Polarity testing for individual loudspeaker drivers is usually easy enough if they are not installed in an enclosure with a crossover wired in, but it becomes much more difficult if you want to verify that they are correct once wired up and inaccessible. A 1.5V cell can be used with (almost) any moving coil driver, and if the positive terminal of the cell is connected to the positive terminal of the speaker (and negative to negative of course), the cone will move outwards (air compression) with correct polarity.
This works with most tweeters as well (the power is under 300mW for an 8 ohm tweeter), but is obviously not possible with ribbon tweeters. Nor is it possible (at least not without an oscilloscope) to verify that the entire signal chain retains the correct polarity. While it's not essential to have a signal that maintains the 'correct' polarity (it's generally inaudible, despite claims to the contrary), it is considered to be the 'right' thing to do.
It can be especially difficult to verify that microphones are correct. Most are, but you generally have no way of knowing for sure. In some cases, it may be due to a miswired mic lead, or the mic itself may have been 'repaired' by someone who was careless about polarity. This is particularly important when two (or more) mics are used on a single instrument, a common practice with pianos and some other larger instruments. An out-of-phase microphone will make the sound very un-natural, with reduced bass output and/ or a 'strange' sound stage (for stereo setups). During testing I discovered that a 'high quality' electret mic (with internal battery) I have is out of phase. I never knew that before, so it shows that testing is worthwhile.
Archimago's Musings: Measurements: Speaker Impedance
The project shown here is fairly simple, and uses the most common signal for this kind of test - a positive pulse, which can be as basic as a simple push-button to discharge a capacitor into a speaker (but there are caveats to this as seen below), or a train of positive pulses with a repetition rate of around 1 - 2Hz or so. Whether you need the continuous pulse provision or not is up to you. If you can get a suitable push-button switch (see below to find out what is 'suitable'), the single pulse option works well and saves a few parts.
There are very few polarity tester circuits described anywhere on the Net, despite an obvious need - especially for live sound and recording environments. There are several you can buy, but some are surprisingly expensive. The polarity of mics can be changed simply because of an incorrectly wired lead or a repair, and reversed mics can play havoc during live sound or recording sessions. Many mixing consoles have a 'reverse phase' switch (by whatever name), but it's always helpful if you know in advance that the 'normal' position really is 'normal'.
Polarity testing is often required with car audio installations as well. The wiring to various speakers may not be colour coded, and it can be difficult to work out which is which. Both leads usually sit at ½ battery voltage (nominally 6V), and neither can be connected to the vehicle's chassis without damaging the amplifier. A simple positive repeating pulse signal can be recorded on a medium that's supported by the car's inbuilt entertainment system. The detector can then be used to determine the polarity of each speaker in the system. For decent sound, the speakers should all be in phase.
How To Test Speaker Polarity & Does Speaker Polarity Matter
Naturally, the pulse generator ('transmitter') is the fist thing needed. This can be in a separate box if you prefer, but mostly it's better if it's in the same box as the receiver (who wants to mess around with two separate boxes?). There are two choices - either a single pulse initiated by a push-button, or an oscillator that provides very narrow (less than 1ms) pulses at a suitable repetition rate. Both are shown below. The opamp used is an LM358 - these are low cost, low power types that allow the output to go to 'ground' (zero volts, within a few millivolts). There are others that may be suitable, but they will be harder to get and usually considerably more expensive. The LM358 is not a hi-fi opamp (I don't recommend it for any circuit that handles audio signals), but it's perfect in this application.
The 'power on' LED (LED3) is shown with a 2k2 current limiting resistor, but if you use high brightness LEDs throughout the series resistor can be increased. Up to 10k is fine with most high brightness LEDs, and this will reduce the current. With the 2k2 resistor each LED (see Figure 3 for the other two) each will draw around 3mA, so even with the 2k2 resistors the total current drain will still be less than 10mA in normal use. Note that all diodes (in all circuits) are 1N4148 or similar.
The simple version simply charges the capacitor (C4) and it will reach close enough to full charge in about 2.5 seconds. When the button is pressed, C4 is discharged into the into the internal (used for mic testing) or external speaker. VR1 provides a variable level signal to the line output connector. During testing, I found that one thing that is absolutely critical is the switch. It must make perfect contact each and every time, with no contact bounce. This is a great deal harder than it sounds, and I tried several different switches, with at least two of them often providing a negative output on the oscilloscope when I knew perfectly well that it should have been positive. This might seem improbable, but a speaker is a reactive load, and even a few microseconds of contact bounce has a huge effect on the results you see. Consequently, I can't recommend most push-button switches because they may be erratic... but only some of the time (which is worse!).
