Will DC power distribution be back soon? 1

[Skogsgurra]

We all(?) remeber when Edison and Westinghouse tried to introduce DC and AC as their respective standard systems for power distribution more than 100 years ago.

We also know that AC won - for obvious reasons. And we all(?) still think that the best guy won. Don't we?

But, time and technology has changed. We have more and more electronic devices and more and more switched power supplies. We also get more and more variable speed drives and the micros already are mostly electronic and the computers all use internal DC and and...

All these devices use rectifiers to produce the DC link power. And the rectifiers distort mains voltage and load the Neutral unexpectedly much and and and... They also have losses. And the DC link capacitors are bulky and costly.

Local DC distribution - within a block or within a house - would make things a lot easier. No harmonics, no hot Neutrals no bulky capacitors. Integral VSDs would be a natural thing and not a special motor. So, there is a lot going for DC again. And there will probably be a time when we regret that AC did win. DC will simply be the better solution for local distribution.

Of course, there are problems as well, but I think that it is about time that we took the AC/DC thing under some consideration.

Comments? Anyone? Are there DC islands already? Work in progress?

 

[tinfoil]

The main reason why AC 'won' was because of the incredible advantages and versatility offered by the AC inductive power transformer. It allowed use of higher voltages and reliable and EFFICIENT change of these voltages to/from the actual device usage voltages.

If DC was to be reused, even at a neighborhood level, what voltage would you use? 5VDC ? 12VDC? Something greater?

Efficient DC/DC voltage conversion is complex and expensive.

The wire size for a 12VDC system to supply a single home would be quite tremendous. Typical peak loads of ~100A at 120/240VAC is ~24kVA. This would be ~1200A at 12VDC (allowing for flat line vs. sinusoid). Big wire. Big Losses. You do not even get the 'cancellation effect' on the neutral that the typical 3-wire 120/240 split service provides.

How could you possibly supply industrial-type load levels of several MVA?

How do you cheaply smooth the torque on the rotor a 100hp DC motor?


I fully expect that AC distirbution is here to stay, whether you are talking about the (inter)national grids or at the neighborhood level.

 

[ijl]

I have heard that A Big Binational, well, actually multinational company has made studies on the DC-distribution for some years ago. One concern was the short circuit current, because there are no inductances to limit the current. But they and others are (probably) still thinking about DC.

 

[busbar]

 
To provide voltage changes and galvanic isolation, AC will long be the necessary intermediate in DC distribution. Search on Asea’s “HVDC Light” program, and if given the chance, tour a converter station. [Unfortunately, I understand both ends of the ±500kV 3100MW Pacific DC Intertie are no longer open for walk-in visits.]

 

[BJC]

I and another engineer did a study on a proposed DC distribution system for a semiconductor fab a few years ago. The owner wanted to evaluate the system in terms of cost and reliability.
The system used two large battery banks with chargers and backup generators to supply a 240 volt ( 475 Volts was also evaluated ) DC system.
Most of the loads were what would have been to replace large UPS banks and motors. The owner was going to use small inverters to supply tools and other special equipment. the DC section of inverters for motors loads and small inverters ( the inverter section of a UPS) for specific tool loads.
A similar plant using AC distribution would have had up to 2 MVA of inverters and a uninterruptable distribution system. Motors loads would have been on an standby bus powered from utility and diesel generators.
This owner was big enough and was a big enough purchaser to get there way. If they speced tools with a DC input and an inverter for AC power in the tool they would have got it. The same went for Cleanroom fans, if they said They wanted AC motor drives with a DC input they would have got them.
It was favorable but that budget went away and at the same time tool makers and the process got smarter. A dozen years ago a "bump" in the power system was a disaster in some semiconductor processes now most tools and processes can live with them.

 

[rbulsara]

for all the above reasons..specailly those mentioned by tinfoil and ijl AC will rule...

key points being

Ease and efficeint transformation
AC motors (DC motors cant beat ac motor in cost and ruggedness)
Short circuit limiting capabiliy
Ease of AC power generaton...DC generators are much more complicated and maintenance prone..and afeter all they 'commutate' AC power to DC anyway..

Also I think AC carries more MW per a given size wire...a 3-phase ac circuit would trasfer power more efficeintly than one or more 2-wire DC circuits...I think (I have not done the math on this)

ALL the DC use you mentioned is in controls/singals and processors..the heavy equipment still uses ans will continue AC power.

