I did a bit of math and a 1% mismatch in the 6k8 phantom feed resistors (in the console) gives about 75mV offset into twin 28k loads. That would be a good thump when the LED is switched on. Since the LED resistors are not matched either, they can cancel the 6k8's mismatch but it means adding a balance pot.
Looks like time for a couple AA batteries?
Looks like time for a couple AA batteries?
The WIRES run straight to the mics but you're forgetting those 6.8K resistors from the 48V supply to the 2 mic leads inside the console. Remember your Ohm;'s law from earlier in the thread? You can break the laws of man but not Ohm's law. As soon as you try ANY current for your LED you will change the drop across the internal resistors and no, you cannot change them. Remember the audio on peaks will be in the millivolts range and for you to have no noise your differential should be as close to 0 as possible - likely in the microvolts range. A dummy load sounds nice in theory but is much easier to say than to do. There is a short time while switching from LED to dummy load that has to be exactly 0. Not easy to do. Whenever I hear a solution start with "all you have to do..." I KNOW it's danged difficult.
I'm pretty sure I can build a mute box but I would use an outboard power supply to do it whether a battery or wall wart. Perhaps John pulled it off without an extra power supply and if so, congratulations to him.
G²
@ John: Thanks for the explanation of the new schematic. I’m still working my way through the implications of this method of muting. It is probably time to do some real world tests. I’ve always disliked the error part of trial and error, so I tend to research til I’m positive I have a viable solution… but in many cases, hands on is the fastest way to go.
@Prariemytic: Thanks, I’ll modify my diagram to incorporate a photo-mosfet.
About the voltage drop… I’m not sure I understand what is going on. Prariemystic mentions a voltage drop based on the leds drawing power…then benb quotes that post and mentions the importance of perfectly matched resistors in the mixers preamp. It almost feels like there are two separate issues happening.
Issue 1
If + and – signal lines are not energized to the same voltage, the difference in voltage would show up as a pop when the mute switch is thrown. If the capacitor isn’t solving this issue I could add a trim pot in the mute box as is mentioned Here…Popless Mic On/Off Switch
(I’m not sure if the trim pot is shown in the correct placement in the new image…should it be at R1 as shown?)
Issue 2
Volts = current * resistance.
I’m trying to predict the voltage at the mic,
so for current it’ll be 7mas in each leg…
and resistance will be the 6k8 resistors in the console?
Then when the leds go on, the current is reduced by something like 2 or 3 mas… and this is what changes the voltage at the mic?
So at rest, the voltage is V= .007 * 6800 47.6 volts in each pin.
And with the leds on the voltage at the mic is V= .0055 * 6800 37.4
Is that it?
So then, provided we can equalize the two legs with a trim pot, it is this 10 volt drop that will cause a thump? Or perhaps it isn’t the voltage, but the 3ma drop that will cause the thump?
I think part of my problem is that I really don’t know what is inside an electret mic. And thus, I don’t know how it will react to given changes in power supply.
I am satisfied at this point to build a unit that runs on batteries (or takes power from a spare channel), but, like a half finished jigsaw, the phantom powering puzzle is still tickling my brain.
@stratus: I hear you on the “all you have to do…” thing. In my family the saying is, “sure, just get a rocket and go to the moon” J
I think I will start building as soon as I figure out where to put the trim pot and find exactly which photo-mosfet to order. I can add the phantom power components whenever I figure out how it needs to work.
Thanks again for all the help, I feel like I'm back in school 8O
@Prariemytic: Thanks, I’ll modify my diagram to incorporate a photo-mosfet.
About the voltage drop… I’m not sure I understand what is going on. Prariemystic mentions a voltage drop based on the leds drawing power…then benb quotes that post and mentions the importance of perfectly matched resistors in the mixers preamp. It almost feels like there are two separate issues happening.
