VFD Troubles Losing 1 Phase waveform from my Hioki 3197

Status
Not open for further replies.
vfd lift station waveform.jpg This is the input waveform reading on a VFD, this is have a phase loss alarm the past couple of weeks. It has been recording for 2 days and I am going to check it and see if it has had any voltage sags. Does that current wave form look normal?
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
View attachment 14542 This is the input waveform reading on a VFD, this is have a phase loss alarm the past couple of weeks. It has been recording for 2 days and I am going to check it and see if it has had any voltage sags. Does that current wave form look normal?

it does not look all that out of the ordinary to me. keep in mind that the diodes only conduct when the peak voltage on the ac side exceeds the dc bus voltage so the current flow comes in little pulses.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The single peak for each half of the 60Hz waveform for the red and yellow line inputs and the double peak on the blue look a little strange to me, but given the apparent severe flat topping of all three input voltage waveforms I cannot say it is unusual or a clear indication of a problem. Nor would it imply any bad consequences.
Certainly high harmonic distortion in the incoming voltage.

I do not know what conditions lead to a phase loss alarm on that unit. But if the same condition that led to the double peaks on the blue waveform could, if exaggerated, lead to zero current instead then I could see it causing an alarm.

OK, here is another observation: The red and yellow current peaks are essentially in phase and of roughly equal amplitude. If that waveform represents, as I expect, the current in the line conductors, corresponding to the voltage waveform on the same conductors (L1, L2, L3) then it is telling you that the bridge rectifier is essentially only conducting between L1 and L2 with little or no contribution from L3. That does seem to show the "normal" situation being disturbingly close to what it would be with the L3 connection completely open.
The only way that there could be current on the L3 line without any corresponding current on L1 or L2 would be if the input is actually wye configuration, with current flowing to the neutral or if there is a significant normal or fault current path to ground if there is no neutral.
Something is definitely wrong with the input to the VFD.

OK: More detail. Hard to be sure from the picture, but the double "blue" peaks may actually be overlapped blue and read for one and blue and yellow for the other.
Possibly one of the CTs for your power analyser is on backwards? The blue one specifically.
 
Last edited:
Thanks for the quick Response.

Thanks for the quick Response.

The single peak for each half of the 60Hz waveform for the red and yellow line inputs and the double peak on the blue look a little strange to me, but given the apparent severe flat topping of all three input voltage waveforms I cannot say it is unusual or a clear indication of a problem. Nor would it imply any bad consequences.
Certainly high harmonic distortion in the incoming voltage.

I do not know what conditions lead to a phase loss alarm on that unit. But if the same condition that led to the double peaks on the blue waveform could, if exaggerated, lead to zero current instead then I could see it causing an alarm.

OK, here is another observation: The red and yellow current peaks are essentially in phase and of roughly equal amplitude. If that waveform represents, as I expect, the current in the line conductors, corresponding to the voltage waveform on the same conductors (L1, L2, L3) then it is telling you that the bridge rectifier is essentially only conducting between L1 and L2 with little or no contribution from L3. That does seem to show the "normal" situation being disturbingly close to what it would be with the L3 connection completely open.
The only way that there could be current on the L3 line without any corresponding current on L1 or L2 would be if the input is actually wye configuration, with current flowing to the neutral or if there is a significant normal or fault current path to ground if there is no neutral.
Something is definitely wrong with the input to the VFD.

OK: More detail. Hard to be sure from the picture, but the double "blue" peaks may actually be overlapped blue and read for one and blue and yellow for the other.
Possibly one of the CTs for your power analyser is on backwards? The blue one specifically.

The system voltage is 480v corner grounded delta.

The idea I had in my head was that it was showing as a single phasing on the input, but it could be the CT is backwards. The other VFDs I have connected my meter to in the past did show a much more even current more in line with voltage peaks. But it has been over a year and I could be wrong on that, The output current was an even 28.1 amps on my fluke TRMS meter between all 3 phases, so that is telling me the output is fine. I need to see if my meter has recorded any problems yet.

Its like going to the mechanic and your car doesn't act up while your trouble shooting it. :)
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
What brand of VFD is it? Most of them are not actually monitoring the incoming phase voltages to determine a "Phase Loss", because adding that ability is expensive and prone to failures. What most of them do is monitor the DC bus ripple percentage. You get the 'Phase Loss" alarm when the DC bus ripple gets too high. They call it "Phase Loss" just because that's the most likely cause; other causes would have additional effects and if they look for those and don't see them, they deduce that the ripple is cause by a phase loss. Allen Bradley drives are all like that. But not all drives are that sophisticated and might give you the Phase Loss alarm for some other reason, such as a failing DC Bus capacitor.
 
Thanks for the info Jraef

Thanks for the info Jraef

What brand of VFD is it? Most of them are not actually monitoring the incoming phase voltages to determine a "Phase Loss", because adding that ability is expensive and prone to failures. What most of them do is monitor the DC bus ripple percentage. You get the 'Phase Loss" alarm when the DC bus ripple gets too high. They call it "Phase Loss" just because that's the most likely cause; other causes would have additional effects and if they look for those and don't see them, they deduce that the ripple is cause by a phase loss. Allen Bradley drives are all like that. But not all drives are that sophisticated and might give you the Phase Loss alarm for some other reason, such as a failing DC Bus capacitor.

I believe it is a westing house or honeywell. The contractor I am working for is just wanting to verify that we don't have a incoming power issue, before selling a new 40hp vfd. This is the worst part of trouble shooting electrical problems, especially when it comes to VFDs as determining the cause to be internal, external, wiring, or the motor. As they are becoming more used in industry I hope the engineers design them to have a better ways to troubleshoot them. After hooking the PQA to my first thought was some sort of failing switching component such as capacitors, SCR or some other item. The only odd items is the voltage is 10 volts low and the distorted current on only the C phase.
 
Status
Not open for further replies.
Top