VSD Reliability - We have 2 duty 1

[BRIS]

VSD Reliability - We have 2 duty 1 stand-by pumps supplying direct into water supply distribution each pump rated at 1800 kW. We have proposed 3.3 kV - VSD drives. The system must have continuous operation and near 100% reliability. The client is concerned over the reliability of the VSD drives and is requesting that we provide standby units.

What is the typical reliability of VSDS and what would the typical down time be in the event of failure ?

 

[EmmanuelTop]

The figures for variable frequency drives (VFD) I have come across in various RAM studies are typically 30,000 to 80,000 hours for MTBF, and 1.1 to 5.4 hours for MTTR. Hydraulic devices (e.g. Voith) have MBTF between 105,000 and 154,000 hours. Vendor data tend to be too optimistic (Voith clames 48 years MBTF for planetary gear), whilst the information listed in reliability databooks can very often be overly conservative.

Is the use of VSD justified?

Dejan IVANOVIC
Process Engineer, MSChE

 

[ashtree]

Reliability can be seriously affectted by how you treat them, in terms of maintenance and the environment they operate in.

A good quality drive well specified and installed should do 7 to 10 years of service but i have seen many do a lot more than that.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[BRIS]

Thanks for the response. In this situation I would have adopted 4 duty and one stand-by and gone for fixed speed, however I have taken over a partially completed design with VSD and cannot go back. The client requirement for pumps and equipment is a minimum 25 year design life. My reading is that the VFD will be lucky to give 10 years (24/7 operation). I have to convince him that he will get the same reliability with VFD as he would get with DOL which he has adopted in all his other stations. (Some pretty hefty switchgear for DOL of 2 MW motors).

 

[MortenA]

Does the VFD need 100% reliability? I dont know so much about electrical systems, but if the VFD could be bypassed and supply the pumps with 50 Hz power for a while until the VFD was repaired then maybe that could be acceptable? This could be the most economical case if the VFD's are installed for powersaving. It might require a BU control vale or by-pass line.

Best regards, Morten

 

[georgeverghese]

Electrical engineers in the electric motor forum may probably assist with more info on the more frequent causes of problems with VFD.

 

[EdStainless]

Critical cooling pumps that I am familiar with are never on VSDs.
Maybe if you could monitor internal temp rise a function of load and irregularities in switching speeds you could have an effective preventive maintenance program, but I have never seen one.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[ScottyUK]

EdStainless,

Slight aside - the reactor coolant circulators in the British AGR nuclear power plants are operated on variable speed drives. 'Over-engineered' is an understatement, but VSD nonetheless.