4 to 4.5 MVA UPS for Kiln Section in a Cement Plant

So is the UPS to be used only to allow time for the backup generator(s) to come on and the transfer switch to operate? If not, how long of a hold-up time are you thinking of? UPS systems with batteries capable of that much power are likely to cost more than you might think. If this is just for short periods, as in seconds, there is a technology called a Dynamic System Compensator (DySC) made by a company called Soft Switch Technologies (SST) that you might consider. I don’t know if they do MV though. Other than that, I would be considering an M-G (Motor-Generator) set with a flywheel, often called a “rotary UPS”.

Ok, it looks as though SST was bought out by Rockwell, but they don’t go that big.

https://ab.rockwellautomation.com/Power-Supplies/Voltage-Sag-Protector#selection

Rotary UPS systems I have seen used in Data Centers.

http://www.iemfg.com/sites/default/files/resources/IEMPS_Critical_Power_Bro.pdf

Dear sir,

Thanks for the reply.We are having a backup DG set to run in case of power failures.But our problem is not power failures.The problem is voltage fluctuations which cause the plant to trip at least 5 or 6 times in a week causing lot of disturbance to the process and production.We were told that in some cement plants in Nepal,UPS is used for such cases.Hence,I am trying to get the information.In case of power failure we need at least 3o minutes back up.Thanks.

So correct me if I'm wrong, but this power fluctuation is causing the motors that drive the kiln to trip the safety protection, or is it a separate power monitoring system? ...also how long can the kiln stop rotating, if at all, during operation, without severe consequences? ...or is there a required minimum rpm that must be maintained? What are your trip rates?...in other words how often...on average...in a month say?

Dear sir,

Thanks for the Reply.The power fluctuations cause breakers to trip on undervoltage and some times motors trip on overcurrent. All DC drives trip on undervoltage. Ultimately all the motors in the kiln section trips because of sequential or process interlocks ,causing the kiln stoppage and hampering the production.kiln drive also will be tripped.The restarting of the process takes 15 to 20 minutes and to get to full production rate,this may take approximately 45 minutes to 1.125 hours depending on the process conditions.The frequency of power fluctuations in 5 or 6 times in a week.

This sounds like a viable alternative....certainly worthy of investigation...

I doubt they will use electricity for process heat, usually gas, but can be oil or coal based. These long slanted rotating kilns need to be turned and the clinker needs to be ground with large electrically driven crushers.

What is needed is a wiring separation for the sensitive parts. The crushers do not need to be on a UPS. By segregating the components needed UPS support and buying one that size for each kiln you can do it with smaller UPS modules.

Is there any reason why you rate your UPS in terms of Apparent Power rather than in terms of Energy? In general the size and the cost of UPS in mainly determined by the Energy that can be stored, unless it is supposed to work for a very short amount of time.

Hi Nvrs...

As quite a few of the others said a batterry back up of a size to run a power supply that big is -- cost prhibitive could be multi millions of dollars ---you never stated the load required just the input power supply voltage.

From my experience rotary kilns had drive motors ranging from hunreds to thousands of hp. Plus control and aux systems.

The only semi cost effective solution I know of is a flywheel MG type ups that was and maybe still is used for international banking Op Centers in Luxembourg. It works by having a big flywheel and genset powered by mains all the time and for a small sag or drop it will coast through and if the voltage / rpm goes below a preset threshold the diesel engine kicks in with no loss of power at any time and a controlled minimum / maximum before startup this system works well for blackout and brownout conditions.

If the firepump type back up engine plus a smallish ( few hundred kva with a short kwh requirement a battery back up could be viable.

Best of luck

Expat07

https://www.qpsolutions.net/2017/07/types-of-flywheels-a-comparison/

As there are <...DG [Diesel generator] sets...> any installation only need to cover the interval between the mains going off and the <...DG...> coming on-line. As the time that this takes has not been shared, any effect at sizing and selecting solutions would appear premature. Also, the effect of this interval hasn't been shared: if everything goes off for this period of time, then what part of the facility would not need energising through the blackout (rhetorical question - NNTR)?

Brainstorming:

• A huge flywheel might be among the options for solving the problem.
• Negotiation with the <...central electricity board...>, with veiled threats of closure or taking the business elsewhere, to provide a more reliable source of supply as a preferential Customer would be a commercial option that falls outside the scope of the original post.

Dear sir,

Thanks for the reply. I like to give a clarification that I need UPS for plant back up essentially during power fluctuations and most of the time it is power dips for a very short span of 50 to 250 m seconds. Power supply is alright after this time. If I can prevent tripping in this time of 50 to 250 m seconds, my process will not get affected. My interest is to avoid plant tripping only during power fluctuations, which stay for a very short time,i.e., in milliseconds.

As far as process materials are concerned, nothing happens in less than 1 second in general terms. The problem is, therefore, not one of supporting an entire plant of some <...4 to 4.5MVA...> waiting for the <...DG...> to come on line; the problem is one of supporting the equipment that operates the trip for longer than <...50 to 250 m seconds...> so as to let the power supply recover. After all, the <...DG...> will not start, synchronise with the mains, support the plant and stop in a controlled manner in <...50 to 250 m sec...>; that is not what it is there for.

So, anything that trips needs the trip to be inactive for longer than the brown-out period to let the plant ride through it. In such circumstances, using an UPS of that rating to support the entire plant would be akin to using several thousand sledgehammers to crack the proverbial nut:

• Imagine the decision-maker's reaction to being asked to support an investment of that size when other solutions of much lower value must surely be available.
• Imagine the applicant's embarrassment were that individual to be unable to answer questions on the alternatives that have been considered, and why.
• Imagine also the applicant's difficulty in finding an economic justification to support that level of investment.

It's time to look closely at the trips that operate in that period of time, rather than the whole plant. The problem needs re-thinking and a re-focus, rather than myopically copying a solution in use elsewhere, as, although the processed materials are nominally the same, other plants' circumstances are different.

"An Engineer is someone who can do the job for 5 bob [25p in the new money] that anyone can do for £1." _{- Anonymous Poster #0}

We had this problem a while back and somebody came up with a solution I think it was at point of use for each piece of equipment....have to search the archive

If I recall correctly it was lots of smaller (cheaper) UPS's distributed on the key computers and sensitive control loads that would cause a site shutdown if they went down.

A distributed UPS arrangement only really works on certain systems and is only as strong as its weakest link.

Go for it.

Some years back, GE was testing a static reactive power compensator, fast enough to react in the time you need. The reactive power support is what you need, your equipment not likely bothered by a small amount of frequency decay. If you can support the voltage during the upset, it is likely the kiln motor will survive the bump. This static compensator must be connected very closely to the load you need to support. It can be used for power system stabilization as well, and economic if applied in smaller kVA blocks.

https://en.wikipedia.org/wiki/Tesla_Powerwall