Power Supply

[chaterpilar]

The power supply used for inputs ( 24vdc) should be grounded or left floating ? I have two panels using 24vdc inputs for PLC inputs and i noticed that the voltage between 24+ and ground is 24 volts and 10 volts in other.

 

This means the power supply (-ve) is grounded in one panel and is floating in other.

 

Now, my question is which is a better method? grounding or floating and why?

 

In case of field sensors failures ( like pressure switches, level switches etc. getting grounded),which system of power supply will be better to handle this failure?

 

Chaterpilar

[marke]

Hello Chaterpilar

 

If the input system is grounded, it should only be grounded in one place.

If the input rack itself is grounded, then all inputs must be floating.

 

If the input system is grounded in two places, there can be major issues with gound loop currents and voltages and this can cause incorrect readings etc.

 

There are different ideas on this subject, and manufacturers have their own preferences.

 

I believe that it is best to leave floating, or to loosely ground through a resistor to limit any current loops.

 

If the inputs are left totally floating, then induced voltage can cause them to drift to a high voltage which may have safety implications.

 

Best regards,

[AB2005]

Dear Chaterpilar,

 

We have upgraded our old machines by removing old electronics and automate them by using PLCs Allen Bradley. This is our normal practice to leave the negative of power supply float. According to JAAKKO POYRY standards, there should be a separate earthing/grounding arrangement for instruments. You shouldn’t ground the PLC rack/instruments with a common ground bar.

 

 

 

[chaterpilar]

Thanks for your inputs.

[chaterpilar]

We are using Telemecanique Modicon Premium plc for our water treatment plants.

 

Recently we noticed that a red LED I/O failure lights up on Input card and the plant stops.

 

However after around 10 mins the red LED goes off and the plant is ready to start.

 

While checking the inputs to PLC i found one input fluctuating from 12 vdc to 24 vdc (normal ).

 

Input is 24 vdc + and is floating.

 

On further investigation, found pressure switch ( glycerine filled had leaked and was making 24vdc to be grounded).

 

Now, whether this can effect the card ? what other conditions can create the red LED (I/O failure) to come ON.

 

 

if the card is problematic why does the plant run without any faults for 4-5 days ?

 

Is it normal for a input card to be erratic as in this case.

 

 

regards

 

Chaterpilar

[AB2005]

Dear Chaterpilar;

 

Perhaps, I can help you but make some things clear.

What is the type of your input card “Sinking” or “Sourcing?”

You told that a red LED I/O failure lights up on Input card and the plant stops and after 10 minutes the red LED goes off and the plant is ready to start. Is that fault occurring continuously?

You told that “one input fluctuating from 12 vdc to 24 vdc (normal )”. I couldn’t understand about “normal”. If this is a digital signal, then it should be 0 or 1 (0V or 24V).

What is the input address of Pressure Switch? Is it same that you have mentioned about one input fluctuating from 12vdc to 24vdc?

Please make these questions clear.

 

[chaterpilar]

Hello AB,

 

The input card is sinking type.

 

All inputs going in card are +24vdc in high state and 0vdc in lo state.

 

By normal i meant it should have been +24vdc but it was fluctuating.between 12 vdc to 24vdc. Yes the pressure switch was causing problem however it has been replaced since but Red I/O failure signal still comes frequently.

 

This failure ( I/O) comes sometimes 2 times a dayother times may b after 3 days. ( erratic)

 

I can change the input card , but i dont want to do that till other causes of this failure are eliminated hence saving the new Input card.

 

regards

 

Chaterpilar

[AB2005]

Hello Chaterpilar

 

Fluctuation between 12 – 24VDC means there is some leakage/noise in the cable which carrying the signal of pressure switch. Have you check or replaced the pressure switch?

If there is leakage in the cable, then I suggest you to connect a resistance in the parallel of sensor’s out put at PLC input card. Calculate the resistance value and make sure that out put current don’t cross the limit of pressure switch max out put current.

If this is a noise, then connect a capacitor. Capacitor should be from 10uf to 100uf.

I have faced many problems like that at our plant. Once, Hydraulic Arms of a Reel Stand of a machine started open by self until limit switch stopped them. A relay was energized these arms. That fault was occurred one or two times a day. Fault was found that there was some leakage in the multi core control cable of relay which some time increased at the limit that it energized the relay. I simply add a load resistance in the parallel of relay coil and removed that fault.

 

[AB2005]

Hello Chaterpilar,

 

I have again read your posts and found that you have replaced the pressure switch. I am sure that the cable of pressure switch is faulted. You should carry the signal of pressure switch through a separate cable up to PLC. These types of signal cables should be a separate duct which should be far off (about 1 feet distance) from power cable ducts.

Changing of input card should be at last stage. Recall, that Mr. Jomega, the member of this forum, once, posted in a forum that;

"Never assume any thing. Always check and make yourself clear"

This is a very good rule in finding any fault. Some times, we left some thing by assuming that it wouldn’t the cause of fault. But it ………..

Please give feed back to me for experience.

[jeff]

On the question of floating supplies, I'd like to add my 2 watts worth.

 

I started floating my supplies after scoping out ground points on a few machines and realizing how much noise there seemed to be. I figured that floating was quieter. I've found that few clients agree however. They'll usually send back schematics with floating supplies and ask (tell) me to, "ground 'em." Maintenance guys like to be able to ground their meters on any common point (door hinge, etc.) and read the whole panel. Fair enough.

 

For the same noise reason, I sometimes use seperate supplies (old-school preference for linears) for inputs and outputs. Solenoids can seriously kick without proper flywheel diodes (best placed at source). Diodes loosen, or die natural deaths and are never replaced. The problem here is that the electricians wiring the panels will occasionally cross ground commons between supplies. If this isn't caught before power-up, it can damage input cards. Some might suggest jumping the grounds (0VDC) between supplies but then, what's the point of having seperate supplies?

 

My rule of thumb for input thresholds is 13 volts but, if you're even close, it's time to look at your design again - something's not right. If you're dropping that kind of voltage due to the length of your cabling, consider remote IO or even changing to 120VAC inputs (yuck!). Sooner or later, questionable levels will come back to bite ya.

 

Jeff

---------------------------------------------------------------

One test is worth a thousand opinions

[AnilSonwane]

The power supply used for inputs ( 24vdc) should be grounded or left floating ? I have two panels using 24vdc inputs for PLC inputs and i noticed that the voltage between 24+ and ground is 24 volts and 10 volts in other.

 

This means the power supply (-ve) is grounded in one panel and is floating in other.

 

Now, my question is which is a better method? grounding or floating and why?

 

In case of field sensors failures ( like pressure switches, level switches etc. getting grounded),which system of power supply will be better to handle this failure?

 

Chaterpilar

 

CTUALLY 24+ is 24 volt with respect to 0volt only thatswhy always we should use 0 volt of same power supply