Issue 1
If + and – signal lines are not energized to the same voltage, the difference in voltage would show up as a pop when the mute switch is thrown. If the capacitor isn’t solving this issue I could add a trim pot in the mute box as is mentioned Here…Popless Mic On/Off Switch
(I’m not sure if the trim pot is shown in the correct placement in the new image…should it be at R1 as shown?)
Issue 2
Volts = current * resistance.
I’m trying to predict the voltage at the mic,
so for current it’ll be 7mas in each leg…
and resistance will be the 6k8 resistors in the console?
Then when the leds go on, the current is reduced by something like 2 or 3 mas… and this is what changes the voltage at the mic?
So at rest, the voltage is V= .007 * 6800 47.6 volts in each pin.
And with the leds on the voltage at the mic is V= .0055 * 6800 37.4
Is that it?
So then, provided we can equalize the two legs with a trim pot, it is this 10 volt drop that will cause a thump? Or perhaps it isn’t the voltage, but the 3ma drop that will cause the thump?
I think part of my problem is that I really don’t know what is inside an electret mic. And thus, I don’t know how it will react to given changes in power supply.
I am satisfied at this point to build a unit that runs on batteries (or takes power from a spare channel), but, like a half finished jigsaw, the phantom powering puzzle is still tickling my brain.
@stratus: I hear you on the “all you have to do…” thing. In my family the saying is, “sure, just get a rocket and go to the moon” J
I think I will start building as soon as I figure out where to put the trim pot and find exactly which photo-mosfet to order. I can add the phantom power components whenever I figure out how it needs to work.
Thanks again for all the help, I feel like I'm back in school 8O
Break this project down into two separate parts.
The muting element you are copying from an established product. This works - and there is no problem with it.
So there is no need to trim the resistor - it can be almost any value you like, but 47K to 100K will work well. The non polarised capacitor similarly can be any value - as high as you can easily find, and low voltage - as each end is at the same potential.
Test this part of the project with a switch of any sort to prove it works.
Then substitute the switch with your photo MOS, which you can regard as a relay. (You need low 'ON' resistance, just like a real relay!)
Drive the opto element with a red LED in series (to show MUTE status) and a suitable resistor from a battery (which will last for ages).
That is it......
Think about Phantom powering the LED another time!
The muting element you are copying from an established product. This works - and there is no problem with it.
So there is no need to trim the resistor - it can be almost any value you like, but 47K to 100K will work well. The non polarised capacitor similarly can be any value - as high as you can easily find, and low voltage - as each end is at the same potential.
Test this part of the project with a switch of any sort to prove it works.
Then substitute the switch with your photo MOS, which you can regard as a relay. (You need low 'ON' resistance, just like a real relay!)
Drive the opto element with a red LED in series (to show MUTE status) and a suitable resistor from a battery (which will last for ages).
That is it......
Think about Phantom powering the LED another time!
As I move forward with part 1, I have a few more questions.
I'm glad to hear the trim pot isn't needed, but I'd like to better understand what the author was talking about here... Popless Mic On/Off Switch.
Where would the trim pot have gone if needed? I think either R2 or R3? It's purpose would have been to balance the voltages on pins 2 and 3? Does it matter if it is R2 or R3, would it fulfill it's purpose either way?
Does anyone have a suggestion of a particularly low on resistance photo-mosfet?
My design had used a 16k2 resistor as the led Rdropper using a calculation based on 48v phantom power...I imagine this needs to be changed to work with a 9v.
With Vled = 2v
and V+ = 9v
and Iled = 2
I get Rdropper = 3500 ohms. This only accounts for the one indicator led. How should I change this to include the power to the photo-mosfet?
Lastly, I have the feeling like the answer to the phantom powering riddle is in power supply circuits. We need to have a component sit in between the power supply and the load to even out changes in the load.... a sort of storage battery.... perhaps a capacitor of some sort. So that the power supply is used to charge this current buffer, which then does the power supply for the actual load.
Again, I feel my ignorance kicking in as I don't know the keywords to search for this "gizmo".
