On the other hand there are times when I get too clever for my own good. It happened the other day as I was inspecting on a service visit. I "caught" the young man taking readings across all the detection circuits without removing the wires. Needless to say I gave him a long lecture about not taking resistance readings with the wires still connected to the control panel. The young man listened in silence and when I had finished and was standing there feeling all smug and self important he casually pointed out that he wasn't taking resistance readings ... he had the meter set to voltage and was reading the voltages instead. Well, that was me knocked off my pedestal.
Here was a new trick on me so I asked about it. The idea was passed to the young man by the technical department of the manufacturer of the control panels he was using. The idea is simple enough. By checking the voltage at either side of the detection circuit it is possible to pick up on a high resistance circuit without having to go through all the routine of taking wires out and measuring the resistance. This has got to be one of the most useful little tricks I have come across in some time. But let's take it from the beginning. BS4737 requires us to take (and record) a resistance reading on all detection circuits at the time of installation. What it fails to do is to tell us why, basically because it is not the job of BS to teach, only to lay down the requirements. I find that a great many installers do not know why we take the readings so that is the best place to start. The idea is to check that we have installed the system correctly instead of dumping the problem on the unsuspecting customer, it works like this: Alarm installation cable has a normal resistance of about 8 ohms per 100-metre loop so if, for example, we install a magnetic door switch at the end of about 20 metres of cable we should have a good idea of the resistance we should find before we measure the circuit. A 20 m loop of cable should give us a reading in the region of 1.6 ohms plus the nominal resistance of the reed switch (around 0.1 ohms). Put them together and we should have an ideal reading of 1.7 ohms. Remember, though, the readings are never fully accurate because the resistances vary from roll to roll and reed to reed. What we are doing is to check or verify that the circuit is a good circuit without obvious defects before putting it into service.
Check for defects
The fire brigade has a whole book full of formulae to work out the requirements of things like how much foam will they need to put out an oil fire of a certain size. Or how long a tank of water will last when using a jet of a specific size. But, obviously, when faced with a raging fire and the flames are flickering round his nether regions the last thing the fireman wants to do is get out the pocket calculator and a paper and pencil to work out the details. He wants answers and he wants them quick. So for that reason the fire brigade used to "round off" the figures to the nearest whole numbers to enable the fireman to calculate the figures in his head whilst "on the run" so to speak. These became known as "fireground procedures".
To make life easier we could adopt some of our own "fireground procedures". We could round the resistances up to 10 ohms per 100m loop. Which would mean we could reckon on 1 ohm for every 10m of cable and 0.1 ohm for every reed switch, so that now, by the time we have put the meter to the wires, we have already worked out in our minds what we should see. Common sense now tells us that if we get a reading of 10 or 20 ohms then we have a bad join or some other high resistance defect somewhere. This is where we sort out the good guys from the cowboys. The good guys will re-trace the wire and find the fault, the cowboys will not even bother to take the readings and leave the control panel to find the fault by going into alarm sometime in the future ... usually two or three years later after the wiring has festered and the guarantee has run out. At this point they tell the customer that the device is worn out and needs replacing and cheerfully charge the customer for a fault that should have been found and rectified on installation. My vote is to bring back hanging!
The next bit of confusion arises at the time of service. A lot of installers think that every circuit has to be checked for resistance at every visit as a BS requirement. This is not true. Once we have verified that the circuit is sound we record the readings so that if we get a fault in later years we can compare the readings to help diagnose a high resistance fault (or at least eliminate one).
It is generally accepted that by taking out the wires and replacing them we actually create more faults than we find so probably for that reason it was not made a requirement of BS to re-take the resistance readings every service visit. On the other hand most good service engineers see the value of finding bad circuits before they become a false alarm so they do the readings every service visit, and this is where the problems start.
Taking all those readings is long, drawn out and tedious, so many service engineers fall into the trap of taking the readings without removing the wires from the control panel and that is totally unacceptable.
For a start the readings are guaranteed not to be accurate and you can damage both your meter and the control panel. To understand this you have to accept that your trusty meter cannot "read" resistance, it has to calculate it. The meter calculates resistance by squirting a small, but accurate voltage and current down one leg of the circuit. It then measures the voltage and current coming back on the other leg and calculates the resistance by comparing the voltage drop. We could say that our trusty meter is a pocket calculator for Ohms Law.
The control panel in the meantime is trying to do the same thing as the meter. It is squirting a known voltage and current down one leg of the circuit we installed and comparing it with what comes back up the other leg. It then acts on the results of those findings, i.e. activation. If we now add our meter to the circuit by trying to read the resistance of the circuit with the circuit still connected, we have two devices trying to feed voltage and current down the circuit at the same time. Result – a false reading and very possibly damage to either the meter or the control panel. If any of you lads reading this have been in the habit of taking reading this way you may have noticed that if you get the meter probes the wrong way round the readings go haywire. That is because of the voltages being applied from two different sources. DO NOT go out and check it ... take my word for it. Tricks like that give David Grant (the ACT Meters faultfinding and circuit-testing Guru) nightmares. Now for the salvation ... By reading the voltages instead of the resistance we may well have a quick answer to our problems. If you set your meter to read 20volts DC, place one probe on the 0 volts of the Aux power supply and then place the other probe on either side of the circuit in turn we can read the voltage at the start and the end of the circuit.
Quick answer to problems
The difference between the two will give us a strong indication if there is a high resistance that shouldn't be there. If the circuit is open (say the door is open) you will get around 5-6v at one side and almost nothing at the other. If the circuit is closed then you should get almost similar voltages (around 4 - 6v) at both sides. If there is a bad connection, i.e. a high resistance, it should show up as a differing voltage to each side of the circuit. If you wish to know more then you should consult the manufacturers of the control panels you use. The voltages will vary from one product to another so the manufacturers will be the best people to tell you of the voltages you should expect to see at each side of the circuit. They will also be able to tell you of the current supplied to the circuit if you want to do your own resistance calculations. If you want to take the idea to greater length then you could record the readings from year to year to spot the discrepancies more accurately.
This reading is not a requirement of BS and should not be considered to be anything more than an indicator of possible trouble. The beauty of it is that the readings can be taken without disrupting any of the wiring and each circuit takes only a couple of seconds to read. If you find something that gives rise to doubt then you can take the circuit out of the control panel and read it properly. Let's face it, if it will enable you to spot the problems and cure them before you get dragged out of bed at 2am for a false alarm then it has to be a bonus and well worth the effort.
Source
Security Installer
Postscript
Mike Lynskey is an independent inspector of security systems , a consultant and tutor. You can email him at mike.lynskey@virgin.net
