Energy Saving and Alternative Energy Forum

The latest design techniques for saving energy or adapting alternative sources

CONVENED AND MODERATED BY JIM HARRISON and PAUL O’SHEA

Jim Harrison : Welcome to our Energy Saving and Alternative Energy Forum. Our Senior Editor Paul O’Shea has joined me today for this discussion and along with Paul we have five industry experts from Texas Instruments, NEC Electronics, Power Integrations, International Rectifier, and Analog Devices. We are going to start right off with the first topic of what are the latest and greatest power semiconductors available for basic ac/dc power conversion.

Eric Persson (Executive Director of Field Applications Engineering, International Rectifier ): What I see our customers using predominantly now for ac to dc is pretty traditional topologies. Very often it is power factor corrected. Many times using super-junction, high-voltage FETs for boost PFC.

Where efficiency is most important and cost isn’t such an issue engineers will use silicon carbide Shockley diodes for the boost diode. And, at the moment, it’s mostly traditional PN junction diodes used for the input bridge although we’re seeing a move towards bridgeless boost topologies. The next layer of efficiency I think is going to be available from getting rid of those diode drops on the input side.

Eric Persson International Rectifier

For upcoming technologies, IR is working on gallium nitride and we think that will be competitive in the same way that super-junction and silicon carbide are now.

Jim Harrison : Eric, I don’t know if you have a handy number for how much more efficient a silicon carbide diode is over a standard silicon diode.

Eric Persson : The main gain you get is you get rid of the switching loss and so you have capacitance still, but you don’t have reverse recovery losses. And the forward drop is a slight penalty there. However, over temperature it’s very stable so you get very similar performance at 100 to 120°C that you do at room temperature, whereas traditional silicon diodes tend to get worse at higher temperatures.

So, you’re going to have several percent efficiency improvement. In the range of maybe 500 W up to 3 kW, running off single-phase 230 V, we see customers all the time getting 97% to 98% efficiency on the PFC front end. It’s pretty impressive.

Paul O’Shea : Eric, you could say the same thing about gallium nitride, only it would be even better than silicon carbide? And, I think you can talk more to that.

Eric Persson : Yeah, I mean, we expect from our early results they should behave similarly. They’re both good at high temperatures. That’s IR’s objective is to be able to make gallium nitride with equal or better performance compared to silicon carbide, but at lower cost because it’s on silicon substrate.

Dave Freeman (TI Fellow and Engineering Manager, Texas Instruments ): Another thing we may want to mention and is that control can also help. It’s not just the power components, right? So another popular thing a lot of people have moved to interleaved PFC, some people have moved to bridgeless rectification and things like that. And all those address the energy efficiency so not only just the semiconductor device, but also the way that you manage them that can improve efficiency.

Dave Freeman, Texas Instruments

Jim Harrison : Yes, absolutely. Are there any new ICs, Dave that aid that aid that advanced rectification?

Dave Freeman : We, at Texas Instruments, and others have come out with interleaved topologies. Some address the lower power levels transition mode like controllers, and other average current mode controllers address the higher current levels.

And by having interleaved topology you can shed one of the boost circuits at lighter loads and that helps you to get efficiency across the load spectrum,.

John Perzow ( Marketing Director, Analog Devices ): I’d like to add to that. Some of these techniques, particularly in the higher currents ac/dc or isolated dc/dc areas, have been employed in analog techniques and have been reasonably well understood. The issue has been in how practical the implementation is. So one power supply operating at one load versus the same power supply operating another load and changing modes in order to maintain efficiency as Dave said over a wide load spectrum is a very complex design if done in an analog domain. It’s doable, it’s understood, but it’s almost impractical because it’s so complex to implement in an analog fashion.

John Perzow, Analog Devices

So the digital techniques that have been coming out lately from a couple of folks make it possible to employ these sorts of techniques much more quickly and in some cases in a way that adapts to real world conditions. So, I think that’s a major step is showing up in systems over the last year.

Dave Freeman : As far as the controllers go, I think as more digital solutions show up I think there is opportunity. At first when people did interleave PFC everybody was pretty much satisfied with just the two channel but now we’re seeing customers asking for three- and four-channel interleave.

And as John was saying, digital makes those things easily expandable, where if you were going to do it with an analog type of solution sometimes you would have to create another IC. And in some cases with digital you just have a new set of firmware or a new configuration of your state machine.