But, considering that there are components that rectify ac power supplies (turning on and off all the time) and put out smooth 5v power...it feels like I've seen this before.
Much thanks for the help on this project, I can almost see it completed now... it's been a rather long road, but interesting and educational.
Thanks!
I'm glad to hear the trim pot isn't needed, but I'd like to better understand what the author was talking about here... Popless Mic On/Off Switch.
Where would the trim pot have gone if needed? I think either R2 or R3? It's purpose would have been to balance the voltages on pins 2 and 3? Does it matter if it is R2 or R3, would it fulfill it's purpose either way?
Does anyone have a suggestion of a particularly low on resistance photo-mosfet?
My design had used a 16k2 resistor as the led Rdropper using a calculation based on 48v phantom power...I imagine this needs to be changed to work with a 9v.
With Vled = 2v
and V+ = 9v
and Iled = 2
I get Rdropper = 3500 ohms. This only accounts for the one indicator led. How should I change this to include the power to the photo-mosfet?
Lastly, I have the feeling like the answer to the phantom powering riddle is in power supply circuits. We need to have a component sit in between the power supply and the load to even out changes in the load.... a sort of storage battery.... perhaps a capacitor of some sort. So that the power supply is used to charge this current buffer, which then does the power supply for the actual load.
Again, I feel my ignorance kicking in as I don't know the keywords to search for this "gizmo".
But, considering that there are components that rectify ac power supplies (turning on and off all the time) and put out smooth 5v power...it feels like I've seen this before.
Much thanks for the help on this project, I can almost see it completed now... it's been a rather long road, but interesting and educational.
Thanks!
That's easy enough (and presuming it turns "on" enough with the 2mA used to run the indicator LED), the LED inside the photo-mosfet probably drops the same 2V (you could look at the datasheet to see an exact value, but this is plenty close enough), so the total across the resistor would be 9V - 2V - 2V, or 5V. Dividing voltage by 0.002A, I get 2500 ohms.
You might calculate how long the mute button is going to be pressed in average or worst-case use, so you could tell how long a battery will last. I'm guessing it will be a few seconds at a time, maybe a couple minutes a month, and a battery could easily last a year. I'd put a note on the bottom to "change battery once a year" with several underlined blanks for "date last changed."
If you want to get a strong theoretical understanding of how voltages add up in series circuits, read Kirchhoff's Voltage Law, and while you're at it, Kirchhoff's Current Law, available through Google.
John Audio shows that such as circuit CAN be made to powered by phantom power, but then he says it "took considerable development time." If you're developing a product yourself it might be worth it, but a one-off or even a small production run I'd consider a battery powered design to be acceptable. And it's pretty much designed now. Get the parts, test it, and if it works as intended, go with it.
@ stratus: If I were to use a plain relay, would the magnetic field have any negative influence and the audio signal? Also, how much power is needed to activate a relay... milliamps? What voltage does it like?
Thanks benb. I had guessed it was simple, but I thought better to check before I make the smoke rise.
I'll read up on the stuff you suggested.
any thoughts on that trim pot?
the Datasheet - Vishay LH1500 photo-mofet shows an on voltage of 20 ohms. Is that low enough to operate as a short?
Gracias
Thanks benb. I had guessed it was simple, but I thought better to check before I make the smoke rise.
I'll read up on the stuff you suggested.
any thoughts on that trim pot?
the Datasheet - Vishay LH1500 photo-mofet shows an on voltage of 20 ohms. Is that low enough to operate as a short?
Gracias
Last edited:
I'm building a power supply selection switch into the part one version to allow for easy experimentation.
Am I correct in thinking that in "Battery" mode, ground is the negative terminal. While in "phantom" mode, ground is pin 1.
I couldn't simply use pin 1 as ground in all cases, could I?
Am I correct in thinking that in "Battery" mode, ground is the negative terminal. While in "phantom" mode, ground is pin 1.
I couldn't simply use pin 1 as ground in all cases, could I?
