Sunday, December 08, 2013

diy garden lamp progress

Clean, simple, solid and bright. Can use recycled parts.

A recent example of the georgesworkshop do-it-yourself garden lamp that I have built here in the new workshop. (click any pic to enlarge)

This lamp is slightly brighter than the earlier ones since it looks after an important safety area where visitors step onto the deck from the grass. It uses C$1.50 LED chips bought on Ebay from China. Six chips can be installed in this lamp. I am using two. I am using slightly more DC current for more brightness but only use 4 watts total. I can adjust these lamps over a wide range of brightness by varying the DC power supply, a much wider range than with an AC dimmer. They don't go out all of a sudden but keep adjusting right down to zero.

I think these might qualify as dark sky lights because all the light is directed downward - none is wasted. This one uses about 4 watts of DC power at 18 volts from a recycled printer power supply. Nothing gets warm, the power supply and the LED chip are not working hard.

My lights are DC powered from an AC supply (they plug in). They are not yet solar powered but they could be. I would put the battery inside the house to keep it warm and the solar panel on the roof for better exposure. Where I want the light is not the best place to gather it in most cases and with more than a few lights, it pays to centralize the battery and panel. I plan these lights to be wired with good quality outdoor wiring into permanent locations but for now they are portable prototypes.

As in the examples last year, I've used plain 2x4 wood as the frame material. I did call this project "making a lamp from a 2x4". Of course, you can use any wood in any thickness or condition that appeals to you. These can be as skillfully and carefully made as you would like. This is a do-it-yourself (diy) project - feel free to customize. I will show you a framework that you can build on.

My lamp is made from the plain spruce 2x4 sold here in North America for building walls in houses. It is not very fancy but is cleanly cut and fastened with four deck screws that you can see in the top surface, pretty basic construction, but strong. While at the lumber yard, I did pick good straight 2x4s without excessive knots. I was able to cut them cleanly at 90 degrees on a saw. I pre-drilled the screw holes to prevent splitting.

The cost of the wood used in one of these lamps is about a dollar. The wood weathers nicely. After a year or two outside it turns grey and it will be solid in this climate for at least five years, maybe ten, even without a finish. For an outdoor lamp, I do feel that the standard 2x4 is a good material to consider.

I've continued with the U shaped frame construction. I have added a combination heat sink and reflector made of thin aluminum (roof flashing). The reflector shape keeps the block in place. The glass block is a commercial product bought at a home center. It acts as a light diffuser and a weather seal for the LEDs. I like the solid glass block appearance but is the most expensive part of this lamp, about $25. Other glass can be used as a diffuser as I show in some samples below.

I am building a number of these lamps for presents and give-aways to get feedback. I have made five this week. The last one took about an hour and a half construction time with about $5 worth of materials, other than the block.

I am using these lamps myself: three light the house back entrance and two are inside as table lamps.

If you have been reading about the project, you know that I had some trouble last year with one of the lamps whose led had failed in a strange way - it was flashing. This year, I am not using the Princess Auto LED described in that report but another which is less expensive and brighter. Hopefully more reliable! It was important to look at a number of sample LEDs to pick the best of what was available. The test bench helped with that.

These lamps use surplus "wall wart" power supplies to convert line power (110 VAC) into what the lamps need (about 9-10 volts DC). By matching each lamp with a suitable power supply, I can set the brightness in a range from downright frugal security light (about a watt used) to very bright, as much as commercial led bulbs using 10 watts but they run way cooler since they are more efficient. These lamps are safe if properly constructed. You can add your own creativity to the basic construction. You can use recycled material like the glass the wood and the power supply.


For this series of lamps, I've settled on a 10 watt LED bought on Ebay from China for about C$1.30ea for 10 with free delivery. They arrived in about a week. They are similar to the ones I showed you here. This is the response curve for these LEDs as measured at my test bench.

To create the graph, I am increasing the power to the LED on the x axis and measuring the light output on the y axis. I am actually increasing the voltage, measuring the current and multiplying the two to get the power (watts). You can see the complete spreadsheet here. You'll see that after about 10 watts, I don't get more light from the LED, just more heat. If I go higher still, I actually get less light. It's interesting that this is offered as a 9 volt to 12 volt LED. I don't think it would be a good idea to run this LED at 12 volts since it would get dangerously hot, enough to burn itself out or cause a physical burn if touched? But they are great at about 9 volts. They give a very pleasing warm white light efficiently with very little heat.