John Perzow : Exactly. So it’s been done in analog and it can be done discreetly in analog without even a different controller, but it’s a complicated circuit and difficult to design and it has much less reliability, therefore. So some of the new multioutput, five to seven output, digital PWM controllers with pretty solid software approaches are now making these things reasonable.

Lance Zheng (Applications Engineering Manager, NEC Electronics America ): I agree with you guy’s comments about digital control. I think there is a place for a microcontroller to kick in here because the standard MCU can have enough intelligence to manage the power and maximize efficiency.

Lance Zheng, NEC Electronics America

Doug Bailey (VP Marketing, Power Integrations ): One of the things I think we’ve been concentrating on is the overall efficiency of power supplies, the other thing worth mentioning is the standby and light load performance. One of the continued pressures that we’re getting from our customers has been pushed on us by the EUP rules and by the OEMs who are looking for 30 mW of no load power from adapters and small consumer electronics power supplies. So the market there is in how to minimize the quiescent power consumption of a power supply.

Doug Bailey, Power Integrations

Jim Harrison : That’s where the interleave design comes in like John was saying?

Doug Bailey : Well that’s in a power factor correction circuit, but in a small consumer electronics device under 75 w you don’t need it. So something like your DVD players, consumer AV, LCD monitors, printers. Most of these products just don’t have them.

Jim Harrison : What technology are we using to get to that low power in those applications?

Doug Bailey : Well the technology we use is based on on/off control so it’s digital it’s not a digital control loop it works in a digital fashion, but it’s still an analog controller, and the Link Switch products that we have manage 30 mW of no load. It’s used in things that need less than about 10 W.

Dave Freeman : Yes, so in those applications, a 75-W-or-below adapter, there’s some solutions that are multimode so you switch from one type of power arrangement to another maybe it’s a quasi-resonant type of converter to a hard-switch converter at some really low power.

But even when you look at the 90-W adapters where you might have PFC, once you sense the fact that it’s just a light load you may even shed the PFC circuit.

Doug Bailey : Yes, I completely agree on the need for multimode operation. Sometimes the right thing to do is to reduce frequency, sometimes the right thing is to be in a traditional PWM and then other times you’ve got to be in a more on/off control mode. So we do that inside what would traditionally be a PWM product and we sell it as a PWM product, but it actually changes operating modes throughout the load range to maximize efficiency.

Eric Persson : Yes, and I see the same thing even in cases where there aren’t controllers available yet to do that. For example, designs using either ASICs or FPGAs or microcontrollers that implement say a two-switch forward converter that changes mode on the fly to a phase-shifted full bridge. So it’s the same electrical topology, it’s just driven in a different way to optimize efficiency over the load range.

Jim Harrison : It sounds like in the last few years we’ve made some real big strides in improving efficiency in all kinds of applications servers and everything else. Do you agree with that?

Dave Freeman : Yes, a lot of it through necessity. When people start looking at the cost of ownership of equipment like servers, it’s not just what you pay for it in the beginning; it’s what you pay for it for the next five years. The other thing about it is that our customers are starting to pay for better efficiency where maybe if you went back three or four years ago and ask them what their priorities were: It was cost, cost, and cost.

John Perzow : I agree very much with what has been said. I think that a lot of technology has been developed in the last few years to improve system efficiencies. I don’t think a lot of it has been adopted yet. We’re just seeing it ramping now in systems on the market.

Part of it is the cost of ownership and cost of energy. It’s a marketing advantage or becoming a marketing necessity. Efficiency is becoming a requirement to even play in the game anymore. But we’re still seeing different sorts of efficiency standards or benchmarks being used to differentiate products to the end user.

Doug Bailey : Yes, efficiency is always a key selling point, but the commercial reality is that people are only going to end up being as efficient as they need to be. Efficiency costs money because it tends to revolve around decreased RDS(on) or decreased switching losses or something where you’ve really got to spend some money to make that happen. It’s the standby-power use that ends up being more or less free.

John Perzow : I’m going to suggest that in addition to just pure cost, it’s also the overall benefit in terms of system or total energy savings. Looking at the ACEEE study that was done about the middle of 2009, it took a stab at quantifying where some of the gains can be made, and residential standby power for example can decrease power usage by appliances by about 4%.

