Saturday, January 11, 2014

georgesworkshop diy weatherproof solar tracking sensor housing

A brief video tour of the solar sensor housing (radome) I made from a standard plastic outdoor electrical box and a hobby store transparent globe.

I had written previously about the housing here.

This sensor housing is mounted on a moving part of the georgesworkshop gen2 solar trough heater. It finds the position of the sun and steers the solar array toward it. A very simple commercially available circuit (US$35) provides all the brains that is required. The entire array uses only a small amount of 12 volts DC and can operate from a gel cel for a week or from a small solar PV panel. I used this array for solar heating of a swimming pool.

In this long view you can see the two locations I tried in the large array (click to enlarge photo).

Of the two spots, I now prefer the one at the top of the motor drive (the right lower arrow). Initially, I had decided to mount the sensor at the top left, since this was the earliest corner of the array to receive morning light and it worked fine.

I could have chosen any of the top or bottom hangers in the array to mount the sensor housing. There was an interesting failure that happened at the top left.

Lessons learned:.In a new appendix for the upcoming book, I discuss various things that went wrong with the gen2 design in the interest of sharing ideas that might improve your efforts. At left is the draft outline. I might add a few more things.

Here is a draft excerpt about melting the sensor housing, for your enjoyment:

"Appendix A - Lessons from Experience
In this section, I share with the reader some of the things that went wrong which were only evident with the passage of time.

Melting the sensor housing I wrote in the sensor section that the sensor housing should be located on a moving part of the array.
On the large swimming pool heater array, I chose to locate the sensor on a hanger at the top west end of the array since this corner received light first in the morning.

Here you can see that sensor backlit by concentrated sunlight streaming out of focus onto the solar sensor and its mount. This happens because I don’t have the array tilted quite high enough for my latitude.

Another view from the other side which shows the de-focused beam heating the side of the box and the deformation of the seal at the top.


Once the top of the Leviton box was removed, I could see that the side wall had deformed as well as looking a bit charred on the outside.

So the moral of the story might be to avoid the focus at the top of the array if possible? Having the sensor on a lower hanger might slightly shade a bit of the reflector but that location keeps the sensor housing out of the concentrated beam. Putting the sensor housing at the top of the motor drive arm, as I did with the Alternative 2 Round Sensor housing also protects the sensor from the beam."

As I said in the last post, I am going to be away from the workshop for a week.

Thanks for your interest.

George Plhak

[to the gen2 intro and reading list]

Friday, January 10, 2014

matching reflector length to the collector

Some video thoughts on matching my gen2 reflector to insulated collector evacuated tubes of various lengths for solar heating and some hardware views.

This is a work in process, actually part of a book I will publish shortly about the work at it's current stage.

If you want more info, you might look at this:

or anything on the solar thermal tab at the top of this page.

It was cold yesterday in the workshop. How appropriate to be talking about a solar heater.

Truthfully, the weather has been awful for two weeks and solar heating would have been useless here. The sun does not shine.

Thanks for your interest. I am able to work steadily on this effort but it seems to expand. I will be taking a short break for the next week to visit sunny california.

Thank you for your interest.

George Plhak

[to the gen2 intro and reading list]

Monday, January 06, 2014

solar tracking motor drive explained

I created this short amateur video to help explain some aspects about the motor drive that I am using for the diy gen2 solar parabolic trough heater.

The actual motor drive in this video has been in service for five years in a Canadian climate running a tracking heater for a swimming pool.

In a different use, the same motor drive served in a series of experiments done to compare performance of a concentrating trough with a flat plate collector where it was used to track both collectors. Here is the report on those experiments.

I have refined and improved the overall design through several iterations until the version published in my book. Other than some small tweaks and techniques, the motor drive shown is what I am recommending for upcoming gen2 book as reliable, inexpensive and easy to construct and maintain. If you built what I described in the book, this is how it should work for you.

There is some rust and wear to give it authenticity - it does work well. I did replace the gear motor two years ago as well as the axle nut which holds the bearings inside the arm. The bronze thread in the axle nut wore out while out riding repeatedly on the steel 1/2" threaded rod which is actually not very smooth.