Another variation which uses an antique telephone pole insulator as a diffuser. The frame is found wood. I was challenged with the insulator to come up with a method to attach it to the frame and a spider wound out of wire and attached through hook eyes into the wood frame. That part has to be improved to get a good weather seal. There are two of these and they are wired in series to get 1/2 of the 18 volts DC from an old HP printer power supply - saved from the landfill.

Here is another variation which uses an Ikea Neglinge candlestick holder, mounted upside down, for a great light diffuser. These clear heavy glass candlestick holders are available at IKEA stores for C$0.79. Some of the details of the construction of this prototype are shown in the insets (click to enlarge).

I have a video of one of these lamps being dis-assembled here at Vimeo that will give you a closer look.

A family portrait of some recent examples all made with the same basic design.

Another family portrait

As always, your feedback is welcome and encourages me to continue working on this project. Thank you.

It is great to be able to make things in the new workshop and to show them to you.

George Plhak
Lion's Head
Ontario
Canada



diy landscape lamp reading list
a very bright 1 watt diy led garden light
making a lamp from a 2x4
best light at least cost - about testing bright diy leds at home
diy testing of led lamps
diy 1 watt led update
diy garden lamp progress - this article
a shielded low power diy garden lamp
diy lamp update

Tuesday, October 29, 2013

reader projects 2

Here are two updates on projects built by readers of my book
How to Build a Tracking Parabolic Solar Collector.

From Andrew Gray in Texas:

Andrew came up with a novel way to swing his array both N-S and E-W and added computer control. He is heating domestic hot water so uses insulated collector tubes.

He has a series of videos about his project posted on Youtube and you can also read more here.

From Bruce Snyder in Florida:

Bruce heats a hot tub so he is able to use uninsulated collector tubes. He writes "It actually produces WAY TOO MUCH heat in the spring and summer".

More about Bruce's project here and here.

As always, comments are welcome.

You can also click the "Reader comments" tab at the top of this blog

George Plhak

[to the gen2 intro and reading list]















reader projects
reader project 0
reader project 1
reader project 2 this article
reader project 3
reader project 4
reader project 5
reader project 6

Tuesday, September 24, 2013

why two of the same is better

You are never without when you have a backup, a source of spare parts and working knowledge. The latest thing that everyone wants is not necessarily better for you.

I am working on my old radial arm saw - cleaning and tuning. I have moved this saw and all my other tools about 200km so it has been partial disassembled, shaken, bumped and otherwise mistreated. I must check all the adjustments so that it is ready to make parts for my projects.

One of the position locks did not lock. I inherited another of the same model radial saw with my new shop but I like mine better so I began looking at both of them to compare.

If I did not have the "spare" machine, I might not have realized that this funny nut lock thingy (click on the picture to enlarge it) has the tang broken in the center. You can see what it should look like on the one at left. The tang goes into the slot in the bolt and keeps the bolt from turning.

Because I have two of the same thing, I have the good spare part from the spare machine to make mine work properly. DeWalt who made the saw I am sure would not sell this part for a 50 year old saw.

I call this the Principle of Two. I wrote about that here:

http://georgesworkshop.blogspot.ca/2013/02/why-used-is-sometimes-better-than-new.html

The Principle of Two worked for me with my old Honda motorcycle also. I bought a parts bike of the same model and year and used it to learn from and to rebuild my own. Lots of good spare parts. Since that Honda was somewhat rare, I was able to sell bits I didn't need and I think paid for my own.

These old saws are almost being given away on Kijiji. Maybe $80 or "please take it away" so I think when I am finished getting what I need from it, it will go to scrap metal. What a shame. Truthfully, I don't have the room for two of them.

George Plhak

Tuesday, September 03, 2013

exploring efficient workshop lighting alternatives

Musings about upgrading the lighting in the new workshop. Newer lighting technologies have some problems, like cost and radio interference. Last year's newer technologies (like compact florescents) and "regular" florescents and their descendants have disposal issues because of the contained mercury..

Hello from my new home in Lion's Head! I am moving in and setting up. It is great to be here.

This is the wood shop in back of the garage. The solid wood walls and the white ceiling make for a pleasant space.

In this room there are 4 ceiling mounted rapid start dual 40 watt fixtures, probably about 10-15 years old. One of the fixtures would not light. I could not find 40 watt bulbs locally. The newer 32 watt bulbs fit in the fixture but they won't light up.

There are four more of these dual 40 watt fixtures in the other rooms, eight in total. Each uses 80 watts of energy so (8x80=) 640 watts. You might have fixtures like these and haven't given them much thought since they are pretty reliable. As of July 1 2012, 40 watt T12 bulbs are no longer manufactured in the US.

