Power consumption involved with driving the refrigerator has been improved by a massive 40x. This is due to data collected from the test setup driving the selection of new electrical components. With additional testing this could be driven up to 100x.
Set up complete and tests underway
Tests are underway. New information is being discovered which will lead to refinement in technique. Outcomes will be listed in a later post.
A number or secondary products have been added to the products page. These were generated during the process of experimenting with solid state refrigeration.
Test Setup 75% complete
An important test is about to be run that will decided how the next stage of the spot cooler will proceed. Custom sensors made for this specific test and have been calibrated. A micro controller with GUI output was also made. This was made to monitor the starting conditions of the test and to prevent a test start unless they have been met. Cables have been made to connect the micro controller to each sensor via an opto-isolator board.
The remaining 25% of work is light soldering, minor circuit testing, and cable making. The only delay is that the wrong diameter coax cable was ordered (will not fit purchased crimp-on BNC connectors) and must be re-ordered.
Spot cooler drive circuit design and theory improvement
Reworked theory of spot cooler. Designing electronics needed to power spot cooler. Calibrating measurement devices.
Electrically Controlling and Creating Air Pressure
For the past two months I have been experimenting with methods of generating/controlling pneumatic pressure that can be turned ON and OFF with electronics. An example of a pneumatic device controlled by electricity would be how a solenoid pinch valve can control the flow of air by varying the power sent to the solenoid. This approach does not work for my product. It is too large and too expensive. A smaller and cheaper method must be created.
A promising technique is being developed to solve this problem. It is in the prototype stage. This technique can be scaled down to the size I need (30x30x6mm)(current prototype is 80x80x80mm for ease of access), generate positive/negative pressures, is very low wear/maintenance, and can be easily/repeatably built.
Work still needs to be done on this technique. There are a number of parts that need to be improved before it can be used in the final product and be tasked with operating responsibility.
I may also build a large version (100x100x100mm) to generate large positive and negative pressures. It is very simple to scale to the size that is needed. It is likely that this will become a stand-alone product.
Rework of how liquid moves through pipes
Have been improving on how liquid moves through pipes and have succeeded in making pneumatically driven switches from stock material.
The reason effort is being put into making these switches is cost, customization, and footprint savings. The only pneumatic switches that have been found that fit my application are costly, come in limited sizes and are large. They are also made outside of the USA (in Germany) which increases lead time. Now I can make switches for low material cost (less than a 50 cents for an individual switch), custom input/output flow, and low footprint 20x20x10mm. The machining of the switch is done by computer program so no human effort is needed.
As a note I have been using 'pneumatic' to describe the switches, Technically they can operate with air or water which means it should be described as a pneumatic/hydraulic switch.
Reworking electrically driven fluid drivers. They are not strong or fast enough to be used in the final product.
Added sensors to fluid pipes - improved fluid drivers
Fluid drivers were blind to the movement of fluid that they were driving. It was clear from early tests with the fluid driver that some sort of sensor feedback was needed. A circuit was made to sense the fluid, and feed data from it into a micro controller.
Fluid driver made, Cooling effect observed
Fluid driver has been completed. It allows allows for the control of liquid as its pushed/pulled through pipes.
I plan to make a couple more of these and connect them together via micro controller to create more advanced liquid movements and increased speeds.
The cooling effect that powers the spot cooler has been observed for the first time. Past attempts did not use powerful enough magnetic fields to generate the effect (which is tiny). The movement of thermocouples in past setups also created EMI (noise) which masked the temperature data collected.
A cooling effect of 0.5C was seen during the experiment. This was done to prove that the cooling effect was real, not to be a product. The final product will be able to have a cooling effect of at least 25C, wattage TBD,
Fluid driver
Currently making fluid driver (electrically driven) for the cooler.
Improvement made to magnetic field sensor (FEA)
Measuring the intensity of changing magnetic field is difficult to do. The changing characteristic adds noise to any metallic sensor that you place in the field (see last post on metallic temperature sensors). To circumvent this a beam of light (Faraday effect) can be used to measure magnetic strength. One of these setups was built and will be referred to as Faraday Effect Apparatus (FEA).
