Post by thread:[OT] 240-volt Power

My parents recently visited some friends who live on an island in the Caribbean which is near Granada. I'd happily mention the name, but I don't remember how it is spelled. The island's electricity is 50 HZ and 240 volts reflecting its British experience, but it got me to wondering. Are fluorescent tubes made for either the 120-volt or 240-volt markets or is all that taken care of in the ballast? Actually my father didn't see much fluorescent lighting where they were but he did notice that incandescent reading lamps in the house seemed to die like flies. I would imagine that good 240-volt bulbs running at the proper voltage probably don't burn out any faster than 120-volt bulbs. The voltage there does fluctuate quite a bit and the bulbs may not be the best money can buy. They do not have any voltage or wattage rating on their base or the glass. It seems like my wife and I bought some bargain-basement 120-volt bulbs for general use and they generally had much shorter life-spans than do the only slightly-more-expensive bulbs with actual names on them.:-) Martin McCormick WB5AGZ Stillwater, OK OSU Information Technology Division Network Operations Group

> I would imagine that good 240-volt bulbs running at the proper > voltage probably don't burn out any faster than 120-volt bulbs First you have to define and find a "good" 240V bulb ! They don't last as long as 120V bulbs. The filament is thinner. Someone did say here once that they used 2 x 120V in series and got pretty good economy. Also, adding a series resistor can increase bulb life significantly http://home.clear.net.nz./pages/joecolquitt/0bulblife.html You'll probably find that most bulbs blow at turn on, because the filament is cold and takes an inrush of current greater than its normal operating current. Particularly if you happen to turn it on at the peak of the mains cycle and the filament is getting old and depleted. So the more often a light is switched on the greater the chance of it blowing As colour degrades with decreasing voltage, this approach might suit some positions better than others ; reading lamp, no - porch light, yes A more sophisticated approach would be soft-start. A zero- cross detecting micro and triac for example to apply power in a controlled fashion rather than a random wallop I'm pleased to report that no bulbs have blown since the last update to that page (June 2002), so that's around 4 years now. Previously, as mentioned, they would blow with annoying regularity

Jinx writes: >First you have to define and find a "good" 240V bulb ! > >They don't last as long as 120V bulbs. The filament is thinner. >Someone did say here once that they used 2 x 120V in series >and got pretty good economy. Also, adding a series resistor >can increase bulb life significantly Well, I had to think a bit about why the manufacturer couldn't simply roll in more tungston wire of the same gauge as the 120-volt bulbs, but it really wouldn't work unless someone figured out how to formulate a higher-resistance tungston wire that was the same gauge as that used in the 120-volt bulbs. A resistor causes the bulb to light more gently because it initially takes the in-rush of the filament current when it is cold. If you put two bulbs in series, they will reach operating temperature faster since they at least theoretically will always have half the supply voltage across them. Martin McCormick

> Have not tried this but someone once said by putting a diode in > series > with the lamp but as this would be half wave rectification I would > think > a full diode bridge would be better. > > But would this help with the turn on burn out problem? > > Maybe a full wave bridge and a capacitor on the DC side would help > to > slow down the turn on pulse to the bulb while the capacitor is > charging. A slow turn on device could be achieved with 2 MOSFETS, a diode, a resistor, a capacitor and a zener. Given the fact that energy saving fluorescent bulbs are selling here for as little as $US2 and they have 2 suitable FETs inside, the cost of such a device in zillions should be able to be well under $US2. Whether it is worth doing is debateable unless you greatly favour incandescent bulbs over fluorescent power savers. If you have a very expensive incandescent bulb (projector etc) then it may well be worthwhile. Two MOSFETS connected Gate to Gate. Source to Source. Drains are input / output. Wire in series with bulb. Capacitor gate to source. Diode in series with R drain to gate. Zener gate to source , cathode to source. Add a second resistor gate to ground for turnoff control A single MOSFET would start up in half wave and switch to full AC - this MAY be enough. RM

{Quote hidden}>-----Original Message----- >From: .....piclist-bouncesKILLspam@spam@mit.edu [piclist-bouncesKILLspammit.edu] >Sent: 08 February 2005 08:59 >To: Microcontroller discussion list - Public. >Subject: Re: [OT] 240-volt Power > > >>A slow turn on device could be achieved with 2 MOSFETS, a diode, a >>resistor, a capacitor and a zener. Given the fact that energy saving >>fluorescent bulbs are selling here for as little as $US2 > >Gee, you are doing well, in Rip Off Britain the best is still >several pounds (but they are down into the single digit pounds mostly). > >I do have one that died, and I believe the semi-conductors >have gone the way it took out the circuit breaker. Seems to >use standard IRF devices inside, so hopefully will get round >to doing something with it one day, or use the tube on another >one when its tube dies (currently the more likely with the >other projects I have). Alan, If you have an Ikea store nearby they sell "own brand" CF's for around £1.50. I have a few in my house, they seem to work fine although they seem to take longer to reach full brightness than some of the brand names ones I have. Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. =======================================================================

>-----Original Message----- >From: piclist-bouncesspam_OUTmit.edu [@spam@piclist-bouncesKILLspammit.edu] >Sent: 08 February 2005 17:57 >To: 'Microcontroller discussion list - Public.' >Subject: RE: [OT] 240-volt Power > > >The point was to slow down the inrush current to the lamp. With just a full bridge and a capacitor, the only thing restricting the inrush current would be the impedance of the rectifiers and mains wiring. You need some kind of resistive element to reduce inrush current. An NTC thermistor would be ideal as it has a high initial resistance to reduce inrush, and as it heats up it's resistance reduces so the lamp gets brighter. However, getting hold of a suitably rated thermistor might be a problem. Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. =======================================================================