Magnetic field would not bother you a bit but too many mAs (5V, 10mA). Johns's suggestion is very good. The 2 LEDs in series (indicator and opto MOSFET) would be about as efficient as you'd get for battery life provided 2 mA is sufficient to saturate the MOSFET.
http://data.energizer.com/PDFs/522.pdf
At 2mA I expect it would run about 2 weeks on.
As far as on resistance goes, put a 20 ohm resistor across the mic leads. I expect it will attenuate it a lot but not completely mute it. Mechanical switch (relay) will kill it.
Of course the simplest solution would be a double pole switch with one pole for the mute and the other for the indicator. It would still mute with no battery at all - just no indicator.
G²
I found a photo-mofet that looks more promising as it has a lower on resistance.
Powered by Google Docs
If I read it right, the R(on) is .7 ohms in the ac configuration. (bottom left of page 2).
Is that low enough?
Powered by Google Docs
If I read it right, the R(on) is .7 ohms in the ac configuration. (bottom left of page 2).
Is that low enough?
After finding my design riddled with switches to provide the various button action combinations (ptt, ptm, etc)
I thought I'd look into microcontrollers. Is there a possibility of running a microcontroller of of the very low mas of phantom power? From this page:
Reducing power usage on an AVR ATTiny4 | Programming | Garlic Software
it looks like the mcu runs on microamps.
The plus side if it does work is that I could eliminate most of the switches in the design.
I thought I'd look into microcontrollers. Is there a possibility of running a microcontroller of of the very low mas of phantom power? From this page:
Reducing power usage on an AVR ATTiny4 | Programming | Garlic Software
it looks like the mcu runs on microamps.
The plus side if it does work is that I could eliminate most of the switches in the design.
You're giving in to 'mission creep'. Next you'll want internet access so muting the mic also pours a beer.
G²
Not really, remember, my original design goal was a phantom powered mute with remote switch and led... current design version doesn't allow that.
You could say I'm hesitant to settle for a half solution. That would be valid.
Although... I am building the simple version now...so I am actually taking john audio's suggestion to the letter.
These extra questions are "thinking out loud" for the second part, the "addon" module to allow for more advanced functionality.
Active questions...
1. On resistance of .7 ohms in photo-mosfet low enough for muting?
2. Where would the trim pot have gone if needed?
3. True or false..."Ground" in battery mode = negative side of battery. "Ground" in phantom mode = pin 1, I can't just use pin1 as ground in both cases.
4. I answered my own question on this one. Common microcontrollers run on microamps, so it's entirely doable to use an MCU to modify the switch behavior. I also looked at a shure mx418 and found they used several mcu's in it.
This is the current design. The value of the Rdropper will be calculated when I decide on the exact model of opto.
I'm still not certain how to configure a simple latching flip flop. I think it is to connect the output from the main switch to both the data and clock lines of a T-flipflop. But I'm still working on that.
Any final thoughts on version 1?
You could say I'm hesitant to settle for a half solution. That would be valid.
Although... I am building the simple version now...so I am actually taking john audio's suggestion to the letter.
These extra questions are "thinking out loud" for the second part, the "addon" module to allow for more advanced functionality.
Active questions...
1. On resistance of .7 ohms in photo-mosfet low enough for muting?
2. Where would the trim pot have gone if needed?
3. True or false..."Ground" in battery mode = negative side of battery. "Ground" in phantom mode = pin 1, I can't just use pin1 as ground in both cases.
4. I answered my own question on this one. Common microcontrollers run on microamps, so it's entirely doable to use an MCU to modify the switch behavior. I also looked at a shure mx418 and found they used several mcu's in it.
This is the current design. The value of the Rdropper will be calculated when I decide on the exact model of opto.
I'm still not certain how to configure a simple latching flip flop. I think it is to connect the output from the main switch to both the data and clock lines of a T-flipflop. But I'm still working on that.
Any final thoughts on version 1?
1: 0.7 ohms is only slightly higher than a switch and wiring. Should be good.
2: I don't see why you'd need a trim.
3: 'ground' in the battery circuit is whatever you'd like it to be. Most will use the negative terminal but it's not set in stone. The only important thing is getting the correct polarity to the LED of the opto.
4: The MCU is actually a good idea as you alluded to with your question about the flip flop. The flip flop is easy but de-bouncing the switch contacts to TRIGGER the flip flop can be tricky. Writing software to de-bounce switches and toggle outputs is not difficult and if you've never done it, it's an excellent learning project. I've used micros to scan MIDI musical keyboards and provide velocity. Switch errors are unacceptable and true N key rollover is a must. It was a fun project.
If you want to use an actual flip flop, a D is easy (as is JK). Tie D input to not Q output and trigger with the clock input. Pay attention to bouncing and rates of change on the clock. Too slow and it won't trigger.
G²
2: I don't see why you'd need a trim.
3: 'ground' in the battery circuit is whatever you'd like it to be. Most will use the negative terminal but it's not set in stone. The only important thing is getting the correct polarity to the LED of the opto.
4: The MCU is actually a good idea as you alluded to with your question about the flip flop. The flip flop is easy but de-bouncing the switch contacts to TRIGGER the flip flop can be tricky. Writing software to de-bounce switches and toggle outputs is not difficult and if you've never done it, it's an excellent learning project. I've used micros to scan MIDI musical keyboards and provide velocity. Switch errors are unacceptable and true N key rollover is a must. It was a fun project.
If you want to use an actual flip flop, a D is easy (as is JK). Tie D input to not Q output and trigger with the clock input. Pay attention to bouncing and rates of change on the clock. Too slow and it won't trigger.
G²
Are there any problems that flare up if I used pin1 as ground when running off the battery?
I had in my head somewhere that batteries won't work unless both terminals are connected. One terminal producing electrons, and the other producing ions. I'll have to test this later. I could be loony there.
I did find a post which indicates that the first version of the pro co cough drop didn't work very well. Pro Co Cough Drop Or Pro Co Short Stop: Mute Pedals Should Be Seen And Not Heard | Gearwire
I haven't been able to find what they changed to fix it. Or whether that change was reflected in the schematics. I'm inclined to think my circuit will work based on your inputs.
I've got a few mcu projects bubbling so I think I might end up toying with a mcu version of this.
Latest version:
Question: The flip flop needs power supplied independently of the momentary switch right? To allow it to keep the mute active after the momentary switch is released? Where do I apply that power...a jumper from +V to...where?
I think I'm about ready to do this thing. At least the version one anyway.
I had in my head somewhere that batteries won't work unless both terminals are connected. One terminal producing electrons, and the other producing ions. I'll have to test this later. I could be loony there.
I did find a post which indicates that the first version of the pro co cough drop didn't work very well. Pro Co Cough Drop Or Pro Co Short Stop: Mute Pedals Should Be Seen And Not Heard | Gearwire
I haven't been able to find what they changed to fix it. Or whether that change was reflected in the schematics. I'm inclined to think my circuit will work based on your inputs.
I've got a few mcu projects bubbling so I think I might end up toying with a mcu version of this.
Latest version:
Question: The flip flop needs power supplied independently of the momentary switch right? To allow it to keep the mute active after the momentary switch is released? Where do I apply that power...a jumper from +V to...where?
I think I'm about ready to do this thing. At least the version one anyway.
There is another solution for this.
First of all a 20ohm across the mic will mute it completely. Even a 60 ohm will mute it. I will not go into explaining this but it will, because of the loading effect.
Try connecting a 60 ohm and see what happens. It will mute but will it do for u.
If so, then use a silonex NSL-32SR3 (IIRC) optic isolator across the mic. Most of your problems will be solved because of its slow behavior.
Gajanan Phadte
Edit: the idea is power hungry.
First of all a 20ohm across the mic will mute it completely. Even a 60 ohm will mute it. I will not go into explaining this but it will, because of the loading effect.