Dave Freeman : One of the examples in consumer electronics is the microwave where it consumes 5 W most of the time telling you what time it is, which is kind of a shame. But, beyond just the cost of energy, one of the guys from a large southern California data centers was telling me that it’s not necessarily just the cost, sometimes it’s the availability of energy. They were told by their energy provider that they are not going to get any more energy in 2010. So, if they want to grow their business, they’re going to have to use energy more wisely. So this motivates people.

Paul O’Shea : So on to the second question. For solar and wind power applications, are there newer semiconductor and designs that aid the efficiency of inverters required to get them on the ac grid, and are the needs of this application different from the typical ac line to dc converters that we use in other applications?

Eric Persson : With solar, we’re seeing a pretty large demand for solar inverter applications. On the semiconductor side, it’s not that there is anything particularly special just for solar, but those applications are more willing to pay a premium for performance.

Also another requirement in solar is they want long lifetime and high reliability. So they’re also looking at ways to, for example, get rid of electrolytic capacitors because those tend to be the lowest-reliability parts in the whole converter chain. And one of them is operating at frequencies where they can get by without electrolytic caps.

Dave Freeman : I’d just like to emphasize on Eric’s point about the solar inverter, efficiency has been important and is more important than in some other applications. You know we have the 80 Plus standard of gold with efficiencies in the upper 80s range. But, solar inverters technology has been shipping with efficiencies 95% to 97% for quite some time. So it shows in that marketplace efficiency means a lot more than in some others.

John Perzow : Right, and it’s easy to understand why in the adoption of alternative energy systems, solar and wind in particular, the key has been to get the cost per watt down to a competitive point and that’s largely impacted by the efficiency of the system and maximum power point tracking for solar.

In wind power systems, the generator output changes its phase, its frequency, and its voltage as a function of wind speed. In order to put that up on the grid that has to be compensated for intelligently and new techniques for that are being very quickly adopted because of the economics involved.

Lance Zheng : Intelligence is needed in order to maximize the efficiency. The MPP tracking needs intelligence and so do the multiple stages of ac/dc or dc/dc conversion. So we need the MCU to monitor the output so you can deliver maximum energy. And also for safety, diagnostics, and communication that gets feedback to a central server. So processors are playing an important role.

Eric Persson : I wanted to refer back to a comment Dave Freeman made earlier about, the use of multiphase. We see the same thing in the micro-inverter solar market. They tend to use multiphase boost converters and they’ll do phase shedding and various techniques to get maximum efficiency over the whole range of solar input.

And I think there’s an opportunity there in terms of controllers because we’re seeing ASICs and FPGAs right now. I think it’s still a fragmented industry and we’ll soon see more unification of control topologies, I suspect. ■

Paul O’Shea : Okay, another question. The alternative energy sources including solar, wind, thermal, hydrogen, and others have generated a lot of interest recently but could use more uptake by product designers and manufacturers to be a viable source.

What industries like automotive, building trades, and others, do you see as most likely to use one of these alternative energies and what are some of the potential hurdles to developing products for entry to the market?

Dave Freeman : Probably my favorite example of this is what California has done for electric vehicles. Supposedly they put 20 recharge stations between LA and the Bay for electrical vehicles – 70 amp type of outputs and 240 volts. The interesting part of that is the recharge stations are all powered by solar.

Lance Zheng : I just want to comment on the smart grid side. The alternative energy sources are going to be integrated into the grid. I think one important thing to do is to establish a standard for communication. Make sure that the power grid can now get the status of the alternative energy source via two way communication. So I see a challenge for this to be adopted.

John Perzow : I agree. The adoption of communication standards and protocols is key to smart grid and alternative energy adoption.

Paul O’Shea : The next question. What incentives, government or otherwise, are in place to increase the adoption rate of alternative energy sources and what more can be done?

Eric Persson : You know, there is a lot of discussion right now about co-generation pricing. You get – whatever energy rate you pay – let’s say it’s 10 cents per KWh, that’s the same rate you’re paid for cogenerating back into the utility. Whereas in Germany, it’s maybe four or five times the rate.

Building new power plants is extremely expensive and if we can offset the need for extra capacity by adding distributed capacity this way, it’s worth it.

Dave Freeman : I would definitely like to see, it’s called a feed in tariff. That would probably be the best motivation.

