I have a power supply that has a floating ground. We had an inspector tell us that we needed to ground the negative output of the supply because the floating ground was dangerous. Anyone have any insight into this? Any pro's con's of floating vs grounded would be helpful.
Grounding Negative of Power Supply
- Thread starter fifty60
- Start date
- Status
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
does the inspector actually have any expertise in this area?
why would it be any more dangerous ungrounded?
in some respects it matters what the voltage is. a 12VDC power supply is not likely to electrocute anyone no matter what, while a 125VDC supply is arguably more dangerous than a 120VAC supply.
all grounding does is ensure that if there is an unintentional ground fault that the OCPD gets tripped. that may or may not make any difference safety wise for your particular application.
It is a 60W 24V power supply. We also use the negative output on the supply for the ground reference on a 0-5V signal, would grounding the negative terminal have any effect on this?
Also, the power supply has internal protection and makes a dinging alarm sound when shorted. It is not a fuse, because the power supply is operable when the fault is removed. What kind of internal protection is this? Here is the supply.
Also, the power supply has internal protection and makes a dinging alarm sound when shorted. It is not a fuse, because the power supply is operable when the fault is removed. What kind of internal protection is this? Here is the supply.
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
a lot of power supplies are internally protected and thus require no ocpd as that protection is built in.
there are several different mechanisms that are used. most are self resetting. i don't know what mechanism is used for this particular power supply and the instruction sheet I found for it does not say.
I would not worry all that much about this particular power supply being grounded or not.
I don't think it will matter any one way or the other as far as it being interconnected with the 0-5 V signal.
It is often conveneint to have one side of a power supply grounded as it is easier to troubleshoot.
- Location
- Bremerton, Washington
There is no code requirement for this, but there is for higher voltages - 48 VDC.
I do a lot of control systems with similar power supplies. Mine all have the DC comm grounded, but its personal preference. Some equipment has the DC comm internally grounded, radios are one example.
I do a lot of control systems with similar power supplies. Mine all have the DC comm grounded, but its personal preference. Some equipment has the DC comm internally grounded, radios are one example.
SceneryDriver
Senior Member
- Location
- NJ
- Occupation
- Electrical and Automation Designer
Unless there is a specific need to ground the negative side of a DC power supply, and I'm referring to 24v units in this post, I try really hard to ensure that there ISN'T a connection between DC- and earth ground. Depending on what you're powering with that PS, the loads may actually REQUIRE a floating supply. Sometimes, especially with PS's that supply 0-10v or 4-20ma analog circuits, grounding the DC- actually introduces unacceptable noise into the system.
When you ground the DC-, you also loose your isolation between input and output. The whole point of a switchmode power supply is that it gives you that isolation. In fact, I believe UL508A requires you to NOT ground the DC- of your power supplies for all of the above reasons. The inspector should quote you chapter and verse of where he believes it to be required.
SceneryDriver
When you ground the DC-, you also loose your isolation between input and output. The whole point of a switchmode power supply is that it gives you that isolation. In fact, I believe UL508A requires you to NOT ground the DC- of your power supplies for all of the above reasons. The inspector should quote you chapter and verse of where he believes it to be required.
SceneryDriver
just the cowboy
Inactive, Email Never Verified
- Location
- newburgh,ny
Low voltage dc I ground for controls but not for instrumentation
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
Not so. In fact UL508a actually requires some (but not all) power supply secondaries be grounded.
Personally, I do not believe there is any real difference noise wise, everything else being equal.
I forgot to mention that this is was a CSA inspector. Does the fact that it is a Class 2 power supply mean that it does not have to have a separate ground? Also, when some one mentions "instrumentation" does that refer to any equipment requiring or producing an analog signal(4-20ma, 0-5v etc.)
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
These days instrumentation can have a digital output instead.
Tapatalk...
Tapatalk...
ELA
Senior Member
- Occupation
- Electrical Test Engineer
For the most part we floated all our 24Vdc power secondaries. I have seen some manufacture's installation instructions directing they be grounded.
In some cases there may be loads that dictate the connection.
For a person that understands anything about EMI - ("Ground is the sewer system of electronics").
Electronics designs often go to great lengths to isolate themselves from noise sources and thus why would you ruin your pristine power source by connecting it to the sewer system?
Unless of course the voltage was high enough that the code or some standard required it.
In some cases there may be loads that dictate the connection.
For a person that understands anything about EMI - ("Ground is the sewer system of electronics").
Electronics designs often go to great lengths to isolate themselves from noise sources and thus why would you ruin your pristine power source by connecting it to the sewer system?
Unless of course the voltage was high enough that the code or some standard required it.
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
People have been blaming "bad grounds" for noise problems for years. Most of the time it just is not the issue.
There are some cases where it is wise to isolate your control signals from earth. As a general practice, it serves no real purpose for most cases these days.
Most industrial communications signals these days are isolated so whether the power supply is grounded or not makes no difference whatsoever. A notable exception is RS232. A common problem with RS232 is ground loops caused by having grounded power supplies at both ends. Often this is something built into the equipment and cannot be avoided short of not using RS232 or adding some kind of isolation. However, RS232 is usually only run a short distance so it is not as big a problem as it might other wise be.
current loop signals are generally very noise resistant, so there is usually no reason to isolate them from ground.
There are some cases where it is wise to isolate your control signals from earth. As a general practice, it serves no real purpose for most cases these days.
Most industrial communications signals these days are isolated so whether the power supply is grounded or not makes no difference whatsoever. A notable exception is RS232. A common problem with RS232 is ground loops caused by having grounded power supplies at both ends. Often this is something built into the equipment and cannot be avoided short of not using RS232 or adding some kind of isolation. However, RS232 is usually only run a short distance so it is not as big a problem as it might other wise be.
current loop signals are generally very noise resistant, so there is usually no reason to isolate them from ground.
Aleman
Senior Member
- Location
- Southern Ca, USA
We usually don't ground our 24V supplies. Usually no need to. Not grounding your DC- presents no safety hazard. On the other hand, it generally will not hurt to
tie the DC common to ground. Noise issues will depend on your devices and cabling. For what I do which is industrial controls and instruments I have no issues with
either method. With 24V controls noise won't be much of an issue. I would say your inspector was full of it.
tie the DC common to ground. Noise issues will depend on your devices and cabling. For what I do which is industrial controls and instruments I have no issues with
either method. With 24V controls noise won't be much of an issue. I would say your inspector was full of it.
- Status