In previous service this motor drive successfully swung 13 parabolic troughs following the sun over about 100 degrees of sky. It is important that the troughs be more or less mechanically balanced for the simple push pull mechanism used to be successful. A single steel rod (I used a 1/4" stainless) coupled to the end of the motor drive arm both pushes and pulls the array into position through control arms attached to the bottom of each reflector.

It is when the arm pushes the control rod that is the problem, not when it pulls. If there is too much unbalanced weight in the array, the control arm will tend to bend when it is pushed and that bending is a problem.

You can see the action on a properly set up array on a hazy day (some wandering as the sun goes behind clouds) in this video:

If you watch carefully, you will notice the vertical post which holds the motor drive moves slightly when the arm pushes the array revealing that some force is being exerted, enough to sway the post slightly. I could have done a better job with the balance and possibly have been able to move more than 13 collectors with the one motor drive but this is the largest that I have attempted.

After making this video I did brace the motor drive support post to the main array frame to give added support.

You can also watch the control arm from the other end of the array in this video:

As always, thank you for your interest in my work.

George Plhak

Other articles I have written about the motor drive:
Reliability and repair of the solar gear motor
Sources for gear motors
Further improvements to the motor drive
Motor drive mechanical 2 (early work)
Motor drive mechanical 1 (early work)

[to the gen2 intro and reading list]

Thursday, January 02, 2014

work light led retrofit

My portable work light until recently used a 500 watt quartz halogen bulb. That was dangerous because it got hot enough to burn.

Now it is an LED light which uses only 10 watts and barely gets warm while throwing lots of light.

I have retrofitted the light to use two 10 watt warm white LEDs. The LEDs run at half their maximum power from a reclaimed printer power supply which now hangs underneath the light. The LEDs get DC from AC via the power supply. Not as bright as with the halogen bulb but plenty bright enough for most situations where it will be needed.

Because of the low power consumption and the low heat produced, I think it is much, much safer than before.

The cost of parts for my conversion was about $6 ($3 for the two LEDs on Ebay from China and about $3 for a used HP printer power supply at the local recycling center, also a bit of aluminum sheet and bar and a few tie wraps and screws, possibly another dollar?). It took a pleasant evening to sort out the method and to do the conversion.

Unfortunately, I did not take a "before" picture. The light that you start with might be different from mine but I think these were very popular five, ten years ago? This is a general guide. Some of my other relevant work with LEDs has links.

The first step in my retrofit was to remove the halogen lamp and its supports and contacts as well as all the high voltage (117VAC) mains wiring from the housing. The parabolic reflector and a mounting bracket were the only parts saved.

This is the halogen lamp and its porcelain supports and contacts, now discarded. (click on any picture to enlarge it)

This pic shows the two LEDs mounted on the reflector. Note that the existing reflector was used as a part of the heatsink.

Another view of the reflector showing the sandwich of two aluminum bars and aluminum sheet with the reflector. The small screws hold the sandwich together and hold the LEDs firmly to the heatsink.

This is how it goes together. The technique is similar to what I am doing with the diy garden lamp. The 1 inch aluminum bars acts as a heat spreader from the LEDs to the aluminum sheet and the existing reflector which provide a large surface area.

I laid out the holes on one bar, drilled them carefully and then used a punch to transfer the hole positions to the other bar. I drilled the second bar and the center hole in both and then used a medium sized bolt through the center hole to clamp everything together to drill the small holes through the existing reflector and the aluminum sheet.

Not shown in the diagram is the heat sink thermal compound which is applied in a thin layer between the parts as they go together. In this closeup, you can see the thermal compound at the top edge of the aluminum bar as well as the mounting and the wiring of the LEDs.

I am using a reclaimed HP printer power supply rated at 18 volts DC at 2 amps. I wire the LEDs in series so that each receives about half the total, or about 9 volts. At this voltage, the LEDs will conduct about 350mA so the power supply has much more capacity than is required. The power supply stays cool.

This is the spreadsheet with the test results I got for the particular LEDs that I am using. The LED tests are described here

This picture is taken auto no flash with my Canon G15. There is no other lighting except for the LED work light. I have found that it is challenging to show you lighting results, but I think that this picture is pretty accurate.

Thank you for your interest in my work.

Please see my other lighting projects at the tab above.

George Plhak