John Perzow : I would concur very much with that.

Lance Zheng : If the utility company or government provides a payback period, that would definitely help out with the technology adoption.

John Perzow : There have been discussions of taxing fossil fuel and the reasons for that are obvious and Europe has done it for a while. And to put a tax on the use of that to compensate for the long term cost of using fossil fuels, as unpopular as that is, strikes me as at least something to debate when alternative energy sources really are viable – economically viable.

Jim Harrison : Well it’s also the consumer rules here. Energy Star of course is voluntary, but, at least the energy usage should be marked on everything. When I buy anything electric I should know how many watts it takes and I don’t understand why that has not been enacted.

John Perzow : It is a bit hard to measure because of the user profile of the television sets, etc.

Jim Harrison : It is complicated but there is a way, John, of getting a unified number.

John Perzow : I agree. I think California’s new rules are mandated, they’re not even voluntary anymore. If you want to sell a television in California you have to hit these new standards.

Yeah I agree. I think that they should be. It is possible to measure it. You can just normalize it. They can and should do that – they don’t.

Jim Harrison : Everything that plugs into a wall should have its wattage on the box and on the on the shelf wherever they buy it from – if they actually buy it on the shelf anymore these days.

Jim Harrison : Another topic. Are we continuing to develop improved methods and semiconductors for efficiently driving motors in home appliances and industry systems, or have we sort of reached a plateau there. Are we done with that?

Eric Persson : This is an area where IR has been working on appliance controls and the inverter power stages for at least ten years. And, as with all of the previous discussions we’ve had on this, economics play a large role in this.

Let’s take the control side where sensorless control algorithms and vector control and methods for optimizing the delivery of energy, to run a motor at its peak efficiency under all operating conditions. The technology hasn’t changed all that much but how it’s implemented and the cost of the processing horsepower to be able to do that has gone down dramatically.

A survey in the appliance trade magazine talked about the average consumer faced with say having to spend $200 more on a refrigerator knowing that it would pay back that amount in less than two years would not opt to do that. They want the lowest price right now. We can make very efficient motors and drives. It’s just that consumers don’t seem willing to pay for that right now.

Dave Freeman : Consumers look for payback, right? If you look at replacing your air conditioning system or something like that, that actually is a pretty long payback. But if you look at some of the simple things that you can do, say, with blower motors, if you go with electronically commutated motors, variable speed, you’ll pay back that motor within a year where replacing an air conditioner system or compressor system, it may be 10 or 15 years to pay that back. But I think consumers are not educated enough to actually know that type of thing.

Dave Freeman : Yes, right. When you first turn on your air conditioner your blower motor runs at the same speed even though the cooling coils aren’t cool yet. So that’s not really doing anything for you. So if you can close that loop and the electronics to close the loop are getting cheaper and cheaper each day.

Eric Persson : You’re exactly right. The other side of that is the compressor motor. For HVAC, having both a variable speed compressor and variable speed blower on the evaporator and condenser coils, that’s kind of the ultimate in efficiency we can get and that technology is available right now. There’s no reason manufacturers can’t adopt that and deliver it.

What I see going on is that the cost of these drives is actually very reasonable, but the way that they’re positioned and branded, and sold, that’s where the margin is made by the OEMs is they want to extract as much money as possible and make it higher priced. And that’s something where I think government and private industry collaboration could help drive a much greater adoption of these and save us all energy.

John Perzow : The technology exists. ADI has done a lot in the last two years with DSP solutions for motor control. Some of the newer, really fancy HVACs might have a three step motor. But there’s no system logic to close the loop, to measure the coil temperature, or to determine how close you are to a set point. You could do a lot of smart things pretty easily.

And people aren’t doing that and I don’t think it’s a cost issue. I do think it’s the fact that most appliances designing and engineering teams are fairly conservative or traditional. They haven’t changed much and they’re not likely to change much. Lance Zheng: I just want to put a comment on adoption rate. I think the motors inside the home appliance are not equal. Some are easy to drive, some are not. A blower is quite easy to drive. That’s why the adoption rate is pretty high. But something like an HVAC compressor, it’s more complicated to drive. So I think it’s very important to come up with some proven and reliable software to drive those motors.