Should I retrofit or replace all of them and with what? I wondered how these older florescents would start in the winter before the wood stove heats the building? Florescents, although improved with the T8 types, use more energy by comparison to LEDs. Florescents, including the CFL type contain mercury so they are a disposal issue. LEDs are instantly at full power in any temperature.

I have found that both the LED and the T8, while giving great light from much less energy, cause radio interference in the FM band. The FM signal is weak here so it is a tough test. I like to listen to the radio while I work. I have not overcome this problem yet. The old fixtures do not cause this interference.

For the florescent, I had to change the ballast and it turned out to be easier to take down the fixture. This gave me a chance to clean it thoroughly which took off years of grime - the fixture is white now and will give better lighting. It was also MUCH easier to work on the bench than on the ceiling.

Any damaged bulb support posts should be replaced. The T8 rapid start, whether the bulbs are large or small diameter, use the same support posts as the T12.

I could not buy new replacement 40 watt tubes for the one that would not start. Only 32 watt bulbs are available now. These new tubes fit (the pins are the same) but will not start in the older fixture unless the ballast is replaced, or the entire fixture. The ballast is the power supply, the block inside the fixture with all the wires. Hard to see in the picture but this one is oozing a black tar like goop. This one is bad.

The fixtures and tubes are dirty so they are not putting out as nearly as much light as they could be.

Cost to retrofit one florescent with newer T8 technology:

  • New 32 watt tubes 2x5 $10
  • New solid state ballast $15
  • remove clean and replace $0 (or have an electrician do this for you)
  • Total $25

With T8 bulbs, each fixture would draw 64 watts or (8x64=) 512 watts if all were retrofitted.

The 17 watt Philips LED bulb however was $44. It is one I have used previously. Would it really compare with only 1/4 the power?

Because the LED is a common screw base (called a T26 base) I had to take down one of the florescent fixtures and temporarily attach a pigtail socket in it's place. I screwed in the LED to the test socket. This was not a permanent installation.

The light quality from the LED was warm white and pleasing. It is very bright. I didn't have a lightmeter so the comparison was totally subjective. The LED certainly was, to my eye and to several visitors, suitable as a worklight. I would probably double them up like the florescents so that there were two LEDs used as a replacement for each of the florescent fixtures. The total energy use per pair would be (17x2=) 34 watts, about the same as one T8 bulb, but at three times the cost! How to mount these will take some study. Perhaps I could reuse the florescent fixtures to mount LEDS (more about that later).

For now, I have removed the LED and retrofitted two of the florescent fixtures with T8. The T8 are on a different switch from the older T12 so if I want to listen to the radio for now, I need to leave off the newer lights. Not a permanent solution.

This is a sample what I hear when I turn on either the T8 or the LED.

Improving the shop lighting will be a work in progress.

Thanks for your interest. If you have ideas about this, I'd appreciate hearing from you.

More about T12 vs T8 Reference

shop lighting reading list
a parabolic workshop light
led household bulbs
exploring efficient workshop lighting alternatives (this article)
work light led retrofit
testing fluorescent light fixtures - the test jig
testing fluorescent light fixtures - the test method (video)
testing fluorescent fixtures - 40 watt
efficient workshop lighting 2
updated bench lighting

Thursday, June 27, 2013

update and moving news 2

georgegesworkshop is moving.

Moving myself and my shop is probably one of the most difficult things I have attempted recently. What a "first world problem" - I have way too much stuff!

It will be worth it once I am set up in the new shop in Lions Head, Ontario on the beautiful Bruce Peninsula. I am excited about my new home.

The workshop is now packed in boxes and moved to storage. I am living in temporary accommodation until my move in on Aug 8. My new home is 200km north and it comes with a built-in shop.

I hope to be continuing my solar heating and LED lighting work in the Fall.

Thank you for your interest and please check back.

I have some interesting things planned.

George

Friday, March 29, 2013

led household bulbs

Tests described which compare three bulb technologies: incandescent (a hot wire in a vacuum), cfl (compact florescent) and led (light emitting diode). The bulbs I tested were all 120 volt AC (alternating current), the type that are used in North America commonly in lighting fixtures. The base is the large base approx 1" (2.54cm) in diameter, known in the industry as an E26.

LED (light emitting diode) lamps have been with us for about 40 years but only recently have they become bright enough to be useful in household lighting and inexpensive enough to have reasonable payback. This is not the first led here. A Philips led has been lighting the dining room table for a year and gives great light using only about 12 watts. The Philips bulb cost about $28 a year ago.

