speaker phase tester for audio installer - Electronic Circuits Projects Diagrams Free

r10069 · 14th June 2004, 10:17 AM

i am a car audio installer and when trying to find out which loudspeaker wire belongs to which loudspeaker, we use a 1.5v battery with a red and a black wire soldered to the positive and the negative terminals of the battery respectively. when touching a pair of speaker wires for a short period (tap) it makes the speaker crackle and therefore you can identify which channel of the amplifiers output you need to use. when you can see the loudspeaker you can also determine polarity because when speaker moves outward the positive lead of the battery is connected to the positive terminal of the speaker, and if cone moves inward the wires are out of phase. the problem is when you cannot see the loudspeaker and need to remove lots of panels to do so, the job becomes complicated and the only way to find out correct phase is to hook up all the speakers and fade to each one and listen for loss of bass and a stereo wide effect. what i need to know is, would it be possible to build a circuit that can generate an audible tone and also light up an led corresponding to the phase in which the speaker has just travelled when injected with the tone? is there any way in which the voicecoils direction of travel can do this? pehaps the circuit can make a 800hz tone and light up a green led when in phase and a red led when out of phase. any help would be greatly appreciated.

audioguru · 14th June 2004, 01:01 PM

Hi, R.
I've been trying to measure speaker phase for years.
There are commercial testers out there, but their circuits are a top secret. I have not been able to take one apart and reverse engineer it.
You can't do it with a high frequency, because the phase changes with distance.
You might be able to do it with a low frequency, where the wavelength is more than the distance. But then most speakers, especially tweeters, don't reproduce such a low frequency.
I tried injecting a pulse into a speaker's amp, then extending the 1st rising edge of the sound, but a speaker's diaphram isn't a perfect piston and therefore the high frequency "tic" arrives first, followed by the low frequency "thump". Maybe that will work with a sharp Low Pass Filter, or using half of a low frequency sine-wave for the pulse.
My idea to detect the pulse and display its polarity is:
1) An amplified mic feeds a sharp LPF.
2) The filter feeds 2 rectifiers and non-retriggerable monostables, 1 for each polarity.
3) Each monostable drives an LED.

My ears are very sensitive to wrong phase. In department or stereo stores I always rewire a speaker to correct the phase, then turn the bass contol back down to normal. Sometimes radio stations broadcast the channels out-of-phase. The poor guys listening in mono don't hear any vocals, just echoes.

Gaston · 14th June 2004, 04:29 PM

whould it work if you used a low frequency tone generator ad placed a microphone in front of the left and right speakers and connected them to each channle of an occiliscope and look athe the waveforms?

stevez · 14th June 2004, 04:50 PM

I am presuming (feel free to correct me if I am wrong) that with no applied voltage the speaker coil sits more or less in the center of travel. If you apply (as you describe) DC the coil moves in or out which assists you in determining the phase.

When the speaker coil moves forward, in the direction of the speaker it would seem that the air in front of the speaker is compressed - the air behind is uncompressed or rarefied. The reverse is also true. Certainly when the power is removed the coil/cone returns to neutral which also might result in compression/rarefaction. If you applied a series of positive pulses - then applied a series of negative pulses - would they sound different (if you were directly in front of the speaker)? Is there enough difference to make out which way? I realize that an enclosure might mix all this stuff up - which is why I say directly in front of the speaker.

If you applied the pulses as described above, and places a mic or speaker in front of the speaker - then sampled the signal from the mic, would that look different on a scope? I was thinking that if you applied a square wave, the resulting waveshape from compression might look different, possibly sharper, that the waveshape as the speaker cone returns to neutral by virtue of the "spring" in the speaker. If they are different you might discover that the harmonic content of the signal from the sampling mic could be radically different depending on whether or not you pulsed positive or negative. You might be able to construct some simple filters to allow comparison of harmonic content as a way of differentiating phase.

I've tried to compress a lot into these paragraphs. I hope it makes sense.

Another thought was to apply a step pulse then hold (limit current so you don't ruin the speaker). Would you be able to see a piece of paper move out or in? If another small speaker were placed in front - the speaker ought to be pushed out - or drawn in, corresponding to the direction of the speaker cone being measured. If you pulsed then held - and had the electronics to note which polarity of voltage the sampling speaker might generate that would give you a clue. Going further - use a pulse that turns "on" quickly yet ramps down slowly to minmize the amplitude of the return.

