Hi everybody
First I'm french so don't blame me for my poor and bad english. I just hope I'll make myself understood.
My problem, as written in the title, is a leaking current in my solid state stereo ampli. It's an Audio Research D100 (vintage yeah !). It work pretty well with my CD but (or even BUT), this leaking burned 2 of the 3 numerical input of my préamp (proton AS2640). As a consequence I stopped using it as the stereo ampli of my home theater set (the other 3 channels were amplified by a 6 ways (bridged in 3*2) Proton AA2480). Now I must use my 6 ways proton even for the stereo ...
So I have to find out where the pb comes from. Is it a "simple" way to rapidly focus on the pb ? Is that kind of pb often linked with capacitances ? transistors ? other ?
Thank you all in advance
JBL, hoping to hear his D100 soon...
First I'm french so don't blame me for my poor and bad english. I just hope I'll make myself understood.
My problem, as written in the title, is a leaking current in my solid state stereo ampli. It's an Audio Research D100 (vintage yeah !). It work pretty well with my CD but (or even BUT), this leaking burned 2 of the 3 numerical input of my préamp (proton AS2640). As a consequence I stopped using it as the stereo ampli of my home theater set (the other 3 channels were amplified by a 6 ways (bridged in 3*2) Proton AA2480). Now I must use my 6 ways proton even for the stereo ...
So I have to find out where the pb comes from. Is it a "simple" way to rapidly focus on the pb ? Is that kind of pb often linked with capacitances ? transistors ? other ?
Thank you all in advance
JBL, hoping to hear his D100 soon...
I never connect an analog device to a numerical input. Here is the connection I made :
PCHC -> pré via rca numerical cable
Pré -> amp 2*RCA
Pré -> CD 2*rca
amp -> loud via usual cables.
I think I did it right, didn't I ?
Damages happened when I unplugged the PCHC to plug the CD in with a numerical connection. Then I think I saw a little (tiny) spark at the numerical wire hot spot at the rear of the pré ... Then the num input was out .....
This phenomenon made me think of a leaking current flowing from my amp ... wrong ? It is very low, but if you plug the amp alone (without any connection to anything) WITHOUT linking it to the ground, then you feel sometimes a little current under your fingers (hen touching it of course ...)
PCHC -> pré via rca numerical cable
Pré -> amp 2*RCA
Pré -> CD 2*rca
amp -> loud via usual cables.
I think I did it right, didn't I ?
Damages happened when I unplugged the PCHC to plug the CD in with a numerical connection. Then I think I saw a little (tiny) spark at the numerical wire hot spot at the rear of the pré ... Then the num input was out .....
This phenomenon made me think of a leaking current flowing from my amp ... wrong ? It is very low, but if you plug the amp alone (without any connection to anything) WITHOUT linking it to the ground, then you feel sometimes a little current under your fingers (hen touching it of course ...)
Hi,
you have made the problem by changing plugs with your equipment still live.
It is always good practice to make changes ONLY when the equipment is switched off and allow time for all the circuits to discharge.
The DC blocker caps at the output of the source and/or preamp and at the input of power amps often do not have a resistor to ground. A charge can build up on the RCA hot connection and when you plug it into a piece of live equipment that charge can damage down stream equipment.
I wonder if this is the spark you saw?
Why are you running the pre into the CD?
you have made the problem by changing plugs with your equipment still live.
It is always good practice to make changes ONLY when the equipment is switched off and allow time for all the circuits to discharge.
The DC blocker caps at the output of the source and/or preamp and at the input of power amps often do not have a resistor to ground. A charge can build up on the RCA hot connection and when you plug it into a piece of live equipment that charge can damage down stream equipment.
I wonder if this is the spark you saw?
Why are you running the pre into the CD?
it should be CD -> pre (2RCA) -> power amp (2RCA) -> Speakers
Earth Leakage currents are quite normal (but not desirable) – you will measure half the AC supply voltage from say the chassis of the amplifier and your SPDIF RCA lead.
The problem with the standard digital (SPDIF) lead is that the centre pin (Signal) gets connected first (before the outer ground connection - that’s why with Audio Signal leads you always hear a loud “Buzz” when connecting or removing), sending the leakage current straight into the sensitive logic gates – 110V even at mA’s current is terminal to voltage sensitive logic… Note – BNC connections do not have this problem because the Ground is always connected first.
What DAC do you use?
The solution is to use optical leads, or an SDPIF Isolation transformer, or add current limiting resistors & protection diodes to the digital inputs – not easy without having a circuit diagram as the input stage will have to be modified to maintain the 75 Ohm environment.
SPDIF Isolation transformers will also improve audio quality, as the 50Hz / 100Hz Earth leakage current modulate the delicate SPDIF signal with heavy harmonics of the mains frequency - these LF products are well outside the Jitter suppression BW of most SPDIF receivers PLL circuits, which start to attenuate Jitter from 1MHz or so (CS8412 etc).