On a recent trip to IKEA, I saw the "LEDARE" led bulb for $15 and I bought one. I'll have to check on a return trip, but I was pretty sure the sign at the display was $12. That is the price on their website. I had a price check at the self serve cash - they told me it was $15. That's what I paid. Still pretty cheap I thought. update Nov 11 2013: The price of the three Ikea E26 Ledare bulbs now available in the Toronto area is C$8.99 for the 4.5 watt (200 lumen), $6.99 for the 8.5 watt (400 lumen) and $14.99 for the 10 watt (600 lumen). I had tested the 8.5 watt bulb.

This is the test setup - a nice corner by the fireplace which is perfect for reading.

The illumination at table height was checked with the sheet of white paper.

The color rendition and detail was checked with the green toy minibus.

The laser thermometer was used to check bulb temperature.

The Canon G15 camera is mounted on a tripod so that all the shots are taken from the same point.

The bulbs were installed in the table lamp one after the other. Time was allowed (about 5 minutes) for the cfl to come to operating temperature before taking any pictures. A cfl, depending on its type and age will not reach full light output until the gas in the tube warms up a bit. The incandescent and the led are at full power instantly.

Full AUTO mode was used on the camera. The flash was OFF. I tried pictures with the flash ON and decided that only the light from the test bulbs would be used to light the scene.

These are the test pictures captured in a screenshot of Photoshop. You can click any picture in this blog to enlarge it.

The pictures are reasonably similar. The camera being on AUTO adjusts the brightness level of the scene to produce a "better" picture but I was happy to let it do that. What is not so apparent in the images is that my own perception of the scene was that the led was slightly less bright than the other two. What I was not able to do was to measure with a light meter. I have only what is built in to the camera.

This closeup of the same pictures shows the color shift in the blue/green turquoise or whatever color the bus is? I remember many years ago trying to match a similar colored woman's wedding dress and the editing suite would just not co-operate. Something about mixed lighting sources at the venue causing the electronics to be fooled as to the color. To do this properly, I would first white balance the camera on the white sheet of paper and then refocus on the scene with the color corrected.

But in terms of this simple test, it is apparent that the brightness from the led is more than adequate for detailed pictures and probably a great reading light.

The color from the led is slightly different from that of the other two. IKEA calls the LEDARE bulb "opal white". Is that like "cool white"? I did a search on "opal white" but the term seems to refer to a frosted white, not a color, similar to the opal mineral.

I checked temperatures of the three bulbs with an infrared thermometer after each had operated for at least 10 minutes.

TYPETEMP F
Incandescent160
CFL145
LED96

Clearly a significant amount of the energy going into both the incandescent and the cfl is being wasted as heat. The led is positively cool by comparison.

Finally I measured the energy consumed by each of the bulbs using a "Kill-a-Watt" meter.

TYPEConsumption W
Incandescent55
CFL12
LED7

The voltage input to the lamps during this test (as read from the Kill-a-Watt meter) was 125 VAC.

So the led uses much less energy than an incandescent but not that much less than a cfl. Still, the reduction from incandescent is truly significant and the led overcomes some of the cfl's weaknesses, such as poor cold weather starting and led do not use mercury (but they use other stuff).

A less technical reader might find the look inside the bulb that follows to be interesting but not essential.

The last part of my exploration involved answering whether both phases of the AC power input were being used for best efficiency and whether there was a fuse. Since I had already packed my current probe for the move I could not look at the input current directly. So I decided that opening the bulb to look at the circuit might help me to answer both questions plus have a look at how the bulb was constructed. This would destroy the bulb and make it unusable. Done in "the interest of Science".

Do not try this at home - use caution if you do!

This is the business end of the IKEA LEDARE with the plastic dome removed. Nicely done circuit and construction. The bulb housing from here down is aluminum which is necessary for heat dissipation, to keep the electronics below some maximum temperature.

Notice the pattern of pinholes around each LED? The manufacturer is using copper plated-through holes to increase the thermal coupling to the aluminum housing below. The circuitry in this led lamp is less heat tolerant (cannot go to as high a temperature) than a cfl which is constructed typically with a ceramic housing.

This led lamp is the heaviest of the three due to the mass of aluminum used. Good recycling potential. Leds should last a long time but when this one dies, it gives up a nice aluminum ingot.