Again, just some thoughts.

audioguru · 14th June 2004, 08:03 PM

Gaston,
Your idea will probably work, but we need a simple, portable solution, with an LED display.

Steve,
Since speakers are very linear, for low distortion, you can't hear the difference between a pulse that is coming or going.
Since most audio amps are not DC coupled, it is very hard to distinguish the polarity of multiple uni-polar pulses.

R,
Another idea is to do what your ears do:
1) Feed a low frequency tone to both channels, like Gaston suggested.
2) Have a single amplified mic placed between the speakers, driving an LM3915 dot-bar LED driver.
3) If the speakers are in phase, the LEDs show a large signal.
4) If the speakers are out-of-phase, the sound cancels, and the LEDs show very little signal.
But you can't do that with mid-range speakers nor tweeters, because the phase changes with distance.
You could probably even do that with program music, if you turn-up the bass and turn-down the treble.

spuffock · 14th June 2004, 08:54 PM

Light a match, hold it in front of the speaker, and connect your battery. See which way the flame jumps.
Sorry it's not electronic, but it's simple and portable, and cheap as well.

TillEulenspiegel · 14th June 2004, 09:41 PM

Hmm well since a speaker is a transducer that is a generator as well as a current to motion converter ( V--> air movement) you should be able to use a small battery to extend the cone and switch it off , then measure the polarity of the subsequent induced voltage........... Ill try to whip up a schematic.

spuffock · 14th June 2004, 10:23 PM

?
surely the induced voltage will be in the same direction as the battery?
I think?
Anyway,it won't tell you anything about the direction the cone moves.

TillEulenspiegel · 15th June 2004, 12:26 AM

surely the induced voltage will be in the same direction as the battery?
I think?
Anyway,it won't tell you anything about the direction the cone moves.

No actually a collapse of a field in an inductor usually produces a V at 180 degrees out of phase with the applied V. Thats why we put ballast resistors on car ignition coils and reversed biased diodes on relay coils.

No again, It will tell you the polarity of the induced V from the generator . Maxwell's equations describe this phenomenon.

JohnBrown · 15th June 2004, 07:41 AM

Quote:

Originally Posted by spuffock

?
surely the induced voltage will be in the same direction as the battery?
I think?
Anyway,it won't tell you anything about the direction the cone moves.

I agree with you(except that the voltage will be in the other direction).
The only thing it will tell you is which way round your battery is(or isn't actually), and you probably knew that already.

TillEulenspiegel · 15th June 2004, 07:17 PM

Researched a bit, instead of a DC pulse, the instrument puts out a 1 sec test pulse consisting of 3 (or so ) 300mS pulses at about A440 and measures the phase of the outgoing pulse against the output of a built in microphone and amp. The bad part is you must be a small distance away from the speaker. I was originally thinking of the DC pulse pushing the cone, measuring back EMF against a Mic that uses a weighted DB SPL scale. The dc will produce a low Fq signal- a thump- that would travel well over distance with less regard for phase shifting then a hi Fq signal. That would work for infinite or enclosed baffles but not ported I suppose for automobiles tho the commercial design would work fine since the distance factor is no problem. The cheaper designs probably utilize PLLs and not too much hard wired logic or any CPUs. Still looking.

john1 · 19th June 2004, 03:41 AM

Interesting.

I think that using the speaker as a sensor is the way
to go here.

How about a single transistor amp feeding a meter,
then give the front of the speaker cover a flick with
something to see which way the meter jumps ...

Its something that has been an annoyance in the past,
but i haven't thought about it till now.

Something simple to indicate the 'phase' of a speaker
would deserve a box of its own in my collection.

I might see about doing that.

John

spuffock · 23rd June 2004, 06:51 AM

A small speaker in a box with , say, a 555 timer and a fat FET to produce a single pulse of maybe 20 or 30 milliseconds. This will produce a single pressure pulse. The distance between this and the speaker under test will not affect the polarity of the pulse, only its strength and time of arrival.
All we need now is something to determine the direction of the first edge of the electrical pulse from the speaker under test. Op-amp, couple of comparators, dual d-type flipflop and a quad nand gate ought to do it.

john1 · 24th June 2004, 08:08 PM

Hi,

I have yet to try this,
but it doesn't look difficult,
i reckon a little whack with a newspaper
should be enough to make the meter jump.

Simple enough to swap wires,
and see if it jumps the other way.

Probably will need a bit of amplifying,
i'll see how it goes.

