Circuit Design Help - Resistance Tester

[[email protected]]

I have a little project I need some assistance on. I have some circuit
design experience, but not enough to know what I need to make this
work.

I have three long wires that I need to test for resistance. If the
wires are corroded, connections are bad, etc. that will be indicated
as a larger resistance than if the wire is intact. Usually the cut-off
is 3 ohms for a bad wire. The wires are under lots of stress, so they
develop breaks and the pins on the ends get pulled out a lot.The pins
also corrode, causing problems.

I need to set up a circuit that will light one of two LEDs or lights
for each wire. A green signal for a circuit resistance of less than 3
ohms, a red signal for more than three ohms. The wires are installed
into a system running at 12 VDC about 0.5 amps.

The idea I have so far is a comparison system. If R1 is less than R2,
then turn circuit three on. If R1 is more than R2, then turn circuit 4
on. Repeat for the other two circuits.

Ideally, the system would have an on/off switch and a single power
source. I plan to use a battery since I intend the system to be
portable. I also plan to have at least one other circuit in the same
system, so the battery would power that circuit as well and there
would be a switch to toggle between test circuit A and test circuit
B.

Any help would be greatly appreciated. I would like to be able to fit
the completed system in a small project box, probably 4 inches by 3
inches or thereabout. I plan to get supplies wherever I can get them,
so the more simple the components are the better.

I can solder just about anything, though I haven't tried surface mount
yet.

Thank you.

 

[John Fields]

I have a little project I need some assistance on. I have some circuit
design experience, but not enough to know what I need to make this
work.

I have three long wires that I need to test for resistance. If the
wires are corroded, connections are bad, etc. that will be indicated
as a larger resistance than if the wire is intact. Usually the cut-off
is 3 ohms for a bad wire. The wires are under lots of stress, so they
develop breaks and the pins on the ends get pulled out a lot.The pins
also corrode, causing problems.

I need to set up a circuit that will light one of two LEDs or lights
for each wire. A green signal for a circuit resistance of less than 3
ohms, a red signal for more than three ohms. The wires are installed
into a system running at 12 VDC about 0.5 amps.

---
Will you have access to the 12V, will the wires be powered up when
you test them, and how do you plan to connect to them?
---

The idea I have so far is a comparison system. If R1 is less than R2,
then turn circuit three on. If R1 is more than R2, then turn circuit 4
on. Repeat for the other two circuits.

 

[[email protected]]

---
The end result of the project will be a stand-alone test device. It
will ideally be a small box with an internal power supply. The wires
to be tested will be plugged into sockets on the face of the box. This
connection method allows the pin-wire connection to be stress tested
(wiggled) and also tested for corrosion damage.
---

---

I haven't been able to figure out how to design the resistance
comparison circuit. I keep thinking that an if-then statement would
work, but that goes to computer logic programming and I don't know how
to relate that to simple circuit design.
The easiest way would be to have three cheap digital multi-meters
mounted in the box and have the user do the check. The LCD readouts
increase the size of the system and you almost need three separate
power sources, one for each meter. While the non-relevant functions
could be removed from the multi-meter, I would prefer to have a simple
visual indication of the status of the wire.

 

[John Fields]

---
The end result of the project will be a stand-alone test device. It
will ideally be a small box with an internal power supply. The wires
to be tested will be plugged into sockets on the face of the box. This
connection method allows the pin-wire connection to be stress tested
(wiggled) and also tested for corrosion damage.
---


---

I haven't been able to figure out how to design the resistance
comparison circuit. I keep thinking that an if-then statement would
work, but that goes to computer logic programming and I don't know how
to relate that to simple circuit design.
The easiest way would be to have three cheap digital multi-meters
mounted in the box and have the user do the check. The LCD readouts
increase the size of the system and you almost need three separate
power sources, one for each meter. While the non-relevant functions
could be removed from the multi-meter, I would prefer to have a simple
visual indication of the status of the wire.

---
If you can get ahold of both ends of the wires and there's nothing
else connected to them this should work:

View in Courier.

.. +6V>----+----------+-------+-------+-----+----------+
.. | | | | | |
.. [118] [1180] +--|---+ [10k] [1800] [1800]
.. |1% |1% | | | | | |
.. | | | | [1M] | [HLMP4740] [HLMP4700]
.. +-->>--+--[1000]--|----+-|+\ | | |K |K
.. | | | | >-+---|-----+--|+\ |
..[WIRE] [1µF] +------|-/LM393 | | >----+
.. | | | | U1A +--------|-/U1B
.. +-->>--+ [30.1] | |
.. | |1% | [3K3]
.. | | | |
.. GND>---+----------+-------+-------+

The current through the wire and the 118 ohm resistor will be around
50mA, so I'd use 4 AA alkaline cells in series for the battery.

