TEC Chiller (Tiny Efficient Copeland)

[Hennels]

Well done !

I wish I could get into this sorta stuff but no funding

[Jack]

Some more questions, if you don't mind

Are you using r-134a refrigerant? (I presume you do, since you're using a r-134a compressor.)

How did you charge the unit? To a specific subcooling/superheat? (That the way people do it with direct die coolers, but I don't know if it works the same for chillers?) And for a specific heatload?

I really need to find some time and build one myself. I really like your build!

[DetroitAC]

Yep R-134a.

I did a charge determination like I would do for a car system. I applied a heavy load (500W) and a first guess charge. I operate the system for a little bit and see how it's doing. You can tell right away if you're over or undercharged by looking at superheat and subcool. In my case I was undercharged which is what you want. Then I let the system stabilize for ~2 hours (to fully warm up the oil and compressor shell) and record data of all the parameters. I add 10g refrigerant, and let it stabilize again (maybe 1/2 hour) and record all parameters. Eventually you will find the optimum charge (for your charging conditions only!) which is usually at a good healthy amount of subcooling. You can also do another degree of freedom optimization study by adjusting the TXV superheat bias screw at each charge level, and in theory find the optimum charge and txv setting. With automotive systems there is a bigger influence of the txv setting since the oil is different. I haven't seen much of a differrence with POE oil so I didn't bother to do that, and my TXV size worked out quite perfectly and I never had to adjust it.

Once I had a charge determined for high loads I checked the operation at all loads, and found that I needed more charge for good performance at low loads. I could have added a bigger receiver (I just have the filter/drier), but a little more charge fixed the low load performance quite well, and I'm just a little past the optimum subcool at high loads.

Then I made sure that the capacity control function with the HGBV worked well with the charge I decided on, it worked just fine, so I called it good.

Good question.

[Jack]

Once I had a charge determined for high loads I checked the operation at all loads, and found that I needed more charge for good performance at low loads. I could have added a bigger receiver (I just have the filter/drier), but a little more charge fixed the low load performance quite well, and I'm just a little past the optimum subcool at high loads.

How come you needed more charge at low loads? You should have been overcharged at low loads already?

And a big thanks for the clear explanation!

[DetroitAC]

Well, it's not in the pictures I posted, but the first prototype had an evap that was inclined so the refrigerant would flow uphill. I also tried adding a second evap to see the effect, as well as trying different size HGBV, and some other things.

For this chiller I started testing with the evaps as shown (parallel and level) and found it inferior to my previous results. I bent the pipes so both evaps would be flowing uphill and the performance returned. At higher loads, I assume the vapor velocity is not letting much liquid pool in the evaps, but apparently at lower loads the velocity is so low that there is a substantial amount of liquid in there. It's a bit backwards from how a system should work if it had a more conventional evap, but these little guys are cheap and work just fine.

[Sgrios]

Beautiful work DetroitAC. You have a lot of good information in here, thank you.

[DetroitAC]

Thanks Sgrios, you from A^2? or just going to school there?

Jack, this is how I've got the evaps oriented now. I'm debating changing the piping around again to keep it compact or just leave it and insulate.

[Jack]

Ah I see! It's clear to me now. Thanks!

[Sgrios]

Thanks Sgrios, you from A^2? or just going to school there?

Jack, this is how I've got the evaps oriented now. I'm debating changing the piping around again to keep it compact or just leave it and insulate.

I live in Sterling Heights, going to school at U of M in Ann Arbor, for Pre-Nursing.

[Gautam]

I live in Sterling Heights, going to school at U of M in Ann Arbor, for Pre-Nursing.

You're not the only U-M overclocker on here.