Marginal power

[Dan5]

I brought this up in the guys comment section. If I have solar cells and nixed 4 kw-hr that would normally be operating from 6-10, ( computer, and substituting lights) and charged my EV instead, where's the increase in marginal load? If you house is still using the same amount, there is no increase.

If environmentally concerned people who buy EVs and have solar cells really want to be a-holes about it, they can set up a battery back up and charge their EVs at night using the stored batteries. That way we aren't increasing the grid or sending our excess clean energy to the grid either.

I can see him complaining about that too, saying that we are making it dirtier by not sending our clean energy to the grid for other people to waste away

 

[jerry33]

If environmentally concerned people who buy EVs and have solar cells really want to be a-holes about it, they can set up a battery back up and charge their EVs at night using the stored batteries. That way we aren't increasing the grid or sending our excess clean energy to the grid either.

1. Storage batteries certainly solve the marginal power argument--there is any marginal power used (other than when you system goes out and you use the grid as a backup power source).

2. Because the utility companies are working on ways to charge people for putting power into the grid under the pretext that it causes extra wear on the grid, I don't see how it's being rude to avoid those charges. The utility companies are practically daring people to do that.

 

[Norbert]

Last I knew LA imported a huge amount of coal power from out of state, something around 40% of usage. So if you are part of the LA grid you may very well be charging from coal. Saying there are no coal plants in CA doesn't provide the whole picture. That's my main point of this whole exercise, that we have solid factual data to use.

According to the study I've been linking to, the coal imports by LADWP, which they called "firm imports", where used for baseload, not marginal power:
ftp://ftp.dvrpc.org/dvrpc_misc/SMegillLegendre_DOEEVActionPlanLibrary_Asof03-01-2012/JournalofPowerResources_MarginalElectricityforPHEVsandFuelCellEVs_2009.pdf

First, renewable and baseload generators, including most firm imports, follow fixed hourly generation profiles that are independent of electricity demand. They do not provide marginal generation for vehicle and fuel-related electricity demand in any scenario in this analysis.

So if you follow the marginal power argument, no problem for EVs in LA. (And meanwhile that surely was reduced altogether, like coal everywhere.)

 

[JRP3]

1. Storage batteries certainly solve the marginal power argument--there is any marginal power used (other than when you system goes out and you use the grid as a backup power source).

At this point storage batteries are an insignificant part of the grid, and additional expense. Hopefully that will change.

 

[JRP3]

As long as the old load being retired is retired at the same rate new loads are added, the baseload is in equilibrium (using the assumption every thing is follows an "established" schedule so the load profiles are the same every night). Even if the rate new loads are added is faster, there's going to be a certain amount of time that a new load being added to the list will fall under the current capacity as old loads are continuously being retired (the line keeps moving, even if it's growing). Then once you factor in new capacity that is being added (and old capacity being retired), which in general is in response to new loads being added faster than old loads retired, then it appears likely there will be enough baseload to cover new EVSEs being installed. Without data on this, you can't really say an EVSE won't be covered by baseload plants without making a huge assumption.

The point is there is no "old" load being retired that compensates for the new load from an EV. Power demand is growing even without EV's, so EV's only add to that demand. The gird dynamics exist independently of EV's. My point was not only that installation order mattered but that load profile did. My "old" water heater is still a marginal load on the grid each night. An EV is a new load on the grid, it's not replacing any other load, and it's not a constant load. If you had an EV that was plugged into the grid and charging 24 hours a day you could say it's part of base load and assign average grid emissions to it. If you have two EV's charging, one at night and one during the day, you could say that combined they are part of base load, and at some point there will be a portion of over lap in charging such that a percentage of EV charging can be considered base load. But if we expect the majority of charging to be done over night then the load it places on the grid will be met by either NG or coal, which means the actual increased emissions from plugging in that EV will not be grid average but directly tied to those two sources.

 

[JRP3]

I brought this up in the guys comment section. If I have solar cells and nixed 4 kw-hr that would normally be operating from 6-10, ( computer, and substituting lights) and charged my EV instead, where's the increase in marginal load? If you house is still using the same amount, there is no increase.

Their argument would be what if you installed the solar and bought a Prius instead of an EV? In their view, using the marginal load argument, the Prius puts out lower emissions, depending on location, and solar power can be installed independent of EV's.

