High Voltage Protection of PLC cards

P

Thread Starter

pulsar

I am looking for a plc card with measuring range: 0 to 10V(DC). But the card should have a voltage protection of minimum 600 Volts(DC) at inputs.

There are cards that offer 200V isolation with the backplane bus. This means that if 200+ DC volts are applied, the CPU does not fry up - but the input card does!

I wonder if it is even possible to get 600V DC protection like this. Can anyone suggest any external devices(transducers/over voltage protection circuits) that I can use with PLC modules. The points to bear in mind are...

1. The process needs highly accurate readings(upto 2 decimal places).

2. Currents up to 13 K amps are flowing in the process (but I am only measuring voltages)

3. I need this info for a Caustic Soda Plant using Membrane technology. Any links or advice related to this industry would be extemely helpful.


Thanks!

Omer
 
M

Michael Griffin

Most of the companies which make terminal block hardware (Phoenix, Entrelec, etc.), plus verious specialists (Action Instruments, etc.) make
analogue isolator or signal conditioner modules. There are models available to either convert between different types of signals (e.g. 4-20ma in, 0-10V out), or just to isolate without any conversion. There are also field configurable models available from some manufacturers.
These are often rated to withstand much higher than 600 volts and will protect your PLC from unintended voltages. Be sure to check their
response rate when you examine the specs. The big market for these things seems to be in process industries where a slow acting device is actually
desirable.


**********************
Michael Griffin
London, Ont. Canada
[email protected]
**********************
 
W

Wayne Shimanis

You can also check out products manufatured by Vortex Technologies at www.vortextek.com. In particular, the Vortex Technologies Iso-Pak Model IP-140 has been utilized specifically for PLC I\O protection applications.

Wayne Shimanis
 
J

Johan Bengtsson

I am not sure I fully understand your question.
Is your concern only insulation betwen the voltage signal and the PLC, or is your problem also that the 0-10V signal can reach 600V?

If it is only the first that is the problem you can easily solve ANY insulation problem using a field bus or similar with communication via opto. Input racks should be possible to find that withstand some 1000V but I don't have any idea from where.

If it is the signal that can reach 600V it is another matter however. That don't mean it is unsolveable but I don't know if there is something off the-shelf availiable, please repost
for a possible solution if this is your problem and you want to solve it by building something yourself.


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
This is exactly my problem! I am concerned about 600 V reaching the input. And am desperately looking for an off the shelf device for this purpose. I haven't found any yet. So, I'll defintely like to know how to make one on my own.

The device should have an error of 5mV or less.


> If it is the signal that can reach 600V it is another matter however. That don't mean it is unsolveable but I don't know if there is something off the-shelf availiable, please repost
 
J

Johan Bengtsson

Ok, it goes like this:

The idea is like this:
1. Use two resistors to bring down the voltage.
2. Use an op-amp to bring up the voltage again.

The two resistors should bring down the voltage to
something the op-amp can handle even when the full
600V appears at the input. The op then amplifies the signal back to the original value, except that it is forced to stay within range.

The simplest possible circuit would be something like this:


----------
|
---
|R|
|1|
--- |\
| | \
*----------| + \
| | \ -----
--- | -----|R3 |---*-------*------
|R| | / ----- | |
|2| ----| - / --- |
--- | | / |R| -----|
| | |/ |4| / \ |
| | --- / \ DZ1
| | | -----
| | --- |
| | |R| |
| ----------------------->|5| |
| | | |
| --- |
| | |
| --- |
| |R| |
| |6| |
| --- |
| | |
---------*-------------------------------*--------------


Dimensioning example:
Make R1 big and R2 something that will divide the voltage down to a nice value, example R1=1M and R2=15k, this will divide the input voltage by 66.7: 10V will turn up as 0.15V and 600V will be 9.0V. The current when applying 600V will be approx 600uA and you will need at least a 0.36W resistor for R1 R4 should be equal to R1 and R6 equal to R2 or other values with the (approx) same scaling factor, like R4=100k and R6=1.5k
R5 is there to be able to adjust for the resistor tolerances, big enough to adjust for worst case values in every direction, low enough to make it easy to adjust.

R3 and DZ1 limits the output voltage to something the input card can handle, for example 11V. Note R3 should be low enough to give current enough for the input card and high enough to not blow itself, DZ1 and the op when the output from the OP reaches Vcc.

Check where you can get the voltage source for this circuit from you might have to use a separate voltage source in order to maintain insulation since the signal GND is the same for input and output.

R2 can be made higher if you put another zener parallel to it, this will give you less noice and higher accuracy since you don't have to divide down the voltage that much and therefore not amplify it that much either.


Suggested values:
without extra zener:
R1=1M
R2=15k
R3=1k
DZ1=11V (at least 0.15W with the given R3)
R4=100k
R5=5k will definitely handle it with 1% resistors, you might not need that big value R6=1.5k

with extra zener:
R1=1M
R2=150k
R3=1k
DZ1=11V
DZ2=approx 11-15V (located parallel to R2)
R4= 100k
R5=10k-20k (I have not checked this value)
R6=15k


You get the general idea, a lot of variations can be done....
The op? most standard op will do, check bandwidth, noice and what the op do when the input signal get close to Vdd (0V input), you might need a double voltage source (providing both positive and negative voltage compared to ground)


Hope this helps

/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
Top