Hello,
The max input voltage for my amp is 3V. What is the best way to ensure the signal coming from the pre-amp/mixer does not breach this threshold? I have a mixer with gains built-in, along with a vue meter but I would like to actively monitor the voltage input and see a notification when it nears the max. I could connect my 4 channel oscope and set triggers, but would like a permanent solution.
Is there a device or voltage limiter I should consider? ( i'm not an audiophile so this is new to me )
Looking at maybe a Rane or DBX.
I've started an STM32 'peak voltage detector' project using ADC out of necessity, but would rather not re-invent the wheel and use a method/device which is reliable and tested.
Thank you!
The max input voltage for my amp is 3V. What is the best way to ensure the signal coming from the pre-amp/mixer does not breach this threshold? I have a mixer with gains built-in, along with a vue meter but I would like to actively monitor the voltage input and see a notification when it nears the max. I could connect my 4 channel oscope and set triggers, but would like a permanent solution.
Is there a device or voltage limiter I should consider? ( i'm not an audiophile so this is new to me )
Looking at maybe a Rane or DBX.
I've started an STM32 'peak voltage detector' project using ADC out of necessity, but would rather not re-invent the wheel and use a method/device which is reliable and tested.
Thank you!
Last edited:
Emphasis added is mine. I'd set up the preceding gear such that it doesn't clip the signal and doesn't output more than +-3V. We'd need to assume that the source does not output a clipped signal. It's clear we're not talking about "ideal" situations, so a non-user-accessible pot may do the trick.
tl;dr - I'd set things such that user (or a kid or a pet or whatever) turned the volume "all the way up" that the signal leaving the gear preceding the amplifier was not clipped nor above +-3V.
I am not disagreeing with pinholer. However, feeding a clipped input signal to a power amplifier at maximum acceptable input voltage isn't ideal or best to me. It's certainly good / much better than nothing, and it is a simple/elegant solution. I'm only pointing out that it may come with some level of risk. If your goal is eliminating/reducing the potential for "damage" vs. just ensuring a voltage isn't exceeded, then you may (or may not) want to try some other things.
A brutal worst-case test of concept for that scenario would be - See if your amp / speakers can handle a +- 3Vp square wave input over the maximum time needed to react to the expected calamity.
Why is that important? Let's say someone (or a pet) can't react or doesn't know how to react when they "cranked it to the max" and it sits for an hour ... will things go poof? Will the speakers ever speak to you again? I don't know your use case, and sadly my brain goes to the worst conceivable situations when doing failure mode and effects analysis.
^^ Thanks! I'll take a look. I think having a voltage peak monitor will help me set levels on my mixer/pre-amp to detect before clamping kicks in.
@ItsAllInMyHead Agree, but I"m looking for a device or way to monitor the input of the amp to ensure the pre-amp is not exceeding desired voltage, well before clipping takes place. ( let's say 2.75V as my set-point max ). How you you envision I do this with let say a mixer with gain controls and a vue display? Voltage could peak briefly without being to detect it by eye using a plain vue meter. Same with my focusrite DAC. I know it won't output more than 2.75V but that is at full volume, potentially clipping.
@ItsAllInMyHead Agree, but I"m looking for a device or way to monitor the input of the amp to ensure the pre-amp is not exceeding desired voltage, well before clipping takes place. ( let's say 2.75V as my set-point max ). How you you envision I do this with let say a mixer with gain controls and a vue display? Voltage could peak briefly without being to detect it by eye using a plain vue meter. Same with my focusrite DAC. I know it won't output more than 2.75V but that is at full volume, potentially clipping.
^ Clamping =/= clipping.
Monitoring the input of the amp is the same as monitoring the output of the pre-amp (unless you stick something in the middle).
You moved the goalposts.
This is what you originally asked. Thus, that's what I answered.
Monitoring the input of the amp is the same as monitoring the output of the pre-amp (unless you stick something in the middle).
You moved the goalposts.
This is what you originally asked. Thus, that's what I answered.
@roboDNA -
Here's where I am coming from with your question.
Really how I'd do it in my shoes in my situation (based on the gear described) is I'd likely not worry about it. I'd use the gear as intended. I'd be wary of any hot sources and set volumes/levels accordingly.
It's tough b/c I don't know what you're actually worried about from an end-user standpoint.
Here's where I am coming from with your question.
