2
\$\begingroup\$

So I'm trying to design a circuit for a situation with the following scenario:

enter image description here

SW1 is representing a cut wire after the access box. I can't tap into anywhere except the access box shown in the picture and I can T in (preferred) or cut into the supply and/or ground lines. I have found one way where I can use a current sensor but that would only solve half the problem. If SW2 is open(chip is good)/closed(chip is bad) and would give false positives if SW1 is open(wire cut). So the problem is needing to accurately detect if it's 27.119V open, open with fault, shorted with fault, shorted.

I've been looking all over and found some similar situations but all had more access than I have. I have found a circuit for doing a time domain reflectometer as a last resort but hoping it can be simpler than that. Thank you for any help.

--EDIT--

To reiterate: for the circuit I need to make it can only affect what's in the square "access box". It's for a senior design project for school and the rest of the circuit shown is a closed system and the red SW1 is a hypothetical we need to be able to test for (wire gets cut/contact in a connector comes loose/etc).

If it helps with the design: from the access box to the chip sensor it's roughly 10' of wire.

--EDIT--

After doing a little more research and with regard to the comments this is a little better representation and now simulatable (I didn't know about the schematic button in the text editor previously... The previous one I made was requiring me to pay for a subscription to share because I made it on their site first so I just screenshot it instead)

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
  • \$\begingroup\$ Everywhere but "access box" is a closed system \$\endgroup\$ – BrownKuma Oct 17 '18 at 2:13
  • \$\begingroup\$ your question only makes sense if it is a school assignment ..... is it a school test? \$\endgroup\$ – jsotola Oct 17 '18 at 2:16
  • 2
    \$\begingroup\$ you said that you are allowed to cut the lines .... isolate the circuit on the right side of the access box in the diagram .... measure every possible passive electrical property of the isolated circuit .... there may or may not be a difference between a good and a failed circuit .... if that testing fails, move onto using active testing like the one that you mentioned \$\endgroup\$ – jsotola Oct 17 '18 at 3:02
  • 2
    \$\begingroup\$ This is why in industrial settings, we like to use "Live Zero" sensors. 2-10VDC or 4-20mA for example. If the reading falls below the normal zero value, open wire. \$\endgroup\$ – R Drast Oct 17 '18 at 12:35
  • 1
    \$\begingroup\$ @Brown: The DS1E-M-DC24V is a Panasonic relay. Why have you referred to it as a sensor chip? \$\endgroup\$ – Transistor Oct 17 '18 at 18:33
1
\$\begingroup\$

One way that could be possible to detect in open-circuit situation, by measuring the capacitance between the two wires, supposing they are within the same cable for most of the way.

To do so, you can inject a high frequency signal and measure the phase shift between the two wires.

If the capacitance changes (is reduced), it is likely the length was changed.

Also the final device input also probably have some capacitive signature that would improve the detection.

If the signal is closed, simply inject a small current as you said.

Here some example.

\$\endgroup\$
  • \$\begingroup\$ The capacitive measuring route seems like a good one. I'm going to explore that a bit. Thank you! Not only are these all built into harness cables but each of these sensor wires pair and go off to the sensor so I can use that provided example to hopefully (need to check if individual wires are shielded and if that would affect my measurements) grab a capacitance measurement when wires wouldn't be cut to compare it against and calculate a good range to send to the processor. \$\endgroup\$ – BrownKuma Oct 19 '18 at 17:35
0
\$\begingroup\$

Your sensor looks like a potential free contact, is this actually the case?

In your comment you say trigger the relay which implies connecting to the relay coil.

If your sensor has some DC resistance you can easily measure the volt drop of the equivalent source resistance (3.9k||120k) in parallel with the sensor R to determine if there is a loop current flowing. The source voltage will rise with a wire break, and fall with a sensor short.

If all this sounds too easy you can use a time domain reflectometer to find any breaks in the wiring but this is not ideal for very short cable runs.

\$\endgroup\$
  • 1
    \$\begingroup\$ I was trying to tease out (in the comments) if there was a quiescent current. It seems not - at least for the exercise - as they are using a relay contact. \$\endgroup\$ – Transistor Oct 19 '18 at 20:24
  • \$\begingroup\$ @duskwuff Thanks for the edit, spell checker offered just one and I took it without noticing that it was not the word I wanted. \$\endgroup\$ – KalleMP Oct 21 '18 at 19:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.