MV VFD Cable Length Concern 5

[Imumahdi]

Greetings.

I haven been looking at all various threads about VFD cable lengths and the insightful responses from all the VFD gurus in this site, however, almost all of these topics were focused at L.V VFDs and very few had discussed this subject in M.V domain.

I have just joined a new company and was briefed to follow an on-going project of ours that involves an MV VFD for our 11kV, 2300kW Export Pump. The unit was already selected before I joined and it's being shipped as we speak; However, upon reading the vendor documents and datasheet, I became concerned that the 500m distance between the VFD room and the Export pump might create a set of issues which might end-up damaging the Export Pump so I raised this concern to the vendor technical team and they assured me that they selected VFD has carefully selected parameters to mitigate any of cable distance issues that might arise.

I'll attach the parameters below and continue addressing my points of concerns:

------------------------------------------------------------
TYPE OF VFD INVERTER (LCI/VSI) : VSI
NUMBER OF PULSES - SUPPLY SIDE/MOTOR SIDE: 54 PULSE / 54 PULSE
INPUT POWER FACTOR AT VFD: 0.85
SPEED RANGE: 1&~150%
POWER CABLE LENGTH FROM VFD PANEL TO MOTOR: COMPLY
HARMONIC STUDY (OVERALL / LIMITED TO EACH DRIVE): COMPLY
CAPACITANCE WITH SERIES INDUCTANCE: Provides avoidance of 3 resonance point
frequencies
REACTOR / FILTERS: NO REACTOR / FILTERS IN VFD
HARMONIC FILTER REQUIRED (YES / NO) : NO
SEPARATELY MOUNTED REACTOR REQUIRED (YES / NO): NO
SEMICONDUCTOR DEVICES (DIODE / THYRISTOR / IGBT / IGC: IGBT
CONVERTER CONFIGURATION (3-PH UNCONTROLLED / 3-PH SEMI: 3-PH UNCONTROLLED
SEMICONDUCTORS PER BRANCH: 36
NO. OF PULSE: 54
TOTAL INPUT RMS CURRENT (IS) (AT 100%,75%,50% LOAD) : 577,433,288A
% CURRENT THD (AT 100%,75%,50% LOAD) : 3%,3.5%,5%,7%
FUMNDAMENTAL POWER FACTOR (DPF): 0.92,0.92,0.92
CURRENT HARMONIC PROFIE :
5th: 2.2A @ 100% DEG
7th: 1.06A @ 100% DEG
11th: 1.42A @ 100% DEG
13th: 1.36A @ 100% DEG
17th: 0.11A @ 100% DEG
19th: 0.07A @ 100% DEG
23th: 0.03A @ 100% DEG
25th: 0.03A @ 100% DEG

DC LINK VOLTAGE: 1800v
MAX. DC LINK CURRENT : 380A
INPUT REACTOR : NO
DC LINK REACTOR/CAPACITOR : NO
SWITCHING FREQUENCY RANGE: 800Hz-2000Hz
Voltage THD: 3%
CURRENT THD: 3%
VFD OUTPUT VOLTAGE SURGE (dv/dt) AT MOTOR TERMINAL:
AH-PH : 1kV(Peak/sec.)
BH-GND: 0.6kV(Peak/Sec.)


-----------------------------------------------------------------------------

My concern is how the vendor achieved the stated values of 1kV,0.6kV for such long cable length without the use of reactors/filters?

 

[FacEngrPE]

A DC link voltage of 1800V is too low for a motor voltage of 11 kV. is there a mistake in one of these numbers or is there a transformer between the VFD and the motor? or inside the VFD as part of the output section?

 

[LionelHutz]

What is the drive topology?

54 pulse sounds like a cascaded H-bridge, but the part about DC link voltage doesn't seem to correspond with one. Cascaded H-bridge topologies don't have many issues with long cable lengths.

 

[iop95]

Very interesting how get 0.85 input PF with uncontrolled rectifier... maybe 18-24 pulse rectifier, but that need dedicated transformers.
1800V DC bus indicate cascaded H bridge, a multilevel inverter or some special topology.
For a VSI cable lenght it's problem even here switching frequency is not high and maybe dv/dt is limited by switching devices/firing control.
I thing is better to request manufacturer at least a simulation for your setup and frequency range / acceleration ramp with motor load from 0 to max allowed.

 

[Imumahdi]

Thanks for reply all. I'm sharing the full datasheet instead of selected data in order no to create any confusion by my lack of knowledge on the subject.

https://files.engineering.com/getfi...6-a462-4fbc-8041-a3fdac68d337&file=VFD_DS.pdf

 

[FacEngrPE]

Line 57 requires a OTHER STUDIES/ANALYSIS RESONANCE STUDY TO BE CONDUCTED
If it was conducted it should be a deliverable. It may have the details regarding how much ringing should be expected.

There are two items to look at
[ul]
[li]resonance resulting from dv/dt from the leading edge of the square wave,[/li]
[li]resonance resulting from the drive excitation frequency.[/li]
[/ul]
A 54 pulse drive output transformer could damp the resonance, depending on transformer design. This is about the limit of my expertise, going further than the vendors resonance report requires experts in fundamentals of EE.

 

[LionelHutz]

It's a cascaded H bridge. The cells are connected in a Y configuration and there are 9 x 600V output sections in series per leg of Y. Using 600V output sections means any switching ringing going to the motor will be limited to around 1200-1400V peak. The motor should be capable of handling that just fine.

 

[Imumahdi]

Thank you everyone for your insightful responses.
I'll be sure to return and verify the data once the VFD arrives from China and will be sure to make a separate post about my findings if any arises.

Regards.

 

[Gr8blu]

A couple of other things on the OEM datasheet to point out. They list a specific harmonic profile, with the "high" single harmonics occurring at 7, 11, and 13. However, they also list that have a 54-pulse system - which means their "profile" should cover all harmonics out to at least 151. The "standard" approach to having a 25th harmonic as the upper limit was predicated on only having 6-pulse supplies. To truly meet the intent of IEE 519 requires that the profile cover at least 3x and preferably 4x the actual pulse count.

Also note that at lower loads, the THD (current) content goes up significantly, as expected, to around 7% at one-quarter load. This may show up as a different "max" harmonic profile compared to the full-load case - which might need additional mitigation.

Lastly - the drive has three "blocked" bands where known resonance may occur based on the system data provided for the drive vendor to quote (presumably the cable type, length of run, and machine impedance connected at the far end). There is also provision within the software to add some (unsure how many) other "blocked" operating points, just in case things aren't quite exactly what the drawings show.

Converting energy to motion for more than half a century

 

[jraef]

To further LionelHutz’s post, typical cable length capability of a Cascaded H Bridge design drive is 1000m from drive to motor without the need for further mitigation.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden