2 Stage: Subcooler cascade

[jlm46]

i was reading about this the other day and a few sites mentioned something like for every extra degree you subcool the liquid refrigerant you gain about 0.5% efficiency for your evaporator? several mentioned that the larger the delta between your liquid refrigerant temp and your evaporation temps, the less efficient your system would be

I was curious about this because I use a CPEV instead of cap tube, so it's difficult to get some heat exchange between my suction line and liquid refrigerant before the CPEV -- since mine is a chiller I actually was experimenting with having a small coil before the CPEV that's submerged in the chilled liquid -- as my chilled liquid temps decrease, having this coil subcooling my liquid refrigerant further right before the expansion valve should keep that delta mentioned above smaller

I haven't seen much of a performance increase from it that I can see, but it's fun to experiment : )

[Starkiller42]

Gomeler: I intend to ghetto up a slhx just like that after seeing a few of the pros doing it here. My question would be, how to do that if you're using a TXV, and not cap tube, as your metering device? Could get away with still using a small amount of larger diameter cap tube?

[Clemmaster]

If it's just a subcooler subcooler then it won't change the discharge pressure. Ball park you might gain 10% capacity with r134a subcooling the r507...

And what about starting? If you start the first stage before the second one, the HX temp will be very low and R507c won't wait its pressure to mach with the ambient temperature to condense, it will happen in the HX first and the liquid is "uncompressable" so you won't get any higher discharge pressure...

[n00b 0f l337]

Star,
I'd put the sense bulb after the SLHX.

[AliG]

So if you are going to make a dual cascade, are you going to just get super temps on the cpu, or are you going to add in a gpu evap too?

[DetroitAC]

It's an interesting idea for cpu cooling since you could shut the "subcooler loop" off when the cpu idles, and reduce the capacity of the main stage.

I think it'll work like you are expecting, except that the flow of very subcooled refrigerant in the main loop will be harder to control. If you subcool so far that after expansion your quality is still zero, then I'm not sure how a captube would work (they work by allowing the liquid to flash inside the tube, but if it's not going to flash ) I don't know of a TXV that small since the flowrate will be very tiny. Could be that the proper combination of refrigerants in the two loops will make it controllable, and I suppose that's also how you "select" the cooling power of:
Main stage only
Main stage + subcooler stage

I suppose you could get these two cooling power settings to match a cpu at idle and load ??? That'd be good, but it's a good bit of thermo homework to work it out.

I don't think there are any efficiency gains by doing this, and really no capacity gains vs. a single compressor with twice the displacement. The ability to have two possible cooling power settings, and reasonable efficiency at both settings makes this a nice idea.

[[XC] 2long4u]

Build it so everybody will stop debating.

[Starkiller42]

Thanks for the response, NOL, but I was asking more if using some cap tube to make a slhx would negatively affect the performance of the TXV; ie if the restriction added by the tube would mess up the performance of the TXV significantly.

[teyber]

Build it so everybody will stop debating.

i second it, Im pretty darn sure you will get the results that you want

[n00b 0f l337]

I don't think there are any efficiency gains by doing this, and really no capacity gains vs. a single compressor with twice the displacement. The ability to have two possible cooling power settings, and reasonable efficiency at both settings makes this a nice idea.

And thats the goal, I think I will try this.

[Cronos]

If you had compressors run in parallel (effectively doubling displacement) you could also have an option to run 1 or 2 compressor at a time. This is more effective.

[AliG]

Have you considered getting two or three weak compressors instead just for the silence of them? Because my water dispenser uses a 1/8hp compressor, and a weak condensor, yet the water is chilled definitely way below the ambient temp, not to mention it's completely silent, so in theory, if you do what you have in mind with a few weaker ones, you would lose some performence, but gain silence instead

Would that be a possible option?

I apologize for n00bness, still learning here, we at daclan spend most of our time primarily at tomshardware instead where it's filled with n00bs with the exception of about 20 people

[n00b 0f l337]

Its hard to expect a high load though. I mostly do quad core units now a days, so I tune for 275 with holding capacity to 300 watts. Hard to do with smaller units. I may try parallel, the problem then occurs with tuning. Youd be way overcharged I think with one compressor...

[AliG]

That's cool, as I said I'm new to this, and us at DaClan mainly spend time at tomshardware where it's filled with n00bs

Sorry if this is a bit off topic, but I plan to upgrade my current am2 cpu to a phenom x4, do you think a 1/4hp/1/3hp compressor would be strong enough for one?

[n00b 0f l337]

Yes, more then enough, I"d suggest making your own thread though.

[AliG]

It's fine, I don't plan to build it for another 2 or 3 months anyways, meaning that by that time we'll have seen k10 benchmarks and power consumption anyways

[SoddemFX]

I think i'm right in what i said and what i said before is the best way i can explain it

I did this for you last night NoL. Dont trust the trend line much below 1:5 compression ratio as i think it will go exponential, hitting 100% vol. efficiency as compression ratio becomes 1:1.

With this, Cool Packs "Refrigeration Utilities" and "Refrigerant Properties" you can design your system to give the best performance in whatever system you choose. This will be a weird little system, so try to use the tools.

Be carefull with the motors in those TF3F's, they're rated at 100W, Danfoss usually go upto 20% over this motor output power in input power (which is what you'll measure) for their other r134a compressors. With high suction pressures and high discharge you easily have the ability to toast the motors, so keep an eye on how much power is going in. If it get much higher than maybe ~150W i'd be worried about the lifespan of them.

Classical cascade is the only way you'll keep it reasonable (it'll still be > 120W), with a powered subcooler on a r507 main compressor at high-ish load you'll be way over.

Best of luck

Tom