Link to home
Create AccountLog in
Avatar of inanton
inanton

asked on

Broadband over powerline equipment costs

I am doing some research on the feasibility of broadband over power lines aka power line communications technology. I cannot find any relevant costs for the equipment online, and I have not got any answers from the equipment manufacturers I approached.

I need information on the prices of the following products or an equivalent system

CT CouplerTM
CT Backhaul-pointTM
CT BridgeTM
Powerline Modems at Premises
CT ViewTM Network Management System

Thanks.
Avatar of The--Captain
The--Captain
Flag of United States of America image

http://www.linux-magazine.com/issue/38/Powerline_Cable_LAN_Modems.pdf

Has a comparison of cable vs powerline networking - It sounds to me like you are looking at trying to do powerline WANs, which AFAIK would not be possible beyond the first high-voltage transformer.

Cheers,
-Jon
Avatar of GinEric
GinEric

Been done before.  Not a good idea.  For starters, anyone can pick it off your neighborhood power lines right at their electrical outlet.  Second, it disrupts other signals used by the Electric Company on their power grid.  Third, would you really want your modem feeding information from your computer directly into a high power line?  Think, there's basically a single diode between the power line and computer frying.  Diode fails, computer goes Poof!

The vendors don't want to discuss their failures.

The research is oldtime, from way back, so how'd you get this research assignment?  Sounds like you got a brush off assignment!

Across the first power transformer, the Block Transformer, maybe, but you step up 110 to around 10,000 volts and little signals, like 5V digital signals, also go up 100 times to 500V.  Also, not a good idea.

Then there's the frequency.  Inductors, of which a transformer is one, don't like higher frequencies: ELI the ICE man, Voltage leads current in and inductor, Current leads voltage in a capacitor, which is another way of saying that the inductor won't pass high frequencies because the current and voltage collide producing a garbled representation of the input frequency.  Electromotive Force [EMF] and Counter Electromotive Force [CEMF] is another way of putting it, the frequency starts to bounce back off of itself and no message gets through.  Even though the Block Transformer is balanced with its own Capacitor, it forms a bandpass filter that restricts all frequencies outside of 60Hz and that's because harmonics can build up to dangerous levels in CEMF feedback and actually melt the transformer! [Exactly what happens on very hot Summer days when you hear one of those Block Transformers go Kapow! right before the whole block's electricity goes out; okay, sometimes their a circuit breaker that prevents it and makes a loud noise too, but they often stick and it melts anyway!]

Do you drink coffee or soda near your computer station?  Better think twice with a powerline nearby!  Modems and computers run on low voltage, with fuses and circuit breakers and stuff, powerline modems do not have a lot of safety features and run at full line voltage.  Their fuses and circuit breakers don't trip until the main trips!  You could be electrocuted by the time that happens.

All in all, broadband over powerlines is a dead horse.  No need to beat it any further.  For research, check any Electric Company Metering service or devices, the phone company tried it once, the cable companies have tried it in he past, a lot of their billing companies have tried it.  It doesn't work, it's costly, it's dangerous, and you won't find any admins that want to work on such equipment, all of which make the cost unfeasible.

The equipment prices are in the $1,000.00 as opposed to the free DSL and, at most, $59.95 DSL modems and telephone modems, so no feasibility there.

Powerline Communications is used mostly in ISA [Instrumentation Society of America] industrial sites, where it is feasible and where power lines are handled every day and are more convenient for carrying feedback systems controls information.  Refineries, Airports, Manufacturing Plants, and the like, where it is more reliable than simple computer and low voltage digital signals are overpowered by the likes of airplanes [a conductor moving through a magnetic field that genereates voltage spikes on nearby wires], industrial 3-Phase Voltages, and even superhigh voltages, such as Hoover Dam Alternators and Power Generation.  It makes sense here because a 250,000 Volt Alternator start up can spike out any nearby CAT5 wire!  The startup surge can rip a spoon right out of your pants!  and it can snap an Ethernet wire right out of its shielding.  So powerline digital information carrying and transmission make sense in such environments, but not in the home.

As for the cable company cables, I've got enough shocks off of them to warrant calling them powerline transmissions already.

Then, every once in a while, the cable company sends a 120 Volt spike down the line to punish would be cablejackers.  If you don't have the filter, you get zapped.  Obviously and illegal act by the cable companies, but until they get caught doing it they figure they're home free.  The do send these signals down their lines every now and then and if you happen to be in contact with that cable connector, you get the full 120V line voltage.