 

[Norbert]

Their argument would be what if you installed the solar and bought a Prius instead of an EV? In their view, using the marginal load argument, the Prius puts out lower emissions, depending on location, and solar power can be installed independent of EV's.

Even if you buy into that argument, you would then have to examine at least two more factors:

a) In how many locations is coal actually used for marginal power, and are those locations the locations where most of the EVs are being added?

The CA study shows that even in a specific utility in LA, where at the time of the study there was large amount of imported coal, it was used according to a fixed schedule independent of any additional load. So the mere fact that some states have a lot of coal, by itself, doesn't mean that it is used for marginal power.

b) How sensitive is the "marginal power" source to the changes of the grid which are currently taking place? (Coal going down actually, NG price changes, added sustainable power (wind now 4.3% in the US), changed requirements and state-specific voter resolutions).

 

[Dave EV]

I was wondering whether some coal plants might get improved to almost NG level (in terms of CO2 and/or other emissions), but it seems that is in general too costly with NG cost coming down, for example.

From a CO2 perspective, it would take CCS to get coal down to the level of emissions of even old NG plants. Way too expensive to happen. Best case is that the upgrade the scrubbers on the coal plants to get all the other pollutants down to acceptable levels.

A fleet of EV's charging at night is a large load on the grid that will increase emissions. Yes it may even out the day/night ratio somewhat, but emissions will increase. Unlike MPT's water heater scenario there probably won't be a bunch of EV's turning on and off over night to compensate and average out, they'll all come on around the same time and all shut off around the same time.

It'd actually be pretty trivial to get EVs to fill in the peak quite nicely. The grid operators know when minimum demand occurs. Simply incentivize EV owners to center their charging around that minimum demand with TOU rates.

Many TOU rate schedules currently run between 12am-6am. Program the car w/these times and have the car aim to put the charging in the middle of that time period. Need 2 hours to charge? Start charging at 2am. If the charging time is significantly less than the charge window - add a bit of randomization to the start time.

 

[stopcrazypp]

The point is there is no "old" load being retired that compensates for the new load from an EV. Power demand is growing even without EV's, so EV's only add to that demand. The gird dynamics exist independently of EV's.

You are still looking at EVs at as a special case (making the assumption it is marginal no matter how the grid/load changes), and not as a black box load. Again, if the order of installation is what matters, then the "old" load will be appliances that are replaced and retired. "new" loads will be newly installed appliances (including your EVSE). Unless you are saying there is never any appliance that is replaced or retired (which I doubt, since I just replaced a couple of old ones in my house recently and retired an old TV since I don't watch TV in that house), then there are always "old" loads being replaced/retired continuously. And since new appliances tend to be more efficient than old ones, that only "adds" more space in the baseload (independent of order), even though capacity didn't change at all. And the point Nobert added about new capacity has to factor into that too.

My point was not only that installation order mattered but that load profile did. My "old" water heater is still a marginal load on the grid each night. An EV is a new load on the grid, it's not replacing any other load, and it's not a constant load.

That's being inconsistent. How do you order the loads in that case (how do you mix installation order and load profile)? As I pointed out, the order of installation and the order of operation can be completely different at any point in time (the "newest" appliance may be the first one to turn on; the "oldest" appliance may be the last one to turn on).

And please define "constant load". There's no load in my house that's constant (24/7) besides from phantom loads. Everything turns on and off at a certain point in time (air con, heater, fridge, water heater, lighting, entertainment, washers/dryers, etc.). If you really care only about phantom loads, it's easy to add an LED to your EVSE (it may already have one and it may already have a vampire draw in the first place) and then count it as "on" 24/7.

 

[JRP3]

Even if you buy into that argument, you would then have to examine at least two more factors:

a) In how many locations is coal actually used for marginal power, and are those locations the locations where most of the EVs are being added?

The CA study shows that even in a specific utility in LA, where at the time of the study there was large amount of imported coal, it was used according to a fixed schedule independent of any additional load. So the mere fact that some states have a lot of coal, by itself, doesn't mean that it is used for marginal power.

b) How sensitive is the "marginal power" source to the changes of the grid which are currently taking place? (Coal going down actually, NG price changes, added sustainable power (wind now 4.3% in the US), changed requirements and state-specific voter resolutions).