John Perzow : I would venture that the barrier to adoption here is either the comfort or the competency of the system manufacturers. I think the technology is pretty well proven and exists. I would say it’s, you know, firmware and so forth.

Jim Harrison : In some cases though we have higher cost of the motor itself, I think.

Dave Freeman : Yes, it’s certainly a higher priced motor and things like that. But I think some of these designs are not looking at a systems solution.

Eric Persson : I completely agree with Dave on this. In order to move forward and make the overall system more cost effective you need three disciplines. You need the machine design side of it, you need the electronics, and you also need the control algorithm. All three of those, you could have three different people all trying to optimize something but they have to work together as a team to solve, you know, the common problem.

John Perzow : I think there is a parallel here in the infrastructure business, interestingly. In say telecom or Ethernet routers where you have large systems, a lot of times the main power supplies are outsourced, specified but usually outsourced. But if the power supply and the system were looked at as one system and there was communication to the processors it could operate more efficiently.

Lance Zheng : I think we have to adapt our solution to the existing infrastructure. We want to change the behavior of the motor designer, it’s tough. So, you might just want to take whatever you have and try to drive it efficiently.

John Perzow : Yeah or, you know, getting the folks that specify the motors or getting the folks that specify the front end power supplies to look at it and require system higher levels of system integration. That’s been a tough – that’s a tough log.

Doug Bailey : I think the observations are very accurate about difficulty in getting a system performance when different people bring different parts of the system together.

Paul O’Shea : Another question: looking forward five to ten years, what application area like automotive, smart grid, AC DC transmission, solar generation, wind generation, LEDs, other lighting, industrial pollution control, home networking, and you can name anything else too, what applications areas do you think will have the most success in environmental improvement?

Dave Freeman : This is Dave Freeman with TI so I’ll tell you my favorite one and it may sound like Mama and apple pie stuff, but it’s actually automotive. If we could get behind a truly practical electric vehicle and not leave it to the novelties of what other people doing this and you look at having synergy.

You would create better motors, right, better batteries, and such. And all of the sudden now we can get off this oil thing that we’re on and it makes one heck of a big difference. So when Kennedy says okay we’re going to go to the moon. Look at all the benefits we got from that challenge. I think that same type of thing would exist from creating a truly practical electric vehicle.

It’s actually a very significant challenge. I don’t think there is a truly practical one right now. I think it’s got a long ways to go. But I think if we get behind that and then take all the components that would be in that car and then apply it to everything else.

Eric Persson : I’m on the same wavelength. I think automotive is the one that will make the biggest environmental improvement simply because power plants and all are reasonably efficient. They may be as high as 80% efficient or something, on that order. But cars are 20% efficient right now running on fossil fuels on a good day.

Paul O’Shea : So both of you are are looking at it from an electric vehicle as opposed to some of the other possibilities for vehicles.

Eric Persson : In my view hybrid, traditional hybrid, I mean, what we have now, has come a long way, but moving towards a more electric car the topology like the Chevy Volt where it’s really an electric car with a stand-by generator that kicks in for long distance would make a difference.

Jim Harrison : Yeah well practical at $40,000 for a Chevy Volt is going to be a problem, I think.

Dave Freeman : Boy practical is the key word.

Paul O’Shea : Yeah, well and that’s where incentives come in from the government.

Dave Freeman : Yeah initially, right? I mean it’s really got to stand on itself to be called practical. And so I think but if we got really behind it it’s probably no more complex than going to the moon. But and we did that in a certain number of years. I think we can get behind it.

And look at the problems we would solve getting off the petroleum bandwagon. All of a sudden it takes funding away from countries that are not necessarily our friends. There’s just so many impacts from doing this. I don’t know why if Obama wants to spend $1 trillion, do it there.

Doug Bailey : I guess our view would be that LED lighting is going to make a big, big difference and the reason being is it’s a technology that can improve in very small steps with individual decision making in the marketplace. So all you have to do is make the technology at the right kind of price and you get a fast improvement in power consumption in what is used everywhere.

Jim Harrison : So it’s easily integrated is what you’re saying?

Doug Bailey : Yeah, the technology — basically it’s just replacing a technology with a lower power consuming one.. I agree that the automotive market offers massive improvements and wonderful things about getting rid of hydrocarbon fuels.