The phone company runs a quiescent 50V, until the phone rings; then it jumps up to 90V, only 20-30 volts under the residential line voltage of the power lines, and this can give you quite a jolt if you happend to be rewiring your phones when someone calls in!  I know from both measurement and the experience of a "call" while working on the line.

It's enough already that the phone line has such power on it.  So, it is possible to send the signal over a powerline, the phone company and all ISP's do it every day, but their powerline is direct to you with no transformers in between.  If there is a transformer [repeater] inline, your connectin will lag and lose bandwidth; often cited by telephone repairmen when responding to reports of "slow" DSL speeds, etc..

You've got the basic information.  You'll have to find the references to pricing data in your own continuing research.  Try the big corporations that make all the wires and stuff, GE, Bell, Belden, AT&T [and their longlines division], GTE, Western Electric, look at their histories.



Avatar of inanton

ASKER

Thanks for the advice, but I am looking at the technology (latest iteration being promoted by Amperion, Current, etc), the regulatory environment (recent FCC rulings, etc), and the cost of inputs (equipment, salaries, etc), and trying to determine what the current level of viability is. The exercise is really about the due diligence that investors (such as venture capitalists) must conduct. The piece of the puzzle that I am missing is the price of the equipment that:

1) provides interface between content provider (fibre, etc) and the electrical grid,
2) Repeats signal along the power line
3) tranfers signal from the main power lines into the home.
4) network management system (usually proprietary)

I was hoping that someone in EE has contacts within a power company that has piloted this, or within an equipment manufacturer, and therefore had an idea of the prices (even an order of magnitude would be very helpful).
Bell Telephone; why don't you call and ask?

A repeater for a phone company is no different than repeating on a power grid.  They're all considered Instrumentation & Control, not digital devices and network stuff.  The box to convert from fibre to electric power is going to be expensive, think in the area of at least a thousand, perhaps ten thousand.  Repeaters are not that expensive, but you must include the labor cost to install them, regulatory fees and oversight that is going to drive costs way up as you pay Linemen now to do the install; sorry, but cable techs are not allowed on power lines.

Western Electric; is who did it for Bell.

Network Management systems are done by really high level power company Professional Engineers.

For a plant, of say 20 acres, whatever the function, you're looking at somewhere around $3,000,000.00 for the first year, mostly because of the design, development, and installation.  Thereafter, you're down to replacing parts and paying full time employees around the clock.  The upkeep cost is about $150,000.00 per employee who starts at $30,000.00 a year.

Then, the whole thing has to be Certified by legions of both Engineers and Electricians.

Electricians, if unionized, as most are, start at about $30.00 an hour; we don't even talk about the yearly salary!

Every salary is multiplied by five to give the upkeep cost of maintaining an employee base, even if that cost is actually only three times the base salary because in a startup you must provide for 80% cost overruns.

And if the plant is nuclear, multiple everything, everything, by about ten!

Tell you what, what's your location and I'll find you another EE.
ASKER CERTIFIED SOLUTION
Avatar of GinEric
GinEric

Link to home
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
See answer
GinEric - do you have *any* sources for the information you have posted?

Cheers,
-Jon
For the simple references of PLC insider business news:
http://www.bechtel.com/newsarticles/97.asp

Government:
http://www.sba.gov/

Universities:
http://www.seas.upenn.edu/
http://www.wharton.upenn.edu/
[it's always a joke a Penn that Harvard's Coat of Arms looks like ours, but more of a 'Rebel Flag' appearanc]
http://www.hbs.edu/
Mitsy:
http://web.mit.edu/catalogue/overv.chap6-crems.shtml

Search also NYSE to find others interested in building PLC Internet:
http://www.nyse.com/pdfs/ChinaSpecial0105.pdf

This works as a wide approximation because of the limits of the author of the calculator at the site:
use V(c)=$8.50
F(c)=$1 Billion [cost of the plant $1,000,000,000.00]
Project sales: 9,250,000  [because calculator only goes to 10M]
S(p)=$25.00  average price of a device, such as modem, nid, etc..

Shows that the loss would be about $847 Million.  You can use this as a first year projection.  Also states that BE is at 60 million units sold.  Use this as a guide to optimize pricing.  Obviously, 10 million customers is not enough, however, at $25.00 per month, you change the sales price accordingly to 12X$25.00=$300.00 for the first year so that revenues now become nearly $3 Billion and profit becomes $1.67 Billion.
Based on 10 million customers for the first year.