Both good questions, which I don't know how to answer. I'd love to have solid data on how much coal power is actually used in which areas at what times.

 

[Norbert]

Both good questions, which I don't know how to answer. I'd love to have solid data on how much coal power is actually used in which areas at what times.

Like the study for California that I have been referring to for a while?

 

[JRP3]

You are still looking at EVs at as a special case (making the assumption it is marginal no matter how the grid/load changes), and not as a black box load.

In a sense I am, because they a special case. I can't think of any other significant load that we expect and want to come on line in large volume in homes across the country. That is our goal right? If they gain popularity and sales volume as we hope, and if the majority of them charge at night as we expect, that will be a huge added load at night. What is going to meet that load?

Again, if the order of installation is what matters, then the "old" load will be appliances that are replaced and retired. "new" loads will be newly installed appliances (including your EVSE). Unless you are saying there is never any appliance that is replaced or retired (which I doubt, since I just replaced a couple of old ones in my house recently and retired an old TV since I don't watch TV in that house), then there are always "old" loads being replaced/retired continuously. And since new appliances tend to be more efficient than old ones, that only "adds" more space in the baseload (independent of order), even though capacity didn't change at all. And the point Nobert added about new capacity has to factor into that too.

Your old TV that you never used is just an old appliance, not an old load, but again old TV's, water heaters, etc, that get removed and replaced aren't the same thing as I described above.

That's being inconsistent. How do you order the loads in that case (how do you mix installation order and load profile)? As I pointed out, the order of installation and the order of operation can be completely different at any point in time (the "newest" appliance may be the first one to turn on; the "oldest" appliance may be the last one to turn on).

That was sort of my point with my "old" water heater putting a "new" marginal load on the grid each night. MTP made a valid point that there would probably be other water heaters in other areas turning off and on at different times to average out the load so that water heaters as a whole could be seen as base load and grid mix. Unless EV charging is staggered and balanced, which I admit could happen in the future, they are all going to come on around the same time and all shut off around the same time, putting a large load on the grid each night which will be met with marginal power.

And please define "constant load". There's no load in my house that's constant (24/7) besides from phantom loads. Everything turns on and off at a certain point in time (air con, heater, fridge, water heater, lighting, entertainment, washers/dryers, etc.). If you really care only about phantom loads, it's easy to add an LED to your EVSE (it may already have one and it may already have a vampire draw in the first place) and then count it as "on" 24/7.

I guess you can look at a certain amount of your household load as constant, which would be the average demand over time of various things turning on and off, and then some of that would be marginal load when it spikes above it's average demand. An EV charging at night is obviously going to spike the load above the average night time demand, hence marginal load. Again, at some point with enough EV's charging around the clock some percentage of that load should be considered base load. We'd have to see actual use patterns to determine how much charging overlap there would be that could be considered constant base load. So I would say that a certain percentage of EV charging load can be attributed to base load at some point. I have no idea how to calculate that but it is a valid point that can be made which cuts in to the marginal load argument.

 

[JRP3]

Like the study for California that I have been referring to for a while?

I'm talking about across the country. It does exist outside of CA you know.

 

[dsm363]

Unless you're going to go to off-grid solar with battery backup, you're kind of at the mercy of the grid. You can buy 'green' power like I do (100% wind energy in Texas) even though that power doesn't directly go to your car. I'd still say in all but the worst areas, you're better off with an EV if CO2 and pollution are concerns for you. Yes, I'm sure there are instances where a highly efficient car like the Prius will do better but on a national basis, EVs are cleaner from what I've read at least. And these cases where a Prius beats all EVs in a geographic area will get smaller and smaller as time goes on since coal market share is decreasing each year in recent years at least. That's the perspective to keep in mind in my opinion.

 

[Norbert]

I'm talking about across the country. It does exist outside of CA you know.

If you say so. This is for New York State 2008:

http://www.nyiso.com/public/webdocs/newsroom/white_papers/fuel_diversity_11202008.pdf

If I'm reading this correctly, NYS uses almost no coal for marginal power.

 

[Robert.Boston]

FWIW, economists consider all load in an hour to be marginal. There's nothing about order or historic operation. Prices on the grid paid by all load are set by the most costly generator. We don't assign the cheapest generation to "old load" and the more expensive generation to "incremental load." Everyone pays the clearing price. The same logically holds true for carbon. If we priced carbon, all loads in an hour, at a place, would pay the same marginal carbon tax.

