Andrew Que Sites list Photos
Projects Contact
Main
   This is my old bench top power supply, and the one I fixed.  I had replaced it because I thought it was done for.  The variable power side no longer worked.  However I didn't get rid of it because it had 3 working fixed voltages: 5, 12 and -12 VDC.  This also meant I could get 24 VDC by connecting across the 12 and -12.  And should I desire such a voltage, 17 VDC by going from 5 to -12.  I have often used it as a fixed power supply.  Now that the variable side is fixed I have a range from 2-20 VDC as well.  This supply is not as versatile as my primary bench supply, but since it is now fully functional is still a good supply.

February 08, 2015

Polynomial Regression Online Interface Additions

   Updated the Polynomial Regression Online Interface and added support for doing three linearizable functions.  These are functions where simple adjustments can be made that allows them to become linear equations.  I have written about how this works in the past and a few people mentioned to be I should add this support to my online interface.  The functions I added those commonly implemented in spreadsheet software.
  • y = a eb x
  • y = a ln x + b
  • y = a xb
   Implementing the functions wasn't the problem.  Doing it in a clean manner and getting all the comment-driven documentation correct took the most time.  I'm not sure I like it fully yet, but I think the functions are ready for beta.  I'm going to clean up the package and release it, but before I do I have some more ideas for things to add.
   Here is the LED driver board for the 10 watt LED for my microscope inspection station.  I used a grinder to time back the edge of the PCB just a bit so it would fit into the existing light housing.  The board still works just fine and my guess is that I didn't even touch any traces with the this trim.
   I have been neglecting to mention all of the lovely snow we have been getting lately at Elmwood Park.  None of the snow storms we had were suppose to be all that bad, and none of them brought blizzard like conditions.  However, they did last a long time and dumped at least 5 inches of snow on us.  This was the first time I actually got to use the snow-thrower this winter.  Even in places where the snow drifted all the way to the top of the intake the machine had no trouble.  The only time I had to slow down was when taking out the snow dumped in front of our driveway by the road plow.  This heavy packed snow presented a challenge, but only required going a little slower.  I even helped two people who were trying to dig out their car in the road after the plow came by.  It's a lot of fun.

February 05, 2015

Machined Heatsink

LED and new heatsink

LED and new heatsink

   My very first machining project.  This is the LED replacement light for my inspection microscope.  As I have written in the past, this light originally had a 10 watt, 12 VAC halogen bulb that was far too dim to be useful.  I initially tried replacing it with a 1 watt LED, but that wasn't good enough.  Then I got a 10 watt LED which worked great.  The 10 watt LED required a heatsink, which at the time I did not have.  Since the LED assembly was able to fit into the existing light body, I wanted to see about keeping this structure.  After talking to a mechanical guy at work about my setup he figured the surface area of the existing structure would be enough to dissipate heat.  I just had to couple the high power LED to the housing.  To accomplish this I picked up an aluminum rod with the idea of machining down a piece.
   I started with a 1" diameter aluminum rod 12" long.  I only needed about 1/2" for this project, but 12" was as short a piece as I could get.  At work I asked for help from one of the machinists who taught me how to setup and operate the lathe and overhead mill.  With a micrometer I was able to cut my 1" diameter rod down to the inner diameter of the light housing.  Then I used the overhead mill to cut two groves in side.  These groves would allow the metal power connectors to pass along side.  After milling this I used a vice and pressed the machined piece into place.
   Trying to reassembly the setup I found a small lip on one of the housing parts that prevented the LED driver board from fitting nicely.  Using a rotary tool I took about 1/32" off the PCB and corrected this problem.  A little bit of heatsink compound and I had the 10 watt LED joined with the aluminum.  When assembled I powered the setup and within about 45 seconds I could feel the body of the light getting warm.  In just a couple minutes the entire housing was quite warm, meaning the heat was transferring from the LED to the housing.  I don't know what the LED junction temperature is and I don't have a good way to check, but I think this setup should run fine. 
   I am pleased with the results.  The LED light is a great alternative, and with the new heat sink should run as long as I need it to be on.  I also learned a little about machining metal parts.  While I wouldn't say I yet know how to do this, I know more than I did.  And I think more projects of this type could be interesting.

February 04, 2015

Weight and calorie tracking

For the past month I have been tracking my weight and calorie intake. It's not that I am fat, or even overweight. But to borrow a term my roomie used, I got “squishy.” So as part of an overall plan to get into better shape I thought I'd start by losing a little weight. I first established a starting point of 165 lbs (75 kg) by weighing in each morning before I took a shower. This weight varied, but 165 was the most typical number. Then I started counting calories. For my age, height and weight I need to take in about 2,200 calories/day. For weight loss it was recommended I take in between 1,200 to 1,800 calories/day. I have been shooting for 1,500/day.

In addition to just taking in less calories, I've started to focus on the kinds of calories I take in. Whole made bread and pasta were the staples of my diet. While I love my cereal grains I've tried to switch away from that. More protein and more vegetables. Technically I should also be doing more fruit, but I don't like most fruit. The best I can do regularly is banana. But I can eat spinach, kale, and broccoli all day. So steamed broccoli and a little lemon juice has become my new favorite snack food.

All this began around December, but I didn't start keeping records until the 9th of January. By this time I had already lost 5 lbs and I decided to keep records to track further progress. Results:

In this 22 day period over the month of January I went from 160 lbs to around 153 lbs (69 kg) with an average drop of 0.41 lbs/day or 2.9 lbs/week (1.3 kg/week). That's a little faster than the recommended 2 lbs/week (2 kg/week) and I need to slow the rate of drop or I risk messing up my metabolism.

My calorie intake has been just about 1,500 calories/day, which is what I was targeting. So why am I dropping weight faster than expected? Exercise. Skating, swimming and short morning routine must mean I am burning a good deal more calories than expected. Swimming alone can burn between 500-700 calories/hour for someone my weight, and we like to swim for at least 2 hours at a time. I had not considered myself a very physically active person, but I am clearly more so than I thought.

With the data obtained so far I am going to adjust my diet a little and take in a few more calories. My initial goal was 150 lbs, and I am almost there. At this weight I will start to focus on maintaining body weight and start dig into doing more exercise. Why? I'm a computer programmer in my mid thirties, and spend at least 8 hours a day sitting in front of a screen. I'd rather not have all the health issues associated with this sedentary lifestyle. I enjoy biking, skating and swimming. So why not use that to counter act the negative effects of my computer dominated lifestyle? Besides, enjoy projects that require me to try new things—especially when they work.

   So the story I am working on is almost 50,000 words long.  By some standards this qualifies it as a novel.  It is far from finished, but I have enough material written to make a novel.  I find that interesting since I neither consider myself a good writer, and I have never read something of this length myself.