The leds themselves are the six small yellow tinted rectangles. You will notice that three (D2,D4 and D6) are a different yellow color than the other three. Yellow/orange indicates the use of a color shifting phosphor in light path of the led. Light is made at one frequency naturally by the led junction and is converted or shifted to another frequency (color) by the phosphor. That two different phosphors are being used shows that there will be a blending of two different light groups or spectra, probably in an effort to smooth the total light output. The yellow orange color is not the color of the light that is produced.

I think we've seen that this led lamp needs some blue but I will do some more work on that.

It is difficult to see but the leds seem to be all wired in parallel - which is a surprise. Leds must be fed a current limited diet and this is difficult to do well in parallel unless the leds are very carefully matched. Otherwise one or more will "hog" current at the expense of the others. Usually leds are wired in series or series parallel.

Here are two views (top and bottom) of the drive circuit which was inside the aluminum housing. The yellow and white wires at the left were soldered to the led circuit board so these are the "output". The resistor (component loose at one end with colored stripes) was connected to the center pole of the base so this is the hot "input". One wire is missing, still attached to the base outer screw shell that was connected (soldered) at the point just below the yellow elipse at the point labelled "B". This is the neutral "input".

I have not analyzed the circuit but I think that I can answer the two questions which brought me this far.

The resistor is the fuse. It is directly in line with the input and has a very low value. I haven't tried to measure but from the color code of the stripes, I would say that it is a 1 ohm resistor. Not the best fuse in the world and certainly not easily replaceable. But no-one will be replacing these fuses I think!

The component in the yellow shaded circle appears to be a full wave bridge rectifier which means that designer is using both phases of the electrical current. With some AC operated leds, this is not so, like with string lights where the lights are only on half of the time. Using the full cycle is twice as efficient.

I am not trying to reverse engineer your circuit Mr IKEA but merely trying to understand what you have done from a diy perspective.

These IKEA LEDARE bulbs are promising. I will have to go back for a few more.

Thanks for your interest.

George Plhak

shop lighting reading list
a parabolic workshop light
led household bulbs (this article)
exploring efficient workshop lighting alternatives
work light led retrofit
testing fluorescent light fixtures - the test jig
testing fluorescent light fixtures - the test method (video)
testing fluorescent fixtures - 40 watt
efficient workshop lighting 2
updated bench lighting

Thursday, March 14, 2013

parabolic light plant stand

My home made DIY parabolic reflector spreads light evenly from a single tube over a whole shelf. A standard IKEA shelf is used.

Using only one lamp with a parabolic reflector cuts my cost of electricity in half compared to the two tube commercial reflector I was using. The plants are thriving and the single tube provides great downward directed light. Very little light is wasted outside the growing shelf. The light goes straight down onto the plants.

The parabolic collector I am using is based on my design for a solar concentrator. Since the parabola can bounce light in either direction, I am using it here to form a beam of light like a parabola is used in a car headlight. The parabolic mirror bounces the light from the line source (the horticultural florescent tube) into a parallel beam aimed downward onto the plant shelf. It is really quite effective used in this way.

I had written about using one of my four foot parabolic reflectors as a workshop light.

Here I am using the same idea to create a plant light stand out of an IKEA IVAR shelving unit.

The growing season here is short (near Toronto). We can't plant outside until about the end of May so we use a plant stand to start seedlings inside in a warm environment. When they can live outside, they have grown to a good size.

An electrical timer allows us to extend the apparent day length slightly. The plant stand is located next to a south west facing window to catch whatever natural light is available.


The normal large IVAR shelf size is 90 x 50 cm. To create a bit more area on the shelf and to match better the size of the light, I cut a standard IVAR shelf in half and attached these two pieces under the narrow ends of a 2'x4' (122 x 61 cm) sheet of plywood to create a 2 foot by four foot IVAR shelf that matched the parabolic light. I also added some brackets underneath to hard mount the shelf to the IVAR vertical supports for a bit more strength.

The light sits on two horizontal strips of wood that mimic the end of an IVAR shelf.

Using the normal pin arrangement of the IVAR shelving, I am able to raise the light as the plants grow.

The pictures are from last year. Since I am moving this year, I won't be growing my own garden plants. We will have a garden but it might be a bit more modest.

When I get settled, I would like to set up this plant stand with an LED light strip. I took this shot of the bottom to show the way I cut the IKEA shelf in half and screwed to the bottom of the plant shelf. I marked the bracket location so that the screws will go back into the same holes. The entire shelf (including the parabolic reflector) will come apart and be packed up flat for the move.

The outdoor LED lights will also get moved.

Thanks for your interest.
George Plhak




[to the gen2 intro and reading list]

Sunday, February 17, 2013

do mice avoid garlic?

