Need help for a simple speaker protection circuit based on voltage

Thread Starter

sam68

Joined Mar 2, 2017
34
I would like to protect my speakers when I use a lot of bass equalization. A slo-blow fuse is too slow and a fast one would blow all the time. My 4 to 8 ohm speaker voltages ranges from 0 to 5 volts.. Could a circuit be designed based just on voltage and when it hits a threshold, it moves the signal over to a resistor?
1. not effect the signal
2. be adjustable from .1 volt to 4 volts
It would be neat to have a meter attached but not necessary
3. when that voltage is reached, it cuts out to a 10 watt resistor.
4. non pc board, just wired

The idea is to listen to a raw speaker driver's mechanical limit, adjust the circuit to cut out. There are boards on Ebay (with relays and??) but they don't say how fast they cut out which is critical.
I'm not even at the level of a novice so you will need to hand hold me if you want to help.
thanks
 

wayneh

Joined Sep 9, 2010
16,102
Items 1 and 3 are mutually exclusive. If you clip the voltage waveform, that's a form of distortion and will affect the output. You could possibly filter out the very low frequency data that the speaker cannot reproduce and/or you can't hear anyway. That would remove a lot of power from what the speaker sees without losing audio data. It might actually improve the output.
 

Thread Starter

sam68

Joined Mar 2, 2017
34
I wasn't thinking to continue with the signal, just reroute it to the resistor therefore the speaker would go quiet. When the volume (voltage) is turned down, the signal is back to the speaker. I would assume this would happen in milliseconds. Would that work?
 

Reloadron

Joined Jan 15, 2015
4,508
If you look at an audio signal using software like Audacity from source forge we see that the audio is a constantly changing voltage signal. You could monitor the audio looking only at the base for example and use a chip like the LM 3915 Dot Bar Display Driver (0 to -27 db) and use one of the output lines to latch a protection circuit at some preset level. That or as wayneh suggest, clip the waveform which I like. Determine a level and when the signal exceeds that level "do something". :)

Ron
 

#12

Joined Nov 30, 2010
18,076
A slo-blow fuse is too slow and a fast one would blow all the time.
That's not how it works with my Onkyo 50 watt amplifier. It has 3 amp fast fuses to protect the speakers and they have not blown in 39 years. You are trying to fix an imaginary problem.
Could a circuit be designed based just on voltage and when it hits a threshold, it moves the signal over to a resistor?
That's a lot harder job than just limiting the power to the speaker.
Audio waves have both positive and negative voltages. No matter what their peak voltage is, they spend most of the time in between the peaks. That means the only way to control their amplitude is to measure the peaks and make a decision based on the peak voltage. At 5 volts into 4 ohms or 8 ohms, you are talking about 3 watts to 6 watts, peak. Or about 1.6 to 3.2 watts RMS. Why not just buy a speaker that can survive at least 4 watts? They aren't anywhere near as expensive as designing a protection circuit.
There are boards on Ebay (with relays and??) but they don't say how fast they cut out which is critical.
A relay operates in 5 to 10 millisecond after it has been given the signal to move. That's around one full cycle of 100Hz to 200Hz.
 

Thread Starter

sam68

Joined Mar 2, 2017
34
That's not how it works with my Onkyo 50 watt amplifier. It has 3 amp fast fuses to protect the speakers and they have not blown in 39 years. You are trying to fix an imaginary problem.

That's a lot harder job than just limiting the power to the speaker.
Audio waves have both positive and negative voltages. No matter what their peak voltage is, they spend most of the time in between the peaks. That means the only way to control their amplitude is to measure the peaks and make a decision based on the peak voltage. At 5 volts into 4 ohms or 8 ohms, you are talking about 3 watts to 6 watts, peak. Or about 1.6 to 3.2 watts RMS. Why not just buy a speaker that can survive at least 4 watts? They aren't anywhere near as expensive as designing a protection circuit.

A relay operates in 5 to 10 millisecond after it has been given the signal to move. That's around one full cycle of 100Hz to 200Hz.
 