- If you ensure it doesn't happen, you don't need to monitor it. At least I wouldn't waste resources.
- It's better to prevent it all-together (if there are no unforeseen issues you haven't described from prevention)
- It's simpler (IMO) and a better plan to prevent it than:
- monitor it,
- actively control it with human intervention once you are alerted through monitoring
Really how I'd do it in my shoes in my situation (based on the gear described) is I'd likely not worry about it. I'd use the gear as intended. I'd be wary of any hot sources and set volumes/levels accordingly.
It's tough b/c I don't know what you're actually worried about from an end-user standpoint.
^^ to be up-front, I'm slightly confused with your answer. ( Not sure what you mean about moving the goal post since I'm still focused on the amp input ) I do appreciate your feedback on this though!! How do you envision monitoring the voltage to ensure it does not go over a target of let say 2.7V? What device would you use between the pre-amp and the amp to see the voltage? I understand a meter or oscope can be used but my question involves coming up with a permanent solution allowing me to easily monitor the voltage. I think I have enough to move forward but wanted to make sure I understand what you mean.
^ Bad phrasing or possible misinterpretation on my part. Apologies. I thought you were changing what you wanted to accomplish and were instead putting up a solution for monitoring vs. control. My fault.
Let me try to do this a bit more linearly. I wish I could do it any other way... but it's how I'm wired (pun intended - groan expected).
I would not monitor it. I would simply ensure it does not go over that target. Therefore, no need for monitoring. Does that make sense? As an example, if you implemented pinholer's clipping or Magician T's clamping solutions, they guarantee you can't go over. So, I would not "monitor" it.
Does that make sense?
Or, do you have the need to know when you're approaching various voltage levels via some sort of a visual indicator?
Again, I admit to not being completely sure of your "root" need, so I'm just trying to solve the problem presented.
Edited to add - if you go with pinholer's / Magician's solution, I would recommend implementing a visual indicator. That way, you know (more importantly) when you're back below threshold Voltage, because they alter the signal if it's over-voltage. Light on.... turn it down until light off.
If you go with a solution that guarantees you'll neither have a clipped / clamped signal nor an over-voltage ... you don't even need a visual indicator b/c it would never light up.
Let me try to do this a bit more linearly. I wish I could do it any other way... but it's how I'm wired (pun intended - groan expected).
Maybe we're using fundamentally different terms, and that's where I caused the confusion.
I would not monitor it. I would simply ensure it does not go over that target. Therefore, no need for monitoring. Does that make sense? As an example, if you implemented pinholer's clipping or Magician T's clamping solutions, they guarantee you can't go over. So, I would not "monitor" it.
Does that make sense?
Or, do you have the need to know when you're approaching various voltage levels via some sort of a visual indicator?
Again, I admit to not being completely sure of your "root" need, so I'm just trying to solve the problem presented.
I wouldn't. See above.
That's where we differ perhaps. I was offering a solution for making sure that the "problem" never occurred in the first place (as were pinholer and MagicianT). No monitoring needed / involved. In the case of using an LED, yes, there could be a visual indicator that the problem has already occurred (or set to just below a threshold) ...
Edited to add - if you go with pinholer's / Magician's solution, I would recommend implementing a visual indicator. That way, you know (more importantly) when you're back below threshold Voltage, because they alter the signal if it's over-voltage. Light on.... turn it down until light off.
If you go with a solution that guarantees you'll neither have a clipped / clamped signal nor an over-voltage ... you don't even need a visual indicator b/c it would never light up.
I hope that's more clear. Apologies. I'm not the best at explanations.
Last edited:
trying to prevent it from happening involves seeing what the voltage peak is. Not all music is the same, voltage could go higher when using drum machines other equipment into the mixer. Using a clipping diode would work but would only allow 1 set point. I want 1 led to light when 2.7-3v is reached and then a red when 3v is reached. I'm confused since I never mentioned clipping. You also suggest you would 'simply ensure it would not happen' without stating how. That is exactly what I"m trying to do by watching the voltage, or 'monitoring the voltage' as I call it.
Check out https://www.akitika.com/documents/BuildingTheAttenuatorRev4.pdf
This is a switchable attenuator to protect the input of a sound card. The LEDs limit the voltage as well a provide visual indication of overvoltage. More LEDs could be added in series to increase the voltage threshold.