I think they do.

This was not a scientific experiment but a demonstration. Try it and let me know if this works for you.

I can't remember where the idea came from. I get hits when I Google "mice avoid garlic" and perhaps you will find better key words but there doesn't seem to be very much about it. I didn't invent this but I wanted to show you what happened here.

Two months ago, I had pulled the toaster and the microwave out of their cubby to clean behind and saw the usual mess left by mice as they feast on the crumbs and the other goodies we leave. That picture is at the end in case it turns you off.

I can't blame the mice but I'd like them to keep away from this kitchen area, messy as it normally gets, until I can get around to cleaning again. Two months ago I left a few of last years garlic along the back of the shelf. I hadn't done that before.

The picture above is how it looks today. I just pulled the microwave out. The toaster had been cleaned earlier in the day so the crumbs are gone but the microwave hasn't been moved until now. No signs of mice on this shelf yet they are elsewhere. I blame the garlic. We'll look again in two months.

In case this helps anyone.

If you have this problem, you don't need anything but some garlic bulbs to try it out. I am not sure if the type of garlic makes a difference and not even sure of what I used, I think Romanian Red? It was organic product for sure, I grew it myself.

You can click any pictures to enlarge it.

Thanks for your interest. Please let me know what you think. I moderate the comments to control the spam.

George Plhak

update Mar 27 - I looked behind the microwave and there is still no mouse dirt!

Monday, February 04, 2013

when used is even better than new

Our economy depends on our collective desire to own more, the newer the better. Whatever the sponsors tell us, we want.

But our behavior creates enormous problems for society in terms of credit and waste.

I don't want to lecture about those topics here. Those are well covered.

Instead, I want to describe how I provide reasonably priced high performance computing for myself while making a tiny individual step to slow the waste stream by staying close, but not on, the bleeding edge. This is my small example of thinking differently about consumption.

Buy used to save money is hardly a new idea. But in this case, I hope to show that I am BETTER OFF buying used while having a small eco contribution as well.

Georgesworkshop Principle of Two:

I am better off to have TWO of a previous product, rather than ONE of the current.
or
Two of last year's technology is less expensive and you have backup. (the slide above)

I use "last year" and "last model/product" loosely and interchangeably. I think you'll understand. The principle can apply to a few classes of goods, particularly technology, but not to everything.

How can this be?

  • I am almost as well off, perhaps better off with last year's model. After all, a year has gone by to see if it was any good. Don't you just hate to be the one who buys the dud product, the one that dies in the market? Wait at least a year for other peoples' experience (at their expense) on which to make up your mind.
  • Less landfill If a few people, or a few million people, started to think and act in this way stuff would stay in circulation longer, like good products used to.
  • I can buy two of the last model for less than one of what is on offer today yet be better off in some interesting ways - in spite of not having the latest and greatest.

I have a special memory of my mother taking me downtown as a boy to buy one of the first transistors that became available in the 1960s. Since then I have spent a lot of money and effort on the quest for new technology which taught me:

  • Buying new is expensive. After a year or so, the used price often falls to half or a third, if it has any value at all. Tech stuff generally does not hold any value over time.
  • New stuff can have yet to be identified problems. Better to wait for at least the first service pack or some software updates?
  • The market can cruelly ignore a new product no matter how "good" it seems and it fails. The company can fail too and customers are left with no support. This is too much like gambling to me.

How to cope? I will illustrate how it works for me with a recent computer addition. It is with desktop computers that I think I can give you a perfect example.

My principle of two computers project - a case study

I needed a new main design computer. One with raw power, a big screen with no travel. It had to be a desktop, heavy metal is ok in this situation, a laptop or a tablet just will not do.

My main machine was getting slow and developing quirky behavior requiring lengthy reboots. You will know the signs that a new computer is needed. I had been running this one (and it's spare) for *seven* years doing useful work (amazing?!) and until recently it had been very stable and reliable. One of two ASUS P4C-800e. Both had been excellent. All my files were backed up elsewhere. I was still able to work with it. I decided to migrate before I was forced to.

Instead of ONE new desktop computer I bought TWO IDENTICAL slightly used, for about half the price. Each received a thorough checkout and a OS software refresh on a clean system drive. Each is a complete functioning backup in all respects for the other. There are some interesting ways to use two computers where otherwise you would have one. Here is the comparison as I see it:

One NEWTwo USED
Cost$1000$550 (both)
OSWin8 64 bitWin7 64bit
Raw speed GHz3.42.66
Cores44
RAM GB126
HD GB1500160
Warrantyone year30 days

The column heads have links to the two types of machines I am comparing. The example New machine is from a national retailer, Staples. I am not picking on Staples and this is only an example. Some of the key specs show important differences. Of course these are not equivalent, I am trying to be macro to illustrate a point.

