Hi there,
This agrees well with my experiences, line level switching via relays usually does not cause problems, correct installation (back diode etc.) and choice of relay provided.
Here in Germany, most PCB-style small signal relays come without an integrated back diode, thats's why I pointed out its importance. Industrial relays usually have one built in, but they are uncomfortably large and not all of them are suitable for switching small audio signals.
Also agreed. But I would never try to relay-switch the high-gain input of a phono preamp stage - why not leave the cartridge permanently connected to it and do the switching afterwards when the signal is up to line level?
Very good practice - but I would short it after a 1k series resistor or something like that to protect the output stage of the sources against the short.
I understand - but, with regard to the original question, you cured this problem by a FET "mute switch" on the signal side, which is a very good idea. You don't mentioned trying to solve the "pop" problem by using a different or high-sophisticated power supply for the relay.
Please don't misunderstand me: I never wanted to give the impression that I do not care about switching "pops" from relays, if they occur, they have to be eliminated. For aesthetic reasons as well as to protect speakers and amplifier circuitry.
I just would not agree that feeding the relay circuit from the amplifier power supply will cause problems. If it pops, the problem occurs and has to be solved in the signal path, not on the supply side. Even if the relay coil causes interference that is not fully shorted by back diode and capacitor, this will lead to a short disturbance of the amplifier B+ rail, which possibly affects the amplifier slightly. But this disturbance will occur in the very instant when the amplifier has no signal to deliver. If this small disturbance on the B+ becomes more audible than the "pop" of the relay contacts switching the signal input, something has gone seriously wrong in the design.
Best regards,
Andreas
PS. As to the FET idea you mentioned: Most hardcore "audiophiles" will not like a semiconductor device permanently connected between signal input and ground. A mechanical switch breaks and makes the contact abruptly and might "pop" as well. Better ideas welcome!
This agrees well with my experiences, line level switching via relays usually does not cause problems, correct installation (back diode etc.) and choice of relay provided.
Here in Germany, most PCB-style small signal relays come without an integrated back diode, thats's why I pointed out its importance. Industrial relays usually have one built in, but they are uncomfortably large and not all of them are suitable for switching small audio signals.
Also agreed. But I would never try to relay-switch the high-gain input of a phono preamp stage - why not leave the cartridge permanently connected to it and do the switching afterwards when the signal is up to line level?
Very good practice - but I would short it after a 1k series resistor or something like that to protect the output stage of the sources against the short.
I understand - but, with regard to the original question, you cured this problem by a FET "mute switch" on the signal side, which is a very good idea. You don't mentioned trying to solve the "pop" problem by using a different or high-sophisticated power supply for the relay.
Please don't misunderstand me: I never wanted to give the impression that I do not care about switching "pops" from relays, if they occur, they have to be eliminated. For aesthetic reasons as well as to protect speakers and amplifier circuitry.
I just would not agree that feeding the relay circuit from the amplifier power supply will cause problems. If it pops, the problem occurs and has to be solved in the signal path, not on the supply side. Even if the relay coil causes interference that is not fully shorted by back diode and capacitor, this will lead to a short disturbance of the amplifier B+ rail, which possibly affects the amplifier slightly. But this disturbance will occur in the very instant when the amplifier has no signal to deliver. If this small disturbance on the B+ becomes more audible than the "pop" of the relay contacts switching the signal input, something has gone seriously wrong in the design.
Best regards,
Andreas
PS. As to the FET idea you mentioned: Most hardcore "audiophiles" will not like a semiconductor device permanently connected between signal input and ground. A mechanical switch breaks and makes the contact abruptly and might "pop" as well. Better ideas welcome!
Last edited:
The question is a sign of lacking knowledge, I'm afraid. You would care if you planned to build something like this. You must be very careful with switching transients and one way to reduce them is to switch slowly.... let me ask a n evil, evil, really evil question:
Who the **** cares about switching noise "coupling into audio circuits" from a relay that switches only to select a different source?
The answer is a sign of lacking reading, one could reply 😉. See long post directly above.
Built several relay-based input switches already, they are correctly protected against switching transients and perform well.
100% agreed, concerning the signal side. Slow switching would be preferable, but is somewhat difficult with relays that throw their contacts instantly. Solid-state switching could be designed soft and slowly, but this is often not appreciated because of influences on the switched signal...
The original question of this thread was about noise or interference caused by powering the relays from the amplifier supply - not about switching transients in the signal path.
Best regards,
Andreas
Last edited:
Quote:"Last edited by Rundmaus; 4th February 2011 at 11:17 PM. Reason: Inserted a smiley. This discussion is meant to be constructive, not emotional or offending..."
Yes, this thread become scaary.. and Mrs. Ayumi disappeared. Together with the remote volume news.
Actually, relay will only affect supply and some very sensitive magnetic sensors. If there pop noise at signal path, the things that need to be secured is the part that noise showing out, avoids ultra small signal transmissions with buffer to amplify it to a safer region, is sometime needed.
Yes, this thread become scaary.. and Mrs. Ayumi disappeared. Together with the remote volume news.
Actually, relay will only affect supply and some very sensitive magnetic sensors. If there pop noise at signal path, the things that need to be secured is the part that noise showing out, avoids ultra small signal transmissions with buffer to amplify it to a safer region, is sometime needed.
connect a half wave rectifier (two diodes) into the AC of the transformer secondary.
The smoothing cap should end up with ~ 20Vdc across it. Stay with the 12V relays.
Add a resistor between the relay and the control circuit to reduce the relay ON voltage to 12V across the coil.
By taking the relay early in the supply, any electrical spikes have to travel back through the relay PSU into the amplifier PSU to get to the amplifier. They will be well attenuated.
If you full bridge rectify the secondary AC then you could use 24V relays and these draw roughly 25% of the current of a 12V relay of the same type.
The smoothing cap should end up with ~ 20Vdc across it. Stay with the 12V relays.
Add a resistor between the relay and the control circuit to reduce the relay ON voltage to 12V across the coil.
By taking the relay early in the supply, any electrical spikes have to travel back through the relay PSU into the amplifier PSU to get to the amplifier. They will be well attenuated.
If you full bridge rectify the secondary AC then you could use 24V relays and these draw roughly 25% of the current of a 12V relay of the same type.
He he he he....
Sorry then, Ayumi.
Andrew's solutions are very good. Relay draw higher current only when start to turn on, the resistor will make it better. Also add a proper(enough) value capacitor after the resistor that help to supply high current demand when switched on. Not too high, because it ~20V, just to supply when switched on, may be 100nF to 500nF.
Um, maybe Suply-R-C-R-Relay configuration is better.
Last edited:
I had this problem many decades ago with a microprocessed appliance driving relays... the solution was the same above. Follow the advice about isolating the relay supply.
Going back to the first message schematic, I think that a separate bridge rectifier for U24 could be sufficient, as Nico Ras has stated on post #7.
Best regards.
Last edited:
- Status
- Not open for further replies.