----- Original Message ----- From: "Jinx" Subject: Re: [OT] 240-volt Power > Also, adding a series resistor can increase bulb life significantly > http://home.clear.net.nz./pages/joecolquitt/0bulblife.html Jinx, that is brilliant, thanks for sharing. I have been thinking along similar lines but using an auto transformer to drop the supply to the whole lighting circuit by a similar amount. The series resistor is such a simple low cost idea that it would be almost daft to not try it. Can I ask you - where did you get the graph at the bottom of the page? Google brings up very little that looks like data sheets, and I would be interested in a copy. Mike. -- Outgoing mail certified virus free. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 265.8.6 - Release Date: 07/02/2005

> And an opinion piece by the editor of Silicon Chip > "Compact fluorescent lights are not economic" > > http://www.siliconchip.com.au/cms/A_102164/article.html That was written in 2004. I'm surprised. He's wrong. Or, at least, my overwhelming experience of CFLs is quite different from his. I can categorically state that I have had some versions last well over the rated 8000 hours as I have a lamp my electrical workshop and another in my front path light that are always on, and the bulbs last typically well over a year. I write the date that I put CFL bulbs into service on their base. I have had some fail long before expected, but overall the results are acceptable, if not as good as some claim. Given the current low cost of CFLs and the substantial savings in energy use, it doesn't take long to justify their cost on energy savings alone. Note that CFLs save money on energy ONLY if you do not have thermostat controlled heating. Any time you use a thermostatically controlled heater energy sources such as light bulbs contribute to the heating and reduce heating power costs proportionately. Simple example of savings. Compare an incandescent 100W bulb and a 25W CFL. (usually 20W are sold as 100w replacements but they don't quite meet 100W equivalent light output so lets be fair). Electricity costs about $NZ1 per watt per years continuous operation. That varies with area, and the price is rising, but it's an OK approximation. If an IB (incandescent bulb) and CFL are assumed to cost equal amounts of capital per hour run then power savings are all that needs to be compared. Even factoring in bulb costs does not make much difference. Energy savings per year run 24/7 are 100-25 = $75 per year. Run a bulb 10% of the time = 2.4 hours per day and the saving is $7.50. As CFL bulbs now often cost under $3 and IBs are down to say $0.50 there MAY be a slight extra bulb cost for the CFLs. But its far less than $7.50/year. I run an often or always on CFL in several locations. Front porch, path light, workshop. There are a few other areas where they are often on. At $1/year/watt and with a small wattage CFL the costs are acceptable. I would not consider doing this with IBs. I just went and checked the ALWAYS on front path light. It was installed on 22 September 2003. That's about 11.700 continuous running hours so far. It's an 8 watt bulb so running costs so far are about $11. ($11.22 at 12 cents per unit.) My workshop bulb was installed in 15 October 2003 after having been moved there from some other (unspecified) location so total hours are unknown but at least as long as the path light. . A significant failure mode is mechanical damage due to careless handling. It's far easier to crack the bulb seal on a CFL than on an IB. Due care is required, There are few places where I would recommend an IB these days. RM

Mike Hawkshaw writes: > ----- Original Message ----- > From: "Jinx" > Subject: Re: [OT] 240-volt Power > > Also, adding a series resistor can increase bulb life significantly > > home.clear.net.nz./pages/joecolquitt/0bulblife.html > > Jinx, that is brilliant, thanks for sharing. I have been thinking along > similar lines but using an auto transformer to drop the supply to the whole > lighting circuit by a similar amount. The series resistor is such a simple > low cost idea that it would be almost daft to not try it. Unfortunately, this is not without it's downside. Not only does it reduce the brightness of hte bulb, but it reduces the *efficiency* of the bulb as well -- the filaments are more efficient at converting electricity into visible light at higher temperatures. While this may be a penny-wise solution in terms of bulb cost, it could very well be pound-foolish in terms of electricity cost. -p.

Russell McMahon writes: > >And an opinion piece by the editor of Silicon Chip > >"Compact fluorescent lights are not economic" > > > >www.siliconchip.com.au/cms/A_102164/article.html > > That was written in 2004. I'm surprised. He's wrong. > Or, at least, my overwhelming experience of CFLs is quite different > from his. My results have been much more mixed. Some of the first CF bulbs I purchased are still working great. Based on my initial good experiences, a bought a six-pack of bulbs "on special sale" to use throughout the house. Two died within the first month. A third started emitting a nasty odor. A fourth only operates at half brightness. I don't know what has happened to the other two. There isn't anything wrong with the technology; there is just a lot of crap out there. If you see a deal of CFs that looks too good to be true, it very well may be... -p.