I've shown only a single channel, so for 3 with all the wires
plugged in and a capacity of 2450 mAH for the battery it'll last for
about 17 hours.

Interestingly, the circuit is ratiometric so changes in battery
voltage will have very little effect on detecting the resistance of
the wire.

 

[Gary Tait]

[email protected] wrote in 57g2000hsv.googlegroups.com:

I would prefer to have a simple
visual indication of the status of the wire.

How about incandescent light bulb, or three?

Or a low ohm resistor as:

Power: o------/\/\/\/\/------x----o Test points o-----> Gnd

Piont x goes into a voltage comparator. A good wire will hold point x low,
and bad wire will let it float to your input power, or at least a
threshold.

 

[[email protected]]

---
If you can get ahold of both ends of the wires and there's nothing
else connected to them this should work:

View in Courier.

. +6V>----+----------+-------+-------+-----+----------+
. | | | | | |
. [118] [1180] +--|---+ [10k] [1800] [1800]
. |1% |1% | | | | | |
. | | | | [1M] | [HLMP4740] [HLMP4700]
. +-->>--+--[1000]--|----+-|+\ | | |K |K
. | | | | >-+---|-----+--|+\ |
.[WIRE] [1µF] +------|-/LM393 | | >----+
. | | | | U1A +--------|-/U1B
. +-->>--+ [30.1] | |
. | |1% | [3K3]
. | | | |
. GND>---+----------+-------+-------+

The current through the wire and the 118 ohm resistor will be around
50mA, so I'd use 4 AA alkaline cells in series for the battery.

I've shown only a single channel, so for 3 with all the wires
plugged in and a capacity of 2450 mAH for the battery it'll last for
about 17 hours.

Interestingly, the circuit is ratiometric so changes in battery
voltage will have very little effect on detecting the resistance of
the wire.

John,

Thank you for your assistance. I am having a little difficulty with
some of the components.

I found the LM393 unit. The HLMP4740 and HMLP4700 are green and red
LEDs. Are there any specific properties that these have? I can't seem
to get those parts specifically, they are obsolete according to my
suppliers (Mouser and DigiKey). Also, I don't know what the 3K3 unit
is. The closest I have been able to come is a power resistor. What is
this component?

Thank you,

Ben

 

[John Fields]

---
If you can get ahold of both ends of the wires and there's nothing
else connected to them this should work:

View in Courier.

. +6V>----+----------+-------+-------+-----+----------+
. | | | | | |
. [118] [1180] +--|---+ [10k] [1800] [1800]
. |1% |1% | | | | | |
. | | | | [1M] | [HLMP4740] [HLMP4700]
. +-->>--+--[1000]--|----+-|+\ | | |K |K
. | | | | >-+---|-----+--|+\ |
.[WIRE] [1µF] +------|-/LM393 | | >----+
. | | | | U1A +--------|-/U1B
. +-->>--+ [30.1] | |
. | |1% | [3K3]
. | | | |
. GND>---+----------+-------+-------+

The current through the wire and the 118 ohm resistor will be around
50mA, so I'd use 4 AA alkaline cells in series for the battery.

I've shown only a single channel, so for 3 with all the wires
plugged in and a capacity of 2450 mAH for the battery it'll last for
about 17 hours.

Interestingly, the circuit is ratiometric so changes in battery
voltage will have very little effect on detecting the resistance of
the wire.

John,

Thank you for your assistance. I am having a little difficulty with
some of the components.

I found the LM393 unit. The HLMP4740 and HMLP4700 are green and red
LEDs. Are there any specific properties that these have? I can't seem
to get those parts specifically, they are obsolete according to my
suppliers (Mouser and DigiKey).

---
They're red and green LEDs with forward voltages around 2V at
forward currents of 2mA.
---

Also, I don't know what the 3K3 unit is.
The closest I have been able to come is a power resistor. What is
this component?

---
it's a 3.3kohm +/-5% 1/4 watt carbon film resistor.

The rest of the resistors (except for the 1% ones) are all standard
5% 1/4 watt carbon film, and the 1% ones (except for the 118 ohm
one) can be 1/8 watt. The 118 ohm one needs to be 1/2 watt.

 

[[email protected]]

John,

One last question and I think I have this down.