The site calculator is okay, but it is better to know the actual Integral formula on which it is based.  That you can find in most any Calculus book or course.

One thing lacking in the calculator is the curvelinear character of cost analysis, which can only be got by integration.  You learn that in either engineering school at bachelor's level, or business school at master's level.

Amperion patents and stuff:
http://uk.biz.yahoo.com/040609/183/evkf6.html

Ambient:  This one gives you the name of Ambient co-founder, 27 years old, can't imagine him not giving an interview, for promotion:
http://www.entrepreneur.com/article/0,4621,312363,00.html

For British Telecom: [for tunnels, you have to find the civil unit that digs them]
http://www.labs.bt.com/

Imperial College:
http://www.ic.ac.uk/P287.htm
http://www.labs.bt.com/research/universityresearch?doc=42996
is part of London College
Specifically, Prof. Midwinter and some friends at Lord Thompson's DBS venture, for the initial phases of DBS, which began in 1986 or thereabouts.

Bechtel/Ambient: see bottom for contact info
http://plugtek.com/PageLibrary/AmbientAndBechtel.htm

For the rest, you'll have to do some searches.  For the cost of building a plant, and R&D on a mainframe, I am the source.  I used them as a comparative cost analysis.  However, these figures are well known for large ventures, corporate and government.  Take the cost of a Nuclear Power Plant, we have a couple of losers in this category, Peachbottom and Limerick.  I don't count 3-Mile Island because it was a complete failure that cost way over the building costs after it melted down.

GE is one of the makers of 250,000V Alternators, such as the ones used at Hoover and elsewhere.  The cost of the dam has to be translated to today's dollars, not hard to do.  The cost of the network management is actually negligible, until something goes wrong, but even then it's less than 1/100th of the cost of construction.

Some ratios you just learn from experience; selling products at ten times to thirty times the cost, The ratio of salary to the actual cost of keeping the employee [1 to 3 post development and 1 to 5 during development].  For a 20 acre plant, the costs are generally found from time to time in most newspapers and magazines, professional journals - sometimes, and public hearings always.

The practice of using existing devices in an initial cost analysis is just common sense.  If the device doesn't exist, you don't know what it costs, but you can base your predictions and calculations on past practices, this is called estimating.  It's a pretty good tool for mathematicians and engineers to get an idea of the predicted behavior of the business model.  The should be teaching this at both engineering and business schools.  All it is is a "substitution of values" type of equation or problem.  Once you have the formula, you can plug in actual values when you discover, usually by experience, what they are.

Again, for the cost of prototyping a device on a small scale, usually at home, I am the source.  From idea, to engineering drawings, making the photoplates, to contacting subcontractors in Virginia to manufacture the printed circuit, and decide on the packaging, then make the prototype, I have done this a few times.  Any good Electrical Engineer should be required to do this to earn his degree.  It's not enough to "just know the theory," you must also have practical experience in the R&D right up to manufacturing of the product you've designed.  Whether it be a little audio mixer or a new logic device made at home, to a mainframe made at a $3 Billion plant.  Of course the latter costs enough and requires more money and employees than you can fit into your home, so you do it at a major manufacturing plant, with pay.  At home, you usually don't get paid unless your device takes off.  Apparently, the designers of some of these PLC technologies had home designs that did take off.

Many of the costs are,indeed, proprietary.  I don't expect Unisys or IBM to tell you what they've spent on some projects.  But there are usually former employees who know because they were responsible for those budgets.

The same applies to PLC, the new technology is something they'd like to keep as a closely guarded secret from their competitors.  However, their costs are quite transparent to someone who knows them all and has done their estimates in the past, including government contracts and costs.  I said they all have this "magic number" that every contractor who makes a bid tries to guess at.  That number is the optimum hourly rate at which they will pay for each manhour.  That number is somewhat guarded by government, but it does seem to leak more often out to the big corporations, moreso than any leakage to a small business.  Knowing this "optimized" number means getting the contract.  It is not the lowest estimate, and it is not the highest estimate.  It is like pitching nickels against a wall: whoever gets closest to this number, and never below it, gets the contract, i.e., no "wall leaners."  The exact same approach as bacarrat or blackjack.

These are really more common sense and street sense issues than documented collegiate "howto's."  It really doesn't matter if it's Wall Street or Canal Street; the method is the same.  And the Wall Streeters are most often taught by the Canal Streeters, not vice versa.  From which, the universities glean their information for texts to publish.