I use such an attenuator when performing distortion tests on amplifiers and when not limiting, I do not notice any additional distortion due to the circuit.
This is a switchable attenuator to protect the input of a sound card. The LEDs limit the voltage as well a provide visual indication of overvoltage. More LEDs could be added in series to increase the voltage threshold.
I use such an attenuator when performing distortion tests on amplifiers and when not limiting, I do not notice any additional distortion due to the circuit.
No, it absolutely does not. You were offered three solutions that 100% would
They differ in their approaches, and they clearly differ from what you seem to want, but was not part of your initial requirement, thus my comment on shifting the goalposts.
I brought it up b/c pinholer's solution would literally clip the signal (if an over-voltage was present), which solves your stated problem directly. However, as I mentioned may have other side effects.
MagicianT's solution would also alter the signal (if an over-voltage was present)
My proposed solution (as mentioned) is attenuation.
Last edited:
@roboDNA - I have an unused one of these (I'm pretty sure) sitting on my desk. Would you like it? It's certainly not as elegant a build as the one shown on the website, but it's yours If I can find it, and if you would find it helpful.
@MagicianT That attenuator looks like almost exactly what I was envisioning, thanks. @ItsAllInMyHead that is very generous of you to offer, thanks. I'll see how far along I get with my STM32 ADC based approach which uses opamps to monitor the voltage. The attenuator gave me some ideas and I also see I don't need to worry too much about introducing noise since @MagicianT knows that circuit works well. ( op-amp will be even higher impedance vs resistor ) Just need to add filter caps and also clipping diodes JUST IN CASE but I'm leaving the clip detection as a separate project. Right now, i just want some type of visual for when I adjust the gain on my mixer to know I have not reached close to 3V.
I'm just trying to figure out which opamp to use. Currently looking at TLV2371, OPA2134, and even LM358. I divide the voltage down to ensure the ADC does not go over 3V, but would like a way to just 'ignore' all voltage above 3V with a Schottky diode or using comparator opamp. Worried the diode may introduce noise, and that is the point I am at. I'll need to build a prototype and test for noise with my oscope I guess.
I'm just trying to figure out which opamp to use. Currently looking at TLV2371, OPA2134, and even LM358. I divide the voltage down to ensure the ADC does not go over 3V, but would like a way to just 'ignore' all voltage above 3V with a Schottky diode or using comparator opamp. Worried the diode may introduce noise, and that is the point I am at. I'll need to build a prototype and test for noise with my oscope I guess.
I found the following doc which gives some good details on the implementation. I can replace the reset switch with a mosfet to drain the cap, and see about using ADC to sense voltage on the OUT but still need to figure out how to best way to limit voltage to 0-3.3V for my ADC input. Has anyone connected this type of peak detector circuit to an ADC? Should a max pre-amp signal of 20V be high enough? ( my mixer outputs 0V-9.4V ) I'm currently planning on just doing a voltage divider and feeding that directly into ADC. I'm considering adding a zener or schottky diode with a series resistor to protect the ADC but have not determined values for those yet.
Peak Detector Circuit at sound-au.com
If anyone wants to beta test my board when I'm done just PM me. I plan on doing 1 with just LEDs and then a second one with an STM32.
Peak Detector Circuit at sound-au.com
If anyone wants to beta test my board when I'm done just PM me. I plan on doing 1 with just LEDs and then a second one with an STM32.
I would think about using a digital potentiometer to do the voltage dividing. Your processor would control the scale factor and also adjust the scaling in real time to limit overvoltage to the ADC.
I have used the DS3502 digital pot which is rated for 15 volts.
You could also use a second digital pot to turn down the input to your amp if the voltage gets too high.
I have used the DS3502 digital pot which is rated for 15 volts.
You could also use a second digital pot to turn down the input to your amp if the voltage gets too high.
This is a power amp? Your ears will tell you if so, no need for any fancy device! It will either be way too loud for comfort, or you'll here clicking on transients that clip. Most audio circuitry will handle input overload without burning out if its sensibly designed.
Which amp is it though?
If the preamp is made so that there can not be overvoltage and power on/off muting is done there is nothing to sorry about. There is not a bogeyman behind every corner.
A buffer would already be a solution when used with standardised sources with 2V rms output level. No way that can be more when using a buffer.