I have three wires to test. Is there going to be a problem if I test
all three at once using three circuits like the one you designed, with
a common ground and DC rail? Or will I need to add a switch to
activate the test circuits one by one? The wire I have to test is part
of a three strand cable with a three-prong plug at both ends. Since I
have lots of these to test, I wanted to be able to just both ends of
the cable into the tester and have the indicators for each circuit
light red or green. If that won't work with your circuit, I can add a
three position switch, but I would prefer the device to be as simple
as possible in operation.

Thank you.

Ben

 

[John Fields]

John,

One last question and I think I have this down.

I have three wires to test. Is there going to be a problem if I test
all three at once using three circuits like the one you designed, with
a common ground and DC rail?
---
No.
---

Or will I need to add a switch to
activate the test circuits one by one?
---
No.
---

The wire I have to test is part
of a three strand cable with a three-prong plug at both ends. Since I
have lots of these to test, I wanted to be able to just both ends of
the cable into the tester and have the indicators for each circuit
light red or green.

---
OK, that should be fine.
---

If that won't work with your circuit, I can add a
three position switch, but I would prefer the device to be as simple
as possible in operation.

---
OK.

BTW, how long are the cables and what's their construction? That
is, are they twisted pairs? Shielded? Etc???

The reason I ask is if there's a lot of noise around the switching
point there may need to be more filtering on the inputs and/or more
hysteresis added to the circuit.

 

[[email protected]]

John,

The wires are about 4 feet long and are 16 or 18 gauge twisted copper
with a plastic or rubberized jacket. Think cheap speaker wire. There
are about 15 different types of wire used, but they are all common in
that they use three insulated strands with the insulation joined to
make a single cable.

There are two main types of cable to test. The first is three wires
running parallel. The second is two wires running parallel with the
third separated. This third wire is about 1/2 the length of the other
two and has a large alligator clip soldered to the other end of it.

The cables are body wires for Sport fencing, the first set are for
epee the second for foil.

There are several units made that test the wires, but they are all
highly expensive and a pain to use. So I am hoping that I can use this
setup to test the wires I repair for my club quickly and easily. The
tester will also be easier to take along to tournaments and the like.

Thanks,

Ben

 

[John Fields]

John,

The wires are about 4 feet long and are 16 or 18 gauge twisted copper
with a plastic or rubberized jacket. Think cheap speaker wire. There
are about 15 different types of wire used, but they are all common in
that they use three insulated strands with the insulation joined to
make a single cable.

There are two main types of cable to test. The first is three wires
running parallel. The second is two wires running parallel with the
third separated. This third wire is about 1/2 the length of the other
two and has a large alligator clip soldered to the other end of it.

The cables are body wires for Sport fencing, the first set are for
epee the second for foil.

There are several units made that test the wires, but they are all
highly expensive and a pain to use. So I am hoping that I can use this
setup to test the wires I repair for my club quickly and easily. The
tester will also be easier to take along to tournaments and the like.

---
Indeed!

Maybe you could even undersell and outsell the highly expensive
competitive product and make a fortune!

If you do, I wouldn't be at all disappointed if you decided to send
a little of it my way!

 

[Fred Bloggs]

The wires are about 4 feet long and are 16 or 18 gauge twisted copper
with a plastic or rubberized jacket. Think cheap speaker wire. There
are about 15 different types of wire used, but they are all common
in that they use three insulated strands with the insulation joined
to make a single cable.

There are two main types of cable to test. The first is three wires
running parallel. The second is two wires running parallel with the
third separated. This third wire is about 1/2 the length of the other
two and has a large alligator clip soldered to the other end of it.

The cables are body wires for Sport fencing, the first set are for
epee the second for foil.

There are several units made that test the wires, but they are all
highly expensive and a pain to use. So I am hoping that I can use
this setup to test the wires I repair for my club quickly and easily.
The tester will also be easier to take along to tournaments and the
like.

You're not looking for a wire resistance tester, you're looking for a
contact resistance measurement, there is a difference. The contact
resistance tester will have limitations on open-circuit voltage and
current, and there may be transient testing, it can get complicated. You
should stick with the factory supplied tester, unless you know better,
before you make a fool of yourself.

 

[John Fields]

If you can get ahold of both ends of the wires and there's nothing
else connected to them this should work:

View in Courier.

. +6V>----+----------+-------+-------+-----+----------+
. | | | | | |
. [118] [1180] +--|---+ [10k] [1800] [1800]
. |1% |1% | | | | | |
. | | | | [1M] | [HLMP4740] [HLMP4700]
. +-->>--+--[1000]--|----+-|+\ | | |K |K
. | | | | >-+---|-----+--|+\ |
.[WIRE] [1µF] +------|-/LM393 | | >----+
. | | | | U1A +--------|-/U1B
. +-->>--+ [30.1] | |
. | |1% | [3K3]
. | | | |
. GND>---+----------+-------+-------+

---
OP built it and reported, via email, problems with the LEDs.

