# On India's power outage

wrote:

Your opinion of "what they should be" and what the individual wants is totally irrelevant. If one wants the temperature either higher or lower than someone else that's nobody's business but the one who pays the bill. We don't have any Temperature Police... yet.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 08 Aug 2012 16:50:19 -0400, " snipped-for-privacy@att.bizzzzzzzzzzzz"

They are trying to reduce PEAK load. Lets say there are 1,000,000 power customers. And lets say they want to reduce the peak load 10%. If all those users AC units were running then turning off 10% of the running AC units would get MORE then a 10% reduction in peak load because the AC while running is probably 50% of the load on each house. So they only need to turn off perhaps 5% of the AC units at any one time. If the meters are really "smart" they would be able to tell if the AC is on or not if by no other way then simply the current draw at that point in time compared to the average for a typical day in the season. So they should be able to target running ac units. And since it's "smart" they should be able to shut down *just the right number* of them to reduce the peak load to whatever their target is. Going back to it probably being 5% or less needing to be turned off at any one time, that means of the 1 million houses, if they cycle thru each house so no house gets hit twice until all the houses with running ac have had a turn, it means they need to turn off 5% of a million, or 50,000 AC's. If they want to keep the off time to 10 minutes (1/6 of an hour), it means that for every hour they want to reduce the peak they need to deal with 300,000 AC's. Given that they have a million of them, and on a really hot day most of them will be running at any point in time, that means they have a 3+ hour time period where they can be turning off AC units without ever hitting the same house twice. Rounding up to make this example worse, lets say I'm off by a factor of two, it still means that at most your AC will be turned off for 10 minutes twice in a 3 hour period and the off time would be 1.5 hours apart. I doubt many people would ever notice those two-10 minute off periods 1.5 hours apart. And for those rare people who do notice it, by the time they notice it more then likely the 10 minutes will be expired and it will be back on.
I just went thru this on the fly so I could have missed something but assuming it's a reasonably good armchair estimate, the use of smart meters to control peak use seems like a very good way to save money (by not building excess capacity to deal with a couple hours of peak use) with essentially zero impact on anyone's comfort and convenience.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

While a quick read of this seems to make sense, the one statement I might have an issue with is this:
"And since it's "smart" they should be able to shut down *just the right number* of them to reduce the peak load to whatever their target is."
If you are talking about the meters, I don't think that they are smart enough to know what's going on across the grid. Some other system would have to tell the meters which AC units to shut down and for how long.
I think. ;-)
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

(You really need to breathe once in a while)
The problem with your theory is that it doesn't work. The peak power is exactly the same (as is the average) unless they're leaving the power off long enough for it to get warmer in the house. Then the difference is only in the delta-T.

Doesn't work. A soon as you turn the block of ACs back on, *every* one will cycle on, increasing your peak. You've actually made it worse because they're all synchronized, now, rather than random.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
snipped-for-privacy@att.bizzzzzzzzzzzz wrote:

Unless the shutdown unit also has a random delay built-in when restored.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Ok, but you've still accomplished nothing unless you've raised the average delta-T. IOW, a ten-minute power off does absolutely nothing.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Mon, 13 Aug 2012 18:14:30 -0400, " snipped-for-privacy@att.bizzzzzzzzzzzz"

Off course it does, otherwise why wouldn't you simply put a timer on your AC to turn it off 10 minutes out of every run hour just to save money. You're not seeing the small picture at the house end nor the big picture at the system end.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Ok, you've just proved my point. If load shedding worked on air conditioners, everyone would just power them off for 10 minutes per hour and save on their power bills. The fact is that it doesn't work. Worse, it synchronizes the loads.
BTW, did it really take you two weeks to read my post?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sat, 29 Sep 2012 17:20:57 -0400, " snipped-for-privacy@att.bizzzzzzzzzzzz"

Yes, there was a 2 week gap.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sun, 12 Aug 2012 22:44:47 -0400, " snipped-for-privacy@att.bizzzzzzzzzzzz"

No kidding. I wrote that a few days ago and just came back to the thread. Wow, almost like a drug induced dream.