 

[rbulsara]

Also zero crossing in AC helps extingish the arc or restrike while opening a contact..DC is much more difficult and requires heavier contactor/switches/breaker..equating to more cost and space..

 

[GOTWW]

With a DC power supply, and many little DC-DC converters, choppers, and etc hanging off it. The DC supply would no longer be DC. It would resemble a bandwidth of Gaussian noise. What new detrimental side effects would be discovered from this Noise Bandwidth Power Buss as opposed to the neat and clean current orderly harmonics that we get at AC?

 

[rsherry]

DC remains a good choice for certain transmission tasks.
Not so much in the US and Europe but in developing countries the distances between load centres can mean that AC lines are well beyond their natural loads and expensive line connected compensation is required, making DC more comparable economically.
DC may also be the only available technical solution in some cases if the idea of undergrounding long transmission lines gains more supporters.

 

[Mstrvb19]

As far as whether or not we would ever get comsumers, such as typical homes, to be powered from DC, I think this is highly unlikely. Everything is too standardized towards AC, and with so many independant companies building products for an AC powered home, it seems an insurmountable task to get those to change. As it stands, there are still many places in the US where homes still have old knob and tube wiring, and many more where home has screw in fuses. Also-take as an example the fact that the US will probably never switch to the Metric system becuase it is just too difficult to consider without having an EXTREME economic advantage across almost all demographics. Further more, if we attempted to replace everything down to subtransmission and distribution with DC without replacing home and other consumer systems with DC, we would have to have converters supplying every load-leading to the issue GOTWW brought up. So, while there are many technical advantages and disadvantages, I think they will all lose to the economic impetus towards the status quo.

 

[lengould]

Some strong arguments on both sides. My sense is that economics dictate AC for regional and local distribution because of the economics of the transformer over equivalent DC equipment. Also as several have pointed out there are distinct economic inertial factors favouring continued use of AC in endpoints, though 50 years ago it was common for off-grid rural installations to use 32VDC to run everything from pumps to vacume cleaners and irons for clothes pressing.

That said, HVDC transmission is definitely the best economically and logically to use for transmission which either a) travels further than about 500 km (300 miles) because the same insulators can handle higher average voltages with DC than with AC
or b) uses insulated cables, e.g. underwater (capacitance I think?)
or c) links distribution systems which would like to be able to maintain and islanding capability. E.g. if the US Northeast had used more DC interconnection at points like Niagra or at all state borders the August blackout would almost certainly have been limited to the one utility which screwed up rather than the huge area it struck. The power transfer of a DC->AC tie can be explicitly set and doesn't much care what happens out on the AC circuit.

In general, a logical modern network for North America should have a continental HVDC supergrid and each state it's own AC grid isolated from all neighbours except through DC connections. Agreed there are some issues re multi-point connections to DC transmission but no theoretical reason this wouldn't be do-able and far more simple and reliable than the present AC network which must be tuned like a longwave radio antena to work at all.


Pechez les vaches.

 

[GOTWW]

I comptimplated about this, a bit more. And came to a revelation. That the raw DC power generated from photovoltaic, and PMDC wind could be directly connected to the DC bus without any additional conversion loss. However, it becomes readily obvious that DC transmission is basically one-way. The wonderful transformer is a reciprocal device. It does not care which direction of the power flow. This is a great advantage for the future (maybe), where distributed generation is the norm, picking up available energy where it is available, from any form available. Suprising that the monopoly of power generation is not advocating DC, but, wait, maybe they are on, the macro-economic scale, really.

 

[logbook]

The increasing availability of low temperature superconductors will encourage the use of DC links.

 

[DanDel]

I believe DC distribution within buildings, concurrent with normal AC distribution, is the most plausible at this point.
A DC system, say at 42V, being used for computer loads and anything with an electronic power supply (all of which now have their own internal rectifiers) can be very practical and will reduce costs.
There could be a large single-unit recifier supplied directly from the main AC switchboard, utilizing the proper filters to isolate any harmonics at the source. This would remove the harmonic issues for most AC systems, while leaving AC power for motors, incandescent lighting, heating, etc.

