I think you're on the right track but I suspect the 5532 is oscillating because of the capacitive loading of the SSM chip. A series 47 ohm resistor should decouple it if it is indeed oscillating. With the SSM out and just a switch it may settle down the 5532 (if it's oscillating) and give a false result. Best test is a scope.
Personally I don't like high value resistors in solid state amps. I'd use 47K max on the input of the 5532 and 10K instead of 470K at the output of the 5532. Higher values just encourage Johnson noise and who needs that? Current is cheap and even wimpy opamps deliver more than tube amps. The 5532 can deliver plenty of current to support a 10K load.
G²
It does seem from all you say that the problem could be down to the logic and you are hearing some kind of breakthrough.
Have you tried disconnecting the logic and manually driving the control pin between 0 and 5 volts with a switch ?
Is is possible that the logic circuitry is drawing a current pulse as it changes state and that this is contaminating the audio grounds ? Signal grounds and "digital" grounds should normally be separate.
Have you tried disconnecting the logic and manually driving the control pin between 0 and 5 volts with a switch ?
Is is possible that the logic circuitry is drawing a current pulse as it changes state and that this is contaminating the audio grounds ? Signal grounds and "digital" grounds should normally be separate.
OK Aucosticraft, I'll connect up the capacitor as you suggest and let you know what happens. I have tried Conrad Hoffman's suggestion of 'softening' the control signal but it hasn't had much effect. I'll get back with the results soon. And yes, I do have a few CD4016 chips in my collection.
Mooly you have raised a good point. I have wired everything together using a common ground; perhaps this is the cause of the problem? There again, how do I create two independant grounds when both circuits are wired to the same battery pack? I haven't tried switching the analogue switch with an indepentant control signal but it's a good idea. I'll try it out and get back to you. I'll be busy this afternoon but I'll be back at the project this evening.
Hi Sratus46, thanks for your input. I can replace the resistors on the splitter circuit if you feel that they are too high; I don't want the circuit to drain too much power though because it uses on-board batteries and needs to be fairly efficient. I'm not building a stratocaster but the instrument in the avotar is an electric solid-body hurdy-gurdy complet with humbucker pick-ups. The output from the melody strings is pulsed by the speed of the crank and comes out stacatto. However, I still have that click in the circuit to get rid of!
Vielle568
Vielle568
Mooly you have raised a good point. I have wired everything together using a common ground; perhaps this is the cause of the problem? There again, how do I create two independant grounds when both circuits are wired to the same battery pack? I haven't tried switching the analogue switch with an indepentant control signal but it's a good idea. I'll try it out and get back to you. I'll be busy this afternoon but I'll be back at the project this evening.
Hi Sratus46, thanks for your input. I can replace the resistors on the splitter circuit if you feel that they are too high; I don't want the circuit to drain too much power though because it uses on-board batteries and needs to be fairly efficient. I'm not building a stratocaster but the instrument in the avotar is an electric solid-body hurdy-gurdy complet with humbucker pick-ups. The output from the melody strings is pulsed by the speed of the crank and comes out stacatto. However, I still have that click in the circuit to get rid of!
Vielle568
Vielle568
Last edited:
The grounds are kept separate just by wiring them back individually to the battery. It's similar to wiring an audio amp and keeping the audio out of the PSU grounds where large circulating "hum" currents flow. You have to think in terms of each wire and conductor as being able to develop a volt drop in response to current flow and ask yourself whether that is going to affect or modulate the signal in any way.
I wish the cap resi arrangement should solve the problem. but in case it dosent work and you need to work on CD4016, then here is the suggested connection.
the drawing is made quick using paint brush. it looks dirty. but this moment here had no alternative.
Attachments
I've been thinking over what you told me Mooly about having separate grounds for the analogue and digital circuits. As I mentioned before, I'd mixed up the audio circuits and the digital circuits using the same power source and ground.
To trigger the analogue switch there's a microprocessor that outputs a series of pulses. These are cleaned up through Schmitt inverting buffers (74HC14s) and fed to the analogue switch. These buffers output either zero or five volts. I tried switching the analogue switch using a seperate power source as you described Mooly but I ran into a slight problem; the switch needed a control voltage that was either ground or something above 2.4V; the pin unconnected allowed the switch to connect.
