I have a computer room fed by 3 UPS systems (10, 20 and 30 KVA) which are fed by 3 separate 480v 3P circuits with individual stepdown/isolation transformers. While looking for possible ground problems that a vender said may have been responsible for a glitch on one of their systems, I happend to notice that there was a 300mA unbalanced AC current on 2 of the UPS feed conduits. Is this in the range of what could be expected from stray capacitance or is this an indication of a possible fault?
Unballanced current on conduit
- Thread starter Dan O.
- Start date
- Status
ron
Senior Member
- Location
- New York, 40.7514,-73.9925
Yeah .... how are you measuring the current?
If the conductors happen to be sitting in some odd orientation near the bottom of the conduit vs the top, I would guess that there would be some additional magnetic Field imposed on one part of the conduit and not another. 300mA is pretty small and may be within the accuracy tolerance/range of your meter.
If the conductors happen to be sitting in some odd orientation near the bottom of the conduit vs the top, I would guess that there would be some additional magnetic Field imposed on one part of the conduit and not another. 300mA is pretty small and may be within the accuracy tolerance/range of your meter.
These currents were measured with a 2" dia. clamp on Amprobe with a 100mA resolution. The 300mA readings were found at several locations in the feed going into and the line coming out of the 10KVA UPS before the distribution panel and 200mA on one conduit out of the distribution panel. Spacing prevents measuring on many of the small conduits out of the distribution panel.
The transformer for the 10KVA UPS is mounted near the celling with a 3"conduit feeding the secondary to the disconnect switch before the UPS. The Amprobe won't fit on the 3" conduit and I don't see a current on either the 480v feed conduit to the transformer or the separate ground conductor. I can't measure too close to this transformer because of EMI and there may be other ground paths through mounting hardware that I haven't been able to test.
On the 20KVA ups I see a 300mA current on the conduit for the 480v feed to the disconnect switch. The transformer is internal on this UPS. The conduits in and out of the UPS are 3" and couldn't be tested. I haven't found the other side of this current on either outputs of this UPS after the distribution panel or the separate ground. The measured current was on straight ridged conduit and well clear of the transformer or wire bends.
I haven't seen unbalanced currents on any of the line cords feeding the computer equipment racks but they could be under the resolution of the Amprobe.
The transformer for the 10KVA UPS is mounted near the celling with a 3"conduit feeding the secondary to the disconnect switch before the UPS. The Amprobe won't fit on the 3" conduit and I don't see a current on either the 480v feed conduit to the transformer or the separate ground conductor. I can't measure too close to this transformer because of EMI and there may be other ground paths through mounting hardware that I haven't been able to test.
On the 20KVA ups I see a 300mA current on the conduit for the 480v feed to the disconnect switch. The transformer is internal on this UPS. The conduits in and out of the UPS are 3" and couldn't be tested. I haven't found the other side of this current on either outputs of this UPS after the distribution panel or the separate ground. The measured current was on straight ridged conduit and well clear of the transformer or wire bends.
I haven't seen unbalanced currents on any of the line cords feeding the computer equipment racks but they could be under the resolution of the Amprobe.
It sounds like you may have some difficulty trying to satisfy yourself what is going on there.
If you are clamping the amprobe around conduits that contain current carrying energized wires, I don't know what you might read doing that.
A 3" conduit seems pretty large to me for a 30 kva 480 v 3 ph UPS. ... seems to me that should be somewhere in the neighborhood of 35 amps.... why so large?
What is the load side voltage / current of the UPS? Is all your computer stuff connected with cords? seems like that could discount the conduit theory.
I wish I could be of more help, but that sounds like something you'd pretty much have to be there to try to figure out.
If you are clamping the amprobe around conduits that contain current carrying energized wires, I don't know what you might read doing that.
A 3" conduit seems pretty large to me for a 30 kva 480 v 3 ph UPS. ... seems to me that should be somewhere in the neighborhood of 35 amps.... why so large?
What is the load side voltage / current of the UPS? Is all your computer stuff connected with cords? seems like that could discount the conduit theory.
I wish I could be of more help, but that sounds like something you'd pretty much have to be there to try to figure out.
Last edited:
If you are clamping the conduit and expecting to measure current on the conduit guess again. The only way I can think of to acutally measure any current flowing in the conduit is to remove a section, or diconnect from a junction so you can break the conductive path, then put a true amp-meter in series between the open sections of conduit.
My guess is your vendor is full of beans, and justifing his maintenance charges vs warranty.
My guess is your vendor is full of beans, and justifing his maintenance charges vs warranty.
What I expect when I place the clamp around a conduit is to read the net current flowing through the core of the clamp wether it is flowing on the wires inside or the conduit itself. If everything is wired properly, the current flowing on the hot and neutral wires will cancle out leaving only the ground currents.
The inductive coupling between wires run in parallel or coaxial will try to induce currents flowing in the opposite direction. So what I see as 300mA may only be the residual of a larger unballanced current on the wires and the counter current on the conduit.
I would expect to see no ground currents if everything was perfect. But I realize that nothing is ever perfect. All I really want to know here is if this level is in the normally expected range or if it's high enough that I should try to chase it down.
The conduit is not actually 3" but just slightly larger than 2" od. What I was calling 2" is actually 1.5" according to my digital calipers.
The inductive coupling between wires run in parallel or coaxial will try to induce currents flowing in the opposite direction. So what I see as 300mA may only be the residual of a larger unballanced current on the wires and the counter current on the conduit.
I would expect to see no ground currents if everything was perfect. But I realize that nothing is ever perfect. All I really want to know here is if this level is in the normally expected range or if it's high enough that I should try to chase it down.
The conduit is not actually 3" but just slightly larger than 2" od. What I was calling 2" is actually 1.5" according to my digital calipers.
brian john
Senior Member
- Location
- Leesburg, VA
It depends on the system loads and size of system, I have seen from 1-15 amps on a properly installed and tested system.
If you want a better measurement go to the neutral/ground bond.
If you want a better measurement go to the neutral/ground bond.
weressl
Esteemed Member
There are always equalizing currents flowing in your grounding system. The currents have various sources, some of it has to do with the capacitive charging of the power sytem, some due common mode noise and/or inductive coupling and some due to the fact that our materials were designed to pure sine-wave voltage and behave diffrently with the harmonic-laden power system that is so characteristic of UPS' rectified front ends and the not-so-pure synthetic sinewave on the output connected to non-linear loads.
- Status