To Compare and Contrast (as they say in high school):

Application software I was not sure if my more important software (Abobe Creative Suite CS3 for example) which I did not have in the latest version (did I mention a tight pocketbook?) would run on a newer OS like Win8. Win7, on the other hand was of the CS3 period and the implication at Adobe was that what I had would not need to be upgraded with Win7 - a possible additional cost of the Win8 approach where I would probably have to upgrade my applications.

I reduced risk by buying only one of the used machines first. I then installed, authorized and tested the bigger packages on Win7 64 bit before deciding to commit to the second used machine.

Performance You will accept perhaps that when the used example was new, only two short years ago (2011), it would would have been then a fine step up from my design computer (2003 3.0GHz P4 running WinXP), the very the same step as I have done now at less than half the cost?

To me, the "new" old machines are a quantum leap from the old XP based P4. I don't honestly think I would have see a difference choosing between the "new new" and the "new old" in performance unless I was really trying to look for it. The contrast with what I am replacing makes it worthwhile.

You might be aware that processor speed, once the ultimate market benchmark, topped out at around 3GHz 10 years ago? Intel realized than it was increasingly difficult to go much faster and instead started promoting other user benefits like "i7-3770"(?). I can tell you that going from a relatively ancient 3Ghz to a much newer 2.66 it is like flying now compared to walking before. There is more to it than just the numbers.

Reliability The used machines have a clean Win7 install patched and field tested in it's 64 bit version. Windows 8 reliability is probably going to be pretty good but do I want to be a tester? Win7 would be safer.

Backup I mean not software but physical backup. Having two identical machines provides an instant available source of backup components - a "hot standby". When you are remote and rely on your computer, having backup is important.

Warranty Having a longer warranty is certainly important for an expensive new piece of equipment. In the used case, the 30 days is long enough to find anything obvious that does not work correctly. Since I have two machines, each can function as a backup for the other. Warranty other than the initial month may be less important to me.

The OS Windows 8 comes on the new computer in the 64 bit version. This must be so even though they don't state since the memory (RAM) provided is 12 GB (more than four, actually three something as the Windows 32-bit sees it).

Windows 7 has been thoroughly tested. I have wanted to try the 64-bit version for some time. Having a 64-bit OS primarily gives the benefit of allowing more RAM. 32 bit versions of Windows (most of them prior to Win8) could address only slightly over 3 GB of RAM. 64 bit allows up to 16 exabytes, (in theory)! I am happy to have the 6 GB offered with the used machines (twice what I had) and I can upgrade each to 16GB for about $100 for both.

At this time, I feel more confident adopting the Win7. I will wait for others to shake down Win8 before I wade in. The change to the Win8 interface is greater for me than to Win7 since I already use Win7 on a laptop.

Cost is a no-brainer for me. Having two of the used is great given the other advantages. Even with add-on RAM and HD to make used and new equal in size and less unequal in cost, buying used is still less than half the price and with a huge performance increase for me.

What to do with the hot spare? After checking the second out, I could quite happily park it on a shelf to keep it handy, un-used.

It could also be set up somewhere as a non-essential spare workstation? A rendering node?

In my case, I will use the second as a non-essential machine for internet browsing and email, functions which I am happy to separate from the main, money making workstaion. I will remove the system hard drive with it's configured and working Win7 and install another system drive and 64 bit SUSE Linux to try that out.

It is very nice to have a spare machine on which to try out new software. You know the kind of software? The kind that you wish that you had never loaded on your important computer? I now have a spare identical machine on which to do tests without any possibility of damage to the operating environment of my main production machine. An off-line hot tester if you like. This is a small but important advantage, and additional insurance of reliability.

Another software benefit: If you are ever wondering about upgrading something fundamental, like the OS, you have an identical spare to test out the upgrade once again without putting production at risk.

The benefit of having an identical hot spare extends to hardware as well. We might all be reluctant to put a main production machine out of service, to pull the wires off or the cover to have a look inside. We would not, with a spare. How many have ruined a perfectly good machine, let's say to try a new graphics card by plugging it in to see if it works? After the smoke clears, rather than fretting about having just killed production, wouldn't it be nice to have just tried the same thing on a risk-free convenient hot spare?