Being unsure if the zero volt output of the Schmitt inverting buffers were the same as ground I replaced the buffer chip with a 4016 configured as a SPDT switching between ground and 3 volts, and I connected this as the control signal to the analogue switch for the audio signal. The click has now diminished considerably, and it's only audible at low signal levels; it did not increase when the audio signal level was increased an so I can only assume that the noise was associated with the control signal and not the audio signal.
I'll try your idea Aucosticraft, it could remove the click completely. It won't take long to add a resistor and see what happens.
Vielle568
To trigger the analogue switch there's a microprocessor that outputs a series of pulses. These are cleaned up through Schmitt inverting buffers (74HC14s) and fed to the analogue switch. These buffers output either zero or five volts. I tried switching the analogue switch using a seperate power source as you described Mooly but I ran into a slight problem; the switch needed a control voltage that was either ground or something above 2.4V; the pin unconnected allowed the switch to connect.
Being unsure if the zero volt output of the Schmitt inverting buffers were the same as ground I replaced the buffer chip with a 4016 configured as a SPDT switching between ground and 3 volts, and I connected this as the control signal to the analogue switch for the audio signal. The click has now diminished considerably, and it's only audible at low signal levels; it did not increase when the audio signal level was increased an so I can only assume that the noise was associated with the control signal and not the audio signal.
I'll try your idea Aucosticraft, it could remove the click completely. It won't take long to add a resistor and see what happens.
Vielle568
Last edited:
Assuming the switch has CMOS type input you could tie the control input pin to zero volts via a resistor of say 100K. That's normal practice and would define logic zero for the pin. Then just apply 5 volts via the switch to turn on.
Thanks for your idea Aucosticraft but it didn't work on my system. The signal being switched by the 4016 is only around 3V and the 10K resistor had the effect of shorting it to ground. With the resistor added to the circuit the second analogue switch being used to pulse the audio signals didn't have a control signal powerful enough to trigger.
Vielle568
Vielle568
I am just confirming if you have done what i suggested ( if i was able to tell you correctly )
option 1.
Keep your old arrangement with 74HCT14 & SSM2412 untouched. Then connect resistor as suggested( resi can be between 1E to 10E), in parallel to the capacitor at the output of NE5532 and input of SSM2412. and check if the click persists.
option 2.
keep your 74HCT14 in place. and replace the SSM2412 with the CD4016 ckt shown in above diagram. and check for click noise.
New option 3.
What is the DC voltage at output of NE5532. If you find DC voltage there, Add 47K in parallel to 1M resistor at the input of Ne5532. and check if the DC voltage at the output pin becomes zero. If yes, then replace input capacitor of 47nf to 47mfd. and see if the noise has disappeared.
Last edited:
OK, a piece of cake (when you know how).
I switched the 2412 using an independant control signal. The results were as follows: First pulse produced a mild click. Second pulse produced a much weaker click. Third and fourth pulses reduced click to an (almost) inaudible level. Subsequent clicks remained at this low level. Increasing audio signal level did not increase the volume of the click, in fact, the clicks were only barely audible at the lowest audio signal levels. At higher levels the audio signal was the only signal that could be heard.
Vielle568
P.S. Mooly, I notice that you live in the U.K. You don't come form Cowes by chance?
I switched the 2412 using an independant control signal. The results were as follows: First pulse produced a mild click. Second pulse produced a much weaker click. Third and fourth pulses reduced click to an (almost) inaudible level. Subsequent clicks remained at this low level. Increasing audio signal level did not increase the volume of the click, in fact, the clicks were only barely audible at the lowest audio signal levels. At higher levels the audio signal was the only signal that could be heard.
Vielle568
P.S. Mooly, I notice that you live in the U.K. You don't come form Cowes by chance?
Hello Aucosticraft, I tried your ideas out on my project.
Option one had little effect on the audio output signal, but this is not surprising as it seems that the cause of the click is due to something in the control circuitry.
The second option I haven't tried as yet but option three had some similarity to the ideas proposed by Stratus46, and so I have gone ahead and changed the resistors as suggested in the signal splitter circuit.
The effect on reducing the switching noise has been only minor but it has diminished slightly. I also tried various values of input capacitors but there was no effect on the switching noise; large caps only created problems at fast switching speeds.
