Electrical Switch

Hello.

I want to make a 'thingy' using my Arduino that I can use a multi-meter in conjunction with, to measure the resistance of two or more conductors. I will be expanding upon the idea, hence why this is so basic to start.

So for instance I have 3 conductors, wires say:

  • Number1 which is isolated from Number 2 and 3
  • Number 2 which is connected to Number 3 via a 10 Ohm resistor say

I want to measure the resistance between each wire. So:

  • Between No1 and No2 (would be open circuit)
  • Between No1 and No3 (would be open circuit)
  • Between No 2 and No3 (would be 10 Ohms)

My first question I want to grind through is:

What electrical component can I use to 'switch' between these conductors I want to measure, instead of manually moving the probes of the multi-meter?

I have done some tests with a MMBT4401WT1G NPN Transistor and have quickly realized this wont work, because when I measure between the collector and emitter I get around 1MOhm. Instead of the short circuit I imagined in my head:

#define test 12

void setup() {
  pinMode(test, OUTPUT);
}

void loop() {
  digitalWrite(test, HIGH);
}

Am I understanding NPN transistors incorrectly here, do they only function as a single conductor when in circuit, say when you use an NPN like this to turn on an LED? Is there a type of transistor that would work in this way?

Will I have to use relays? I dont want to ideally.

I dont want to be applying any voltage to my conductors. I've seen those projects with the voltage dividers but I dont want to go down that route.

I think that could be done with 4 relays, but I don't have the circuit completely laid-out in my head. Note that the Arduino can't directly drive a relay coil so you need a relay driver circuit or a "relay board" with built-in drivers.

Transistors could be used as long as you are measuring open/short and not actual-accurate resistance, but I don't have that circuit figured-out either.

I dont want to be applying any voltage to my conductors.

Of course, an Ohmmeter applies a small voltage & current and there's no other way of testing for resistance/conductivity.

There is something called a [u]Walking One's Test[/u] which is often used to check regular to check for opens/shorts on "one-to-one" cables, but it might be adaptable for your application.

It works by connecting inputs to one end of the cable and outputs to the other. For example it takes 3 inputs and 3 outputs to check a 3-wire cable. Each I/0 pin represents a bit so if you write 001 you expect to read 001, and when you write 010 you expect to read 010 etc. The outputs do have to be short-circuit protected, or it's common to use open-collector (or open drain) connections with pull-up resistors.

Thanks Doug!

Yeah, the small voltage from the meter itself is fine. Not sure if that is just the same as a voltage divider?

Ideally I would like to design this with the constraint of only one side of the conductor being accessible. Imagine u have taken the socket off ur wall plug and u have the live, neutral and gnd but have no idea where the conductor goes. Crude example haha!

I've been a lecky for over 50 years and I can't even imagine why or where one would use such a "starwars" device.

Thanks :slight_smile:

Firstly, your strange spelling of "you" and "you're" is very irritating.

You can do multiple tests for short circuit by connecting the conductors via current limiting resistors to the Arduino digital pins. A simple program will allow you to set highs and lows. Set one pin high at a time and look at the other pins to see what the status is. You can't measure continuity without connecting the other end in the same way.

So, this isnt a formal areana. People can use whatever abbrevations, text lingo they want. Save ur opinions that people are ittitating. I bet a lot of newbies find ur tone ittitating.

U dont even understand the problem. Of course u can measure the continuity between conductors like wires without both ends. How do you check that each conductor in a connector is not shorted to another? You only need one end of the wire...

DrDaveyG:
So, this isnt a formal areana. People can use whatever abbrevations, text lingo they want. Save ur opinions that people are ittitating. I bet a lot of newbies find ur tone ittitating.

U dont even understand the problem. Of course u can measure the contonuity between conductors like wires without both ends. How do you check that each conductor in a connector is not shorted to another? You only need one end of the wire...

OK I'm out. If you are so inconsiderate that you refuse to use coherent English or American, you will not be getting any assistance from me.

Have you ever read "How to win friends and influence people"? Have you ever heard "Don't bite the hand that feeds you"?

One last point... Sentences do NOT start with the word "So".

Hello.

Is there an alternative to relays? In the sense that when you operate the coil on a relay, the resistance between the input and output to and from the relay is ~0 Ohms because the relay just completes the circuit just like a switch would.