Thread Starter

sam68

Joined Mar 2, 2017
34
I appreciate the feedback and you'll know so much more than I do about circuits. I may not have explained it well so let me try again.
I have already bought a bunch of speakers so I would like to work with them. I only used 5 volts as an example but many of my speakers are rated to handle hundreds of watts which means the voltage could be 50 ac volts or so.
In my mind, I'm not trying to "control their amplitude" but as soon as a positive voltage hits an adjustable voltage, the circuit would send the signal to a resistor. Obviously, I don't know anything about circuits: I just assumed that there is some electronic part that would monitor voltage and cut the signal or could route it somewhere else. If not, no need to reply and I will figure out something else.
thanks
 

Thread Starter

sam68

Joined Mar 2, 2017
34
Would the bulb effect the quality of the signal? Also, wouldn't the bulb lower the volume across all frequencies which really isn't what I want to do. Usually the higher voltage is in the deep bass range and that is where a small speaker will hit its mechanical limit especially with bass boost. But maybe you are right - how would I find the right voltage / wattage bulb?
 

Thread Starter

sam68

Joined Mar 2, 2017
34
Since I have many speakers and amps, the voltages/power will vary. That is why I thought we needed something to adjust the voltage. The way I would adjust the voltage would be to hook up the speaker outside the box, play the lowest sine wave with the bass boost I want, turn up the volume until I hear the voice coil hitting its limit, then adjust the "pot" to mute the speaker by moving the signal to the resistor which protects the amp. Seems somewhat simple to me BUT "nothing is impossible to the person who doesn't know what he is talking about."
 

blocco a spirale

Joined Jun 18, 2008
1,535
I think you would find that, if you were to build such a circuit, you wouldn't be able to set it as precisely as you imagine and for it to operate reliably it would have to be set at a lower threshold than would be acceptable. i.e. it would cut in at levels that you perceive to be safe.

If the amplifier is supplying a "legitimate" audio signal and the loudspeaker is hitting its mechanical limits the solution is to use a better loudspeaker. Loudspeakers are only reasonably linear of a small range of movement (+/- a few mm) so operating one at close to its mechanical limits is likely to put it well outside of its designed specification and, apart from sounding bad (very audible distortion), it probably won't last very long.
 
Last edited:

wayneh

Joined Sep 9, 2010
16,102
I'll throw out an idea. I think this circuit would accomplish the goal. It's a shunt regulator. R4 is the speaker. The rest of the circuit does nothing unless the zener D1 is forced into conduction by a voltage above it's value.

The set point voltage is established by choosing D1, R1 and R2 to give 0.7V on the transistor base when at the set point voltage. The sum of R1 and R2 is enough to protect the diode against over-current. R1 + R2 could probably be, for instance, a 200Ω pot. I think you might use a cheaper (lower power) pot if you replace Q1 with a darlington and then target 1.4V at the base. R3 protects the transistor as necessary.

Note that this circuit increases the load on the amp by lowering the speaker's impedance, but that's what you described as the goal.

Screen Shot 2017-03-03 at 9.49.06 AM.png
 

Thread Starter

sam68

Joined Mar 2, 2017
34
Thank you wayneh,
OK, I, being less than a novice, and the first time I have read and want to build a circuit, I have some questions if you would be so kind to answer:
0. I'm not sure about you last comment "lowering the speaker's impedance". I didn't mean to imply that: what I meant was: The circuit will "watch" for a threshold voltage, and when it sees it, it routes the full signal to a 10 watt (or more) resistor (speaker is now mute until that short time passes when the circuit sees less than the threshold, thereby, routing the signal back to the speaker. Unless a sine wave signal, it might only cutout less than a second of music.) - Will the circuit above do that? Now that I look at it, I don't see the 10 watt resistor which is needed so the amp sees a load all the time.
1. I want it to be adjustable from .1 to 50 or so volts: I will take your suggestion to replace R1-R2 with a Pot. Would this be the one to buy:
2 x 2K OHM Linear Taper Potentiometer Round Shaft Solder Lugs - USA Seller
2. What would be the specs for the R3 (I see the 5 but what does that mean?) Does the 8 by the R4 mean 8 ohms? Some of my speakers are 4 ohms and if I hook them in parallel, then 2 ohms or sometimes, 16 ohms. Will that change any of the specs?
3. What are the specs for D1 and Q1?

4. Where could I put two voltage meters in the circuit? one= 0 to 5 v and the other 0 to 50 like this one AC 5V Analog Panel Voltmeter Volt Voltage Meter Gauge 85L1 Class 2.5 AC 0-5V I would rather have analog vs digital to see what is happening.