It is hard to get sources from any successful CEO.  Many of them are success secrets which are not given out for free.  But observation of a CEO over time gives away those secrets.  Sometimes, they are just too much fun not to publish.  Sometimes, they're left as a legacy or in an autobiography.  Usually, they tend to "close ranks" on these issues.

Some sources are governed by a promise not to reveal the source.  This is a biggie in business.  This is overcome by the personal interview, if you are given permission to publish the source.

I think I posted at least two strategical contracts, India and China.  Which fit well with the Ambient/Bechtel strategy of third world type markets.  I must say though, I don't think of either India or China as third world anymore.  Their computer and electronics manufacturing is now something like 1,000 times the volume of the U.S. and as far as I'm concerned, the U.S. is no longer a competitor in the industries which it invented.  Thank you very much Mr. Congressman, Mr. Senator, and Mr. President!

The U.S. is more of a third world chip maker, computer maker, electronics manufacturer, than any of these countries and at a strategic disadvantage thereby.  For example, if China were to sanction the U.S. computer market, the computer economy in the U.S. would begin to collapse.  The same for electronics.  You can see from the worldwide markets of PLC and other technologies that the U.S. is not even a player in this, but to be used for testing as a third world country!

Amperion and Ambient have to lean on big brother Bechtel in Germany.  And New York is the test area.  Con Ed, specifically.  You won't find the pilot program at Con Ed's site:
http://www.coned.com/
but you'll find it here:
http://www.plugtek.com/PageLibrary/ambient_fcc.htm

You have to, of course, tie the loose ends together to see that the pilot project is funded by the State of New York, and the money goes to pay for the German Bechtel company to build the pilot in Manhattan and glean the profits back to Germany.  Interesting, New Yorkers would then be paying the people of Germany for their Internet, and maybe eventually for their electric power.

Which brings us to feasibility and PUC Regulations.  The F.C.C. apparently has approved, but will the voters of New York?  Luckily, Ambient and Amperion seem to be U.S. interests, but will they remain so with foreign options under their financing belts?

You may find in any university library the low down on how this game is played.  Edison was ousted early from his own company.  It became Con Ed and GE under the new consortium's ownership.  One of them was Marconi.  Marconi's own RCA was litigated into Marconi, America, RCA.  
This was brought on by Americans who didn't want a foreignor owning all the broadcast and communications industry in America.

The same could happen to Con Ed, Ambient, and Amperion.

The source for that are common records and books at every university, and at various historical records places.  The history of most companies and inventors also tells of these events.  They're common knowledge among investors, or at least "informed" investors.

I think I posted the source for the White Paper in the other place where this question was asked.

Having had not a whole lot to do, testing these question areas, and preparing for publishing a book, I've particpated in order to discover this process of Experts Exchange question/answer sessions because I wanted to find out what drives people to take so much time to answer other people's questions.  Like the man says, "It ain't the money."  I've gotten at least that far.  It doesn't seem to be the points either, unless someone wants to be an EE superstar, for whatever reasons.  I think I'm now leaning toward seeing it as a necessary component of social interaction.  That only means that people like to talk to each other.  I really don't know why so many people take so much time to answer questions and to help others.  Good deeds or somehow profit motivated; I'm not sure.

I can write a lot because that is what I do, I write. [among other things]

I also read a lot, a very lot.  I can't shut my mind off; it's always going, looking for more information, for decisionmaking.

I may wind up contradicting my first statement, "not a good idea," after considering the problem; executive prerogative.  There's always the adjunct market, and it's always best to play devil's advocate first, before investing or advising investment.  A good lawyer or businessman does this all the time.

So it may turn out to be a good idea.  I suppose that will depend on the final research paper "inanton" presents.  By the way, the cost analysis can be found in Howard Anton's books, "Calculus {I and II" which made the Drexel University Professor a millionaire.  [Source]

Drexel is just one place where I studied engineering, math, and business, accross the street from another alma mater, University of Pennsylvania (under a tutorship).
Avatar of inanton

ASKER

Thanks for the info. I have been tearing my hair out trying to get some idea of the costs of the equipment. I have emailed and called Current, Amperion, etc with little feedback. I am actually getting a lot of help from the local utility company who saw the pilot in Cinncinati. BTW, they still have not been given a quote for the cost from Current. Current offered to run a small pilot for them, but refused to divulge cost information... really strange.

I will post my report and add the link here if anyone is interested.