John
The problem with the standard digital (SPDIF) lead is that the centre pin (Signal) gets connected first (before the outer ground connection - that’s why with Audio Signal leads you always hear a loud “Buzz” when connecting or removing), sending the leakage current straight into the sensitive logic gates – 110V even at mA’s current is terminal to voltage sensitive logic… Note – BNC connections do not have this problem because the Ground is always connected first.
What DAC do you use?
The solution is to use optical leads, or an SDPIF Isolation transformer, or add current limiting resistors & protection diodes to the digital inputs – not easy without having a circuit diagram as the input stage will have to be modified to maintain the 75 Ohm environment.
SPDIF Isolation transformers will also improve audio quality, as the 50Hz / 100Hz Earth leakage current modulate the delicate SPDIF signal with heavy harmonics of the mains frequency - these LF products are well outside the Jitter suppression BW of most SPDIF receivers PLL circuits, which start to attenuate Jitter from 1MHz or so (CS8412 etc).
John
Hi, and thanks for those first replies.
For Peranders, PCHC is a anglo french acronym for HTPC (Home Theater PC)
For AndrewT : Yes I plug/unplugged the devices without switching them off ...
... I hope I'll learn from my own mistakes.
Of course the CD is connected to the pre and not the opposite...
When I plug the amp only (without ground connection), there is as 20 to 30 mV tension between chassis and ground. Is is significant enough ?
As a consequence, I'm looking for
1) a way to have a good connection between elements and ground but which also avoids 50 Hz noise (=no mass loop)...
2) Or a way to repair my amp ...
JB
For Peranders, PCHC is a anglo french acronym for HTPC (Home Theater PC)
For AndrewT : Yes I plug/unplugged the devices without switching them off ...
... I hope I'll learn from my own mistakes.Of course the CD is connected to the pre and not the opposite...
When I plug the amp only (without ground connection), there is as 20 to 30 mV tension between chassis and ground. Is is significant enough ?
As a consequence, I'm looking for
1) a way to have a good connection between elements and ground but which also avoids 50 Hz noise (=no mass loop)...
2) Or a way to repair my amp ...
JB
Be carefull, as I expect you will find 110Vac
IMHO the problem is the HTPC. Most of your installation has no connection between it's grond and the earthing of the mains - EXCEPT the HTPC. For some stupid reason, PC grond = mains earth, which is not always a good thing, because mains earth may be connected to mains neutral at the distribution box, rather than being a true earth. After that, all it takes is one reversed mains plug and you have a problem.
To make things worse, I have yet to see a HTPC, even with an expensive audio card, that has a SPDIF isolation transformer. True, this is not required in consumer SPDIF but then you must be very careful with ground vs earth, with a transformer, this is not an issue at all. Alternatively, you could use optical SPDIF, though that tends to have worse jitter problems (this may or may not be a problem depending on actual DA conversion components in use).
Also, beware of secondarty connections - for isnatnce, your monitor may be connected to HTPC and to some sort of audio output on the rest of the installation, that again connects ground and 'earth', since VGA and DVI ground on the HTPC is again connected to mains earth on the HTPC.
The problem you describe is caused by establishing a connection between a grounded and non-grounded device. Non grounded devices often have a 'fake earth' created by a capacitive divider across live and neutral, with two equal caps. At any point, except ehrn the mains AC passes through zero, both are charged to about half the mains voltage. The grond on connectors on this sort of devices are usually conencted to this fake ground - normally it should be through a resistor, but very often there is a just a wire! When you connect this to a real earth (like through a HTPC), earth is very near the potential of neutral. So, suddenly, one of the caps in the capacitive divider is discharged through the connection, which usually also casues a similar voltage difference between the signal wires - and your input (and sometimes output) blows. Even if not, there will be an AC current resulting from the remaining cap in the capacitive divider being charged and discharged through the ground wire of your interconnect cable, so you get hum.
To make things worse, I have yet to see a HTPC, even with an expensive audio card, that has a SPDIF isolation transformer. True, this is not required in consumer SPDIF but then you must be very careful with ground vs earth, with a transformer, this is not an issue at all. Alternatively, you could use optical SPDIF, though that tends to have worse jitter problems (this may or may not be a problem depending on actual DA conversion components in use).
Also, beware of secondarty connections - for isnatnce, your monitor may be connected to HTPC and to some sort of audio output on the rest of the installation, that again connects ground and 'earth', since VGA and DVI ground on the HTPC is again connected to mains earth on the HTPC.
The problem you describe is caused by establishing a connection between a grounded and non-grounded device. Non grounded devices often have a 'fake earth' created by a capacitive divider across live and neutral, with two equal caps. At any point, except ehrn the mains AC passes through zero, both are charged to about half the mains voltage. The grond on connectors on this sort of devices are usually conencted to this fake ground - normally it should be through a resistor, but very often there is a just a wire! When you connect this to a real earth (like through a HTPC), earth is very near the potential of neutral. So, suddenly, one of the caps in the capacitive divider is discharged through the connection, which usually also casues a similar voltage difference between the signal wires - and your input (and sometimes output) blows. Even if not, there will be an AC current resulting from the remaining cap in the capacitive divider being charged and discharged through the ground wire of your interconnect cable, so you get hum.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.