My only concern is trying to support the infrastructure build out required to support a very large number of electric vehicles in an urban environment might make a government agency a little bit reluctant to embrace that quickly. I mean, imagine the power requirement on the grid from everybody plugging a car in at 5:00 pm when they get home from work or 6:00 or 7:00 or Silicon Valley 9:00 pm when they get home from work.

Dave Freeman : That’s the reason we need the smart grid.

Doug Bailey : Yes, certainly it would have to built around other technologies as well.

Dave Freeman : Yes, one more incentive.

Lance Zheng : I would pick smart grid. Smart grid is a wide range of things, right, including plug-in electrical vehicle and lighting and indoor lighting as well as home appliances. So I talk about demand response smart grid, right? So you can shut down some of the inefficient plants back up plants if you can achieve a smart response.

John Perzow : I really like the focus on electric vehicle and transportation and all of the benefits to society and cross technologies that would result in. In terms of energy savings, total BTUs, it would probably be the number two possible impact.

The number one area according to the DOE is improving the efficiency of distributing energy, electricity in particular. So a lot of energy is lost before it’s ever consumed and just in the distribution transmission of electricity. So smart grid initiatives attempt to improve that and address that. In terms of the benefit to society or fossil fuel usage or decreasing investments in generation, that would be a major impact.

Paul O’Shea : Do you think that can happen in five to ten years?

John Perzow : I’m out of my depth here. Probably not, it’s such a huge change. Our grid system at least in North America is so antiquated, I mean, worldwide it’s really incredible, it’s so 19th century. So no, it took 100 years to build it up, I don’t know how long it would take to fix it but not five or ten. But unfortunately that can get used as an excuse to delay the investment in that area. It’s essential that we invest in that area too.

Lance Zheng : I can provide a number. The transmission loss is about 8% according to one report. Part of the smart grid is the communication technology. So once the appliance inside the home is talking to each other and smart meter talk back to the utility you can do all kinds of things. And I’ve been seeing smart meters being deployed in new homes here in California.

John Perzow : The smart meter is nice in that it helps the consumer monitor things and make decisions on consumption and it also helps troubleshoot and diagnose transmission issues. But the simple conduction losses over the line is a problem that we have to fix.

Paul O’Shea : John if you could go to micro-grids where you have distributed power, if you could get that transmission distance down and put energy sources closer to the point of load then that goes a long way, and especially if you can manage them.

Jim Harrison : Yeah I think that’s the key. The big advantage for the smart grid really is in leveling the power because you have to make these coal fired plants big enough to handle the peak load. If you can pull the peak down and level it out a little bit, you save a lot in pollution.

Lance Zheng : That’s right, that’s right. Actually there are a lot of backup plants that are standing by, for a peak. Those are not efficient plants. If you can reduce the number of those inefficient backup plants you’re going to reduce the impact a lot.

Jim Harrison : I think the distributed power is a really big potential as well.

Lance Zheng : That’s right, yeah. I actually read an article China has been spending tons of money on the ultra-high wattage transmission. I heard they are considering their strategy. They see the smart grid initiative here in North America. They feel like the smart grid might be a little bit better than the ultra high wattage transmission.

Lance Zheng : I want to talk about LED lighting as well. We are seeing a lot of customers starting to retrofit LEDs to existing infrastructure, for example street lights with power line communication and lights that work with dimmers. So we see that happening. You know, the lighting itself consumes probably 20% of the entire energy usage. And an LED lamp can be five times more efficient than incandescent bulb, right?

Jim Harrison : Yeah of course the only thing there is that the LED lamps are just now getting to the same efficiency as the tube fluorescent lamps.

Lance Zheng : That’s true, but I would argue down the road the LED component the cost is going down. And the performance of LED is going up. You have color rendering.

Dave Freeman : One of the things that everybody always compares LED to incandescent but most buildings, they’re not incandescent, they’re the long tube fluorescent so that efficiency is already there. And, if you look at the number of fluorescent lamps, other than the decorative lighting, that most people have already done. So the real challenge for LED is to replace fluorescent.

John Perzow : Absolutely. Yeah and there is a lot of opportunity for things like changing color temperature and mood in addition to efficiency that LEDs bring. But they’re going to have to compete on efficiency with fluorescents.

Jim Harrison : And price.

