Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

VFD fed from UPS-Harmonics issues?

Status
Not open for further replies.

ryaneewgi

Electrical
Apr 7, 2004
1
I have a situation that requires 125HP pump motors (currently fed from VFD's) to be provided with uninterruptible power, they must run during an outage. One proposed solution is to use large UPS's (~250kVA) to feed each VFD. There are some concerns brought up by myself and others regarding possible harmonic problems and/or unknowns associated with feeding a VFD from a UPS (both IGBT) for this large of a motor load. The distance between the UPS and VFD can be adjusted to any length pretty much, but some concerns that have been mentioned are the UPS and VFD "fighting" each other in terms of voltage regulation, harmonics unknowns, and responding to load changes. Most UPS manufacturers that I've spoken to don't necessarily have a problem providing UPS's, but they seem hesitant to guarantee a solution, even with the option of providing new VFD's with their UPS's. Also, motor starting has been raised however I feel with soft start capability in the VFD's that this may not be as much of an issue. I'm still learning when it comes to harmonics but some colleagues of mine have concerns with regard to how the UPS and VFD will react/respond to each other. If anyone has experience with this type of scenario or could offer any information on the possible harmonics issues it would be appreciated.

Thanks!
 
Replies continue below

Recommended for you

 
If the pump operation is that critical, it may be appropriate to base the procurement verbiage for the UPS system on a mandatory local-site visit for load-profile measurements—basing release from the UPS-gear manufacturing facility based on your preliminary acceptance of demonstrated satisfactory performance when operated under the field-measured load characteristics. Then, base final signoff on acceptable installation-site performance under actual operating conditions.
 
I think a lot of the answer depends on the quality level and sophistication of the UPS system. If it can output a reasonably clean sine wave at all load points, I would think this would work, although I have no direct experience.

If this really is a critical system, I would recommend requiring factory testing of the UPS-VFD-Motor as a unit before shipping to the site. If it's really critical, you should have a spare VFD you can ship to the UPS supplier or test lab.

There is no substitute for an actual test.
 
I have been involved in a similar project for radar tower sites where the VFDs were on HVAC systems critical to the site's function. We used Toshiba for both the UPS's and VFDs since they make both. No problems whatsoever, but to be honest, the coordination of the two systems took place behind the scenes at Toshiba Houston so I necver saw or heard what they did to make it work.

"Venditori de oleum-vipera non vigere excordis populi"


 
That sure seems like a waste from a power conversion standpoint. AC -> DC -> AC -> DC -> PWM AC

UPS systems that size have a DC bus voltage near the DC link voltage in VFD's.

Would it be possible to work with a VFD manufacturer to have a large battery network paralleled directly to the VFD DC link. A dedicated battery charger would have to be used in conjuction with the VFD rectifier for bulk battery charging. Much simpler and more reliable. I don't know if this exists on the market however.

The large UPS would still require an absurd amount of batteries.

How about a steam turbine for backup?
 
An additional comment for using batteries as reserve power for critical applications:

Stay away from sealed (VRLA) type batteries unless your customer will agree to replacing the entire battery network every 2-3 years. These batteries are very unreliable over the long term regardless of maintenance/testing performed and regardless of design life.

They are notorious for open circuiting when placed under sudden load; meaning you just lost critical power when most needed.

Flooded cell type batteries are more reliable and predictable as long as proper water level is maintained. We have used them successfully for periods up to 25 years in non-climate controlled buidlings for both UPS and battery charger (substation control power) service. They do take up more space and require periodic checks and cost more initially.
 
What about a diesel generator (tested regularly)? These generally give good 'sine' waveforms and have pretty good reliability if maintained.

Can you withstand a short outage while an automatic transfer switch/controller starts up a genset? If not, get a UPS with enough capacity to 'ride through' the genset startup.

Then, so long as you have fuel, you have AC. Many of these units allow for filling the fuel tanks while the generator is running.
 
Rotary UPS units engine coupled through shift on the fly clutch and there you go. UPS supported when the power dips and if it dips for longer that you wanted the diesel will kick in and run the rotary unit. Job done.

This way you dont have to worry about harmonics as these are being supported from the motor generator rather than some fancy thyrisor switch pack.

But in saying that UPS units are getting more reliable and able to deal with complicated loads. Computer centres using racks and all sorts depend heavily on SWPSU's and there supported pretty well by these units....

Rugged
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor