class-II system than to do the required testing.
The relevant standard clearly says that safety relevant parts need approval (VDE or CE marks).
Electrical installations are single fault proove and the second fault should do no harm.
In your house you may have control of the installation and you are free to do what you want, but don't assume too mutch about other installations.
It is NOT necessary for a correct gounding scheme to compromize safety. Many sucessful companies have proven this for many years.
In small installations where amplifier, Cd-player and other parts are connected to one circuit outlet different ground voltages and ground loops are not a big problem. Otherwise go balanced (XLR).
Think what may happen if there is a second fault (e.g.bridged main fuse or production fault of the not-for-safety-build bridge rectifier)...
If the bridge rectifier is open-circuit: The 230 V goes over a non-isolated RCA connector to the connected CD-Player..
A lot of amps flow down the RCA cable and produce an interesting smell...
A solid wire connection between signal ground and metal housing is the way to go in a Class-I scheme. This is a long proven and accepted system.
A solid wire connection works even without a main fuse as the PE resistance is so low that the voltage on chassis stays below 50 Volts.
Deviation from this scheme need a lot of testing and expertise.
It is mutch easier to build a Class-II system than to do the required testing.
The relevant standard clearly says that safety relevant parts need approval (VDE or CE marks).
Electrical installations are single fault proove and the second fault should do no harm.
In your house you may have control of the installation and you are free to do what you want, but don't assume too mutch about other installations.
It is NOT necessary for a correct gounding scheme to compromize safety. Many sucessful companies have proven this for many years.
In small installations where amplifier, Cd-player and other parts are connected to one circuit outlet different ground voltages and ground loops are not a big problem. Otherwise go balanced (XLR).
Have you tested the Disconnecting Network?
I have tested it direct to the Mains LIVE to PE.
It works.
Nonsense.
Most of the problems we have reported on this Forum are either broken amplifiers or interference loops.
Inadvertent LOOPs are a big problem!
Last edited:
All our mains powered equipment is either ClassI or ClassII.
We cannot design, build, test and guarantee ClassII projects.
We MUST comply with ClassI standards.
The two most important are:
1.) The Chassis must be permanently connected to Protective Earth (PE).
2.) All exposed conductive parts should be connected to the protected Chassis.
If you built an active speaker with ALL the electronics inside the insulated speaker box, then the second requirement is met without needing to connect any conductive parts to the protected chassis.
But you still have to meet requirement 1.) PE to Chassis.
tranformes break
The Law is clear .
But for ideal system I think is to havy voltage stress for transformer with some output connected to the Protective Earth .
On ideal for me signal ground is to safe as completely insulated and only metallic chassis of audio chain one by one connected to the Protective Earth (PE).
Currents cant' flow from arm to arm for any fault topologies
The Law is clear .
But for ideal system I think is to havy voltage stress for transformer with some output connected to the Protective Earth .
On ideal for me signal ground is to safe as completely insulated and only metallic chassis of audio chain one by one connected to the Protective Earth (PE).
Currents cant' flow from arm to arm for any fault topologies
I have tested it, and it does not work in the SECOND FAULT case (no fuse, or defect bridge rectifier). The automatic Secutest system shows a fault too. Inadvertent LOOPs are not a big problem TODAY, as almost any CD-Player or other sources are CLASS-II ->> so no Ground LOOP problem! Better to be safe than to reduce the noise floor by a few MicroVolts, nobody can hear that anyway.
Absolutely no one here is proposing that the equipment chassis is not earthed.
The ground lifter works - just use a decently rated rectifier (I use a 25A coninuous device or a 35A chassis moint type). Any short of mains either through the transformer to the secondary, or from the mains to the chassis will blow the fuse and/or trip the MCB. Also, please note that in many countries, no mains wiring may be exposed inside equipement - this is rthe rule I generally follow in my build as well - I am a big user of heatshrink for exactly that reason.
In my view, relying in any way for a safety earth through an interconnect is extremely dangerous.
If equipment is not fused, that is illegal in many countries. If consumers bridge out a fuse then they are breaking the law and risking their lives - either way, there is little a supplier can do to stop them from being stupid.
Last edited:
Udok.
EVERY two channel (analogue) source creates a loop when connected to a two channel receiver.
It does not matter that they are ClassI, or ClassII, or a mixture of both.
and it applies to balanced interconnections as well. The difference with balanced impedance interconnections is that the wiring is designed to keep the LOOP out of the audio signal.
EVERY two channel (analogue) source creates a loop when connected to a two channel receiver.
It does not matter that they are ClassI, or ClassII, or a mixture of both.
and it applies to balanced interconnections as well. The difference with balanced impedance interconnections is that the wiring is designed to keep the LOOP out of the audio signal.
What?
You are saying that the Disconnecting Network FAILS when the mains fuse is absent/removed !
You must be talking about something else !!!!!!!
- Status
- Not open for further replies.