Go Down

### Topic: Arduino digital input, 80Vdc (Read 20011 times)previous topic - next topic

#### Sergegsx

#45
##### Oct 19, 2013, 11:20 pm
1st part ---> Completely clear!
I will leave the 1n4148

2nd part --->

This is my first project with optocouplers and all the tutorials on the internet are much simpler than all this. they dont go into CTR, they just talk about the optocoupler as a led diode, current limiting resistor and not much more.

#### Pelleplutt

#46
##### Oct 19, 2013, 11:38 pm
Do not belive everything on the net.
CTR are important to calculate with. Look at the 4N27, 10% up to what I have seen as max, 600%.
The solution with the zeners are not optimal when the input signal varies 20 volts (60-80)
Was it nessecery to let voltage in the span 0-50 volts be absolutly off.
With only a resistor, 0-10 volts can be off and over 60 on.

Pelle

#### Sergegsx

#47
##### Oct 19, 2013, 11:48 pm

Do not belive everything on the net.
CTR are important to calculate with. Look at the 4N27, 10% up to what I have seen as max, 600%.
The solution with the zeners are not optimal when the input signal varies 20 volts (60-80)
Was it nessecery to let voltage in the span 0-50 volts be absolutly off.
With only a resistor, 0-10 volts can be off and over 60 on.

Pelle

It was initially suggested to use zener diodes and it seemed a good idea, thats why i went that way. I also found it interesting to be able to make sure if the channel was really on or off, and by using the zener I would know for sure if the voltage is above 57V.
Apart from that, yes, the voltage can vary between 60V and 80V but I guess using the zenner diodes we are dropping the voltage enough to be save after that, right?

What is your suggestion? I am always open to learning new things, maybe there a much better way of doing all this.

dc42 I am still trying to figure out all the CTR things

#### Sergegsx

#48
##### Feb 02, 2014, 06:28 pmLast Edit: Feb 02, 2014, 06:39 pm by Sergegsx Reason: 1

Do not belive everything on the net.
CTR are important to calculate with. Look at the 4N27, 10% up to what I have seen as max, 600%.
The solution with the zeners are not optimal when the input signal varies 20 volts (60-80)
Was it nessecery to let voltage in the span 0-50 volts be absolutly off.
With only a resistor, 0-10 volts can be off and over 60 on.

Pelle

Hello Pelle,
I am looking into this again after some real tests i have done and I am starting to understand you point on why this could be not a good way to go.

With voltages varying so much, I am having problems choosing the components. Either they use too much power, or the resistors do not suit to all cases.

Please could you suggest another method ? Anyone else?

I will post now some simulations which are very close to what I have been seeing in real tests.
thank you very much.

#### Sergegsx

#49
##### Feb 02, 2014, 06:38 pmLast Edit: Feb 02, 2014, 06:42 pm by Sergegsx Reason: 1
Option1H - Supply of 74V
Works ok. Problem that at 50V it will not work and that the Led D3 can not be very bright due to power dissipation in 10k resistors

Option1L - Supply of 50V
Does not work due to low voltage after zenner

Option2H - Supply of 74V
Works ok, but if voltage lowers then leds do not behave the same due to different resistors

Option2L - Supply of 50V
does not work, not enough current through resistor VR1, if I reduce VR1 value, then the Zener need to dissipate too much current.

I wanted to send this online to make the PCB and build my first SMD pcb, so power dissipation of components is crucial.

#### dc42

#50
##### Feb 02, 2014, 07:57 pm

With voltages varying so much, I am having problems choosing the components. Either they use too much power, or the resistors do not suit to all cases.

Please could you suggest another method ? Anyone else?

You need to state your requirements better. You have said that you want to detect the difference between (1) 60 to 80V, and (b) 0V. Anything else? e.g. do you require a particular behaviour when there is more than 0V but less than 60V? Do you need to allow for more than 80V input, if so what is the maximum voltage? And what is an acceptable power consumption at 80V input?
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

#### Sergegsx

#51
##### Feb 02, 2014, 08:06 pmLast Edit: Feb 03, 2014, 11:24 am by Sergegsx Reason: 1
Hello dc42 !

