Are you confirming that the increase in volume when both attenuating relays are operating in parallel is the "pop" you hear?
What about adding a transistor type mute on the output signal during the relay switch over period?
What about adding a transistor type mute on the output signal during the relay switch over period?
Hi Andrew,
since it isn´t DC I thought the pops I hear are the jump from -46.5dB to 0 db to -48dB going down and the somewhat lesser pop going from -48dB (in 5 steps) to -93.5dB (in 1 step) to -46,5dB going up. It is at the moment the only explanation I can think of.
I will try to change the steps since I don't really need -93.5dB (-75dB would be enough) and maybe make the biggest step with two relays each attenuating half the value.
The only thing that doesn´t really fit is that it also pops going up. I did a few measurements with a digital scope but could not really find something. Maybe I´ll have to take another look.
I was hoping someone had the same problem (and a sollution.......)
William
since it isn´t DC I thought the pops I hear are the jump from -46.5dB to 0 db to -48dB going down and the somewhat lesser pop going from -48dB (in 5 steps) to -93.5dB (in 1 step) to -46,5dB going up. It is at the moment the only explanation I can think of.
I will try to change the steps since I don't really need -93.5dB (-75dB would be enough) and maybe make the biggest step with two relays each attenuating half the value.
The only thing that doesn´t really fit is that it also pops going up. I did a few measurements with a digital scope but could not really find something. Maybe I´ll have to take another look.
I was hoping someone had the same problem (and a sollution.......)
William
I don't think a "pop" would result from an excessive mute as you have just described.
And I obviously misunderstood since you are NOT confirming what I said.
But, I still think the "pop" could be due to excessive output.
Could that excessive output be happening during the relay change overs? There must be a time when both contacts are "open". What is the output when a relay is in this mid state?
And I obviously misunderstood since you are NOT confirming what I said.
But, I still think the "pop" could be due to excessive output.
Could that excessive output be happening during the relay change overs? There must be a time when both contacts are "open". What is the output when a relay is in this mid state?
Hi Andrew,
there are 6 relays. The pops only occur when changing from 011111 to 100000 or visa versa. Maybe it´s the 5V power supply or the earth connection which isn´t up to it when 6 relays are changing position in very short time. I will have some time to look at it next week.
(there are actually 12 Relais switching as it is a balanced attenuator)
william
there are 6 relays. The pops only occur when changing from 011111 to 100000 or visa versa. Maybe it´s the 5V power supply or the earth connection which isn´t up to it when 6 relays are changing position in very short time. I will have some time to look at it next week.
(there are actually 12 Relais switching as it is a balanced attenuator)
william
Last edited:
Are there any news about this issue?
What is the meaning exactly, which is this position when 5 relays on?
What is the meaning exactly, which is this position when 5 relays on?
Hi Lemon,
haven´t found the time to have a look.
5 relays switch on or off going from 100000 (1 relay on 5 off -48dB) to 011111 (5 relays on, one off -46,5dB) or visa versa.
William
haven´t found the time to have a look.
5 relays switch on or off going from 100000 (1 relay on 5 off -48dB) to 011111 (5 relays on, one off -46,5dB) or visa versa.
William
Thanks for the reply.
I have find another similar thread in this forum. It is here.
According to information that presented there, the issue summarized to:
1. Routing ground on PCB in a right way.
2. Recalc all Res values to make Rin and Rout lower down to 600 Ohm (for example)
3. Use RC circuit for each relay (for coil contacts)
4. Make a delay for relais switching
5. Avoid capacitive load
My concern is what is the meaning with (2).
My system is dac->R2R->DCB1->Final Amplifier.
The dac output is 23 Ohm, the output of R2R almost 10K. I tried to removed the DCB1 buffer and feed direct the R2R to Final Amplifier (47K input) but the same.
I have find another similar thread in this forum. It is here.
According to information that presented there, the issue summarized to:
1. Routing ground on PCB in a right way.
2. Recalc all Res values to make Rin and Rout lower down to 600 Ohm (for example)
3. Use RC circuit for each relay (for coil contacts)
4. Make a delay for relais switching
5. Avoid capacitive load
My concern is what is the meaning with (2).
My system is dac->R2R->DCB1->Final Amplifier.
The dac output is 23 Ohm, the output of R2R almost 10K. I tried to removed the DCB1 buffer and feed direct the R2R to Final Amplifier (47K input) but the same.
Hi Lemon,
interesting that they have the same problem (going from 100000 to 011111) with a completely different setup.
william
interesting that they have the same problem (going from 100000 to 011111) with a completely different setup.
william
Hi,
tried beefing up the earth connection from the 5V power supply for the relays. This had no influence (as expected but it was easy to do).
Next I will change the control from 6bit to 5bit thus using two relays for the biggest volume step. So instead of -1.5,-3,-6,-12,-24,-48 the steps become something like -3,-6,-12,-24,-24,-24 where the two last steps are alway on or off at the same time.
By switching them not directly after another but 2 other steps in between I hope to soften the impact a bit.
At the same time I will make the maximum attenuation smaller as -93 is too much, something like -75 should fit well giving ca. 2.5dB steps x 32.
So I only have to calculate the new values and desolder 48 resistors from a double sided board
William
tried beefing up the earth connection from the 5V power supply for the relays. This had no influence (as expected but it was easy to do).
Next I will change the control from 6bit to 5bit thus using two relays for the biggest volume step. So instead of -1.5,-3,-6,-12,-24,-48 the steps become something like -3,-6,-12,-24,-24,-24 where the two last steps are alway on or off at the same time.
By switching them not directly after another but 2 other steps in between I hope to soften the impact a bit.
At the same time I will make the maximum attenuation smaller as -93 is too much, something like -75 should fit well giving ca. 2.5dB steps x 32.
So I only have to calculate the new values and desolder 48 resistors from a double sided board
William
- Status
- Not open for further replies.