Vielle568
Option one had little effect on the audio output signal, but this is not surprising as it seems that the cause of the click is due to something in the control circuitry.
The second option I haven't tried as yet but option three had some similarity to the ideas proposed by Stratus46, and so I have gone ahead and changed the resistors as suggested in the signal splitter circuit.
The effect on reducing the switching noise has been only minor but it has diminished slightly. I also tried various values of input capacitors but there was no effect on the switching noise; large caps only created problems at fast switching speeds.
Vielle568
Analog Devices: Analog Dialogue: Ask The Applications Engineer — 26
I think you may be suffering from charge injection from the control circuit this is minimised in audio parts but is still present. See link above and look for the section on charge injection. It has some hints on minimising it as well.
I think you may be suffering from charge injection from the control circuit this is minimised in audio parts but is still present. See link above and look for the section on charge injection. It has some hints on minimising it as well.
Its ok Vielle, I was just cross checking by re-framing the words. Yes probably it looks like a problem in a switching pulse.
My third option was actually was to see if any DC offset voltage appears at the output. and trying to minimize it. So checking if the presence of DC in the output of NE5532 is the reason for charging of capacitors, and a huge pop.
also I will try to post a block diagram how a switch should be implemented.
this is based on my previous experience in cleaning off the pop noise. I have just formatted my laptop. need to re install tools. give me some time.
Not from Cowes
On pulsing the switch. You may have two problems here. The control signal from the logic breaking through is one. A separate issue could be this,
If the clicking gets less and less then that suggests there is a slight DC differential across the switch that is becoming less as time goes on. That would fit in with any coupling caps charging and slowly settling to a steady state value.
Looking at the circuit in post #5 the switch is AC coupled at the input side with the 2.2uf cap and 470k to ground. Try reducing the value of the 470K by a factor of 10 or so. That will allow the cap to charge that much quicker.
The same things apply on the output side of the switch. It would help to see what this is connected to but as an experiment you could repeat the 2.2uf and resistor/s (47K) on the output.
The polarity of the caps depends on what is actually measured in the real circuit. Whether the final 47K is needed also depends on what is connected here.
Attachments
EUREKA!
Last night I reconfigured the splitter/buffer circuit feeding the audio signal to the analogue switch. I used the component values suggested by stratus46 but I have had very similar advice recently from Mooly and Aucosticraft. The rework did reduce the switch noise slightly, but it did not remove it completely.
Your latest message Mooly hit the nail on the head. The output from the analogue switch will be connected to an external amplifier and as such I've been simply measuring the output signals directly from the IC pins; no additional componects have been included. With the extra cap and resistors not only the clicks have disappeared but also the hum that was present when the analogue switch was in the off state.
I'm certain that many of the modification along the way have been beneficial but this final addition was something I obviously overlooked and should have included. Aucosticraft kindly sent me a schematic for an improved circuit and the same output configuration is also shown on his design.
Thank you Mooly and thank you Aucosticraft for all your help on getting this circuit to function correctly. It would have taken me quite a while to have found the solution without your aid; I am very grateful. The sound is now clear and there's silence between the pulses; there are more no clicks or pops to worry about.
Vielle568
Last night I reconfigured the splitter/buffer circuit feeding the audio signal to the analogue switch. I used the component values suggested by stratus46 but I have had very similar advice recently from Mooly and Aucosticraft. The rework did reduce the switch noise slightly, but it did not remove it completely.
Your latest message Mooly hit the nail on the head. The output from the analogue switch will be connected to an external amplifier and as such I've been simply measuring the output signals directly from the IC pins; no additional componects have been included. With the extra cap and resistors not only the clicks have disappeared but also the hum that was present when the analogue switch was in the off state.
I'm certain that many of the modification along the way have been beneficial but this final addition was something I obviously overlooked and should have included. Aucosticraft kindly sent me a schematic for an improved circuit and the same output configuration is also shown on his design.
Thank you Mooly and thank you Aucosticraft for all your help on getting this circuit to function correctly. It would have taken me quite a while to have found the solution without your aid; I am very grateful. The sound is now clear and there's silence between the pulses; there are more no clicks or pops to worry about.
Vielle568
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.