In my head a transistor might work like this but then again the whole reason a transistor works is down to the current going through the NPN/PNP regions right?

I dont want to put any voltage across the transistor/relay, I just need it to measure ~0 Ohms (Short circuit).

I might have to purchase ~200 of these relays which would be expensive. I am wondering if there is some kind of IC what has miniature relays, or a transistor type component which behaves slightly differently?

Is this really a different question from this one?

I'm thinking a logic level mosfet would do the trick. I don't know if they come in ic packs of more than one at a time.

I thought I would get more knowledgable replies on the general electronics section.

Logic level mosfet, right I will study up on those now. Thanks!

I thought I would get more knowledgable replies on the general electronics section.

Learn not to cross-post please....

Lets be clear here, you want to measure resistance between various combinations of nodes in a circuit?

If that's the case you are wrongly fixating of having a zero-ohms method of switching.

What you can do is inject a fixed current between two of the nodes and measure the voltage drop across
them, neither of which require low-resistance switching. Basically do a 4-wire resistance measurement for
each combination of nodes.

To do that you need to be able to switch a current source onto one of the nodes, switch another node to
ground, and then measure the voltage difference between those two nodes.

The voltage signals can be routed via an analog multiplexer chip easily enough, but for larger currents
you'd probably need to use transistors to switch.

So please indicate what range of resistances you need to measure, and how accurately you want to measure
them.

Can we assume no other source of voltage is present on the circuit/wires under test?

MarkT:
but for larger currents you'd probably need to use transistors to switch.

FETs that is, logic-level FETs. :sunglasses:

OK, read this thread and the previous thread. What is needed is a clear and concise explanation of exactly what you want to do. Do you, for example, wish to make a cable test fixture? Do the wires under test have connectors? While I can appreciate that likely in your head you can clearly see your objective we can't.

Ron

Duplicate topics moved to the same forum section and merged

If you

thought I would get more knowledgable replies on the general electronics section.

then you could have asked for the topic to be moved

Cross-posting is against the rules of the forum. The reason is that duplicate posts can waste the time of the people trying to help. Someone might spend 15 minutes (or more) writing a detailed answer on this topic, without knowing that someone else already did the same in the other topic.

Repeated cross-posting will result in a timeout from the forum.

In the future, please take some time to pick the forum board that best suits the topic of your question and then only post once to that forum board. This is basic forum etiquette, as explained in the sticky "How to use this forum - please read." post you will find at the top of every forum board. It contains a lot of other useful information. Please read it.

Thanks in advance for your cooperation.

MarkT:
What you can do is inject a fixed current between two of the nodes and measure the voltage drop across
them, neither of which require low-resistance switching. Basically do a 4-wire resistance measurement for each combination of nodes.

To do that you need to be able to switch a current source onto one of the nodes, switch another node to
ground, and then measure the voltage difference between those two nodes.

The voltage signals can be routed via an analog multiplexer chip easily enough, but for larger currents
you'd probably need to use transistors to switch.

So please indicate what range of resistances you need to measure, and how accurately you want to measure them.

Can we assume no other source of voltage is present on the circuit/wires under test?

Thanks for the kind words Mark. I've thought a bit more about this and you've put me on the start of the right path I think.

The accuracy of the measurement isn't too important, the ability to differentiate between 10s and 1000s of Ohms would be enough. I want to know if its between 0-10 Ohms, 10-100, 100-1000, not exacts.

No other voltage source is present.

I think what you are suggesting is how multi-meters measure resistance right? They use the constant current technique? How is resistance measured? | Tektronix

The reason I was apprehensive before is because I dont want to potentially damage what could be on the end of the conductors I could be testing against one another. They could be connected to an IC, an LED etc. But now I understand that is how a Fluke works I am ok. To be clear, I just want this element of design safety in my 'device', so I can measure the resistance between conductors without causing any damage to what they could be connected to.

How low is the current that they/other OEMs use to measure unknown resistors and not damage the units under test (if known)? How low were u going to suggest?

UKHeliBob:
you could have asked for the topic to be moved. Cross-posting is against the rules of the forum.

Also, thanks for the cross post lesson Bob, my apologies.

Sorry if I am butchering this Mark, but is this how you imagine the circuit (very basically). PSA.

With constant current and measured voltage you just calculate the resistance?

See #14!