Interposing Relays on output cards

Erik vdH

Member
Join Date
Sep 2006
Location
NZ
Posts
42
Hi all,
I have a contractor trying to convince me to install 'interposing relays' on the output of PLC cards. The supposed benefits being
1) Easier to replace one failed relay rather than PLC output card
2) Less stress on the PLC output card
3) Can divert the output to a different relay in case of failure on no IO re-mapping
4) different voltages possible on outputs by using different relays and supplies
5) Can use high density cards (more outputs because can use lower voltage versions)

Disadvantages could be
1) More wiring - more possible points of failure
2) Higher initial cost

anyone got any other considerations or thoughts?
Cheers
Erik
 
I agree with your contractor.

When I did use high density i/o, I always used interposing relays.

However, nowadays I use more and more finegrained distributed i/o (ET200S), and less and less high density i/o.
With finegrained i/o I get all the advantages and none of the disadvantages. I can in many cases save both interposing relay and a separate terminal strip. It is much more flexible. I can place the distributed i/o where it is needed in stead of centrally. It saves time and costs.
 
I am with you Jesper on this one. I find myself using more field distributed I/O (like the ET200S). Also on output cards most of the time I have protect the card with a terminal block style fuse on each output. I typically stay away from high-density cards as the wiring gets to be a mess and I hate having to use break-out/wiring harness type blocks to wire to. I agree with your position on interposing relays, but in my case I hate having to create more drawings and giving up real-estate in my panel for these relays.
 
I use high density transistor output modules and terminal blocks with relays built in. Same effect, the relays are outside the PLC and easily replaced. And the transistor cards last longer and cost less typically.

ABB has terminal blocks and pre-wired connectors for many many PLC I/O modules.
 
for 'me'

I think it depends... I would agree with others, but I have also just removed several (20) from a system, the relays were failing, not the outputs, these were discrete inputs to a few servos, these being the only thing that is failing on the system over and over, I have decided to remove the relays

I know its not much but there is also a time lag that you must consider, so again I think it would depend on the application
 
I hate interposing relays. They are a waste of my time, panel space, and money.

I'm a believer in getting the right PLC output card to drive the outputs. This of course is extremely difficult nowadays with PLC's getting so small they can't drive even the smallest of solenoids. So in the end, sometimes you end up with interposing relays because the PLC outputs are too wimpy. And, the fact that PLC outputs are too wimpy is the cause of the whole problem.

So what Erik must do is evaluate whether interposing relays are necessary based on the PLC output ratings and the size of the loads they are driving. And, if you are using PLC relay outputs check the number of cycles per year to see how long they will last. If none of that pans out with room to spare then you will need interposing relays.
 
I should moderate that
"I always use interposing relays with high density i/o"
to
"I always use interposing relays with high density outputs for either inductive loads or interface signals"

and if the load is both inductive and has to switch frequently, the interposing relays would be SSR's not regular dry contacts.
 
geniusintraining said:
[...]I have also just removed several (20) from a system, the relays were failing, not the outputs, these were discrete inputs to a few servos, these being the only thing that is failing on the system over and over, I have decided to remove the relays.
What was the application exactly ? Relays failing for transmitting discrete signals sound odd - unless of course they were regular dry contact relays, and they had to switch way too often. We use (interposing) opto-couplers in such cases.
 
I prefer interposing relays on all my outputs. I don't agree with all of your contractors reasons, but he is right. My main reason is that I got burned with triac outputs too many times. But, I also like the fact that if there are interposing relays, I can just use 24vDC output cards, and it does not matter what the actual devices in the field require. We're often given the wrong information on some of those devices. Usually its a device that has options, and the supplier tells us what they bought LAST time.


-jeff
 
Gerry M said:
I hate interposing relays. They are a waste of my time, panel space, and money.

I'm a believer in getting the right PLC output card to drive the outputs. This of course is extremely difficult nowadays with PLC's getting so small they can't drive even the smallest of solenoids. So in the end, sometimes you end up with interposing relays because the PLC outputs are too wimpy. And, the fact that PLC outputs are too wimpy is the cause of the whole problem.

👨🏻‍🏫

Here here. I use interposing relays when the load demands it. Whenever possible, I select a card rated for the load and unless the card has its own output protection I fuse each of the outputs.
 
JesperMP said:
unless of course they were regular dry contact relays, and they had to switch way too often..

Hi Jesper,

Yes and Yes... I agree that I could of changed them out with a opto's, but why? why not just remove them? 24v output 24v input to the servo

These were a output from a S5 going to a SEW Movidine (spelling) servo (enable, etc) several thousand cycles a day, the S5's are still running strong and I have had very few failures with the outputs from them, but the relays would start to fail after a few years, so when I run across a bad relay, I just wire it straight or when I have time and the line are not running, I change them over
 
I have done similarly before. I had to tie the grounds together.
Theoretically, an electrical problem can travel accross both control systems. In reality it is not so big a threat, in particular if the distances between the two systems are not so great.
 
Thank you Gerry M!

I can't believe I'm reading this, one-size-fits-all does NOT work!!! This is why everyone
has more than one kind of output card for their PLCs. Your contractor doesn't seam to
understand the difference between different types of outputs (*)and only considers max load
and max number or points.

Interposing relays have their function. using them for few selected loads is part of the job
(it's part of design process and involves reading specs and sizing...).
but putting blindly one relay on every single output is rather extreme. I call this
waste of resources and it's not just overkill, it could be very poor design too...