How about using an NTC thermistor, just like the inrush current limiter in low cost switching supplies? These low cost switchers have a problem when you get a power interruption for less than a second, as the filter capacitor discharges but the thermistor does not cool off. When the power comes back, there is no inrush current limiter. I've had a bunch of rectifiers blow out due to that. In those cases, I just stuck a resistor in series with the input to the supply, as has been suggested here in series with an incandescent lamp. However, the lack of inrush current limiting with the incandescent is not as serious as it is for the switching power supply (we're back to the way the lamp has always been run). Another approach to limiting inrush current on incandescent lamps is to use a consumer type triac dimmer that looks like a normal wall switch. These dimmers are off when the handle is down, on when the handle is up, and 50% when the handle is horizontal. If you just "flip the switch on," the lamp has a little time to warm up as the handle is brought up, limiting the inrush current. You also get the benefit of having a dimmer! Several years ago I did current measurements on flashing broadcast tower lights and captured the current waveform. These flashing beacons have a pair of 620 watt lamps in each beacon. The inrush current is amazing! I just measured the cold filament resistance on one of the lamps. I get about 1.8 ohms. So, at 120VAC, the inrush would be 67 amps. Once warmed up, the current is 5.2 amps. When flashing on a tower, the lamps do not cool off all the way between flashes, so there is inrush current, but it's not as bad as it could be. Harold -- FCC Rules Updated Daily at http://www.hallikainen.com

>> > http://www.siliconchip.com.au/cms/A_102164/article.html >>... my overwhelming experience of CFLs is quite different >> from his. > My results have been much more mixed. Some of the first CF bulbs I > purchased are still working great. Based on my initial good > experiences, a bought a six-pack of bulbs "on special sale" to use > throughout the house. Two died within the first month. A third > started emitting a nasty odor. A fourth only operates at half > brightness. I don't know what has happened to the other two. > > There isn't anything wrong with the technology; there is just a lot > of > crap out there. If you see a deal of CFs that looks too good to be > true, it very well may be... Even the el cheapo ones have mostly worked OK for me. When prices are about the same I try to buy name brand ones. I find Philips & Osram have been OK. They may all come out of the same factories but name brand sellers have a reputation to protect and are perhaps more likely to keep quality reasonable. Comparison with older and newer ones shows a reduction in parts count and general complexity and cutting corners. eg power devices seem to be of smaller physical size with about no attention to heat dissipation. RM

>>>> http://www.siliconchip.com.au/cms/A_102164/article.html >>> ... my overwhelming experience of CFLs is quite different from his. >> My results have been much more mixed. >> >> There isn't anything wrong with the technology; there is just a lot >> of crap out there. > Even the el cheapo ones have mostly worked OK for me. > When prices are about the same I try to buy name brand ones. I've had very mixed results. The commercial grade CFLs ($8-10 each) have run without a problem. The cheapos (4 pack for $15) have dropped like flies (4 dead out of 5 in less than a year). However, we don't run any lighting 24 x 7. Maximum continuous on time for any location is ~8 hours per day. Switching events seem to take a toll. Since I switch lighting off to conserve energy (can't beat 0KWh), I'm not very enamored of the slow light output ramp-up (instant on bulb is well under half the rated light output at first and takes 1-2 minutes to reach full brightness (I've measured it) ). One dead one was an inexpensive 3-way CFL. I haven't been able to find a replacement. While looking for same, I did find a recall notice for the 3-way CFL units due to dangerous failure mode (fire hazard, I think). So the corner cutting has been very tight -- some might say, too tight. Here in southwest US, I haven't seen name brands at anywhere near the same price as cheap ones. By my estimates, price break-even point due to energy saving is only a few months. A couple weeks ago, I was looking for a pair of CFLs to put in some roof fixtures. Access is via ladder with a long stretch out to get to the bulbs. Old 75W reflector flood incandescents would last for years. New CFLs saved energy. But I do not want to change them every 6 months. Previous unit was cheapo CFL. I went with better commercial grade units this time. I did compare in the store. Mechanical construction of more expensive units (rivits) was better than cheapos (crimped metal). Only time will tell how reliable the new units are. Short CFL lifetimes may be giving them a bad reputation with the general public. I doubt that most people do power consumption calculations. All they see is the more expensive CFL dying at the same rate (or faster) than the older incandescent bulbs. Lee

>> Any time you use a >> thermostatically controlled heater energy sources such as light >> bulbs >> contribute to the heating and reduce heating power costs >> proportionately. > But you'd still have to figure into account the efficiency of > generating > electricity, compared to the efficiency of generating house heating. > In > my country the latter is ~ 90% using high-efficiency gas heaters. > The > former is much lower. True. This will vary with country and, as far as the consumer is concerned, the cost of the energy sources. In NZ a large amount of our domestic heating is by mains electricity. A large proportion of our electricity is from hydroelectric sources. A limited number of coal and oil fired stations exist. Reticulated LPG gas (mainly from gas fields off the west coast of our North Island) is substantially cheaper than electricity but not available in all areas. Bottled LPG gas is available nationwide but is dearer per energy content than mains electricity. Then, there's always gas lights ... :-) RM

>-----Original Message----- >From: KILLspampiclist-bouncesKILLspammit.edu [RemoveMEpiclist-bouncesTakeThisOuTmit.edu] >Sent: 08 February 2005 23:38 >To: Microcontroller discussion list - Public. >Subject: Re: [OT] 240-volt Power > >Unfortunately, this is not without it's downside. Not only >does it reduce the brightness of hte bulb, but it reduces the >*efficiency* of the bulb as well -- the filaments are more >efficient at converting electricity into visible light at >higher temperatures. > >While this may be a penny-wise solution in terms of bulb cost, >it could very well be pound-foolish in terms of electricity cost. Only if you add more lamps to compensate for the decreased brightness, which wasn't the idea of the scheme. As Joes web page shows, dropping the voltage by ~5% significantly increases lamp life whilst still giving acceptable light output. Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. =======================================================================