Lance Zheng: I just want to comment I do not doubt the efficiency of fluorescent tube and the infrastructure. But in California we have title 22 or 24. One is for efficiency, the other one is for hazardous materials. So with that kind of regulation in place I see LED lighting technology going to be deployed quickly.

Jim Harrison : That’s a good point, and the reliability of course is better than fluorescent. And the compact fluorescents, which are what we use in the home, are not nearly as efficient as the tubes. They’re about half the efficiency.

Dave Freeman : And not as reliable.

Eric Persson : Definitely not.

Dave Freeman : Yeah, there are bugs to be worked out.

Doug Bailey : Also pure efficiency comparisons can be a little misleading because one of the nice things about LED lighting is the directability of the light source. The efficacy of the lighting is 100%. With fluorescents there is a certain amount of power loss in light going in all directions and has to be redirected.

So starting with street lighting, one of the reasons that LEDs is preferred, the actual light output efficiency isn’t particularly exciting. But nearly all the light is used.

Dave Freeman : So, there was an interesting thing on the news that I never really considered and I guess people that installed LED traffic lights didn’t think about is that in because we’re coming up on wintertime and stuff like that. So when you get snow blown into your traffic light the old incandescent lamps used to melt it. And with the new LEDs…

Jim Harrison : Now we’re going to have to put a heater in the damn light!

Paul O’Shea : Make it be less efficient.

Dave Freeman : That’s right.

Jim Harrison : Yeah, okay what else? Thinking about the next five to ten years, that would be really significant, technology wise.

John Perzow: I mean, if these three things we talked about were actually invested in to the degree appropriate, I mean 80% of the problem is solved.

Paul O’Shea : That’s what you need.

Dave Freeman : So one of the things that we talk about around here is if you really could have practical communication, where most peoples office environment would work from home. And you really got that picture phone working or the images working out and we could adopt that way of communicating, then that is another thing that can influence our life and help save energy in that you wouldn’t have to necessarily leave your house to get your job done.

Jim Harrison : It could make a big difference, you’re absolutely right.

Paul O’Shea : And it’s funny that you would mention that because we were just talking about that, we editors, about picture phones and camera phones. And yeah, they’re not quite there yet.

Dave Freeman : No they’re not.

Dave Freeman : So when we’re talking about renewable energy technologies, really what makes them very attractive is the combination where the load requirement goes down, with higher efficiency. The combination really makes things practical.

But most of the time people really think about solar panels offsetting the electrical cost and stuff like that when really it’s that combination.

Jim Harrison : So are the computer power supplies, is that a big business, or is industry sort of ignoring that right now?

John Perzow : It is big. We didn’t go and quantify this stuff, it has been done. The ACEEE report does take a stab at quantifying the returns based on potential improvements that can be made. And infrastructure in general, computing and communications infrastructure is a huge one. It’s way up there.

Jim Harrison : I’ve often wondered whether the communication structure takes a bunch of energy. Let’s just say the Ethernet overall, right, there’s 10 zillion ports of Ethernet running around right now, 24/7. If we lower the energy there, is that significant or not?

John Perzow : It’s going the other way with POE.

Dave Freeman : Yeah it’s going the other way. I mean, even you look at wireless access points. You know, I think going from G to N doesn’t make it more energy efficient. I mean, you try to get the data rate up, you try to get the distance up, and that takes more energy.

John Perzow : Yeah and POE it is running whatever, take a number, 54 volts DC on little wires and the limitation there is the wires get hot. I mean, really wasteful. It’s very practical, it’s very convenient, but it’s not going the right direction in terms of improving the efficiency of the infrastructure.

Paul O’Shea : So that would be another area that we could improve on then.

Dave Freeman : Yeah, actually we’ve talked about looking at the protocol in the type of data and the way that you do the data across wired devices. Maybe it’s time to look at it again and kind of come up with an energy efficient method of TCIP. Because at first we just did it so that it was practical, something you could implement with the circuits you have. But maybe it’s time to start looking at more energy efficient things.

John Perzow : Like optical.

Dave Freeman : It could be optical, it could be a different way of modulating things. You know, just look and see what you can do.

Jim Harrison : Okay, I think it’s about time to wrap up. Any final comments anybody?

Dave Freeman : No, I think it was a good conversation and brought up some good points. You know, if we could just had that level of influence from this to get more people active in this area, I think that would be great.