We like the sex appeal of renewables, but we don't much like to own them -- remember, we don't have any financial incentive to prefer low-cost generation.

OK, granted, but it sounds like they don't have any financial incentive to avoid low-cost generation either.  From what you write, it sounds like they are largely indifferent to fuel prices.

So my question: Solar PV has a zero fuel price, but it also has one of the highest capital costs of any generation system.  So by the logic you describe above, utility companies should be all over large-scale PV.  But large-scale PV deployments are relatively few and far between.  That's probably just as well IMO: PV makes most sense distributed in small-scale systems.  But it still begs the question of why.

Sean, you keep implying that there are ways to dramatically improve the fuel efficiency of traditional fossil-fuel power plants. Please provide some specific examples and how much improvement in efficiency they would give!

The only things I'm aware of that could at all improve the efficiency of fossil fuel generation are:

(1) Combined cycle gas turbine technology - that works for natural gas, but does it greatly improve the efficiency of coal to electricity?

(2) Fuel cells - but none of the current fossil fuels are in a form that can be readily fed into fuel cells.

Both of those involve capital spending too, but according to you that's not a problem. I really don't understand your argument Sean, and it would be nice to have some more specifics (with numbers) on why you think there's a lot of room for improvement here (and why it would be a better buy than investing in wind and solar).

The biggest opportunity is capturing (otherwise wasted) heat for space or process heating.  This is hard to do with the traditional large central plant, unless there happens to be a big industrial user nearby.  But if the regulations were changed to incentivize efficient generation, then the utilities would be encouraged to look at distributed generation options, and would also work toe encourage large industrial users to locate near their already-existing large power plants.

Fuel cells are in fact one of the best ways to do medium-scale distributed generation.  And the types you would use in that application (e.g. molten carbonate) can eat natural gas (or biogas, or coal gas) directly -- they reform them to hydrogen internally, using their own operating heat.

One of your statements was "Burning less fuel to make a kWh lowers power costs -- and therefore, we don't focus on it."

Can you state a specific example where a utility could literally "burn less fuel" in a way that actually "lowers power costs", with numbers? co-gen may have niche applications, but it's not going to significantly change power plant efficiency, especially when you account for all the thermodynamic components of efficiency correctly (heat, by itself, is a poor use of fuel energy because of the randomization of thermal processes).

co-gen may have niche applications, but it's not going to significantly change power plant efficiency, especially when you account for all the thermodynamic components of efficiency correctly

Please explain what you mean here.  Space heating uses water at 150-180 F and domestic hot water runs around 120F -- well below the exhaust temperatures of most generation systems.  Between electricity generation and heat recovery, you can usually get 75-85% of the fuel heat energy, of which 30% - 60% is electricity.  So your comments don't make a lot of sense to me.

AP:

There are massive opportunities, but none of any significance involving existing power plants.  It's sort of like saying that you could boost fuel efficiency if you made lighter cars: the statement is true, even though it isn't practical to boost the fuel economy of the existing fleet.  The difference in the power sector though is that the existing fleet exists largely because of existing regs, and therefore in spite of economic incentives to conserve that exist in unregulated sectors.  

Thus, while one could argue that the economic dislocation of shifting to more fuel efficient cars might be unacceptably painful, the same logic doesn't apply to the power sector - just because a plant was built  not to conserve doesn't mean it should be given a perpetual license to operate.  After all, any other factory that was built without consideration of long-term operating costs might find it difficult to remain competitive, and no one would argue that it is societally disadvantageous for it to do so.  The difference on the power side is that we've built these plants under the guaranteed-equity-return rubric that is really bad societally in terms of both economic and environmental consequences.  In a different paradigm, those plants never would have been built.  So change the paradigm.  What we would actually then build is a fleet that looks much different from the current fleet, and much more like the one we had in 1910, before we passed all those guaranteed equity recovery laws.

Specifically:

The DOE estimates that there is a technical potential for 135,000 MW of cogen in the country, all of which would - if deployed - cause other, less efficient plants to be idled.  This analysis is based upon calculations of thermal loads at existing industrial and commercial facilities.  As  a practical matter, it isn't possible to convert existing central stations to cogen of any consequence because (a) they're too remote and (b) they make way too much heat.  (Remember that it's a heck of a lot easier to transport electricity than heat.  As a result, the economically rational power that has been built in this country was sized to the thermal load and is considerably smaller than the current central, power-only fleet.)

In addition to that, the EPA has estimated that there is another 65,000 GW of potential power generation from waste energy (heat, opportunity fuels, etc.) at industrial facilities.

Our company has looked at both sets of data pretty closely and concluded qualitatively that they both almost certainly understate the total, for reasons that largely have to do with the fact that it is difficult for DOE/EPA types without industry expertise to really understand the opportunity.  We are working on getting this published, but in the meantime, just look at those ~ 200 GW worth of potential.

That's 20% of the power fleet, but closer to 40% of the total MWh since these are all plants that will tend to operate baseloaded (as compared to the large portion of the existing power fleet that runs as peakers).  Run that math, and that works out to something on the order of a 20% reduction in all US GHG emissions (including those from transportation), all of which could be built just based on economic self-interest.  And to your point, it would not enhance the efficiency of the existing fleet, but shut it down.  To the benefit of everyone except the investors in those plants.  And before we shed a tear for them, let's not lose sight of the fact that that would impose a discipline on those investors that utility investors have never had - but which has been shown to be pretty effective at preferentially targeting capital investments towards societally-beneficial activities in every other sector.

Your comment that "heat, by itself, is a poor use of fuel energy because of the randomization of thermal processes" makes an all too common mistake that confuses basic thermodynamics with economics.  (Indeed, one of the things that I tell all new engineers is that they will not be productive until they learn to get over the idea that the laws of thermodynamics are inviolable.)

Which I realize is a rather fiesty thing to say, so let me explain.  Thermodynamics is inviolable.  But when building a power plant, it is economics that trumps.  And specifically, the economics of fuel use.  And those economics know nothing of the second law.  

Let's take a very simple cogen plant as an example.  Suppose I have a 10,000 Btu/kWh heat rate on the power side, meaning that I release 10,000 - 3,413 = 6,587 Btu/kWh of heat.  Let's say I can recover 70% of that heat in a useful form as steam (pretty typical for a heat recovery steam generator), so I get a total of 6,587 x 70% = 4,611 Btu/kWh of steam out of that cogen plant.

What's the net economic and environmental consequence of that plant?  Thermodynamics-trained  engineers will lean (too heavily) on the first law to say that I have a heat rate of 10,000 - 4,611 =  5,389 Btu/kWh actual heat rate, or (3413/5389) = 63% overall efficiency.  But here's the catch: that calculation is irrelevant.  Irrelevant economically, irrelevant energetically, and irrelevant environmentally.  Here's why:

Whichever E (econ, env, energy) lens through which I view the world, what matters to me is the net change in raw energy use per kWh.  And that number is not a function of how much steam I recover, but how much fuel I displace.  Invariably, that steam was previously being produced in a steam boiler that was burning fuel to make (low grade, high entropy) heat.  The fact that that's a shi*&y use of fuel is not my issue: what is at stake here is how much of an improvement I can make in the prior situation.  So now let's factor in the fact that the 4,611 Btu/kWh of heat I am recovering as steam would have otherwise been produced in an ~80% efficient boiler.  That means that my recovery of these 4,611 Btu's as steam allows me to eliminate 4,611/80% = 5,763 Btus/kWh of fossil fuel.  

Now let's walk through the math: before the cogen plant was installed, the plant was burning fuel and buying power.  After it was installed, they have to burn an additional 10,000 Btu/kWh of fuel, but they avoid the need to burn 5,763 Btu/kWh for heat, giving them a net economic heat rate of (10,000 - 5,763) = 4,236 Btu/kWh, or what is in effect a (3413/4236) = 81% efficient power plant.

This is not simply an academic calculation.  If I build that plant, it is in my economic interest to keep it running so long as the electric price is greater than my fuel cost divided by 0.81.  The fact that I am converting some fuel to heat and some to power doesn't matter: only that the net change in my fuel purchase and power generation (solely as a function of the cogen facility) enables me to realize effective economics of an 81% efficient power plant.  Same deal from an environmental side, or energy use side - and all result from the fact that the first mistake is to convert high-value fuel into low-value heat.  But having made that mistake, there is nothing "wrong" with fixing it, and in so doing for the corrective investment to take credit for the fix.  

