Post of the Day
February 25, 2000
Rat's Broadband Bandwagon
Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light.
Re: Clean Stable Reliable Power
[Fool Note: This thread references The PowerChip Paradigm, an article available for free download in PDF format from George Gilder's Website. The "9's" in this post are a reference to reliability, as in 99.999%; the more nines, the better.]
The PowerChip Paradigm ?
After reading through "The PowerChip Paradigm", I though I might add a few comments on the propositions put forth by the author's of this article.
I am a chronic lurker, but since my background is in the area of large-scale power systems analysis, hopefully I can provide some additional insights (but then again...maybe not since I am not a "power quality" expert). However, I am associated with the planning of new power transmission facilities based on "PowerChip" technology. These systems are based on IGBTs (insulated gate bipolar transistors) controlled with PWM (pulse width modulation). These devices switch very fast in order to produce the correct waveforms at the desired power levels. The use of this technology allows for a high degree of control over the power flow (in the 100's of MWs) and the system voltage. These facilities are intended for the bulk transport of power (in broadband terms this would be described as a backbone facility).
Having just spouted off about expensive technology, I feel the article in question focuses almost exclusively on expensive band-aids while largely ignoring the potential of distributed generation as a more permanent solution for higher levels of "9s". Essentially, distributed generation is a collection of smaller power sources, usually serving one load, as opposed to the the large bulk facilities of the existing power system which serves many loads. In other words, solutions to the problem can be looked at in two ways; "fix" the existing power, or develop ways to generate your own "clean" power. The authors have chosen the "fix" route.
In my opinion, the "clean power" and "ride-through" products which are discussed in the article are essentially band-aids for the "dirty" and unreliable utility grid power. Non-utility sources, such as stand-alone generators (isolated to the load) were presented almost exclusively as back-up sources for total loss of utility power, and not as primary sources.
So why don't they use these generators as their primary source? Well, these sources (principally diesel generators) are expensive to run, expensive to maintain, noisy etc. Having one of these supplies as the main source is also likely to be more unreliable than the grid itself (which is supplied by thousands of sources). In other words, even if you have your own supply, you would still need the aforementioned band-aids to meet your "9's" need. Plus your own power would cost much more than the grid power.
Now suppose that fuel cell technology (H2FC) or micro-turbine technology advances to the point where the $/KW-hr is not a great deal more expensive than grid prices. This would provide the impetus for a paradigm shift towards distributed generation as the solution for "9s" problem. Under this scenario, the H2FC generation is the primary source of power and the utility grid may become the back-up(?). It would appear likely that these devices are also inherently more reliable than diesels in that they have minimal moving parts and would require less maintenance. Futhermore, they could be networked through power electronics to provide increased reliability. Fuel cells are also likely to have other advantages over their predecessors such as being "urban" friendly as a result of reduced noise and pollution. And not to forget the PowerChip, which in this scenario are the devices which convert the H2FC DC power to very clean AC power as well as provide the fast sensing and switching for networked units.
So, the way I see it, the potential for distributed generation (H2FC) to emerge as a solution will depend on a number of factors:
1. The kWs produced are economically competitive with grid power or at least not significantly more expensive.
2. The reliability of an individual unit is very high.
3. Networking schemes utilizing very fast power electronics can be devised to provide a high level of control.
4. The ability to handle large commercial loads, specifically large changes in instantaneous demand.
These are big IFs, but the need to find economical and permanent solutions to the "9s" problem should drive development. I see point (4) as a potential hurdle (one that could bring us right back to the expensive band-aids). The distributed generation sources will need to respond instantly to demand changes which can be an issue at high power levels such that the system voltages remain unchanged and therefore do not interrupt sensitive system processes. To explain, you see this problem in your house nearly everyday. For instance, when the refrigerator or vacuum cleaner is turned on....the lights may dim momentarily. When the device turns on, it demands a high current (for a motor it is usually much higher than its operating current). This high current, when flowing through the impedance (resistance) in the circuitry in your house, creates a drop in voltage, which in turn, affects your lights. Likewise there are certain situations which may create the opposite scenario...a surge. Sensitive devices such as may be found in semi-conductor fab plants may not be able to withstand this type of a disruption. At low power levels it is easier to fix but at high power levels it is more difficult problem. This is the issue addressed by AMSC's SMES device (Superconductor magnetic energy storage system).
Can a network of H2FC supplies handle this type of situation at high power levels? I don't know, but I would guess that it will depend on how they are networked, how robust their integrated batteries are, the design of their controls and how fast the power electronics can respond.
