UPS for a plant
UPS for a plant
(OP)
One of my clients is asking about an uninterruptible power supply. They have about twenty five CNC machines. They're in Silicon Valley (land of vaunted power dependability) where the power blinks with alarming frequency that seems to be getting more frequent all the time.
Any blink that's long enough to upset these machines is a disaster. Often they have to run 30+ hours on one part. When the power blinks the machines freeze without hurting anything but on their longest running parts they cannot pick-up where they left off because these are grinding operations not machining operations. This means they have to start over on the part and as Murphy will have it the power blinks tend to always be near the end of long runs.
They're asking about short term interruption protection. They believe a couple of seconds of hold-up would weed out 90% of their power disturbances. Last week's and the week before the disturbances were classic brown-outs surrounding a poor utility breaker trying to reconnect into a short on some other distribution line. Three one second long brown-outs separated by about 2 seconds each.
This essentially crashed the whole facility causing several hours of technical work getting the place up again and of course several 'start-overs' and that's if the technical people don't have to return from home.
The place is fed 480V and maximum load is around 150kW.
Typical loading is about 90kW.
About 60% of the machines are 240V 3ph.
What do you folks suggest we should look into for carry-thru.
Is this a whole facility type application?
Just UPS single panels?
Single machines? (could be about 20)
Battery?
Rotary?
Fuel-less rotating generator?
Suggested suppliers for this?
Any blink that's long enough to upset these machines is a disaster. Often they have to run 30+ hours on one part. When the power blinks the machines freeze without hurting anything but on their longest running parts they cannot pick-up where they left off because these are grinding operations not machining operations. This means they have to start over on the part and as Murphy will have it the power blinks tend to always be near the end of long runs.
They're asking about short term interruption protection. They believe a couple of seconds of hold-up would weed out 90% of their power disturbances. Last week's and the week before the disturbances were classic brown-outs surrounding a poor utility breaker trying to reconnect into a short on some other distribution line. Three one second long brown-outs separated by about 2 seconds each.
This essentially crashed the whole facility causing several hours of technical work getting the place up again and of course several 'start-overs' and that's if the technical people don't have to return from home.
The place is fed 480V and maximum load is around 150kW.
Typical loading is about 90kW.
About 60% of the machines are 240V 3ph.
What do you folks suggest we should look into for carry-thru.
Is this a whole facility type application?
Just UPS single panels?
Single machines? (could be about 20)
Battery?
Rotary?
Fuel-less rotating generator?
Suggested suppliers for this?
Keith Cress
kcress - http://www.flaminsystems.com






RE: UPS for a plant
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: UPS for a plant
RE: UPS for a plant
There are obviously a lot of motors in the load, but are they directly connected or through servo or VSD controls? It makes a big difference to how the load looks to the UPS.
If there are a lot of directly-connected motors then a electromechanical rotary UPS from (e.g.) Piller or Eurodiesel would be a good fit. If the loads are mainly rectfier-type loads at the front end of drives and controllers then a conventional static-electronic type is likely a more economical solution. I'm not overawed by the CAT rotary-electronic types which in my mind incoprorate the weaknesses of both rotaries and static electronic types.
Keep away from VRLA cell types, they are false economy because they give a capital saving up front, test OK so as far as you can test them, then fail the first or second time they are called on in anger. As far as possible, avoid very high voltage batteries with large numbers of cells in series - the more cells you have in series the greater the chance of one of them being a bad one which tales down the entire string. Obviously there is a trade-off to be had there - you can't realistically supply 150kVA of load from a 24V battery, but equally you don't need to go up to a 600V battery either. Flooded lead-acid recombination cells would be my first choice if very low temperature wasn't a concern.
RE: UPS for a plant
davidbeach; Thanks.
Scotty; motor/rectifier... About 20 of the machines are rectifier front ends. That's all the CNC machines naturally. But about 5 or 6 of the machines are straight up surface grinders that are motors that just DOL and are pretty big. ~50hp. Sounds like maybe a two solution case where we use a different UPSs for the two load species.
Thanks for the VRLA heads-up, that pretty much describes my personal VRLA experiences.
Care to educate me on the "incoprorate the weaknesses of both rotaries and static electronic types"?
Keith Cress
kcress - http://www.flaminsystems.com
RE: UPS for a plant
The smallest rotaries I'm familiar with are in the 150kVA - 200kVA region so big enough for your entire facility. They use a fairly unusual rotating machine with a dual rotor and a stator. The inner rotor is a reasonably conventional synchronous generator rotor, and the outer rotor is a high speed, high mass energy store running at twice synchronous speed. A clutch decouples the inner rotor from a diesel engine which picks up the load within a few seconds by driving the inner rotor through the clutch and re-accelerates the storage rotor to full speed.
The CAT UPS weaknesses? It uses a conventional rectifier / inverter but instead of a battery it uses a motor to drive a very high speed rotor as an energy store (in a vacuum chamber IIRC) and the motor acts as a DC generator when needed. You get the same weaknesses of the much cheaper dual-conversion static UPS such as poor fault-clearing capability on inverter power and also get the mechanical maintenance / lifetime problems with vacuum pumps etc, plus relatively short duration energy storage compared to even a modest-sized battery. The main rotor itself runs in magnetic bearings IIRC so should be decent life from them, but very few things with moving parts beat a good-quality flooded cell battery for reliability or operational life.
With both types above, if the generator (or engine) doesn't start on second or third try then you drop the load. A storage battery at least gives you time to race to site like a maniac, curse like a pirate with Tourettes at the dead genset, and maybe even have time to kick it a couple of times in frustration before it drops the load while you watch, helpless.
RE: UPS for a plant
On the VFD machines a battery feeding the DC bus may give enough time for an orderly shutdown.
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: UPS for a plant
They had a big room full of giant glass storage cells to take care of short glitches,
and a bigger room full of really big engines to take care of longer glitches.
They started one of the engines for us, just for fun.
It was at speed almost instantly, and it was LOUD.
Clearly, both the battery room and the engine room were fully staffed all the time.
Mike Halloran
Pembroke Pines, FL, USA
RE: UPS for a plant
This problem calls for a rotary UPS solution. 90kW is pretty big for a double-conversion type of static UPS (with battery). A few seconds of power is a perfect application for a rotary UPS.
GG
ps I have never had issues with VRLA batteries in float duty. Assuming they are sized correctly and made by a reputable manufacturer.
"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)
RE: UPS for a plant
RE: UPS for a plant
GG; Got a rotary brand I should look at?
Keith Cress
kcress - http://www.flaminsystems.com
RE: UPS for a plant
Scotty: I have yet to have a bad experience with GNB (Gould National Battery) type Absolyte VRLA batteries. (Touch wood)
"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)
RE: UPS for a plant
I would give theses guys some consideration. Seems like they have been around the block a time or two;
http://www.pscpower.com/what-we-do/products/uninte...
"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)
RE: UPS for a plant
and
http://www.piller.com/en-GB/262/uniblock-ubtd-rota...
Definitely check on support in your area, that's as important as the equipment itself.
RE: UPS for a plant
After looking at total cost of ownership for each route, we based our design on a Piller UBTM+2150 (2400 kVA rotary system). Unfortunately the customer hasn't decided to move forward with installing the system, so I can't speak to the quality/performance of Piller.
RE: UPS for a plant