Try connecting a 60 ohm and see what happens. It will mute but will it do for u.
If so, then use a silonex NSL-32SR3 (IIRC) optic isolator across the mic. Most of your problems will be solved because of its slow behavior.
Gajanan Phadte
Edit: the idea is power hungry.
Last edited:
The battery circuit is completely independent of the mic circuit. It's probably a good idea to have your mute ground tied to the mic shield but is functionally unneeded. The flip flop cannot drive LEDs directly. The easiest LED driver would be something like a 2N7000 MOSFET. The flip flop needs power continuously to remember the last state but the CMOS power consumption is so low in the 'off' (not muted) mode that the battery life will be shelf life. You still need to process the mute trigger to avoid faulty triggering and that WILL give you a bit of grief. Using a JK rather than D flip flop might be possible to use the other flip flop as the debouncer but I've never done it that way. Should be able to find some contact debouncing circuits on line. Using digital devices in the real world is always a challenge for the interfacing. You'll also need a Power On Reset (POR) circuit so that it always turns on in the same condition rather than random. BTW all those functions can be done in an MCU except for LED the driver.
G²
G²
Last edited:
Thanks for the input ghphadte. I have a few questions, if you have time...
I'm interested to hear that you believe a 20ohm connection will satisfactorily mute the mic. I expect I'll order the lowest on resistance photo-mosfet that I can find that is reasonably priced.
I was not aware of such a thing as a opto-isolator. What difrences are there between the opto-mofet and the opto-isolator? Both drive an led, and use that light to lower the resistance on the output to a functional short. I seems like they would both do the job.
What problems with the opto-isolator improve by having a "slow behavior"
Which Idea is power hungry? The photo-mosfet idea shown in the diagram? or the idea with the opto-isolator?
Did you mean that I could use a schmitt nand/inverter to debounce the flipflop trigger?
@stratus:
Active questions:
1. What pin of the flipflop is connected to the always on powersource. [EDIT: I think I found this, most tutorials on flipflops use a block diagram with only 5 pins shown, data, clock Q, not Q. There are also always V+ and Gnd pins on the chip. So I'd have to run a jumper from V+ to V+pin and from Gnd to Gnd post-led. I think that is the way anyhow.]
2. are you saying that... the output from the flipflop cannot provide enough power for my 2 leds, and so it should be connected to a transistor used as a small amplifier? that is, gate one, power flows to leds, gate off, no power to leds.
I'll look into debouncing circuits and see what I can learn.
I tend more and more to like the mcu idea.
Thanks again
I'm interested to hear that you believe a 20ohm connection will satisfactorily mute the mic. I expect I'll order the lowest on resistance photo-mosfet that I can find that is reasonably priced.
I was not aware of such a thing as a opto-isolator. What difrences are there between the opto-mofet and the opto-isolator? Both drive an led, and use that light to lower the resistance on the output to a functional short. I seems like they would both do the job.
What problems with the opto-isolator improve by having a "slow behavior"
Which Idea is power hungry? The photo-mosfet idea shown in the diagram? or the idea with the opto-isolator?
Did you mean that I could use a schmitt nand/inverter to debounce the flipflop trigger?
@stratus:
Active questions:
1. What pin of the flipflop is connected to the always on powersource. [EDIT: I think I found this, most tutorials on flipflops use a block diagram with only 5 pins shown, data, clock Q, not Q. There are also always V+ and Gnd pins on the chip. So I'd have to run a jumper from V+ to V+pin and from Gnd to Gnd post-led. I think that is the way anyhow.]
2. are you saying that... the output from the flipflop cannot provide enough power for my 2 leds, and so it should be connected to a transistor used as a small amplifier? that is, gate one, power flows to leds, gate off, no power to leds.
I'll look into debouncing circuits and see what I can learn.
I tend more and more to like the mcu idea.
Thanks again
Last edited:
- Status
- Not open for further replies.