Some AutoC History...
The history of mixed-refrigerant technology, to my knowledge, starts in 1936 when W.J. Podbielniak of Chicago obtained a U.S. Patent (Patent Number: 2041725). His single-compressor low-temperature refrigeration system contained a mixture of hydrocarbons. Although Freons were not then in common use, he knew that they would work in his system. His invention was notable considering he developed it over 80 years ago. Earlier, in the 1920s, work done on gas separation laid a foundation for mixed refrigerant systems. The same physical principles apply.

It is interesting to note that Mr. Podbielnaik was limited primarily to using hydrocarbon based refrigerants, something that is starting to make a come back due to the increasing restrictions on Freons because of ozone depletion and global warming potential. In my upcoming project, I too will be focusing on using hydrocarbon refrigerants where ever possible.

A Little Theory...
The diagram at the top of this entry shows a very basic AutoC design utilizing as few as two refrigerants in the mixture. This is as simple as it gets for AutoC's, and in most practical real-life designs additional stages and refrigerants are needed. The principle of operation is to separate warmer boiling condensates from the pressurized mixed refrigerant stream in what are called phase separators. These condensates are then evaporated in the low pressure side of the cascade condenser that follows, to further cool the pressurized stream as it passes through. And in the case of our single-stage AutoC, evaporate the higher boiling components that are left at the tail end, within the evaporator itself.

Note: It is quite possible to obtain temperatures approaching liquid nitrogen by chaining together multiple phase separators and cascade condensers, while still be driven off of a single compressor. Of course additional refrigerants would also be required.

My New Project
My AC-4 project will be based on a 2-stage AutoC with an Auxiliary Condenser preceding the 1st Phase Separator, and a Sub-Cooler following the final Cascade. It is the forth in a series of small AutoC's I have prototyped over the last several years, and some good information on these prototypes can be found at: Morphing Air Conditioner into AutoCascade System over at the Xtreme Systems Forum.

In this new project my goal is to build the smallest fully functional -140C AutoC ever. Having a foot print of only 9" x 9". Stay tuned as more information about this project is revealed.

Note: The AC-3 Prototype successfully hit -140C about 3 weeks ago. In doing so, it has served its purpose, and laid the ground work for the AC-4 project. The refrigerant charge research that was done will certainly be integrated into the AC-4 Prototype. And at this time, no further development work is anticipated on the AC-3 Prototype.

With the fabrication, painting, and installation of the hangers out of the way, the Load Controller is now completed, and ready to ship out.


The purpose of the hangers is to allow the Load Controller to be "hung" from the square tube frame of the unit under test. It is something that was traditionally done starting back in the Polycold test area in the mid 90's. The portability aspect of this controller is also good when you only have one such device, and need to use it in several different test bays, or locations (which is the case with my client).

I always prefer having enough Load Controllers to allow for fixed installation in a respective test bay. This way my creaky old back doesn't get abused by having to lug around a heavy steel box (a completed Load Controller is nearly 30 lbs).

And here is a close up of the finished unit's front panel showing the labeling, and also what happens when there is no thermocouple plugged in for the crycoil return temperature. Basically it fail safes to an over temp condition, and disconnects the Load Controller from the test load's heating elements via the internal safety contactor.


Well I hope my client really enjoys their new Load Controller, and it should make their life a lot easier than using a Variac Auto Transformer, volt meter, amp meter, and calculator to derive a wattage reading. Not to mention the inevitable drift that has to be compensated for as the test load heating element's resistance changes as it warms up. With my Load Controller, you can just set it, and forget it (I think Ron Popeil used to say that).

I just got done updating the electrical schematic to better represent what was done on this custom portable version of the older series Heat Load Controller. I call this the older series because it is based upon a no longer to be used method of fabrication (labor intensive), and a no longer to be made Load Controller PCB which is limited to just over 4000 watts. A newer design that is in development will have at least 6000 watt capability, and is designated for 19" rack mounting.

Basically the reason for this build was to use up what was left of my old stock. Of which I only have enough parts to build 2 Load Controllers of this configuration, and they are both spoken for. So no more Load Controllers after this, at least not until I finish the development work on the RLC-6000. When will this happen? Not any time soon, due to other higher priority projects in process.

Since my customer for this portable version wanted to be able to plug it into an already available 3-phase receptacle in their test facility, I had to make a compromise on the required source for the 120 VAC neutral. So don't be surprised when looking at the electrical schematic that you see the ground connection also doubling as the neutral connection. I know it ain't exactly code, but the required amperage at the low voltage selection is relatively low and should be well within the rating of the ground wiring.

View: V1B-Portable_schematic.DXF

6:00pm Update:

Connected and tested the RS232 communications port, calibrated the Temperature Controller, and ran the Load Controller through its paces. Everything checks out OK.

Here's a peek inside at the finished wiring.


And here are a couple of views with the box buttoned up and labels applied.



Tomorrow I'll be fabricating the Hanger Brackets which will be the last thing left to do.

View: V1B-Portable_Hanger_Bracket.DXF

Yesterday turned out to be a real bummer day for me

When I was wiring the 12 VDC power into the Load Controller Board, I had accidentally shifted the location by one pin position without knowing it. This turned out to be quite a catastrophe, since what I had inadvertently done, was to tie the 12 VDC bus into the 5 VDC Logic Supply.


Although there wasn't any smoke, I did end up frying a couple of chips and the LCD display, amounting to about $30.00 in parts. Luckily I didn't burn out the ADC chip, which is $28.00 all by itself (and something that I don't have a spare for).

Anyway after about 2 hours of disassembly and reassembly, I was able to fire up the Load Controller and do a preliminary test with great success.

Today I decided to do something better then black pin stripe tape around the LCD in order to create the look of a bezel. Instead I masked it off and used some textured spray paint to give it a cleaner look (see below).


This really ended up looking a lot better and tends to match in with the rest of the front panel components.


As you can see, we are powered up and operational, and that is the actual ambient temperature reading in my garage this morning (17C = 63F). I've got the controller set point at 25C (for test purposes only, and definitely higher than recommended for the actual application). When the big red indicator lamp is off, we are considered good to go, with the main contactor engaged. However normally I would have the controller set point at -50C, because when it comes to Polycold refrigeration units (which is what the end customer will be testing), being this warm on the cryocoil outlet would be a sign of trouble. But because I am bench testing the Load Controller, I need to have it set above my ambient conditions so that the load output power is connected thru the contactor.

Things left to do:
1) Install RS232 cabling and connector.
2) Fabricate steel Hanger Brackets and attach them to the box.
3) Run a complete set of tests, including exercising all aspects of the RS232 communications.
4) Add some labeling.
5) Clean it up and get it ready to deliver to my customer.

Today I am running a test of a free DXF Viewer on my site. It appears to work fine in FireFox and Goggle Chrome Browsers, but seems to need a missing add-on for MSIE (always something with this one). I thought it would be cool to have some of my CAD drawings online in a better form than just an image. Something more like viewing it in an actual CAD program, thereby allowing zooming, layer selection, and even printing.

Give it a try below (clicking on link opens pop-up window and executes Java DXF-Viewer).

View: V1B-Portable_Front_Layout.DXF

View: V1B-Portable_TBS_Layout.DXF

No updates today, besides the fact that I am up to my ears in wires, and trying to get them all connected to something