Brief Description of the Problem:
The FEA I made had a bad light source. After troubleshooting the setup the light source got fixed and is now providing much better output data. This means I can measure the intensity of changing magnetic fields better.
Details:
In the FEA it is very important that the source of the light stay at the same intensity so that it doesn't foul the output data. This data fouling happened in my setup and showed itself as a continuously rising voltage in the photosensor amplifier output. The data I needed still showed itself in the output but it had to be separated from the fouled data post measurement. This was not a good use of time and hurt repeat-ability.
After trouble shooting the FEA it was found that the light source, not the photosensor/amp, was the problem. It was constantly changing it's light intensity (the intensity change was small enough to not be perceptible to human eye). In order to make it stay at the same intensity a current mirror was placed inline to prevent the light source (405nm 5mW laser) from drawing additional current as it heated up (diode forward voltage drops more the hotter the diode. This means more power gets provided to the laser if using a constant voltage source) and thus change its intensity.
The FEA output is now stable thanks to the addition of the current mirror to the light source.
Conclusion: Do not use voltage regulators to power LEDs and laser diodes in applications where a very controlled light intensity is needed. Use current regulators.
Limitation of thermocouples and other metallic temperature sensors in a magnetic field
Thermocouples and other metallic sensors will pick up EMI from changing magnetic fields. This adds noise to my data. Placing thermocouple length in SS braid does not prevent EMI for the fields it is exposed to, I think the frequency I'm exposing it to is just too low (less than 10Hz). Maybe soft iron foil wrapped around the length of the thermocouple will help.
To circumvent the EMI issue I have been looking into buying (or developing) some non metallic temperature sensors. No metal would mean no EMI. First and second test of home brew fiber optic temperature senor did not work.
Worked more on the magnetics of the cooler. Made progress, Will run tests tomorrow.
Need to make a thin glass tube filled with oil so that the FEA can interface with the new magnetics that I made
July 2018
-Trouble shooting with the FEA revealed a long standing problem I had been having. The photo sensor output kept climbing even though it appeared to have a constant laser light source. Replacing the laser with a LED confirmed that it was in fact the laser's light intensity that was fluctuating and not an issue with the photosensor's amplifier like I had assumed. A constant output laser will have to be purchased and used or an active light intensity filter will have to be placed in front of the existing laser.
June 2018
-Experimented with making custom heat pipes out of soft copper tube and later borosilicate tube. Results inconclusive.
-Attempted to flatten MC metal with hammer, anvil and propane torch. Rusted due to heat.
-Made nichrome wire heater to MC metal into smaller parts. Heater did not reach hot enough temperatures. Exploring other options.
-Bought MC metal and iron foil from metal supplier.
May 2018
-Moved to new location in Houston TX.
-Developed step diagram of how cooler will function. Simulated operation in excel to 500 steps
-Experimented with non contact IR thermometers to measure cooler temperature. This was to prevent EMF pick up through wired themocouple and thermistor temperature sensors. The IR thermometer did no work to the standard I was looking for and will use thermocouples that have been EMF shielded using EMF braid.
April 2018
-Made one-way valve and ON/OFF valve for fluid system. This allows for the one directional flow of liquid and the ON/OFF flow of liquid. Also made one-way pressure driven pump to move liquid.
-Made numerous magnetically driven pumps to drive liquid. None of them worked to the standard I would like. Exploring other options. Still using 'push pull' pressure driven pump.
March 2018
-First tested Faraday Effect Apparatus (FEA) on light table to measure strength of magnetic fields. This was done as a safegaurd to prevent EMI from creating errant data points on my datalogger. Something that might have happened if I was using wired Hall effect sensors.
-Made ruggidized stand for FEA so that future tests are repeatable.
-Made current supply for FEA
February 2018
-Fernbrake LLC created
-Developed method of making fluid pipes
-Made 'push-pull' driver (mechanically driven) to force liquid through fluid pipes.