Wayneh, you are nice to spend the time with a neophyte like me. I volunteer at a public school 2 times a week maintaining 100+ laptops and develop free software for the teachers so I appreciate when others help me.
Thanks
 

recklessrog

Joined May 23, 2013
988
I really can't see why you want to dump the power into a resistor, why not use a simple compressor and if you only want it to work on low frequencies, use a low pass filter. It is the more accepted way of limiting power to speakers. Last year I fitted one into my bass players 300 watt amplifier, It has a bargraph display and allows limiting from 1 watt right up to full power.
With Very high power P.A systems, they are mandatory unless you want to spend hundreds of £'s ($'s) on replacement cones, especially the high frequency drivers.

I live in the U.K and Maplins sell a kit for under £20.00
 

wayneh

Joined Sep 9, 2010
16,102
Will the circuit above do that? Now that I look at it, I don't see the 10 watt resistor which is needed so the amp sees a load all the time.
No, and I've probably misunderstood your goals. I didn't know you wanted to shunt the entire signal. My shunt idea is probably not a good strategy for this application, because it assume that sending power in to the shunt will reduce power to the load. This would be true if the power source was, say, a solar panel, but a powerful amplifier would simply drive the speaker and the shunt and the shunt would not affect the load to the speaker.
1. I want it to be adjustable from .1 to 50 or so volts: I will take your suggestion to replace R1-R2 with a Pot. Would this be the one to buy:
2 x 2K OHM Linear Taper Potentiometer Round Shaft Solder Lugs - USA Seller
http://www.ebay.com/itm/2-x-2K-OHM-Linear-Taper-Potentiometer-Round-Shaft-Solder-Lugs-USA-Seller-/222359354092
I'll answer, despite the likely dead end for this approach. For such a wide range, you'd want a log taper (also called audio taper). The shunt transistor should be a darlington if you're going to use a 2KΩ pot. The current to the base won't be enough to drive the transistor with a that high a resistance. That's probably a good idea anyway.
2. What would be the specs for the R3 (I see the 5 but what does that mean?) Does the 8 by the R4 mean 8 ohms? Some of my speakers are 4 ohms and if I hook them in parallel, then 2 ohms or sometimes, 16 ohms. Will that change any of the specs?
All the numbers I show are guesses, and are in ohms. The current through Q1 has to be limited to some safe value and that is the function of R3. It would potentially be dissipating a lot of power and needs to be rated as such. R3 might not be necessary at all with a small amplifier, in which case Q1 dissipates all the power. And yes, this circuit would be matched to a particular load. The specs would change with the speaker.
3. What are the specs for D1 and Q1?
For the wide range of voltages you want, this won't work. D1 would be chosen for a particular target voltage. The resistors R1+R2 (a pot) only tweak that. You'd need a different D1 for each voltage

I think you need a different approach. This problem has been solved before and using the solutions others have invented is a good idea.
 

Thread Starter

sam68

Joined Mar 2, 2017
34
Thank You Wayneh for hanging in there. I notice that others gave their opinion, which was nice, but you were the one that took a stab at it. It will be a small amp, probably not over 20 Rms but I would like to design the circuit to handle a 100 or so rms. I realize that the idea is not as good as a real compressor/limiter but I don't need that sophistication , just something to cut off the voltage. I'm willing to give it a try if you think you have the time to help me. Just to clarify:
1. Voltage is going to the speaker and at some point, too much will cause the speaker to reach its maximum excursion (Xmax)
2. While one of my many 2, 4 8 or 16 ohm small low power bookshelf speakers is out of the box, I will use a sinewave and bass boost, and gradually increase volume until I hear that maximum soft "thump". I will note the voltage using the voltage meter (optional) and move the volume control back a little.
3. Then I will turn the audio taper pot until the pot transfers the signal from the speaker to the resistor. I will then turn the pot about 10% further to "lower" the threshold voltage to kick in a little sooner. Or I can turn the pot all the way so no music comes out, then turn it until I hear music and keep turning a little further.
4. Since I noted the voltage for that speaker, then when I use that speaker again, I will know where to set the voltage.
5. When listening to the music, the low end will cut out for milliseconds while the resistor is handling the load of the amp.

What do you think, is it doable?
Thanks
3. Then in the future, when that voltage
 
Top