I think you miss the whole point of the system. It *IS* going to reduce the peak because that's the driver determining whether to turn off *any* houses. If the algorithm says "we're peaking according to the powermeter" it starts to turn off AC units and continues to do so until the algorithm says "you've turned off enough and the power output meter is now back down where we want it". It will continue to iterate that cycle till it starts to drop off the peak and then it turns AC's back on. Of course it's not going to pick YOUR system and keep it off for 3 hours, it's going to spread the *off* time over the millions of houses, 10 minutes at a time. And yes, the houses turned off will each get a tiny bit hotter, ... if that were not the case their AC would not have been needed.

Doesn't matter, the algorithm will have taken that into account and each time one it turned back on, another house somewhere else gets turned off unless the "master power meter" says it's started dropping off the peak. Not only will another house get turned off as a replacement for the one that was just turned back on but if the system is still "peaking" an additional house will get turned off. That's the beauty of this kind of system, it's able to spread the pain out over such a large number of houses, for so short a time at each individual house, that no one house even notices the 0.3 degree rise in temperature for that one hour slot.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

No, I didn't miss anything. The system doesn't work unless they're going to shed loads long enough for the *average* temperatures to get significantly higher. Ten minutes isn't going to do anything except synchronize the load (the worst possible outcome).

Then you've done exactly nothing. The house you're just turning back on *will* come on. It's duty will not have changed. Its total power will not have changed, except for the 10-minute lag.

Except it doesn't work.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sat, 29 Sep 2012 17:26:18 -0400, " snipped-for-privacy@att.bizzzzzzzzzzzz"

All I can say is we will have to disagree. The power company seems to think it works and they are the ones who know how much power is being produced and consumed as well as when and where.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

No, the power company has convinced you that there is a free lunch. You're willing to go along with their scheme to control you. Reality won't be so "free", as that is what you will be giving up.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
snipped-for-privacy@optonline.net wrote:

Maybe all the cumulative inrush current happening randomly is part of the problem. If they shut off compressors in whole neighborhoods, they stand to route it elsewhere. Then just rotate the "rolling AC blackout" around the city at different times.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 08/08/2012 11:39 AM, snipped-for-privacy@optonline.net wrote:

If you have to run your AC 100% of the time to be comfortable, you have an incorrectly sized AC. Doesn't the power company take this into consideration when they decide if you are eligible for the discount?
Jon
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Or, the AC you have is running far less than capacity.
Have it serviced.
If you have to run your AC 100% of the time to be comfortable, you have an incorrectly sized AC. Doesn't the power company take this into consideration when they decide if you are eligible for the discount?
Jon
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 15 Aug 2012 17:58:28 -0400, "Stormin Mormon"

Actually, for maximum efficiency and maximum dehumidifying, an air conditioner that has to run constantly on the hottest anticipated day to maintain the desired temperature is sized "just about right". Mine is too big - just like my furnace - and runs about 8 hours a day when outside temps hover in the low 90s F - which is as hot as it usually gets around here. Running only roughly 30% of the time, humidity is not as well controlled as it should be - an additional dehumidifier extracts about a gallon or more a day with outside humidity in the 73% range - which is on the low side of normal for a sothwestern ontario hot summer day.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 15 Aug 2012 19:28:25 -0400, snipped-for-privacy@snyder.on.ca wrote:

Some decades ago, my father was the maintenance supervisor for a printing plant. The plant was air conditioned to provide for better handling of paper and printing. To control the humidity on damp cooler days, they would run the heat at the same time as the AC.
I don't know the capacity of the AC, but I remember walking into the ducts where the filter system was. .
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 8/15/2012 9:12 PM, Ed Pawlowski wrote:

The air conditioning for data centers where something like a super computer is operated on a raised floor has air handlers which blow air down under the raised floor so cooling air can come up through open sections under equipment to cool it. The air handlers often have pans of water with electric heaters to add humidity to the air and will control humidity using a method called reheat which will reheat the cooled dehumidified air the bring it back up to room temperature. ^_^
TDD
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 15 Aug 2012 22:58:34 -0500, The Daring Dufas

Are any "supercomputers" air-cooled? The ones I've worked with were water- cooled, with heat exchangers under the covers for those parts that use air. Only peripherals, smaller computers, and test equipment used air from the floor as the cooling source.