Russell, On Wed, 09 Feb 2005 12:10:17 +1300, Russell McMahon wrote: >...< > Note that CFLs save money on energy ONLY if you do not > have thermostat controlled heating. Any time you use a > thermostatically controlled heater energy sources such as light bulbs > contribute to the heating and reduce heating power costs > proportionately. You're making some hefty assumptions here: 1. That heating is electrical, so costs the same regardless of the device supplying the heat. Mains-fed gas heating is usually much cheaper (ironic, since electrical heating is 100% efficient in itself - all of the energy turns to heat, which is what you want! :-) 2. That heating is always used (whenever there are lights on), ignoring the time when the temperature is fine without heating, and that air-conditioning isn't used. As soon as you add air-con to the equation it turns round dramatically, because you are paying for the electricity to create the heat, then paying again to extract it from the building. I realise you're from a similar climate to me so heating is what you think of, but if you look at somewhere like Southern California the time of maximum electrical demand is high summer due to air-con. > Electricity costs about $NZ1 per watt per years continuous operation. Good grief, that's cheap! Can you send some to me and we'll split the difference? :-) I'm on a split meter, 7 hours overnight "cheap" at 3.09p/kWh, 17 hours at 12.84p/kWh. That comes to 7.895p + 79.67p = UK£0.876 per watt per year or NZ$2.32 or US$1.63 at today's exchange rates (this is without tax, and without volume discounts). If I was on a single rate meter it would be dearer, at UK£0.95 (NZ$2.53, US$1.77). I hadn't realised we were being fleeced quite so much! As for the CFL discussion, I don't have any lights on 24/7, but several on for 8 to 12 hours a day, depending on the time of year. I've been doing this for some years, since the original Philips units came out, that had a tough glass cover over the tubes (and they weighed a lot!). Those lasted for about 4-5 years. Being in a cold environment (garage, unheated room) seemed to cause them problems, with failure to strike being a common failure mode (flicking the switch repeatedly would usually get them going) . The more modern ones with exposed tubes, and some with a decorative globe cover, seem to last at least as long. I must start recording when I start using each one, so I'll know for sure! Prices of CFLs vary dramatically here, from about 2.5 watt-years' cost to about 4 times that, with no obvious reason for the difference. Cheers, Howard Winter St.Albans, England

From: "Peter Johansson" Subject: Re: [OT] 240-volt Power > Unfortunately, this is not without it's downside. Not only does it > reduce the brightness of hte bulb, but it reduces the *efficiency* of > the bulb as well -- the filaments are more efficient at converting > electricity into visible light at higher temperatures. > > While this may be a penny-wise solution in terms of bulb cost, it > could very well be pound-foolish in terms of electricity cost. True, but the problem lights at our house are for mood lighting, and have 25W bulbs in them. These things are wall mounted and have small bayonet type holders - the fittings are rated for 40W bulbs. The fittings are mounted such that the lamp is connector side down, yet the bulbs typically last 4 months. I wouldn't want to use CFLs in these lights as the quality of the light is too harsh and the lamps would stick out from the top of the fitting, so reducing the voltage to the bulbs is really the only way to go, and will not impact on electricity useage at all. Interestingly, another direction this thread took was CFL life span. We use about 8 of these where I work, to provide just enough light to navigate round the building at night with the normal office lights off. We find that they typically last between 18 and 24 months (13,000 -> 18,000 hrs) however they are never ever switched off. The lamps end up very black just before they fail though! This probably saves money in the long run, as it causes less wear and tear on the office lights when ever someone gets called in to a fault. Mike. -- Outgoing mail certified virus free. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 265.8.6 - Release Date: 07/02/2005

From: "Jinx" Subject: Re: [OT] 240-volt Power > Honestly don't remember. Not to worry it gives the right idea! > Here's a graph showing CFL vs filament > http://www.susdesign.com/cfl/cfl-spreadsheet-1.xls > > And an opinion piece by the editor of Silicon Chip > > "Compact fluorescent lights are not economic" > http://www.siliconchip.com.au/cms/A_102164/article.html Both these are very interesting. We have a mixture of CFLs and fillaments at home for reasons of mood / convienience / efficiency. I posted another reply with my experiences of CFLs at work. We get the cheepest we can find, and get about 2 years continuous life (24/7). From this, I suggest that the electronics probably fails long before the tubes if the second article is correct. Cheers....Mike. -- Outgoing mail certified virus free. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 265.8.6 - Release Date: 07/02/2005

Thanks for the link...I'm playing around in HiCalc...and on the efficiency graph, it seems that if I supply my lamp with more than 100% of it's rated voltage...it becomes more than 100% efficient? The only think I can think of is that they are referencing it back to the rated output...so with higher voltage, you get a higher than rated output, so they say it's more efficient? That's a bit confusing...I can't think of a better way to describe what I'm thinking though. Ideas? Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams On Wed, 9 Feb 2005 17:35:32 -0000, Mike Hawkshaw <spamBeGonemikespamBeGonespikey-mike.com> wrote: > I've just found this software on GE's web site. It plots graphs and > calculates life etc given lamp and socket voltage. > www.gelighting.com/apo/specoem/hicalc.html

I wrote: > >Unfortunately, this is not without it's downside. Not only > >does it reduce the brightness of hte bulb, but it reduces the > >*efficiency* of the bulb as well -- the filaments are more > >efficient at converting electricity into visible light at > >higher temperatures. > > > >While this may be a penny-wise solution in terms of bulb cost, > >it could very well be pound-foolish in terms of electricity cost. Michael Rigby-Jones writes: > Only if you add more lamps to compensate for the decreased brightness, > which wasn't the idea of the scheme. As Joes web page shows, dropping > the voltage by ~5% significantly increases lamp life whilst still giving > acceptable light output. If you need less light, it then becomes more cost effective to simply purchase a lower-wattage bulb. Sure, you'll need to replace the bulb more frequently, buy you will save money in the long run. -p.

> >From: "Josh Koffman" > > Thanks for the link...I'm playing around in HiCalc...and on the > > efficiency graph, it seems that if I supply my lamp with more than > > 100% of it's rated voltage...it becomes more than 100% efficient? From: "Bob Ammerman" > They are expressing a relative efficiency using lumens per watt. When you > raise the voltage the lumens go up faster than the watts, and thus the > efficiency goes up. Definately confusing....but correct, I think, in Lumens / Watt. It must be the devils own work to design these things...no wonder they are so expensive. Cheers....Mike. -- Outgoing mail certified virus free. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 265.8.6 - Release Date: 07/02/2005

Ok, good explanation. Now how about the "Graph Voltage" part? You end up with a graph that has Percent Rated Lamp Voltage on the X axis...and Percent Rated Lamp Voltage on the Y axis. At least the X values are equal to the Y values :) Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams On Wed, 9 Feb 2005 19:42:33 -0000, Mike Hawkshaw <TakeThisOuTmikeEraseMEspam_OUTspikey-mike.com> wrote: > From: "Bob Ammerman" > > They are expressing a relative efficiency using lumens per watt. When you > > raise the voltage the lumens go up faster than the watts, and thus the > > efficiency goes up. > > Definately confusing....but correct, I think, in Lumens / Watt. It must be > the devils own work to design these things...no wonder they are so > expensive.

Jinx wrote: >>Can I ask you - where did you get the graph at the bottom of the page? > > > Honestly don't remember. It must have been off the web because it's > so crookedly scanned, which I wouldn't have done. I have made a > small attempt to make it readable > > Here's a graph showing CFL vs filament > > www.susdesign.com/cfl/cfl-spreadsheet-1.xls > > And an opinion piece by the editor of Silicon Chip > > "Compact fluorescent lights are not economic" > > www.siliconchip.com.au/cms/A_102164/article.html > They must not have CFLs that are as good as the kind I get here in the states for a few US$ a piece that have been used for several years. I don't think I've seen one fail in my house. -- Martin K http://wwia.org/sgroup/biofuel/

My own experience with CFL's for the past couple of years at home has been good, with one exception. I have had up to four at a time in a (poorly vented) bathroom, where they only last a few months. I suspect the humidity kills them, so I've went back to incandescent there. I have also started writing the install date on the base of the CFL bulbs. Very useful spreadsheet. Gives me the urge to purge any other incandescents I can find. GC {Quote hidden}> > > Here's a graph showing CFL vs filament > > www.susdesign.com/cfl/cfl-spreadsheet-1.xls > > > > And an opinion piece by the editor of Silicon Chip > > > > "Compact fluorescent lights are not economic" > > www.siliconchip.com.au/cms/A_102164/article.html > > Both these are very interesting. We have a mixture of CFLs > and fillaments at > home for reasons of mood / convienience / efficiency. > > I posted another reply with my experiences of CFLs at work. We get the > cheepest we can find, and get about 2 years continuous life > (24/7). From > this, I suggest that the electronics probably fails long > before the tubes if > the second article is correct. > > Cheers....Mike. > > > > -- > Outgoing mail certified virus free. > Checked by AVG Anti-Virus. > Version: 7.0.300 / Virus Database: 265.8.6 - Release Date: 07/02/2005 > > --

>> >While this may be a penny-wise solution in terms of bulb cost, >> >it could very well be pound-foolish in terms of electricity cost. >> Only if you add more lamps to compensate for the decreased >> brightness, >> which wasn't the idea of the scheme. As Joes web page shows, >> dropping >> the voltage by ~5% significantly increases lamp life whilst still >> giving >> acceptable light output. > If you need less light, it then becomes more cost effective to > simply > purchase a lower-wattage bulb. Sure, you'll need to replace the > bulb > more frequently, buy you will save money in the long run. Bulb costs, especially of fancy ones, are liable to be an average of $NZ2. (Std bulb $0.50 - $1. Fancy bulbs $2 - $5). Bulb lifetimes in some applications can occur within months and are liable to normally be 1 to 2 years. Adding a series resistor can, according to Joe's experience, extend lifetime indefinitely - say 4 years. Savings for a 6 months to 4 year life extension are $2 x (2 - 0.25) = $3.50 per year. I can imagine that a suitably constructed resistor base or inline add in (terminal block and resistor mounted in junction box) would meet that cost. Labour cost of installation is very well offset by bulb changing labour. In fact, based on labour costs alone, the resistor solution wins hands down. RM

gacrowell@micron.com wrote: > My own experience with CFL's for the past couple of years at home has > been good, with one exception. I have had up to four at a time in a > (poorly vented) bathroom, where they only last a few months. I suspect I've had mine working just fine since 4/98. My wife and Kid take mega long showers without a vent fan (they like the sauna effect) so it's isn't the humidity. I have GE 23W straight tubes that do not appear to have any special sealing at the tube/case junction. > the humidity kills them, so I've went back to incandescent there. I > have also started writing the install date on the base of the CFL bulbs. Ditto. And the warrantee expire date so I know whether to dig out the receipt from the 'warrantee' files. Unfortunately the shipping cost is more than the cost of replacements. Nice scam. > Very useful spreadsheet. Gives me the urge to purge any other > incandescents I can find. I went completely CFL over a period of a year or so. Payback was only a few years since the other bodies in the house don't seem to understand the concept of 'lights off when not in use'. R