How To Adjust The Phase Of The Subwoofer? ― 130.com.ua
In particular, avoid 'snap-action' switches (especially those that you might expect are specifically intended for this type of application!). Toggle switches are generally no better, although one (very old and overly large) worked reasonably well. Mini 'tactile' switches (as seen in the inset in Figure 1) seem to work well, despite the snap action which is reason for the 'tactile' nomenclature. Unfortunately, you'll have to be a bit inventive to mount this style of switch. 'Ordinary' push-button switches that close with no 'snap' should also be fine. These rely on simple metallic contact and close at the same rate as you press the button. A firm press ensures good contact with no contact bounce.
R23 (10Ω) has been included because without it, the pulse will be far too long as C4 would be discharged into 10k (around 2.2s discharge time, instead of a more sensible 2ms with 10 ohms). I recommend that you use an oscilloscope to measure the signal going to a loudspeaker... not the resistor, as that isn't reactive and you may not see the signs of problems. Switching must be clean - a single pulse with no 'hash' at the beginning which is indicative of contact bounce. It may be necessary to test a few different switches until you find one that works reliably... every time!
The following circuit may be preferable for a variety of reasons. In particular, there is no issue with switch contact bounce. It sends a continuous stream of pulses that should make setup easier. The pulse duration is consistent (around 2ms), regardless of load impedance. In either case, ensure that the line level control (VR1) is set to minimum before connecting it to an amplifier or preamp input, because the pulse is a fairly high level (at least 6V peak!).
Phase Switch Doesn't Solve Phase Problems!
This version generates a pulse train with a repetition rate of about 0.6 seconds (1.67Hz), which also drives the speaker and 'line' output connector. This is a good option, because you don't need to worry about contact bounce with a simple switch, and the output pulses are of entirely predictable duration and polarity. Of course, it's also more complex which will deter many constructors. Although I did build it, I'd be perfectly happy with the version shown in Figure 1, which I used for all initial testing.
I have to leave the internal speaker to the constructor. It will typically be a small (around 50mm diameter) type, and the availability of suitable types is somewhat variable, depending on where you live. Something that's easy to get here may not be available elsewhere (and vice versa). The impedance isn't particularly important, but less than 8 ohms is not advised. One option that I tried is a small car tweeter (these are often available as an after-market accessory, shown in Figure 9). The one I used is small, about 35mm total diameter, and it works very well (see Figure 9). The high frequency pulse is sufficient to get a 40mV peak from a run-of-the-mill dynamic microphone.
The connector used for the 'line level' output depends on the type of system, so will typically be
Reverse Polarity Speaker Cabinet Wiring
In particular, avoid 'snap-action' switches (especially those that you might expect are specifically intended for this type of application!). Toggle switches are generally no better, although one (very old and overly large) worked reasonably well. Mini 'tactile' switches (as seen in the inset in Figure 1) seem to work well, despite the snap action which is reason for the 'tactile' nomenclature. Unfortunately, you'll have to be a bit inventive to mount this style of switch. 'Ordinary' push-button switches that close with no 'snap' should also be fine. These rely on simple metallic contact and close at the same rate as you press the button. A firm press ensures good contact with no contact bounce.
R23 (10Ω) has been included because without it, the pulse will be far too long as C4 would be discharged into 10k (around 2.2s discharge time, instead of a more sensible 2ms with 10 ohms). I recommend that you use an oscilloscope to measure the signal going to a loudspeaker... not the resistor, as that isn't reactive and you may not see the signs of problems. Switching must be clean - a single pulse with no 'hash' at the beginning which is indicative of contact bounce. It may be necessary to test a few different switches until you find one that works reliably... every time!
The following circuit may be preferable for a variety of reasons. In particular, there is no issue with switch contact bounce. It sends a continuous stream of pulses that should make setup easier. The pulse duration is consistent (around 2ms), regardless of load impedance. In either case, ensure that the line level control (VR1) is set to minimum before connecting it to an amplifier or preamp input, because the pulse is a fairly high level (at least 6V peak!).
Phase Switch Doesn't Solve Phase Problems!
This version generates a pulse train with a repetition rate of about 0.6 seconds (1.67Hz), which also drives the speaker and 'line' output connector. This is a good option, because you don't need to worry about contact bounce with a simple switch, and the output pulses are of entirely predictable duration and polarity. Of course, it's also more complex which will deter many constructors. Although I did build it, I'd be perfectly happy with the version shown in Figure 1, which I used for all initial testing.
I have to leave the internal speaker to the constructor. It will typically be a small (around 50mm diameter) type, and the availability of suitable types is somewhat variable, depending on where you live. Something that's easy to get here may not be available elsewhere (and vice versa). The impedance isn't particularly important, but less than 8 ohms is not advised. One option that I tried is a small car tweeter (these are often available as an after-market accessory, shown in Figure 9). The one I used is small, about 35mm total diameter, and it works very well (see Figure 9). The high frequency pulse is sufficient to get a 40mV peak from a run-of-the-mill dynamic microphone.
The connector used for the 'line level' output depends on the type of system, so will typically be
0 Response to "Speaker Phase Test Guitar"
Posting Komentar