AMSC's solutions to the millisecond problem at high power levels was certainly touted in the article. AMSC has been around since the discovery of HTS in the mid-eighties (high temperature superconductors), which requires liquid nitrogen cooling (low temperature superconductors LTS require liquid helium, a much lower temperature). So far, it appears to me, that this technology has met with limited commercial success. A few large scale semi-con fab plants and a few utility transmission system applications for voltage stability during disturbances. Why is that? My guess is that the technology, while certainly amazing, is too expensive. Quite possibly, other manufacturing facilities or large data centers have sought more cost effective solutions or maybe their supply feeds are strong and stable enough that they can't justify the premium. I don't know. It could be that their new micro-SMES devices will open many new doors if it is cost-effective. As far as utility applications are concerned, very little of anything has been added to the bulk transmission systems in the last 15 years. So they have either just gotten caught in the downdraft or there are too few situations where it is an economically justifiable solution. Some years ago when FACTs (Flexible AC Transmission systems) devices were first being introduced, of which SMES fits in to this category (among a number of other more widely known PowerChip-type advanced devices like the Universal Power Flow Controller or the Static Compensator), the industry buzz was that these were "solutions looking for problems". The makers of said devices pushed very hard, but to this day there are still very few commercial applications on utility systems. It could be that their day has finally come, but then again, maybe not. The impetus nowadays, at least from a utility transmissions systems standpoint, is that all of these devices can be used to alleviate constraints in the power system - given that it is very difficult to build new transmission. From a pure technical standpoint, this is true, but I think there are two strong opposing forces looming on the horizon - which may serve to alleviate some large constraints without touching the existing transmission (or distribution) system.
The first is the coming onslaught of new large scale merchant generation (over the next 5 years or so). De-regulation has created an environment where all of the new generation is being added by free-market merchants looking to "bid" their supply into the system (the theory is that this is a competitive environment which will reduce prices). There are tens of thousands of planned MW in the Northeast alone. On the surface, this would look like an environment that will create further transmission constraints. In localized situations, this is true. However, most of this new generation is likely to be sited in such a manner as to be closer to the load centers (metro areas) thereby off-loading much of the constrained transmission (interfaces) that is presently used for long distance transport from existing generation facilities to load centers. The second factor is the (potential) future onslaught of distributed generation (H2FC/Micro-turbines) sited AT the load (as already discussed). This has the effect of reducing the amount of power which must be transported via the transmission system.
Of course the argument could be made that this will have the effect of "freeing" up the existing transmission which will then become over-utilized again for long distance economic transactions (cross-country). In other words, since it is constrained now, many long distance economic wholesale transactions can't be accommodated. The on-set of distributed generation may free up capacity so that these transactions take place. However, long-distance transport is expensive and the supply price differentials would have to be high to support this. Anyway, if this is the case, then FACTs may be back in business (as well as keeping my own job). It is too early to tell as the industry is still evolving (at a snail's pace).
Present environment for adding new devices to the Grid.
On the other hand, the "9s" problem could also be attacked from the other direction, that is, by "fixing" the utility supply system itself through PowerChip enabled technologies as discussed in the article. In a word: forgetaboutit. While the generation supply side of the industry is de-regulated, the transmission and distribution networks are not. While some forward progress have been made (it depends on the part of the country), in my opinion, clear incentives for investing in network upgrades are not there.....nor are they likely to be any time soon. For instance, if Company X wanted to install "clean power systems" at key nodes in the transmission network, how would they get a return on their investment? In most cases it is not clear. On top of that, obtaining the siting and the permitting requirements can often be a YEARS long process. And even if we did install these devices throughout the bulk system, you are still at the mercy of the squirrel who has decided to commit suicide on your local distribution feeder. New things will still get built, but until there is some substantial changes, it will only be a trickle.
The point is: In the present climate, solving the "9s" problem is going to have to be driven from the load (customer) end of things. This points strongly towards either the H2FC concept or economical means of cleaning grid power at the customer end. My long term bet is on distributed generation. In the interim, the players in the grid power "cleaning" game may do well if they are providing cost effective solutions AND the wide-spread drive for extreme "9s" truly materializes (as the author's believe). Most likely, it will be some combination of both.
Just as a side note:
Some years back, a group of us (bulk power system analysis consultants) attended a presentation given by GE/PlugPower on the emergence of H2FC technology.
On the way back, the most respected member of our party (a legend in the industry), turned to the rest of us and said "I think we've just been made obsolete".
Easy for him to say...he's retired.
Anyway, back to lurking.
|Liked this post?
Read more posts by this author.
More Recommended Posts