1) Easier to replace one failed relay rather than PLC output card
I don't think so - it's not easier! It may be cheaper but only if:
a) relays ware already installed
b) failure is on relay or relay base (and not on the card for example)
c) you can identify failure fast enough (production downtime is ticking bomb costing much more than new card)
(sorry for splitting hairs)

2) Less stress on the PLC output card
If card is designed to handle the load, WHO CARES? There is no guarantee that relay will not be
overloaded either, shall we all start using PLC card+interposing relay+interposing contactor+bunch of
paralleled contactors - on each point? Maybe everyone should be afraid to sit on the chair or use stars without fall protection gear?

3) Can divert the output to a different relay in case of failure on no IO re-mapping.
someone still lives in the past and prefers costly wiring change versus simple program change?
sounds like someone who doesn't get along with technology, maybe a brainstorming session on why the
PLCs ware invented in the first place? (Hint: it's not that there was shortage of relays...)

4) Different voltages possible on outputs by using different relays and supplies
nothing that different output cards would not handle, in fact there are relay output cards,
some have each output isolated and can handle bigger loads.

5) Can use high density cards (more outputs because can use lower voltage versions)
So... you cannot use high density cards unless you add bunch of relays? How is that?
i've been and still use number of different platforms and some of the output cards allow as little as 100mA
per point (plus some additional limitations such as max group current, max % of energized outputs etc.)
none has failed yet... (i did have few cards with dead output but so far every single one was
result of poor quality control - bad out of box and exchanged since new and under warranty)

Btw. there are also high density output cards with protection so you don't have to sweat
for each short made by contractors...
Sure there are heavier loads one can encounter (some valve types take more than 2A etc.)
This is what designer should look into and decide what kind of output is needed, and maybe
add relay or contactor - when needed.



I agree on disadvantages:
1) More wiring - more possible points of failure
2) Higher initial cost
3) Take lot's of space (even the TB style)
4) Relays are slow and don't produce clean signal, they have a contact bounce - see (*) and makes them useless
in so many applications (anything high speed, packaging, triggering cameras etc.)


Another note: The only saving on rack based system when using high density versus regular cards is the
rack space (used number of slots), overal panel space saving is usually only marginal (you still need
to put those TBs somewhere), there is no money or time saving (TBs and cables also cost, labor for
32 point is still same regardless if they are on one or two cards etc.). This is one of the reasons
other I/O systems have emerged.
 
Last edited:
panic mode said:
Thank you Gerry M!

I can't believe I'm reading this, one-size-fits-all does NOT work!!! This is why everyone
has more than one kind of output card for their PLCs.

There is nothing wrong with having a standard goto setup. Of course the same setup will not work everytime, but when you are intimately familiar with how particular cards or a particular scheme works, its easier on you when you do have to deviate, and there are fewer opportunities for gotchas.

panic mode said:
3) Can divert the output to a different relay in case of failure on no IO re-mapping.
someone still lives in the past and prefers costly wiring change versus simple program change?
sounds like someone who doesn't get along with technology, maybe a brainstorming session on why the
PLCs ware invented in the first place? (Hint: it's not that there was shortage of relays...)

That sounds elitist. So, if someone does not know how to program, they should not be working maintenance in a facility that has a PLC? Many of my customers do not have a programmer, heck, most dont even have the software, but they do have some darn good electricians. For them, the wiring change is simple and easy, and the programming changes are costly.

panic mode said:
4) Different voltages possible on outputs by using different relays and supplies
nothing that different output cards would not handle, in fact there are relay output cards,
some have each output isolated and can handle bigger loads.

But then you get into the problem of keeping spares for everything. Use 2 different kinds of cards instead of 1, and thats twice as many spares you need on the shelf (for some PLCs that can be $1k a pop). Plus, as I said before, sometimes we get the wrong specs on equipment and do not know until we're out there. So we have a $10k piece of equipment with a leadtime of months that was accidentally bought 120V instead of 24V ... we could buy 2 new cards for the 1 input and 1 output we need from the thing, and hope we have room in the rack and go ahead and double the spares the customer has to keep on the shelf not to mention the fact that even though it wasn't our fault we will be blamed for the delay because they are waiting on us ... or ... no wait a minute, all the inputs and outputs are already wired up to relays, so no big deal. At worst we have to change out 1 relay for the input! Hooray for us, the fact that we have a standard way of doing things that helps compensate for other contractors deficiencies has saved the day. We'll just pat ourselves on the back because no one else will. That scenario, or versions of it, have been pretty common in the 10 years I've been doing this. I'd guess 25% of the jobs get jacked up, or almost jacked up in similar fashion.


-jeff
 

Similar Topics

Just wondering what other folks have used on their digital output loads when using this digital output module. There will be a number of...
Replies
5
Views
1,046
My apologies for revisiting a subject that has probably been beaten to death BUT some input please....I have had problems with PLC outputs failing...
Replies
18
Views
5,676
I've read two or three post on this PLC Talk where it was stated that 24 VDC powered relays were preferred for interposing relays. A few...
Replies
23
Views
12,170
What is the use of Interposing relay, Is it only required for solenoid Valves because of different requirement Voltage (PLC is restricted to...
Replies
14
Views
9,847
  • Poll
Derived from.... http://www.plctalk.net/qanda/showthread.php?t=31776 there have been a lot of comments on that thread... but just to see how...
Replies
3
Views
3,675
Back
Top Bottom