On 10 Feb 2005 at 12:16, Russell McMahon wrote: > Bulb costs, especially of fancy ones, are liable to be an average of > $NZ2. (Std bulb $0.50 - $1. Fancy bulbs $2 - $5). Bulb lifetimes in > some applications can occur within months and are liable to normally > be 1 to 2 years. Adding a series resistor can, according to Joe's > experience, extend lifetime indefinitely - say 4 years. Savings for a > 6 months to 4 year life extension are $2 x (2 - 0.25) = $3.50 per > year. I can imagine that a suitably constructed resistor base or > inline add in (terminal block and resistor mounted in junction box) > would meet that cost. Labour cost of installation is very well offset > by bulb changing labour. In fact, based on labour costs alone, the > resistor solution wins hands down. > I guess the electricity cost can be reduced changing that resistor by an AC capacitor. The power meters don't care about reactive power. Mark Jordan

Russell McMahon wrote: {Quote hidden}>>> >While this may be a penny-wise solution in terms of bulb cost, >>> >it could very well be pound-foolish in terms of electricity cost. > > >>> Only if you add more lamps to compensate for the decreased >>> brightness, >>> which wasn't the idea of the scheme. As Joes web page shows, >>> dropping >>> the voltage by ~5% significantly increases lamp life whilst still >>> giving >>> acceptable light output. > > >> If you need less light, it then becomes more cost effective to >> simply >> purchase a lower-wattage bulb. Sure, you'll need to replace the >> bulb >> more frequently, buy you will save money in the long run. > > > Bulb costs, especially of fancy ones, are liable to be an average of > $NZ2. (Std bulb $0.50 - $1. Fancy bulbs $2 - $5). Bulb lifetimes in > some applications can occur within months and are liable to normally > be 1 to 2 years. Adding a series resistor can, according to Joe's There used to be a US product called "WattSaver" which did EXACTLY that. It was a 1" diameter disk that you stuck in the base of the Edison socket to drop the voltage and hence 'save watts'. Of course it killed efficiency. > experience, extend lifetime indefinitely - say 4 years. Savings for a > 6 months to 4 year life extension are $2 x (2 - 0.25) = $3.50 per > year. I can imagine that a suitably constructed resistor base or > inline add in (terminal block and resistor mounted in junction box) > would meet that cost. Labour cost of installation is very well offset > by bulb changing labour. In fact, based on labour costs alone, the > resistor solution wins hands down. In which case you just install 240V bulbs in 120V sockets as is often done here on campus in the fire exit signs. They're a bit dimmer, but they don't need replacement, EVER. (and much cheaper that switching to LED lights). I believe you can also get 384V bulbs for commercial lighting circuits, so those of you in 240V countries may have that option. R

Just a silliness in the program IMHO. Bob Ammerman RAm Systems ----- Original Message ----- From: "Josh Koffman" <RemoveMEjoshybearTakeThisOuTgmail.com> To: "Microcontroller discussion list - Public." <piclistEraseME.....mit.edu> Sent: Wednesday, February 09, 2005 3:05 PM Subject: Re: [OT] 240-volt Power {Quote hidden}> Ok, good explanation. Now how about the "Graph Voltage" part? You end > up with a graph that has Percent Rated Lamp Voltage on the X > axis...and Percent Rated Lamp Voltage on the Y axis. At least the X > values are equal to the Y values :) > > Josh > -- > A common mistake that people make when trying to design something > completely foolproof is to underestimate the ingenuity of complete > fools. > -Douglas Adams > > On Wed, 9 Feb 2005 19:42:33 -0000, Mike Hawkshaw <EraseMEmikespikey-mike.com> > wrote: >> From: "Bob Ammerman" >> > They are expressing a relative efficiency using lumens per watt. When >> > you >> > raise the voltage the lumens go up faster than the watts, and thus the >> > efficiency goes up. >> >> Definately confusing....but correct, I think, in Lumens / Watt. It must >> be >> the devils own work to design these things...no wonder they are so >> expensive. > --