After looking the circuit over, it turns out I had U1B's inputs
wired backwards, duhhhh...

After rewiring it:

.. +6V>----+----------+-------+-------+-----+----------+
.. | | | | | |
.. [118] [1180] +--|---+ [10k] [1800] [1800]
.. |1% |1% | | | | | |
.. | | | | [1M] | [HLMP4740] [HLMP4700]
.. +-->>--+--[1000]--|----+-|+\ | | |K |K
.. | | | | >-+---|-----+--|-\ |
..[WIRE] [1µF] +------|-/LM393 | | >----+
.. | | | | U1A +--------|+/U1B
.. +-->>--+ [30.1] | |
.. | |1% | [3K3]
.. | | | |
.. GND>---+----------+-------+-------+

the OP reported that it worked perfectly.

 

[Frank Buss]

. +6V>----+----------+-------+-------+-----+----------+
. | | | | | |
. [118] [1180] +--|---+ [10k] [1800] [1800]
. |1% |1% | | | | | |
. | | | | [1M] | [HLMP4740] [HLMP4700]
. +-->>--+--[1000]--|----+-|+\ | | |K |K
. | | | | >-+---|-----+--|-\ |
.[WIRE] [1µF] +------|-/LM393 | | >----+
. | | | | U1A +--------|+/U1B
. +-->>--+ [30.1] | |
. | |1% | [3K3]
. | | | |
. GND>---+----------+-------+-------+

the OP reported that it worked perfectly.

It works perfectly, until the OP accidently feeds the 6V, which you are
using for measuring, into some low-voltage logic input, e.g. some expensive
and unprotected IC with 2.5V input.

 

[John Fields]

. +6V>----+----------+-------+-------+-----+----------+
. | | | | | |
. [118] [1180] +--|---+ [10k] [1800] [1800]
. |1% |1% | | | | | |
. | | | | [1M] | [HLMP4740] [HLMP4700]
. +-->>--+--[1000]--|----+-|+\ | | |K |K
. | | | | >-+---|-----+--|-\ |
.[WIRE] [1µF] +------|-/LM393 | | >----+
. | | | | U1A +--------|+/U1B
. +-->>--+ [30.1] | |
. | |1% | [3K3]
. | | | |
. GND>---+----------+-------+-------+

the OP reported that it worked perfectly.

It works perfectly, until the OP accidently feeds the 6V, which you are
using for measuring, into some low-voltage logic input, e.g. some expensive
and unprotected IC with 2.5V input.

---
Well, Frank, now that you've unknowingly and awkwardly skewered
yourself by not having read the earlier parts in the thread before
casting aspersions, I guess it's too late for you to save face, eh?

Clue: The [WIRE] and the connectors on its ends are the devices
under test, LOL.

 

[Frank Buss]

Well, Frank, now that you've unknowingly and awkwardly skewered
yourself by not having read the earlier parts in the thread before
casting aspersions, I guess it's too late for you to save face, eh?

Clue: The [WIRE] and the connectors on its ends are the devices
under test, LOL.

I have read the rest of the thread and your schematic is perfect for
testing sport fencing wires, but it is very easy to build a safe tester,
which can be used without fear to damage electronics. There was an article
in de.sci.electronics some time ago and it is possible to measure with as
low as 2mV instead your 6V at 118 ohm:

http://www.autometer.de/files/Tester.gif

The low testing voltage has the advantage that you can use it for detecting
pcb track shorts, even on populated pcbs without damaging the parts.

 

[John Fields]

Well, Frank, now that you've unknowingly and awkwardly skewered
yourself by not having read the earlier parts in the thread before
casting aspersions, I guess it's too late for you to save face, eh?

Clue: The [WIRE] and the connectors on its ends are the devices
under test, LOL.

I have read the rest of the thread and your schematic is perfect for
testing sport fencing wires, but it is very easy to build a safe tester,
which can be used without fear to damage electronics.

---
Who cares?

The OP asked for a device to test sport fencing wires, and that's
what he got.
---

There was an article
in de.sci.electronics some time ago and it is possible to measure with as
low as 2mV instead your 6V at 118 ohm:

http://www.autometer.de/files/Tester.gif

---
Again, who cares?

That device doesn't even support the two LEDs the OP asked for.

 

[neon]

The way I see it and it might be inpractical because the info is not complete. but anthow using those three wire and borrow some ua to set up a comparator at each end to monitor the voltage and output status this way you will be able to tell where the problem is at one end or the other or in between.