Non-intrusive continuity test

I've decided to russle up a simple electrical tester I've had in mind for some time. I'll say at the outset I'm intending this to be simple and built with a bare minimum of design effort and circuitry.

The basic idea is for a tester with no mode or range switching (from logic levels up to mains voltage) and delivers all information audibly via different tones. The tests I want to run are polarity (subject to a minimum threshold of 1-1.5V to make it logic compatible), AC/DC (which in terms of detection is simply a fast polarity test) and continuity.

Detecting voltage between the probes is a no-brainer but this continuity test is giving me pause for thought. Remember there's no mode switching, so anything we put on the probes mustn't materially affect the vast majority of circuits even when they are operational

- you'd never want to test voltage on a dead circuit after all. The obvious way would be to place a signal on one of the probes at high impedance so it can be detected by the other probe, or easily over-ridden by the circuit under test if the probes are not connected.

However, a high impedance output would seem to make it difficult to ensure that the "continuity" in question is a good one - I'd consider a resistance up to a few tens of ohms as continuity, but hundreds of K is intervening circuitry or even leakage. I don't see how you can discriminate between the two when the output resistance is by necessity 1M or more.

Therefore I'm thinking I must be heading down completely the wrong path. Does anyone know how commercial gear deals with this kind of issue?

--
Andrew Smallshaw
andrews@sdf.lonestar.org
Reply to
Andrew Smallshaw
Loading thread data ...

As you're no doubt aware, this is all about threshholds. Is 10 ohms continuity? 100 ohms? 10K? 1 ohm plus the inductance of a transformer winding?

I'm not sure the problem is at all simple. You really need to know something about the setup you're going to be measuring and the objectives in order to specify and build the device, to use it safely and to apply its results safely.

Personally, I'm very wary of these go-nogo type testers, for anything big, hazardous or critical, I always fall back on basic measurements, together with a good amount of background investigation. Even apparently benign devices such as common DMMs can produce very misleading results under some conditions.

Reply to
Bruce Varley

Without some form of electronic range/gain/attenuation switching you will have problems. Each measurement has to start by assuming mains (depending on what your mains is (volts and frequency).

1-1.5V will turn a lot of diodes on, so you have diode test as well but may not be able to detect the results depending on Vf.

Are you trying to measure an analog level or just a 'logic' input with no buffer/op amps on front end.

What Over voltage/current Protection are you going to put on your inputs? Especially if measuring upto mains (110/230V for some people

415V).

If doing no mode switching you will permanently add to one of the probes a signal for continuity testing to be subtracted from your measurements that must be small enough not to turn on a diode or affcet any other part of live circuit under test.

Sounds like you will have to do auto mode switching to determine if NO voltage AND current detected on several ranges to then inject a signal for continuity.

How will you know just by tones that it is continuity not a voltage measurement? Let alone is there just some path 0 to 100k and not how good the continuity is?

Tone ohms (for getting different tones at differing milli ohm ranges) have to be continually adjusted for range as you get closer to shorts.

How will your unit detemine it is in voltage mode not continuity mode, because the probes are not on the same signal run and you have misplaced a probe or other reasons.

If you have a low impedance o/p then you could supply a current to the circuit which the unit does not know the probes are on different parts that turn on or OVERDRIVE INPUTS,

Reminded of incoiming goods inspection that did standard tests on fuses with a meter for continuity, and always failed the 100 micro amp fuses!

One of my ASIV testers does continuity testing by electronic mode selection to test all pins are connected to the socket, by inserting a

100k into commoned VCC and GND to device under test and measuring current in 100k when each pin is individually grounded.

Continuity testing is deifferent things to different folks.

I am not fully sure I fully understand what you are trying to acheive.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul

I don't think that is strictly needed here - we're only interested in a yes or no result after all. That means ultimately we're only interested in whether a transistor turns on or not. Set the maximum current draw at a 340V mains peak at something manageable (e.g.

3mA) and we'll still have 10uA at 1.5V. That kind of range seems reasonable enough to me, especially with Darlingtons.

That is a little ambiguous. The input is analog but the detection is digital, i.e. whether some threshold (whose actual value may be fairly loosely defined) is crossed or not.

To be honest I wasn't planning on doing any beyond double insulation for safety. My attitude here is that if you kill the device that's your own fault. That's part of the reason for going all the way up to the mains - you should know straight away if a device under test uses higher voltages than that. I strongly suspect the imput impedances will be high enough that a dead device is all you need to worry about even up to quite high voltages, as opposed to e.g. fire.

Possibly. My initial thoughts were that the continuity signal could be pulsed at ~100Hz so the measurements could be differentiated. That will of course trigger false positives through capacitors, especially at the kind of currents we're taling about here. However, this project is all about optimising the common cases rather than guaranteeing correct treatment of corner cases. I'm still musing whether that is acceptable or not.

That is the kind of corner case I am referring to. This is essentially intended as a quick and dirty test for the most common cases. I don't know about you but the most common tests for me are simply:

1) Is there continuity here? That tends to be mostly checking wiring and by far the single more common test. 2) Is there voltage here? That is usually in two forms: i) Is there power here? ii) What is the static logic level here? AC/DC is basically an extra since I guess it will be almost "for free". If those three basic questions are answered I'm happy. There are other devices for precision measurements.
--
Andrew Smallshaw
andrews@sdf.lonestar.org
Reply to
Andrew Smallshaw

Depends how many volts you are allowed across that 1M. eg 10V injects 10uA, which will give 100uV on your 10 ohms. It is possible to sense/test to 100uv, with a little care.

eg ADA4528-1, could allow you to set a threshold-component.

-jg

Reply to
Jim Granville

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.