Requirements.
1) Voltage can vary between 50 to 80Vdc (although it should be nominal at 75Vdc)
2) For each input, I need to put 1 led onboard, and 1 led in a case (that means there will be a 7 pin connector in the PCB to connect the external leds)
3) Voltage supply (75Vdc) has no power supply restrictions, (it can use far more than this board will need)
4) Reverse voltage requiered (in case its connected wrongly, or to prevent a certain channel feeding another one)
5) Negative for the 75Vdc will be common to all inputs
6) At first I wanted to use zener diode to make a clear distinction between voltages under around 50Vdc and over it. The reason was that it should be around 75, so anything lower than 50Vdc does not make sense. However, I wouldnt mind if it can detect a range like under 50Vdc (logic 0) over 50Vdc (logic 1)
7) Using SMD components (0805, 1206) so 250mW for the latest as max power dissipation I guess.
One of the two leds (the external mounted) will need to be hocked afterwards directly to ground. That way for 6 inputs I only need to extend a 7 wire cable (6 leds+1ground). If the led is between the circuit I would have to extend 2 wires for each input. Also that would make this external led necesary for the rest of the circuit to work which I dont want.

please let me know if more details are required, and thanks for looking into it.
screenshot Option1H is working correctly, but as I mentions has some problems when voltages go under 60Vdc, also the 10K resistors can not be replaced to allow more power to the led as they will get too hot.

Thanks again !

#### polymorph

#52
##### Feb 03, 2014, 12:15 am
I think I'd probably use a comparator with high value resistors.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8

#### dc42

#53
##### Feb 03, 2014, 01:32 am
One more question: how many 72V inputs are there that you wish to monitor?
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

#### Sergegsx

#54
##### Feb 03, 2014, 05:28 amLast Edit: Feb 03, 2014, 11:24 am by Sergegsx Reason: 1

One more question: how many 75V inputs are there that you wish to monitor?

Initially I need 6 or 8.
Ideally for the future, I would like to have the possibility to expand to around 10 to 20.

Thanks dc42!

#### Sergegsx

#55
##### Feb 03, 2014, 05:31 am

I think I'd probably use a comparator with high value resistors.

hi polymorph and thank you.
I will try to figure out what you mean cause I have no idea right now. Time for google. please let me know if you can develop a bit your answer.

thanks !

#56
##### Feb 03, 2014, 10:17 am
72V nominal, is this for railway work?

Here is a circuit I designed a while back, originally for 50v (that was tested and worked) but the values in this version are for 100v (so far not tested).

I don't think it will suit you as is because it doesn't have a remote indicator LED and also the threshold is very low at just a few volts because in our application anything from about 4-5V to 100V is "on". But it may offer a different approach worth looking into.

Basically the constant current source (20mA) fixes the large range of currents over a 0-100v range you get using resistors, after that it's just a matter of choosing LEDs/diodes that are happy with 20mA.

______
Rob

Rob Gray aka the GRAYnomad www.robgray.com

#### Sergegsx

#57
##### Feb 03, 2014, 11:34 am

Its not, I just wrote 72 as an example. As I said, it can be anywhere around 70 to 80V, although normally is around 75V.

I am starting to think it might be a bit difficut to find an easy solution for requirement

Quote
6) At first I wanted to use zener diode to make a clear distinction between voltages under around 50Vdc and over it. The reason was that it should be around 75, so anything lower than 50Vdc does not make sense. However, I wouldnt mind if it can detect a range like under 50Vdc (logic 0) over 50Vdc (logic 1)

If its much more easier, I dont mind if the circuit will be activated as soon as there is any voltage on the positive. But as I say, it needs to be prepared to withstand 0 to 80Vdc.

Thank you very much all.

#### polymorph

#58
##### Feb 03, 2014, 05:10 pm
Perhaps you could distill your requirements into one message. We've asked a lot of questions, you've given a lot of answers, but it is now spread out over 4 pages.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8

#59
##### Feb 04, 2014, 12:27 am
Quote
If its much more easier, I dont mind if the circuit will be activated as soon as there is any voltage on the positive.

That's exactly what my circuit does, and I would say it's a lot easier or this thread wouldn't have got so long

That said maybe two voltage dividers and a window comparator would do the 60-80v detection you really want.

______
Rob
Rob Gray aka the GRAYnomad www.robgray.com

Go Up