A short overvoltage will likely be tolerated too but your ears will have already have given severe panic pulses to the brain to do something immediately (like turning down the volume) when the level was way before even 2V. You can trust on digital trickery or use your ears and common sense.
A buffer would already be a solution when used with standardised sources with 2V rms output level. No way that can be more when using a buffer.
A short overvoltage will likely be tolerated too but your ears will have already have given severe panic pulses to the brain to do something immediately (like turning down the volume) when the level was way before even 2V. You can trust on digital trickery or use your ears and common sense.
Last edited:
@Mark Tillotson it's the F5m and I use a Soundraft Signature12 mixer as a pre-amp. ( I also sometimes use my focusrite scarlett i2i but it does not output more than 3V. The mixer outputs up to ~9V ) I'm using brand new Wharfedale Super Lintons and don't want to damage them.
@jean-paul I've been playing my amp and speakers at lower volumes and have not turned them up yet since I've been trying to understand the amps' input voltage better. I plan on connecting a meter or oscope to check the voltage. Is there a 'buffer' device brand or model I can look at since I have not heard of this before. ( Do I build one or buy one? ) I think studios use 'limiter/compressor' units but not sure. I'm surprised there is not more focus on this critical part of the signal path for others... I wonder how many actually know what the voltage is, and how many just play it by ear. Even if I adjust the pre-amp to limit voltage, would it not need to be adjusted for different types of audio since some are louder than others? Also, if I limit the voltage coming out the pre-amp too much, do I not degrade the audio signal when I try to play it louder? I realize there is a sweet spot but am confused as to how everyone monitors this voltage on a permanent basis, without connecting temporary digital meters or oscopes. I'm guessing that is where the limiter/compressor units come it. @ItsAllInMyHead also suggested I limit the pre-amp but not sure how to do this on a permanent basis considering the signal may be louder with some audio. I know my roland drum machine sounds much louder than youtube for example, even when the volume/gains are not changed.
I'm also seeing many voltage peak detection circuits from other diyadio members ( and other sites ) and see companies like Rane and DBX have solutions but either discontinued or combined with other features/functions I do not need. ( like recording or mic inputs ) I know this is a problem which has been addressed in many different ways for many years, looking at some early peak detector circuits.
I'm still planning on finishing my own version of a uC based peak detector which I have made good headway on. It will have presets for voltage division with either jumpers or digital POT as @pinholer suggested.
@jean-paul I've been playing my amp and speakers at lower volumes and have not turned them up yet since I've been trying to understand the amps' input voltage better. I plan on connecting a meter or oscope to check the voltage. Is there a 'buffer' device brand or model I can look at since I have not heard of this before. ( Do I build one or buy one? ) I think studios use 'limiter/compressor' units but not sure. I'm surprised there is not more focus on this critical part of the signal path for others... I wonder how many actually know what the voltage is, and how many just play it by ear. Even if I adjust the pre-amp to limit voltage, would it not need to be adjusted for different types of audio since some are louder than others? Also, if I limit the voltage coming out the pre-amp too much, do I not degrade the audio signal when I try to play it louder? I realize there is a sweet spot but am confused as to how everyone monitors this voltage on a permanent basis, without connecting temporary digital meters or oscopes. I'm guessing that is where the limiter/compressor units come it. @ItsAllInMyHead also suggested I limit the pre-amp but not sure how to do this on a permanent basis considering the signal may be louder with some audio. I know my roland drum machine sounds much louder than youtube for example, even when the volume/gains are not changed.
I'm also seeing many voltage peak detection circuits from other diyadio members ( and other sites ) and see companies like Rane and DBX have solutions but either discontinued or combined with other features/functions I do not need. ( like recording or mic inputs ) I know this is a problem which has been addressed in many different ways for many years, looking at some early peak detector circuits.
I'm still planning on finishing my own version of a uC based peak detector which I have made good headway on. It will have presets for voltage division with either jumpers or digital POT as @pinholer suggested.
A buffer is a preamp without gain. Feed it 2V and its output will be 0.99 x 2V. Even without a buffer and without a preamp your sources have maximum output of 2V rms. There is no problem. There can be a problem when adding a gain block. An unnecessary gain block. Introduce it and get possible issues in return.
What can’t happen can’t happen. What can’t happen needs no monitoring or worrying.
What can’t happen can’t happen. What can’t happen needs no monitoring or worrying.