Frequency drives burning up ug splices?????

Status
Not open for further replies.
G

GWElect

Member
Location
Grants pass. oregon US
I do allot of underground well wire fault locates and repairs. Lately I've had a number of uf splice kits fail. The well contractor I work with have had a number of wire spices down the shaft fail. The common denominator seems to be they are on frequency drives. The spices are burning up evenly inside the uf kits as though the sealant is conducting. Anyone seen this before? Any Ideas?
 
I suppose that it is possible that the high frequency properties or the transient high voltage properties of the sealant are not up to the job.
Two other possibilities that come to my grasping mind are:
The long run of non VFD cable could be causing standing wave patterns that are causing higher than rated voltages or capacitive currents in the sealant, and the standing wave pattern just happens to hit at the splice.
Or the mechanical characteristics of the splice itself (varying conductor diameter, etc.) are actually causing an impedance jump at the splice at high frequencies forcing a node in the standing wave pattern to occur there.

Tapatalk!
 
GoldDigger said:
I suppose that it is possible that the high frequency properties or the transient high voltage properties of the sealant are not up to the job.
Two other possibilities that come to my grasping mind are:
The long run of non VFD cable could be causing standing wave patterns that are causing higher than rated voltages or capacitive currents in the sealant, and the standing wave pattern just happens to hit at the splice.
Or the mechanical characteristics of the splice itself (varying conductor diameter, etc.) are actually causing an impedance jump at the splice at high frequencies forcing a node in the standing wave pattern to occur there.

Tapatalk!
Click to expand...
I agree, although I don't think the standing wave pattern hitting the splice is even necessary. It's much more likely that the splice system design did not take the possibility of the higher voltage spikes into consideration at all. There are several papers I have read recently regarding the type of motor lead insulation that should be used, all espousing that THHN should NOT be used between a drive and motor (RHHW is better for that). One of the arguments is that the PVC used in THHN can have voids (bubbles) in it, and the presence of those voids makes it much much easier for the standing wave pulses to punch though. If microscopic voids will make it easier, think about the voids that might be present in a field made splice!

Someone in the submersible industry should think about this more I think.
 
So, other than adding load reactors, can you think of a solution that would allow that type of splice kit to be used?
What are the chances that a different manufacturer's splice would hold up better?
Just never use splices in well casing when using a VFD?

Tapatalk!
 
Oil wells, geothermal wells, or plain water wells?

Am surmising water wells? Depth probably not sufficient even at 50 kHz to cause standing waves, so possibly the culprit is skin effect.
A 12 AWG wire at 50 kHz is equivalent to only a 22 AWG at dc. Cannot directly compare as the 50 KHz (or down to 20kHz, depending on vfd PWM frequency) is only a ripple, not the fundamental. Still could contribute to overheating of the splice, but should not due to being 'water cooled'. Splice above water level could overheat/oxidize and fail.
 
GoldDigger said:
So, other than adding load reactors, can you think of a solution that would allow that type of splice kit to be used?
What are the chances that a different manufacturer's splice would hold up better?
Just never use splices in well casing when using a VFD?

Tapatalk!
Click to expand...
I would use a sine wave (dV/dt) filter, not just a load reactor. Like a KMG series from TCI or a Series A filter from MTE. Not every installation needs to have this level of protection, but once you KNOW it does, then you have to deal with it, and this is the simplest way, assuming changing the cables to be shielded is not an option (which is likely the case on an ESP).
 
Is there a common denominator like all the times this happens is on a 480 volt drive/motor, or does it happen on 208-240 volt as well?

Or are these single phase pumps driven by something like Franklin's "Mono Drive" or Berkley "Intellidrive"?
 
GWElect said:
I do allot of underground well wire fault locates and repairs. Lately I've had a number of uf splice kits fail. The well contractor I work with have had a number of wire spices down the shaft fail. The common denominator seems to be they are on frequency drives. The spices are burning up evenly inside the uf kits as though the sealant is conducting. Anyone seen this before? Any Ideas?
Click to expand...

