Some preparation before you start testing
Whether you are commissioning, fault finding or maintaining a PAVA system you need to check whether the loudspeakers and cabling is correct. To do this you’ll need your:
- Impedance meter, Zircle or another
The following are always worth doing before you start testing:
- Estimate the power (load) you expect on each loudspeaker circuit, more on doing this below
- If your batteries are low on one or both of your meters, change them, you can’t trust those readings otherwise.
How to estimate the loudspeaker circuit load
Count the number of loudspeakers. If all are tapped the same multiply by that tapping. If some are tapped differently then you’ll need to do a bit more adding up.
Remember it’s an estimate.
25 ceiling loudspeakers each tapped at 2 watts
25 x 2 W = 50 W.
- 10 ceiling loudspeakers tapped at 2 watts
- 10 wall cabinets tapped at 5 watts and
- 5 music horns tapped at 20 watts.
10 x 2 W + 10 x 5 W + 5 x 20 W = 170 W.
The 7 tips
Important: Disconnect the loudspeaker circuit from the amplifier. Otherwise your readings will be odd and if the amplifier’s powered up you are risking your meter.
Connect the Zircle (or other impedance meter) test leads to the loudspeaker cable at the start of the loudspeaker circuit.
Note: Polarity is not important although it is good practice to test all the circuits with the leads in the same orientation.
- On the Zircle press and hold down the TEST button
- On other meters you’ll need to choose the test range and possibly the test frequency (use 1000 Hz)
- Wait about 1 second
- Note down the impedance readings
- With the Zircle note down the watts reading too
- For other impedance meters use the equation in Tip 2 to calculate the watts.
Zircle does this for you, but for other impedance meters or if the loudspeaker circuit is not 100 volts read on.
You need to know:
- The voltage of the loudspeaker circuit. Often 100 volts, but also 70, 50 or even 25 volts
- The impedance meter reading you just made.
The power is calculated by squaring the voltage, then dividing this by your meter reading.
Power (W) = voltage2 / impedance meter reading
If your mental maths calculations are different to that which you measured, then something is wrong. If you are confident your maths is correct and your meter is reading correctly then the following might be helpful:
- If the measured power is high, check for short circuits
- If the measured power is low, check that the circuit is not broken anywhere.
Hint: If you can hear the tone through the loudspeakers then leave your impedance meter connected and walk round,
listening to all the loudspeakers on that circuit.
Where the tone stops is probably the location of your problem.
- Check the dc resistance (using a multimeter) between an electrical earth and each conductor; it should be open circuit – more in Tip 5 below
- If the cable has a third wire make sure that it is not connected by accident anywhere.
If all that does not resolve the problems then double check you are connected to the right loudspeaker circuit.
If you aren’t using a Zircle:
- are you squaring the voltage before dividing it by the impedance?
- are you using the correct frequency?
4 common faults that can derail your efforts:
- Loudspeakers sometimes have tappings for different voltages as well as powers; make sure the loudspeaker tapping is right for the voltage of the circuit
- Make sure all your loudspeakers have transformers; a loudspeaker without a transformer is low impedance and won’t work on a 100 V, 70 V or 50 V system
- Make sure there is no amplifier connected to the circuit it will have an effect whether or not it is switched on
- For a dc monitored loudspeaker circuit check every end-of-line resistor is connected and you have the correct number of resistors for that circuit.
If the results are still wrong give us a call on 01273 034630, we’ve find just speaking to someone else can trigger you in to finding the issue.
Because every 100 volt loudspeaker has a transformer fitted, the multimeter dc voltage will pass straight through that transformer. It’s similar to you shorting out your meter probes – pretty unhelpful.
Note: Loudspeaker circuits with dc blocking capacitors in each loudspeaker would be seen differently by a multimeter. This means you can do a handy extra test, see Tip 5 and 6.
However, you can use it for one helpful test: checking the resistance between each leg and earth.
To do this test find some metal that you know is earthed.
Then with your meter set to read resistance touch one of your probes on this earth point and the other probe on one of the loudspeaker wires.
Your meter should read open circuit or at least 50 000 ohms (50 kohms).
Anything below this means something is amiss. Possible issues are:
- Water (particularly if the reading keeps changing).Check junction boxes and any kit or cabling outside
- Loose wire: if the reading is close to zero ohms then its likely a wire has worked free and is touching some
metal. Ceiling loudspeakers are notorious for this.