Regards, John

Fleabay · 28th November 2006, 10:48 PM

Just to let you know I own such a device you are speaking of, I also install car audio and have used it for years it works great, I did a Google search for it and I could not locate it. It's called speak-rite by Smart products. It works with a green led and a red led and has a mic at the top for detection. It has a red and black leads (probes) to test the wire and a rca hookup also.........I've owned this item before the internet was born.........well about 25 years ago...David

14th June 2004, 10:17 AM	(permalink)
r10069 New Member	speaker phase tester for audio installer i am a car audio installer and when trying to find out which loudspeaker wire belongs to which loudspeaker, we use a 1.5v battery with a red and a black wire soldered to the positive and the negative terminals of the battery respectively. when touching a pair of speaker wires for a short period (tap) it makes the speaker crackle and therefore you can identify which channel of the amplifiers output you need to use. when you can see the loudspeaker you can also determine polarity because when speaker moves outward the positive lead of the battery is connected to the positive terminal of the speaker, and if cone moves inward the wires are out of phase. the problem is when you cannot see the loudspeaker and need to remove lots of panels to do so, the job becomes complicated and the only way to find out correct phase is to hook up all the speakers and fade to each one and listen for loss of bass and a stereo wide effect. what i need to know is, would it be possible to build a circuit that can generate an audible tone and also light up an led corresponding to the phase in which the speaker has just travelled when injected with the tone? is there any way in which the voicecoils direction of travel can do this? pehaps the circuit can make a 800hz tone and light up a green led when in phase and a red led when out of phase. any help would be greatly appreciated.

14th June 2004, 01:01 PM	(permalink)
audioguru Experienced Member	Speaker phase tester Hi, R. I've been trying to measure speaker phase for years. There are commercial testers out there, but their circuits are a top secret. I have not been able to take one apart and reverse engineer it. You can't do it with a high frequency, because the phase changes with distance. You might be able to do it with a low frequency, where the wavelength is more than the distance. But then most speakers, especially tweeters, don't reproduce such a low frequency. I tried injecting a pulse into a speaker's amp, then extending the 1st rising edge of the sound, but a speaker's diaphram isn't a perfect piston and therefore the high frequency "tic" arrives first, followed by the low frequency "thump". Maybe that will work with a sharp Low Pass Filter, or using half of a low frequency sine-wave for the pulse. My idea to detect the pulse and display its polarity is: 1) An amplified mic feeds a sharp LPF. 2) The filter feeds 2 rectifiers and non-retriggerable monostables, 1 for each polarity. 3) Each monostable drives an LED. My ears are very sensitive to wrong phase. In department or stereo stores I always rewire a speaker to correct the phase, then turn the bass contol back down to normal. Sometimes radio stations broadcast the channels out-of-phase. The poor guys listening in mono don't hear any vocals, just echoes.

14th June 2004, 04:50 PM	(permalink)
stevez Experienced Member	I am presuming (feel free to correct me if I am wrong) that with no applied voltage the speaker coil sits more or less in the center of travel. If you apply (as you describe) DC the coil moves in or out which assists you in determining the phase. When the speaker coil moves forward, in the direction of the speaker it would seem that the air in front of the speaker is compressed - the air behind is uncompressed or rarefied. The reverse is also true. Certainly when the power is removed the coil/cone returns to neutral which also might result in compression/rarefaction. If you applied a series of positive pulses - then applied a series of negative pulses - would they sound different (if you were directly in front of the speaker)? Is there enough difference to make out which way? I realize that an enclosure might mix all this stuff up - which is why I say directly in front of the speaker. If you applied the pulses as described above, and places a mic or speaker in front of the speaker - then sampled the signal from the mic, would that look different on a scope? I was thinking that if you applied a square wave, the resulting waveshape from compression might look different, possibly sharper, that the waveshape as the speaker cone returns to neutral by virtue of the "spring" in the speaker. If they are different you might discover that the harmonic content of the signal from the sampling mic could be radically different depending on whether or not you pulsed positive or negative. You might be able to construct some simple filters to allow comparison of harmonic content as a way of differentiating phase. I've tried to compress a lot into these paragraphs. I hope it makes sense. Another thought was to apply a step pulse then hold (limit current so you don't ruin the speaker). Would you be able to see a piece of paper move out or in? If another small speaker were placed in front - the speaker ought to be pushed out - or drawn in, corresponding to the direction of the speaker cone being measured. If you pulsed then held - and had the electronics to note which polarity of voltage the sampling speaker might generate that would give you a clue. Going further - use a pulse that turns "on" quickly yet ramps down slowly to minmize the amplitude of the return. Again, just some thoughts. __________________ stevez

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

14th June 2004, 04:29 PM	(permalink)
Gaston Experienced Member	whould it work if you used a low frequency tone generator ad placed a microphone in front of the left and right speakers and connected them to each channle of an occiliscope and look athe the waveforms?

