Currant imbalance submersible pump 1

[nj01]

We have a 130kw submersible pump on a VSD that is showing nearly 9% currant imbalance. It also shows induced current on the earth rising from the motor.

The pump is 100m deep and the riser pipe is 8 inch and the well casing is 12 inch. Pump is 12 inch.

95 mm2 cable is run down to the 6 wire 130kw motor (three conductors up the riser, 6 out of motor).

Insualtion resistance starts @ 35mOhm and rises to 80mOhm

Winding resistance is .09 .09 .09 - so balanced.

From memory and would have to check , the DAR test was 1.1

We have tried all three hook ups and the imbalance follows the blue phase in each hookup.

Amps

212 185 208

voltage

r-w w-b b-r
449 450 444

Earth currant is 26a (3.5V) when connected to the well head imbalance is there, when disconnected imbalance is still there.



Any ideas on what is happening and how to fix?

Cheers

N

 

[waross]

blue phase in each hookup.

The blue phase in the VFD, the cable or the motor?

 

[EdStainless]

New motor?
New VFD?
New Cable?
or has it been different in the past?

 

[nj01]

The blue phase in the VFD, the cable or the motor?

drop cable from the well head to the motor, imabalance stays with that conductor in all 3 hook up combinations.

 

[nj01]

New motor?
New VFD?
New Cable?
or has it been different in the past?

New pump/motor and cable - existing vsd. we have no way of knowing if there was an issue with the old pump in terms of imbalance as the old paperwork from comission only noted drive amps as average of three phases.

 

[waross]

It sounds like flat cable.
The inductive reactance per phase cable may not be equal.
We see that on long un-transposed distribution circuits.
The reactance per phase cable may not be equal.
I am taking a chance and assuming that you did roll the connections on the VFD.

 

[nj01]

It sounds like flat cable.
The inductive reactance per phase cable may not be equal.
We see that on long un-transposed distribution circuits.
The reactance per phase cable may not be equal.
I am taking a chance and assuming that you did roll the connections on the VFD.

We did not change the connections on the VFD, we assumed as the low current heading down the well that the issue was with the drop cables/motor or join.

 

[Marke]

This is a common problem.

When the pump was lowered down the well casing, the four conductors (three phase plus earth) need to be dead vertical and not rotate around the riser pipe.
If any one phase conductor is allowed to move around the riser by a partial turn, you will get a transformer effect inducing current into the earth conductor.
Typically in NZ, the earth conductor from the motor is bonded to the top of the well casing and this creates a shorted turn and induced current flows, but the induced voltage is very low.
I have seen earth currents as high as 120Amps and current imbalances as high as 20% at full load.
In extreme cases, we advise to lift the pump and be extra vigilent about how the cables go down the hole.

If the motor earth is bonded to the top of the well casing, and this is then earthed back to the supply earth bar, try running the earth from the motor directly back to the supply earth bar without connecting to the top of the well casing, and run a separate earth from the earth bar to the top of the well casing.
This has proved beneficial in some cases.

 

[EdStainless]

Flat cable or round?
Flat is a PIA since the inductance and capacitance is different for each conductor.
Here is a tip going forward, if you are running centrifugal pumps record the POWER for each phase.

 

[Gr8blu]

OP writes: resistance of each phase is similar (to within 5 milliohms, assuming the 0.09 readings given is for OHM, and accounting for rounding error).
Presumably, the resistance is measured with each phase disconnected at the drive and includes both the cable run to the motor and the motor winding itself.

However, as waross noted, chances are excellent that the INDUCTANCE per phase is not identical, hence the unbalanced currents. It is unlikely that there is significant change in the motor winding inductance without a corresponding change in the resistance - but the cable parameters may be dominating the equation. most likely is that the cable has wrapped itself at least partially around the riser pipe.

To alleviate, do as Marke mentioned: pull motor and cabling up. Check each length of cabling for damage/faults and replace as required. Re-insert motor and cabling being careful to ensure cables (and neutral) drop STRAIGHT down the shaft with no twisting.

 

[nj01]

OP writes: resistance of each phase is similar (to within 5 milliohms, assuming the 0.09 readings given is for OHM, and accounting for rounding error).
Presumably, the resistance is measured with each phase disconnected at the drive and includes both the cable run to the motor and the motor winding itself.

However, as waross noted, chances are excellent that the INDUCTANCE per phase is not identical, hence the unbalanced currents. It is unlikely that there is significant change in the motor winding inductance without a corresponding change in the resistance - but the cable parameters may be dominating the equation. most likely is that the cable has wrapped itself at least partially around the riser pipe.

To alleviate, do as Marke mentioned: pull motor and cabling up. Check each length of cabling for damage/faults and replace as required. Re-insert motor and cabling being careful to ensure cables (and neutral) drop STRAIGHT down the shaft with no twisting.

We have lifted the pump, no damage at all. The cables look to have one wrap arcoss the total of the 100m of riser.

The pump went back to the supplier as the supllied the cable and made the joins, insualtion resistance was low @ 30Mohm, we thought that there may have been moisture in that conductor ( the imbalance follow that conductor with all three hookps)

When I went into see the supplier about the issue, they had dientifed a cable that was suspect - the same one that we had identified. They have now rejoined the cables and run the pump to their satisfaction 2.5-3% imbalance.

Could moisture cause increased impeadance in that conductor, no water run out of the cable - but considering the low insulation resistance and imbalance issues it could be the problem?

 

[Marke]

We recently had a case of quite bad imbalance, one phase current high and earth current.
Pump 100M down the hole.

Pump was lifted and reconnected using the same cables, in a test well. All cables were just coiled up on the ground.
Almost perfect balance, so not cables.

Pump was reinstalled down the hole and extra care was taken with the cables, absolutely no wrap around the riser.
Current in earth was much lower and imbalance reduced, but still not perfect.

No change to cables, just a better control over cable layout as the pump went down.

 

[waross]

It could have been a manufacturing defect.
I found a defect in a new cable.
Apparently, as they were drawing the bare cable through the extruding machine to apply the insulation, they came to the end of the cable reel. They mounted a full reel and used a butt splice to draw the cable through the extruder behind the last of the previous cable.
They forgot to cut out the butt splice and we got a new cable with a butt splice inside the extruded insulation.
I am not saying that that is the case here, but anything can happen.

 

[EdStainless]

When we made cable we used a special splice for end of reel.
It added a lot of resistance so it was easy to find if the marking was ever lost.
We did have some cables where butt welds were allowed.
After all we were making >12,000' spools.
We would run symmetrical round cable most of the way down.
We had fixed lengths of flat cable (100' and 250').