 

[lengould]

DanDel: Actually by assigning motors to the AC subnet you're missing some great benefits of DC. These days it might make sense for all present AC motors to be supplied DC at the distribution point. The job of the motor starter can be re-asigned to an inverter which, at least in smaller sizes, these days is almost cheaper than the comparable AC starter. Remove the rectifier / filter section by providing a DC supply at say 120 or 480 volts and you could likely beat the cost of the AC equipment, and the improved management functionality would enable a lot of added benefits, including energy efficiency. Though the cost of the central rectifiers might add some hurt.


Pechez les vaches.

 

[Bung]

Modern day electronic drives can do some stunning control feats just using your bog standard el-cheapo squirrel cage motor. Electronics has eleiminated any advantage of ac vs dc to drive motors.

So which is "best" for energy transfer? AC is great if you want to keep the control and switching cheap and simple. DC is great if you want to transfer raw energy large distances. With dc, just don't try to switch it, interrupt it under fault conditions, or otherwise mess with it - it is very unforgiving.

DC traction systems give enough headaches with earthing, electrolysis, etc. Imagine all that in trumps from a power distribution system!

I don't think either will supplant the other - better by far to use the right one for the right application.


Bung
Life is non-linear...

 

[mc5w]

Some advantages to both systems, I agree.

I noticed early on at Carnegie-Mellon University that the homopolar motor/dynamo was the only true direct current rotating machine. All others, including the Van DeGraff generator, are AC machines with synchronous switching power conversion.

One possible DC system would be a 2-wire system suported by a 180 volt or 192 volt lead acid battery. 4x36=180 volts and 4x48=192 volts. You could use forklift truck batteries either way. Would need a multiple pole isolation switch that would break a bank into its 36 volt or 48 volt pieces for maintenance safety.

For 180 or 192 volts you could run 180 volt motors off of it. 120 volt DC relay coils would need a ballast resistor. The Emerson Maxi-Maestro servo amplifier (5,000 watts nominal) can direct connect and with a small sine wave oscillator connected to the velocity input would give you 120 volts AC - would have to be passed through a 7.5 KVA 120 volt primary 120 volt secondary transformer. The maxi-maetro's cut in voltage for the dynamic braking resistor is 240 volts DC and can withstand a DC bus voltage of about 300 volts DC so even fast charging of the battery will not hurt it.

Mike Cole, mc5w@earthlink.net

 

[lengould]

So how long until one of the major MCC mfgrs. comes out with an integrated plugin DC MCC? With master frequency signal distribution included? Might make some sense in places like auto factories, food processors, paper mills etc. where large groups of equipment all co-operate on a single task.


Pechez les vaches.

 

[peebee]

A few random thoughts:

+ Big DC power distribution systems never completely went away. All the phone companies switching centers, and I think all the CO's, all use MASSIVE battery installations and HUGE copper conductors to distribute DC power to their equipment. They never did quite trust those generators to start (and they might have had a point -- how does the reliability of your phone service compare to your ISP?).

+ Alluded to above, but I don't think anyone actually said it -- voltage drop at 12 or 24 or 48 volts is a big pain in the neck. If you need to go any distance at all, the Vdrop can be a much bigger problem than the ampacity, requiring even HUGER conductors.

+ Train systems (like the NYC subways) use DC. I'm not sure why, but I think primarily for historic reasons. Lots of big rectifiers in NYC for those trains.

+ DC power over Cat-5 & Cat-6 is becoming more commonplace every day. Maybe that will start to take over as the default residential DC power distribution system, at least partially supplanting the ubiquitous wall wart DC power supplies (how many of those things do you have plugged in around your computer?).

+ A friend of mine is a big solar nut. One of his biggest pet peeves is the big leakage current running through all the wall wart transformers even when the associated utilization equipment is shut off. Pain in the neck for him because if he really wants to max out his solar system he has to unplug stuff whenever he's not using it. Obviously this is true regardless of the source of 120vac, but it's even more critical when your source is (relatively) undersized and the cost per kW is so much higher. Just one more reason to look for something better than a plain-jane wall-wart as a DC source.

+ DC power transmission was mentioned above -- my understanding is that AC-DC-AC converters are often used to interconnect the national power grids (I'm pretty sure that all connections between California & other grids all use AC-DC-AC).

 

[peebee]

What's the distance limitation on USB? Maybe I should have thrown that in with the Cat-5/6 above too.