High sensitivity continuity tester

[mattyboson12]

TL;DR Summary

Freelance product design project

Hello,

I'm a mechanical designer so have limited experience on circuit design. I'm designing a product in my freetime which requires a high speed continuity signal to trigger my system - the faster it is, the more accurate the output so ideally looking at sub 1ms.

• I'm hoping to power it with a 12v power supply and when it triggers (when there is continuity) a 12V signal is emitted
• Looking at sub milli second response from the tester
• Looking at testing high resistance objects e.g. continuity through a large machine so i assume it would be in the +kΩ

I was hoping someone could point me in the right direction/ any resources which could help me design this circuit. What would the limitations be when trying to design a high sensitivity tester? Thanks, Matt

 

[Baluncore]

Looking at testing high resistance objects e.g. continuity through a large machine so i assume it would be in the +kΩ

You can get 10 usec times from a microcontroller without problems. The resistance path will depend on the type of machine, and what you are trying to measure.

Machinery is a huge field. You need to provide more information so as to get good advice.

 

[hutchphd]

In particular what is creating the "continuity signal" and what are the expected open and closed impedances.

 

[mattyboson12]

Apologies I should have explained myself more clearly. I'm trying to build my own tool setter for a CNC router/milling machine so when the cutter contacts the tool setter, an electrical connection is created and then a 12V signal is sent to the machine to record the coordinates of the tool

 

[berkeman]

Apologies I should have explained myself more clearly. I'm trying to build my own tool setter for a CNC router/milling machine so when the cutter contacts the tool setter, an electrical connection is created and then a 12V signal is sent to the machine to record the coordinates of the tool

Is the tool setter insulated from the rest of the machine?

 

[mattyboson12]

Yes I'm intending on insulating it from the machining table- having one test point connected to the machining table and the other test point connected to the contact pad of the tool setter. So when the tool contacts the contact pad, it detects the continuity through the machine and issues a 12v signal

 

[Baluncore]

Are you using it for X and Y position, or only for height Z ?

 

[mattyboson12]

Just for setting the z height

 

[mattyboson12]

(tool length)

 

[mattyboson12]

I found this circuit online but I was wondering how I could modify it to send a 12V signal rather than lighting the led

https://www.homemade-circuits.com/make-this-simplest-continuity-tester/

 

[berkeman]

I found this circuit online but I was wondering how I could modify it to send a 12V signal rather than lighting the led

https://www.homemade-circuits.com/make-this-simplest-continuity-tester/
View attachment 270018

That looks like the same circuit you showed in your earlier thread:

Hello, I'm building a homemade continuity tester from the schematic below, and I want to bump up the sensitivity even further so it can detect high resistance objects. What part of the circuit would I have to modify to allow this? Would it be a higher voltage?

Thank you
Matt

How you generate the 12V output signal depends on what it needs to do. If a simple CMOS logic output signal is enough, you can use a CMOS logic IC from the CD4000 series, powered with a low-power 12V rail.

But if you want to use the 12V output to power a motor or something, you will need a higher output amplifier to drive that signal.

 

[DaveE]

... you can use a CMOS logic IC from the CD4000 series, powered with a low-power 12V rail.

I would avoid that series of ICs for machine tools. They are really easy to break with ESD and such; or you have to build in so much protection that the complexity is huge compared to a couple of transistors.

 

[DaveE]

You can try this circuit. It should be pretty rugged. There are 100's of variations possible. The power supplies for the input and output can be different; if you want different voltages, then you'll have to change some resistor values. R4 is optional.

Edit: This circuit may not work well if the output is connected to a low impedance input downstream, especially if that input has a pull-up resistor. It's easy to modify to accommodate, ask if you think that's a problem.

 

[mattyboson12]

Thanks all! I'll have a go at building the above circuit this week and will let you know how it goes! :)

 

[DaveE]

I actually prefer this version with the opposite polarity. Either will probably work.

 

[mattyboson12]

Hi Dave, if I want to get a 24V output instead, would I simply just drive the circuit using 24V supply instead? Thanks for your help, I really appreciate it!

 

[DaveE]

You'll want to adjust the R1 value for different power supply voltages. I didn't calculate it right the first time, although it still would have worked.

The transistors are only rated for 30V maximum; so 24V is OK but no higher. I've also added some parts for protection against transients, which is more important at higher supply voltages and a really good idea for machine tools anyway. This is about reliability, not function. You only need 3 parts to make this work initially.

Ask if you want to see the protection circuit for the other polarity version.

edit: BTW, practically any diode will do, it doesn't have to be 1N4148. The capacitor can be any ceramic cap rated 50V or more. The resistors are 1/4W (metal film preferred).

 

[DaveE]

Summary:: Freelance product design project

Looking at testing high resistance objects e.g. continuity through a large machine so i assume it would be in the +kΩ

Oops! I just noticed (recalled?) this requirement. The circuit should work up to at least 100Kohm contact resistance. But you may have to increase R4 (1M - 10M) or remove it if the circuit isn't sensitive enough.

 

[mattyboson12]

Thanks dave I'll try both combinations! I've just ordered the parts so while hopefully build it up by the end of this week . Just to clarify in terms of powering the circuit, would what I've scribbled be the correct way to power it, and is the switch referring to the cutter/tool setter coming into contact?