An analogy I like to use is the guy who's been hitting himself on the brick for 5 years, and keeps buying bigger helmets.  If he suddenly decides to stop hitting himself in the head with a brick, he is going to realize a real economic gain in avoided helmet purchase & maintenance costs.  Does helmet purchase have anything to do with bricks?  It shouldn't - but if that's your alternative, it's a legitimate benefit.  Similarly, the fact that we've been throwing away heat in cooling towers and then recreating it in boilers may be a bad idea - but it doesn't mean that the guy investing in cogen doesn't realize a benefit from both.

And, re: your niche comment, see my prior response about the 135 GW.  Compare this to the present installed cogen base of 83 GW, and then compare that aggregate (135 + 83 = 218 GW) to the ~100 GW of installed nuke capacity.  Hardly niche!

I heard Peter Darbee, CEO of CA's PG&E utility, give a talk a few months ago that included some key elements of Sean's wish.

He said not to expect anything useful to happen until other state PUCs changed to incentivize efficiency, and even then, getting PG&E to change wasn't easy, although replacing 28 of 35 key executives helped.

-John Mashey

The cogen percentage in the US has gone down, not up, in recent decades, as major industrial production has left the country. I don't see it as a long-term solution at all. Are you seriously saying that 100% of that 200GW is 30+-year 100% capacity-factor heating need? That seems very unlikely.

The usual definition of cogen (the 135 GW piece) does not in any way reduce the quantity of fuel needed to produce electricity - in fact it typically has lower electric power output relative to what you get from combined-cycle natural gas systems, for instance. Cogen does reduce the quantity of fuel used for heating, but then we should talk about it as a way to improve the efficiency of commercial/industrial heat in the nation, not as a way of improving the efficiency of electricity production. It has no effect on the efficiency of electricity production.

Now what you're doing (the 65 GW piece) does reduce fuel input to electricity by pulling electric power out of industrial process heat. That's fine.

But your calculation that 200 GW of potential results in 20% reduction in GHG emissions is just wrong because you're claiming it's an efficiency improvement for electric power, when it's not.

You might at best get a 5-7% reduction in GHG emissions going the route of replacing 20% of our electric generating infrastructure. But spending the same amount of money (something like $500 billion) on solar or wind would result in at least a 15% reduction in GHG emissions for the US - 2 to 3 times the benefit for the same cost.

But we've had this cogen argument before.

Do you have anything else specific, besides cogen, to offer on improving the efficiency of electric production, as any kind of proof that utilities really are dragging their heels due to the current regulatory environment?

The usual definition of cogen (the 135 GW piece) does not in any way reduce the quantity of fuel needed to produce electricity - in fact it typically has lower electric power output relative to what you get from combined-cycle natural gas systems, for instance. Cogen does reduce the quantity of fuel used for heating, but then we should talk about it as a way to improve the efficiency of commercial/industrial heat in the nation, not as a way of improving the efficiency of electricity production. It has no effect on the efficiency of electricity production.

What we're really talking about here, if you want to be excruciatingly precise, is improving the efficiency of fuel utilization in the electricity generation sector.

If you insist that we "talk about it as a way to improve the efficiency of commercial/industrial heat in the nation", then what we're talking about is restructuring the electrical generation infrastructure to do double duty, to satisfy a need using a resource/waste stream that is currently untapped.

Whatever.  If recovered heat offsets the need to burn additional fuel, then that's a win.

You're focussed on commercial/industrial applications.  Don't underestimate residential applications, particularly in high-density urban contexts.  Individual water heaters are a silly way to supply domestic hot water in an apartment complex.  Same for individual furnaces, for that matter.  A unified, building-level heating and domestic HW system using heat recovered from onsite cogen can save a hell of a lot of heating gas.

Similarly, don't overlook the importance of heat recovery for COOLING apartments/malls/industrial/ office spaces -- low pressure steam can be just dandy for supplying the heat needed to drive chillers ...

Save the world: Reduce greenhouse gas emissions 5% annually.