I didn't realize I had started such an active thread.:-) A couple of comments: Someone in N.Z. suggested putting a diode in series with a lamp and a couple of others described the diode buttons that fit in the lamp socket. In one of my past lives, I was an electronics technician with the Audio Visual Department at Oklahoma State University and one of our staple fix-it jobs was to revive a certain name brand of overhead projector that, for some reason, was designed with a diode in series with the projection lamp. I think they even called it an 85-volt bulb but it's been 15 years so some of my memory may have been dimmed by the sands of time. Anyway, the diode sometimes burned out to a short and put the full mains voltage across the bulb. The energy of a half-wave rectifier is .7 the energy of full-wave sinusoidal AC so the bulb ran somewhat brighter but not for long. Those bulbs were usually in the $20-$30 range so the customers used to get pretty peeved when they replace the first bulb, thinking its time had run out only to find the next bulb's time ran out a few minutes later. We'd plug in a new diode and sell them yet another $20 bulb and the projector was whole once again until that diode overheated and fused in to a fancy jumper once again. I never understood the value of that useless bit of over-engineering at all. Plenty of models of overhead projectors ran on similar bulbs except they were optimized for 120-volt service so no diode was needed. It would have made sense if they were 120-volt bulbs and one could throw a switch to un-short a diode in series with it for slightly dimmer (= longer bulb life) operation. The Kodak Slide projectors did have a Dim position on their power switch, but it was a ceramic tubular wire-wound resistor in series with the bulb, not a diode. Someone on this list also suggested putting a full-wave bridge ahead of the lamp and a filter capacitor across the output. Another lister responded that it would make the lamp brighter which is quite true. Just as a half-wave rectifier supplies .7 of the energy of the full sine-wave AC, the filtered bridge will supply the peak, not RMS voltage of the AC wave. This is the square root of 2 or about ~1.4 times the RMS voltage. On a 240-volt lamp, this will be hitting it with a steady 340 volts assuming that the filter capacitor is large enough to fully charge. The bulb will run brighter, whiter and die much sooner just like those ill-fated projector bulbs did. A bit of useless trivia is in order. A lot of projector bulbs have what is known as an iodized quartz filament. They actually must run at a certain range of voltage for optimum bulb life. I don't totally understand what is happening, but these type bulbs have a process at which the filament vaporizes when run at low voltage but recovers somewhat at the optimum temperature. The lamps even have a collection grid in them which is not an electrical component, but a chemical one that is involved with this process. No matter how one cuts it, those bulbs are like rase cars. They run extremely hot and die young with expected lives of a few tens of hours at best. It's not uncommon for 16-millimeter movie projectors to use a 30-volt lamp since the filament can be made smaller which makes the optics a little easier to produce. Those systems have a monster transformer and a few of them even have a Dim option on the lamp switch which selects a slightly lower-voltage tap on the transformer. Martin McCormick WB5AGZ Stillwater, OK OSU Information Technology Division Network Operations Group

"Wouter van Ooijen" writes: >The suggestion was to put a diode in series with the lamp. No capacitors >(or inductors or other storage elements). You want to say that this >istuation the energy in half the sine (say positive halve only) is 0.7 >times the energy in the full sine? So the energy in the positive + the >negative halves must be 1.4 the energy in the full sine ???? I think you >should apply for a patent immediately! I get an Email from time to time offering an incredible number of good Email addresses I can use to spam people with for an amazingly low price of $99.95 or other such nonsense so maybe I should. One person did suggest both a diode bridge and capacitor and that's what I referred to. With a capacitor, you get the peak voltage and if the cap has enough Joules in it, that will supply the load with a steady DC voltage that is about 1.414 times the RMS value of the sine wave. If I wasn't clear, that is what I meant. A resistive load like a light bulb should burn almost the same brightness when placed at the output of an unfiltered bridge as it does when it is directly across the line. Martin McCormick WB5AGZ Stillwater, OK OSU Information Technology Division Network Operations Group

> "Wouter van Ooijen" writes: >>The suggestion was to put a diode in series with the lamp. No >>capacitors >>(or inductors or other storage elements). You want to say that this >>istuation the energy in half the sine (say positive halve only) is >>0.7 >>times the energy in the full sine? So the energy in the positive + >>the >>negative halves must be 1.4 the energy in the full sine ???? I think >>you >>should apply for a patent immediately! > A resistive load like a light bulb should burn almost the same > brightness when placed at the output of an unfiltered bridge as it > does when it is directly across the line. A single diode supplying half wave will give the filament time to cool for substantially longer. Bulb filaments have time constants which are roughly in the order of a half cycle - ie they do give mains hum when a photodetector is used with them but do not give deep flickering on the half cycles. So you would expect a reasonably greater degree of cooling when left off for a whole half-cycle. A warm filament has a higher on resistance than a colder filament so MAYBE one might expect half cycle AC to give MORE than half the watts of full cycle AC :-) RM

> One person did suggest both a diode bridge and capacitor and > that's what I referred to. > > With a capacitor, you get the peak voltage and if the cap has > enough Joules in it, that will supply the load with a steady DC > voltage that is about 1.414 times the RMS value of the sine wave. So where does that factor 0.7 come from? Rectifying (half or full) with sufficient capacity yields sqrt(2) times the voltage, hence 2 times the power (neglecting non-linearity of the lamp). > A resistive load like a light bulb should burn almost the same > brightness when placed at the output of an unfiltered bridge as it > does when it is directly across the line. Now we agree. And roughly half the brightness with a single diode if the characteristics of the lamp were linear. But as they are not, a single diode will reduce the brightness much more. Plus (in my experience) introduce and irritating flickering. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products docent Hogeschool van Utrecht: http://www.voti.nl/hvu

> > Now we agree. And roughly half the brightness with a single diode if > > the > > characteristics of the lamp were linear. But as they are not, a > > single > > diode will reduce the brightness much more. > > Intuitively I would expect brightness to go down below 50% of full. > > But what did you think of my argument that power consumption should be > ABOVE 50% of full due to the colder than average filament? Yeah I follow, but looking from another angle... if the power consumption were above 50%, how could the filament remain colder than average?!! (PS, don't listen to me - I'm not even thinking about it) :-) -- Brent Brown, Electronic Design Solutions 16 English Street, Hamilton, New Zealand Ph/fax: +64 7 849 0069 Mobile/txt: 025 334 069 eMail: RemoveMEbrent.brownTakeThisOuTspamclear.net.nz