 

[DaveE]

Thanks dave I'll try both combinations! I've just ordered the parts so while hopefully build it up by the end of this week . Just to clarify in terms of powering the circuit, would what I've scribbled be the correct way to power it, and is the switch referring to the cutter/tool setter coming into contact?

yes

 

[mattyboson12]

So I've tried to build a test version of the circuit- apologies its a bit messy!
-powered it with a 3V battery and R1 was 360 ohms (voltage might be a bit low, but wanted to test without a powersupply first)
-the 4N33 terminals were in different positions on the datasheet so I had to rearrange the schematic a bit
- I connected the output signal to the negative terminal of the battery and measured in parallel with a multimeter to detect the output

Unfortunately I couldn't detect an output when the switch was closed- instead I was was getting a rather hot battery. I've doubled checked all the components, the circuit routing, any shorting but no luck unfortunately!

 

[DaveE]

So I've tried to build a test version of the circuit- apologies its a bit messy!
-powered it with a 3V battery and R1 was 360 ohms (voltage might be a bit low, but wanted to test without a powersupply first)
-the 4N33 terminals were in different positions on the datasheet so I had to rearrange the schematic a bit
- I connected the output signal to the negative terminal of the battery and measured in parallel with a multimeter to detect the output

Unfortunately I couldn't detect an output when the switch was closed- instead I was was getting a rather hot battery. I've doubled checked all the components, the circuit routing, any shorting but no luck unfortunately!

It looks like you have the output shorted to ground and you have the power supply connected to the collector (U1-5), also the out put. So you have a direct path from the power supply to ground.

Also your voltmeter is connected across the same wire (short circuit). You want your output to be from the collector to ground.

Some conventions with drawing schematics:
1) wires that connect always have a dot at the intersection. No dot = not connected. Dots aren't required when a wire terminates at a component.
2) wires that connect are always "T" intersections. They always only have 3 wires connecting. So, for example, if you want to have 4 wires connect on the schematic, you draw that as 2 "T" intersections. "X" intersections are always crossing (not connected) wires.

Rule #2 is frequently violated because people don't know it. Rule #1 is uniformly followed except some people use small dots, which isn't great.

 

[mattyboson12]

Thanks dave. So I just realized I ordered 4N32 by accident rather than 4N33 (just re-ordered some) - is one normally closed and the other normally open?

However I built it with the 4N32 for the time being. I made the modification to the capacitor routing you mentioned, and then probed between the output and ground to detect the trigger signal. I noticed that when the switch was open, the multimeter was reading 3.62v and when it closed (tool contacting the setter) it was reading 0.8V? I've only got less than 3v powering the circuit so is the additional voltage coming from the multimeter?

 

[DaveE]

4N32 is ok, it's essentially the same part. I don't know why you measure an output higher than the power supply voltage, that doesn't make any sense for this circuit.
https://www.mouser.com/datasheet/2/308/Fairchild-4n29-33-1191850.pdf

 

[Tom.G]

For the 3.62V reading, check what it reads when measuring the batteries.

If it still reads 3.62V either the meter is out of calibration or you have some very unusual batteries. Post a photo of the batteries showing as much of the labels as possible.

If the batteries read close to 3V then there is probably stray pickup (outside interference) entering the circuit. It could be from nearby power lines, nearby flourescent lamps, or a radio or TV station.

I see you have a prototype board that is only being used for battery power. Try building the circuit on that using shorter wires.

Cheers,
Tom

 

[mattyboson12]

I've rebuilt it on the breadboard and the multimeter looks like it's behaving now! Not sure what happened there! What I'm a bit confused about is when the switch is open I get 2.82V from the output and when it's closed I get 0.63V which was the opposite to what I was expecting. I was expecting to get 0V when the switch was open and 2.82V at the output when the switch was closed?

 

[DaveE]

I've rebuilt it on the breadboard and the multimeter looks like it's behaving now! Not sure what happened there! What I'm a bit confused about is when the switch is open I get 2.82V from the output and when it's closed I get 0.63V which was the opposite to what I was expecting. I was expecting to get 0V when the switch was open and 2.82V at the output when the switch was closed? View attachment 270649View attachment 270650View attachment 270651

The polarity you described is correct for this version of the circuit. When the switch is closed current flows through the LED. The light from the LED which is generated from the current through it illuminates the base of the transistor(s) and turns them on. When the transistors are on they divert the current flowing through the 10Kohm resistor ("pull-up" resistor) so the output voltage is low.

My original configuration in post #13 has the opposite polarity (use the R1 values from my later posts, that one was a bit too high).

Either your original construction or this one, on the breadboard, will work to quickly demonstrate functionality. Neither is acceptable for any long term usage, they both lack the mechanical strength you'll want and are prone to short circuits. Be careful handling the current version it's really easy to displace some of those components, which will make you confused for a few minutes until you figure out what happened.

If you are unclear about the basic function of these circuits, then I would encourage you to partner with an electrical engineer if you want to continue to use this in real products. You will eventually need someone with more experience to avoid implementation mistakes.

 

[DaveE]

For some subtle interface reasons, I prefer the later version (low output with input continuity). So if the polarity in your CNC program is arbitrary, I would choose this one. But the other circuit will also work.

I have made some assumptions about the machine input that this will connect to, since there isn't any interface data available. You may need to do some troubleshooting if you have problems when it's integrated into the system, but I expect it will work ok as is.