Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Can someone explain isolated and non-isolated circuits?

Status
Not open for further replies.

lompa

New Member
Can anyone explain why some plc's/outstations have isoloated I/O and some have non-isolated I/O's

What does the term isolated mean

and why are there two methods for PLC/RTU/Outstations?

Thanks
 
lompa said:
Can anyone explain why some plc's/outstations have isoloated I/O and some have non-isolated I/O's

What does the term isolated mean

and why are there two methods for PLC/RTU/Outstations?

I don't know anything about PLC's, but the terms are simple enough.

If devices are isolated, then there's no direct connection between them, the connection is made by some 'non-contact' method, such as optical isolation, or a transformer - or even a radio link!. Whichever method is used, the key point is that there's no direct connection.

To go back to a long recurring discussion, about earthing scopes!.

Assume you have an earthed scope, and you attempt to connect it to the chassis of a switch mode mains PSU - there will be a big BANG, because there's a direct connection between the scope ground (which is probably earthed - but not at my house!) and the live mains in the PSU.

To prevent this you can do one of three things:

1) Isolate the scope from ground by disconnecting the earth and using an isolation transformer (the scope probably already has a transformer anyway).

2) Isolate the PSU from the mains by adding an extra mains isolation transformer on it's input.

3) Isolate the connection between the scope and the PSU - this is the best in theory, but technically VERY difficult for a scope.

For your PLC's, option 3) will be the one the inputs use, the inputs are fairly basic (unlike a scope) and pretty simple to isolate, mostly using opto-couplers.
 
Nigel Goodwin said:
To go back to a long recurring discussion, about earthing scopes!.

Assume you have an earthed scope, and you attempt to connect it to the chassis of a switch mode mains PSU - there will be a big BANG, because there's a direct connection between the scope ground (which is probably earthed - but not at my house!) and the live mains in the PSU.

You are again assuming that the PSU is at fault and therefore its chassis is at live potential. Hovever this may be a rare case if you take precautions to ensure that the chassis of the PSU is also at earth potential by using an earthed chassis PSU. If you are 100% sure that this is the case then I would not worry about isloation at all !
 
Electrix said:
You are again assuming that the PSU is at fault and therefore its chassis is at live potential. Hovever this may be a rare case if you take precautions to ensure that the chassis of the PSU is also at earth potential by using an earthed chassis PSU. If you are 100% sure that this is the case then I would not worry about isloation at all !

I'm NOT assuming that, and NEVER have - the primary side of a SMPS is ALWAYS at live potential (unless you feed it through a conventional isolation transformer).
 
Nigel Goodwin said:
Electrix said:
You are again assuming that the PSU is at fault and therefore its chassis is at live potential. Hovever this may be a rare case if you take precautions to ensure that the chassis of the PSU is also at earth potential by using an earthed chassis PSU. If you are 100% sure that this is the case then I would not worry about isloation at all !

I'm NOT assuming that, and NEVER have - the primary side of a SMPS is ALWAYS at live potential (unless you feed it through a conventional isolation transformer).

ok.. i was unaware that the chassis of the SMPS is always at live potential... :?

if thats the case, then ur right.. u just can't afford to connect an earthed scope the chassis.... :D
 
Electrix said:
Nigel Goodwin said:
I'm NOT assuming that, and NEVER have - the primary side of a SMPS is ALWAYS at live potential (unless you feed it through a conventional isolation transformer).

ok.. i was unaware that the chassis of the SMPS is always at live potential... :?

if thats the case, then ur right.. u just can't afford to connect an earthed scope the chassis.... :D

You'll notice that I didn't mention 'chassis', I only mentioned the primary side of the SMPS, the chassis will normally ONLY be connected to the secondary side - the primary should be isolated via the switch-mode transformer.
 
late but nvertheless...you wanted a clear distinction between isolated and nonisolated...u got it !

Check this page and run the animaton...

**broken link removed**
 
in industrial enviroment you often get to work with different devices
which can be powered from different sources. this is why isolated i/o
exist. they cost more and since non-isolated version can also be
packed more densly, non-isolated versions are available too.
for example, isolated inputs will have optocoupler for every point.
although optocuplers can be small and inexpencive, don't expect them
in high density cards (32, 64 point or more).
isolated outputs are usually relay type. depending on size of the card
and number of outputs they can be grouped in different arangements.
truly isolated output cards usualy have only 4 or 8 outputs.
16 point output cards are usually grouped 2x8 or more frequently 1x16
to save on terminals.
if you have 2x8 arangement it means that eight outputs share same comon (power terminal). if you decide to use some of them in a certain circuit (say 120VAC), you cannot use remaining outputs for 24VDC for example. you may add relays for example to isolate them on point per point basis but this complicates panel wiring and also increases cost.
small plcs ("bricks") usually have outputs divided into few small groups
to fix this problem (grouped like 1+1+2+4+2+2+4 instead of 1x16).
some input cards have all inputs isolated but in a same group.

you also have to check polarity. some input cards can be used with either sink or source circuits (pnp or npn output from sensor). same goes for transistor outputs but this is another story...

since you mention RTU, i guess you might be using fieldbus type of network (such as modbus). use of fieldbus can dramatically reduce wiring
by bringing the i/o points to the place where they are going to be used as oposed to keeping i/o at the controller and running multiconductor cables areound the equipment. fieldbus cables are usually only shielded 2-3 conductor cable (or 5 conductor cable if power is transfered as well).
if you have plc rack loaded with bunch of 32-point cards, you will need one conductor for every input or output so cable trays or conduits will have to be significantly larger. this also quickly becomes bottleneck if the wiring is supposed to be flexible (motion system).
isolated i/o are also used when simple interlocking between different systems is required (handshaking).
note that signal isolation doesn't only apply to digital I/O. it can be required anywhere (power, communication, etc.).
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top