There are alot of these being installed and I had not heard of this yet. Time to talk to my local well guy and see what his experience is.
 
Thanks for the information!! These are 240 volt water wells on residential properties. Some have splices in the wires before the control boxes where changed to VSDs. Most are thhn in the ground.
 
Am I understanding correctly that there is cable especially for the run between the VSD and the pump and that thhn may be a problem in itself even without splices? its uf cable
 
Last edited:
Jraef said:
I would use a sine wave (dV/dt) filter, not just a load reactor. Like a KMG series from TCI or a Series A filter from MTE. Not every installation needs to have this level of protection, but once you KNOW it does, then you have to deal with it, and this is the simplest way, assuming changing the cables to be shielded is not an option (which is likely the case on an ESP).
Click to expand...

After your feedback I researched vsd cable online. Seems that pvc wire like thhn will arc and heat up between layers. Voltage at terminals can become twice nominal voltage due to capacitance and standing waves. I find it amazing these are being sold to the pump companies to install on existing systems!!!

What are the odds the dv/dt filters will correct all this and allow use on existing thhn systems?
 
GWElect said:
I stated thhn in the ground. I mean uf cable with thhn in it.
Click to expand...
When you say "in the ground" are you talking about in the well casing" If so you should be using "submersible well cable" instead of UF cable. They use different conductor insulation then THWN. That could still not matter if the splices are what is failing though. Usually there is a short lead on the pump motor that will be spliced onto longer leads as every install will have a different length, and is usually insulated with heat shrink.
 
GWElect said:
After your feedback I researched vsd cable online. Seems that pvc wire like thhn will arc and heat up between layers. Voltage at terminals can become twice nominal voltage due to capacitance and standing waves. I find it amazing these are being sold to the pump companies to install on existing systems!!!

What are the odds the dv/dt filters will correct all this and allow use on existing thhn systems?
Click to expand...

Pretty good actually. The standing / reflected wave phenomenon is not quite as rampant as the filter people would like you to believe, but when you have it, you have it, and that is one of the possible cures.
 
Jraef said:
Pretty good actually. The standing / reflected wave phenomenon is not quite as rampant as the filter people would like you to believe, but when you have it, you have it, and that is one of the possible cures.
Click to expand...
I too don't think it is near as common as the filter people say. I have installed at least 100 drives with motor lead lengths of between 75 and 500 feet without any type of filtering and we have not seen conductor or motor failures. The conductors were all THHN/THWN. Many drives say to limit the motor lead length to 50' but most of the time I can't even get out of the MCC room without using more than 50' of conductor.
 
Jraef said:
Pretty good actually. The standing / reflected wave phenomenon is not quite as rampant as the filter people would like you to believe, but when you have it, you have it, and that is one of the possible cures.[/QUOTE


Thanks for the info! I'm now finding the technical data sheets for some drives are recommending output reactors if the line to the pump is 100'. Then saying they are mandatory over 300'. Then there saying to change to a dv/dt filter over 500'.( Lenze-actech.com. "When to use line or load reactor")

Is it safe to say the dv/dt filters would be better even at a lesser length?
Click to expand...
 
GWElect said:
Is it safe to say the dv/dt filters would be better even at a lesser length?
Click to expand...
Let's put it this way, the only harm to using a dV/dt filter when something less expensive might have worked, will be to your pocketbook. If you don't know and/or don't want to have another call back, then that might be the way to go.

As to why some drives are fine, others are not, it has a lot to do with the drive and/or supplier. Some VFD mfrs design filtering into the VFD and/or use transistor firing techniques to allow very long distances between the drive and motor without special attention. Those drives are more heavily engineered and tend to be more expensive, but are designed for industrial installations. Some mfrs are more interested in selling cheap drives to people who are looking for the least expensive way to say "We supplied a VFD", but don't really know or care about the details. Those are the ones that say "Use a reactor whenever the lead length is XX feet or more as a CYA statement. That length by the way varies by which way the wind blows, because it's a SWAG anyway. If a mfr. shows a longer lead length, it is usually a tested lead length, not a SWAG.
 
Status
Not open for further replies.
Top