 

[stopcrazypp]

In a sense I am, because they a special case...What is going to meet that load?

The growth of EVSEs will be slow at best. The official goal is 1 million plug-ins by 2015 (announced since 2008). That's about 3300-6600MW (3.3kW-6.6kW * 1 million) of load staggered over 7 years. There was 1695MW of wind power added in first quarter 2012 alone (6815MW in 2011). The growth of EVSEs doesn't appear to be any special case to the grid in general (heck, even wind power is greatly outpacing the growth of EVSEs in the US), so I don't see any need to treat it as such (and that still ignores the point about how to determine what counts as marginal and what doesn't). By the time the growth of EVSEs pick up steam, the power companies would have been prepared for them.
http://www.awea.org/learnabout/industry_stats/index.cfm

Your old TV that you never used is just an old appliance, not an old load, but again old TV's, water heaters, etc, that get removed and replaced aren't the same thing as I described above.

The old TV is certainly an old load, since the owner before me used it in the house, now it's completely unused (no more load from the TV)! A old water heater etc. is an old load too, being replaced by a new load from a new water heater (which likely is more efficient too). My point is there has to be an order to things (and the EVSE is a part of that order) or you will be stuck always assigning the EVSE as a marginal load with no consistent reason.

That was sort of my point with my "old" water heater putting a "new" marginal load on the grid each night. MTP made a valid point that there would probably be other water heaters in other areas turning off and on at different times to average out the load so that water heaters as a whole could be seen as base load and grid mix. Unless EV charging is staggered and balanced, which I admit could happen in the future, they are all going to come on around the same time and all shut off around the same time, putting a large load on the grid each night which will be met with marginal power.

But the thing is if, you turn on the EVSE early in the night, the marginal power will be meeting the needs of other appliances that turn on later than your EVSE (back to operation order being more important). Your EVSE would already take up the baseload slot, so it's 100% baseload at that point. Again, you can't just arbitrarily say the EVSE is always going to be the load to be marginal, just because it might be large (there's other large loads on the grid too, like heaters during cold nights and air con during hot nights; there's even more during the day).

 

[Norbert]

FWIW, economists consider all load in an hour to be marginal. There's nothing about order or historic operation. Prices on the grid paid by all load are set by the most costly generator. We don't assign the cheapest generation to "old load" and the more expensive generation to "incremental load." Everyone pays the clearing price. The same logically holds true for carbon. If we priced carbon, all loads in an hour, at a place, would pay the same marginal carbon tax.

That sheds new light on these questions. However that makes it even stranger. It seems to mean that a utility can charge a lot of money to its customers, if its current marginal power cost is high, even if most suppliers charge much less. Or are all suppliers raising their price at the instance a more expensive marginal power comes into play?

It certainly doesn't mean that CO2 emissions are as high as the "marginal carbon tax" would suggest in this scheme.

 

[stopcrazypp]

FWIW, economists consider all load in an hour to be marginal. There's nothing about order or historic operation. Prices on the grid paid by all load are set by the most costly generator. We don't assign the cheapest generation to "old load" and the more expensive generation to "incremental load." Everyone pays the clearing price. The same logically holds true for carbon. If we priced carbon, all loads in an hour, at a place, would pay the same marginal carbon tax.

For economists, it's easy because all you have to look at is the cost to generate the last kWh (marginal cost) and then charge everyone the same price since that's going to be the highest price. They don't care at all what load counts as marginal, they just look at what it costs to generate that last kWh. In general the loads are all mixed (like my original argument) and everyone pays the same and gets treated the same (which supports the argument of just looking at the power plant mix at a certain point in time).

They aren't assigning the loads to different power plants, like we are doing here with the "marginal load" arguement. In this case, there has to be an established order or we will never be able to assign ANY loads to the baseload powerplants (which doesn't make sense).

 

[Jason S]

Wouldn't it be fair to compare the average carbon output per mile of EV versus the average carbon output per mile of gasoline when all portions of each fuel are taken in to account? Transport + refining + burning...

Then the mythical marginal source wouldn't matter at all. Might as well blame Facebook for high usage in the evening hours because of all those computers running...