14th June 2004, 08:03 PM	(permalink)
audioguru Experienced Member	Gaston, Your idea will probably work, but we need a simple, portable solution, with an LED display. Steve, Since speakers are very linear, for low distortion, you can't hear the difference between a pulse that is coming or going. Since most audio amps are not DC coupled, it is very hard to distinguish the polarity of multiple uni-polar pulses. R, Another idea is to do what your ears do: 1) Feed a low frequency tone to both channels, like Gaston suggested. 2) Have a single amplified mic placed between the speakers, driving an LM3915 dot-bar LED driver. 3) If the speakers are in phase, the LEDs show a large signal. 4) If the speakers are out-of-phase, the sound cancels, and the LEDs show very little signal. But you can't do that with mid-range speakers nor tweeters, because the phase changes with distance. You could probably even do that with program music, if you turn-up the bass and turn-down the treble.

14th June 2004, 08:54 PM	(permalink)
spuffock Experienced Member	Light a match, hold it in front of the speaker, and connect your battery. See which way the flame jumps. Sorry it's not electronic, but it's simple and portable, and cheap as well.

14th June 2004, 09:41 PM	(permalink)
TillEulenspiegel Experienced Member	Hmm well since a speaker is a transducer that is a generator as well as a current to motion converter ( V--> air movement) you should be able to use a small battery to extend the cone and switch it off , then measure the polarity of the subsequent induced voltage........... Ill try to whip up a schematic.

14th June 2004, 10:23 PM	(permalink)
spuffock Experienced Member	? surely the induced voltage will be in the same direction as the battery? I think? Anyway,it won't tell you anything about the direction the cone moves.

15th June 2004, 12:26 AM	(permalink)
TillEulenspiegel Experienced Member	surely the induced voltage will be in the same direction as the battery? I think? Anyway,it won't tell you anything about the direction the cone moves. No actually a collapse of a field in an inductor usually produces a V at 180 degrees out of phase with the applied V. Thats why we put ballast resistors on car ignition coils and reversed biased diodes on relay coils. No again, It will tell you the polarity of the induced V from the generator . Maxwell's equations describe this phenomenon.

15th June 2004, 07:17 PM	(permalink)
TillEulenspiegel Experienced Member	Researched a bit, instead of a DC pulse, the instrument puts out a 1 sec test pulse consisting of 3 (or so ) 300mS pulses at about A440 and measures the phase of the outgoing pulse against the output of a built in microphone and amp. The bad part is you must be a small distance away from the speaker. I was originally thinking of the DC pulse pushing the cone, measuring back EMF against a Mic that uses a weighted DB SPL scale. The dc will produce a low Fq signal- a thump- that would travel well over distance with less regard for phase shifting then a hi Fq signal. That would work for infinite or enclosed baffles but not ported I suppose for automobiles tho the commercial design would work fine since the distance factor is no problem. The cheaper designs probably utilize PLLs and not too much hard wired logic or any CPUs. Still looking.

19th June 2004, 03:41 AM	(permalink)
john1 Experienced Member	Interesting. I think that using the speaker as a sensor is the way to go here. How about a single transistor amp feeding a meter, then give the front of the speaker cover a flick with something to see which way the meter jumps ... Its something that has been an annoyance in the past, but i haven't thought about it till now. Something simple to indicate the 'phase' of a speaker would deserve a box of its own in my collection. I might see about doing that. John

23rd June 2004, 06:51 AM	(permalink)
spuffock Experienced Member	A small speaker in a box with , say, a 555 timer and a fat FET to produce a single pulse of maybe 20 or 30 milliseconds. This will produce a single pressure pulse. The distance between this and the speaker under test will not affect the polarity of the pulse, only its strength and time of arrival. All we need now is something to determine the direction of the first edge of the electrical pulse from the speaker under test. Op-amp, couple of comparators, dual d-type flipflop and a quad nand gate ought to do it.