As mentioned above a dc voltage passes straight through a loudspeaker transformer like it isn’t there. That’s because it provides little resistance, the dc voltage sees the transformer as a short circuit.
Can a dc voltage damage a transformer?
If there’s any power behind that dc voltage then that short circuit will generate high currents. This will heat the transformer wire and as those wires are thin, they act like a fuse and melt. Which is the end of that transformer.
This suggests dc monitoring is a bad idea
Not at all. By fitting a capacitor in each loudspeaker, you can block the dc from passing through the transformer. The diagram below shows each loudspeaker with a capacitor fitted in series on one leg of the 100 V transformer.
You’ve now got one of the best forms of loudspeaker circuit monitoring
dc monitoring works reliably with any type of cable. Whereas ac monitoring has to be suited to the cable type.
With ac monitoring:
- MICC type cable may not pass higher frequencies, so 20 kHz (ultrasonic) fault monitoring can be problematic
- FP type cable is susceptible to ac voltages being induced on circuits running parallel to each other. This can fool the fault monitoring detection in to thinking a faulty circuit is OK.
The blocking capacitors mean that the monitoring circuit would see an open circuit. However, by adding a resistor on the end of the circuit that’s across the wires we can measure that end of line resistor from the start of the circuit.
To monitor the whole loudspeaker circuit, that resistor must be at the end of the main run and on the end of any branches too.
Now the resistance reading you see on your multimeter will tell you how many resistors are fitted.
The manufacturer requires a 10 000 (10 k) ohm resistor be used.
You connect your multimeter to the loudspeaker circuit at the rack (remember to disconnect the amplifier) and it reads 2500 ohms.
As the resistors are all in parallel you can divide the one resistor value (10 000 ohms) by this 2500 ohm reading which means there are four resistors.
In this example this suggests there is one main run and three branches or spurs.
Are you thinking that’s a bogus assumption?
Well you’d be right.
- doesn’t tell you if any branches have become disconnected
- doesn’t tell you if connected branches have no resistor fitted
- doesn’t tell you if two resistors have been used on the same branch
- doesn’t confirm that those resistors are really at the end.
The following are also possible but less likely assumption errors:
- There is just one 2500 ohm resistor fitted
- There’s a fault that just happens to cause the resistance to read 2500 ohms.
However, all these assumptions are equally true for all unknown circuits, whether dc blocked or not. You are just more in the dark when dealing with circuits without capacitors fitted.
Go and break the circuit midway along its length and measure the resistance in both directions
Now you read 3333 ohms in the direction of the amplifier and 10 000 ohms to the end of the line.
This tells you there are three resistors nearer to the amplifier and just one in the other direction.
Do this test without blocking capacitors and all you’ll see is a short circuit in both directions. That could be a fault or it might mean its healthy.
A short circuit on a loudspeaker circuit where all the loudspeakers are fitted with capacitors means you know you have a fault. The only time it might not be a fault is if one or more of the loudspeakers have no capacitor fitted, although that is still a fault.
You’ll get even more insights by breaking the circuit again
The important point is that in less time you will be able to find the cause of a fault. You can then quote for the repair confidently, that pleases your client and increases your income.
There are two general ways to monitor a loudspeaker circuit using an ac or dc voltage.
Most monitoring system use an ac voltage, to them a capacitor is invisible and dc monitored systems need the capacitor.
You’ll find the price difference between those loudspeakers with and without capacitors is tiny often they are the same.
5 advantages to using loudspeakers fitted with
- You don’t need to stock two types of loudspeakers
- Fault finding a loudspeaker circuit is similar to fault finding a fire alarm loop
- You have two ways to fault find and test loudspeaker circuits: using your multimeter and impedance meter
- If you are called out and don’t have your impedance meter with you, you stand a better chance of finding
- Only using one type of loudspeaker means you can buy in larger quantities increasing your discount.
Remember without capacitors in the loudspeakers your multimeter is close to useless for fault finding, testing and commissioning. You will need an impedance meter.
You still need an impedance meter for loudspeaker circuits where loudspeakers are fitted with capacitors, but your multimeter makes everything easier and quicker.