In the event of a gross failure of any of the components of the "new old" main machine, I have an identical working replacement only 5 minute away, by removing it from the spare.

All of this at half the cost of new? To a small business owner or an individual who makes their living from one key machine, backup and near zero failure downtime (not to mention significantly lower cost!) are not just large company IT concerns.

Summary I don't regret not having taken the "new" computer option. I think that in this case, having two of the previous model is perfect - for me.

But I am biased. I do like finer old stuff which can still be useful (like me?:) Please see In praise of older machines. Some of this writing shows it's age and mine. Later in the piece, the ASUS P4Cs I was using in 2005 are the "old" machines I am replacing here. May your own computers be useful to you for at least seven years! But other than fixing some spelling, I left it the way it was in 2005. Dust from the archive.

Thanks for your interest. Your comments are always welcome. Comments here are moderated.

George Plhak

Friday, February 01, 2013

diy 1 watt led update

Two home made garden lamps are faithfully lighting our home walkway, coming on at dusk and off at daybreak for one year now.

The lamps are suitably bright so as to almost eliminate the use of the outdoor floodlights except perhaps when company is expected.

The untreated wood frame of the lamps is weathering nicely. The glass block looks bright and clean and is pretty much indestructible. Interestingly, most visitors don't remark on these lamps, which might be a good thing? Others don't have to notice that I am saving money while lighting their way.

Now that the days are shorter and the lamps are coming on earlier, their glow welcomes automatically.

These lights, with the exception of the LED bulb and the glass block I have built myself at minimal cost. The power supply for the lamps and the auto turn on is being done with a reclaimed plug-in transformer unit from a very popular system around here 20 years ago called the Moonlight from Noma. I built a small rectifier/filter to improve the light quality and to improve operating efficiency.

Two short videos to give you an idea of the illumination, taken well before sunrise.


This video is the properly working light

I was impressed with my iPhone as a camera on this cold morning. In the first video, the wind is blowing strongly from behind me so the audio is pretty good. In the second, I am facing into the video so the wind noise is bad.


This light developed a regular flashing in one segment of the LED array.

Sorry about the audio! I think you can still understand?

Since the video was taken, the flashing segment has failed and is dark although the rest of the LEDs in that lamp still light.

Thanks for your interest

George

diy landscape lamp reading list
a very bright 1 watt diy led garden light
making a lamp from a 2x4
best light at least cost - about testing bright diy leds at home
diy testing of led lamps
diy 1 watt led update - this article
diy garden lamp progress
a shielded low power diy garden lamp
diy lamp update

Sunday, January 27, 2013

update and moving news

Dear reader,

I am sorry that I haven`t written here recently. There are distractions. I am busy in the workshop with packing and home renovations rather than projects.

I must move. My partner of 32 years and I have decided to split. We will sell our home so I must relocate myself and george`s workshop. Not fun. There are boxes everywhere.

We have given ourselves until spring to get the house ready for sale and to organize our stuff. The kids are grown so the family is going separate ways hopefully on good terms. By fall 2013 I will be settled somewhere in a new place, perhaps Lion's Head, Ontario?

On a brighter note, I am learning some new tools to improve my notes here. For example, I have not posted much video and I intend to change that.

There were reasons. My current rural location has limited internet bandwidth. The old Canon Optura camera was SD only and used DV tape so it was a slow process. The main work computer was acting up and getting slow. Even though I'd spent much of the last twenty years helping people to make video, it had become difficult to do it for myself. My tools were woefully outdated.

I am getting back into video to better show my own work to you. I intend to redouble my efforts on the web and to my writing.

I hope to move to a small community where hard wired internet is available. I have new camera, a Canon G15 which shoots HD 1080p very nicely. Also, I have added a new old computer, a Dell Optiflex with four cores running Windows 7 Pro 64 bit.

My new home will accommodate my projects. There is a lot of work to do and to tell you about moving forward as they say.

For a sample of my Canon G12 1080p HD tests, please see this video preview. The video is not very exciting. I am showing the playback of a Zoom H4n recorder which I am also learning for upcoming projects.

In the video what you hear is captured by the Canon G15 as playback sound from the H4n bouncing off the table top. The speaker on the H4n is about 3cm in diameter and is on the back (bottom) of the device. It sounds like a telephone compared to the next two.

This is the same audio but as captured by the H4n and another sample showing the stereo image. These are the sort of tools that I hope to be using to tell you about upcoming work.

Thank you as always for your interest and support.

Please consider buying my book. Your support helps the work. I will have another (the Gen2 version) out next year once I get settled.

George Plhak