>> Intuitively I would expect brightness to go down below 50% of full. >> But what did you think of my argument that power consumption should >> be >> ABOVE 50% of full due to the colder than average filament? > Yeah I follow, but looking from another angle... if the power > consumption > were above 50%, how could the filament remain colder than average?!! It doesn't seem intuitive initially, but it's possible. A bulb is probably more efficient in terms of watts per lumen (or watt-ever) as the filament gets hotter. The practical limit for hotness is a tradeoff between temperature and lifetime. Also, as temperature rises emission moves to shorter wavelengths and moves more out of the invisible infrared into the visible wavelengths. So a bulb run at low temperature will draw more power than the same bulb at a higher temperature but MAY radiate less of its energy in the visible spectrum and also may be less efficient. I can feel a simple experiment coming on. Defining and measuring "power" here may be trickier than is at first apparent. RM

Russell, On Wed, 16 Feb 2005 01:19:14 +1300, Russell McMahon wrote: > So a bulb run at low temperature will draw more power than the same > bulb at a higher temperature Indeed, a bulb that's severely under-run may emit heat but no light whatsoever. > but MAY radiate less of its energy in the visible spectrum and also may be less efficient. But hang on - surely efficiency of a lightbulb is the ratio of electrical power used to useful light emitted? So emitting less of its energy in the visible spectrum is a definition of lower efficiency! All of the power it uses is emitted (unless it creates matter per E=MC^2, and we won't get into that :-) but a lot of it is heat (and IR) which is no good to us for the purpose of seeing. That's why fluorescents are more efficient, because more of their output is light rather than heat. > I can feel a simple experiment coming on. Defining and measuring "power" here may be > trickier than is at first apparent. Well input power is dead easy: P = IV ! Measuring the power of the light emitted is tricky because it's spread all round, and collecting it to measure it will involve losses. It's actually easier to measure its heating effect and subtract that from the input - what remains is (almost) the light that we wanted in the first place. I've always thought it ironic that electric heating is (in the UK at least) very expensive, when an electric heater in a house is 100% efficient - all of the energy it is fed is emitted as heat, which is what you want! Cheers, Howard Winter St.Albans, England

{Quote hidden}>-----Original Message----- >From: EraseMEpiclist-bouncesspamspamBeGonemit.edu [RemoveMEpiclist-bouncesKILLspammit.edu] >Sent: 15 February 2005 13:00 >To: Microcontroller discussion list - Public. >Subject: Re: [OT] 240-volt Power > > >Russell, > >On Wed, 16 Feb 2005 01:19:14 +1300, Russell McMahon wrote: > >> So a bulb run at low temperature will draw more power than the same >> bulb at a higher temperature > >Indeed, a bulb that's severely under-run may emit heat but no >light whatsoever. > >> but MAY radiate less of its energy in the visible spectrum and also >> may be less efficient. > >But hang on - surely efficiency of a lightbulb is the ratio of >electrical power used to useful light emitted? >So emitting less of its energy in the visible spectrum is a >definition of lower efficiency! All of the power >it uses is emitted (unless it creates matter per E=MC^2, and >we won't get into that :-) but a lot of it is >heat (and IR) which is no good to us for the purpose of >seeing. That's why fluorescents are more efficient, >because more of their output is light rather than heat. > >> I can feel a simple experiment coming on. Defining and measuring >> "power" here may be >> trickier than is at first apparent. > >Well input power is dead easy: P = IV ! Measuring the power >of the light emitted is tricky because it's >spread all round, and collecting it to measure it will involve >losses. It's actually easier to measure its >heating effect and subtract that from the input - what remains >is (almost) the light that we wanted in the >first place. > >I've always thought it ironic that electric heating is (in the >UK at least) very expensive, when an electric >heater in a house is 100% efficient - all of the energy it is >fed is emitted as heat, which is what you want! Whilst the conversion into heat is pretty well 100% efficient, converting it from coal/oil etc. in the first place is not! The you have the transmission losses to take into account. Whilst gas costs some money to pump around, at least you get out of one end what you put in the other (more or less). Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. =======================================================================

>> I've always thought it ironic that electric heating is (in the UK at least) very expensive, when an electric heater >> in a house is 100% efficient - all of the energy it is fed is emitted as heat, which is what you want! The high cost of electricity is due to it's less that 100% efficient generation process, as well as market conditions meaning that the electricity companies can get away with their price. I always find it amusing that Argos, etc, advertise "high efficiency" etc electric heaters. I'd be impressed if they managed one less that 100% efficient. -----Original Message----- From: piclist-bouncesSTOPspamspam_OUTmit.edu [spamBeGonepiclist-bouncesSTOPspamEraseMEmit.edu] On Behalf Of Howard Winter Sent: 15 February 2005 13:00 To: Microcontroller discussion list - Public. Subject: Re: [OT] 240-volt Power I've always thought it ironic that electric heating is (in the UK at least) very expensive, when an electric heater in a house is 100% efficient - all of the energy it is fed is emitted as heat, which is what you want! Cheers, Howard Winter St.Albans, England