Chiller temp vs BTU output

[dannym]

Odd, amateur question.

I have a couple of lasers with water cooling I need to hack together a building chilled water system for. The laser creates a heat load of 2KW, which is 6800 BTU of cooling.

However, the laser is not cooled much below room temp. It cannot go below the dewpoint as condensation would be a catastrophe. Using a car radiator without a refrigerated cooling unit would mean that the laser would operate at ~85F which will significantly decrease its output power and output consistency. So, it needs to be water cooled to 65F. The laser temp can be presumed to be the water temp, need 65F water. The design goal is to have minimum delta-T across the laser, its water inlet and outlet temp should both be near 65F. So there is no point in installing a passive radiator to cool or pre-cool its outlet water, it is already below room temp and a radiator would only warm the water.

I have seen a lot about hacking an AC window unit by pulling out the evap core and getting it dipped in a water tank without cutting the lines, which sounds great. I would get a new R410a unit.

My question is about performance calcs at different conditions. I know BTU output decreases with increased temp differentials. Reduced evap airflow produces very cold air but relatively few BTU. Higher outside temps (higher condenser temp) reduces BTU output for a fixed room temp.

The "normal" case for AC might be a 72F room temp but the evaporator itself might be 55F. In my case the evap temp may be fixed at 65F due to water cooling. Additionally the condenser temp expects ~95F, this unit will be indoors with a 75F condenser inlet temp.

So, this is a critical question since it affects the choice of AC size and sizing the electrical circuit accordingly: how does the BTU performance increase with a 75F condenser airflow and a 65F evap temp? If there's 2 lasers operating, that's 13,600 BTU. Question on my mind is, under these more favorable operating conditions (higher suction temp, lower condenser airflow temp) will, say, a "9000 BTU R410a window unit" pump 13,600 BTU? It would be good to settle this before buying units and spending many hours to modify and test.

[lynn comstock]

The design goal is to have minimum delta-T across the laser, its water inlet and outlet temp should both be near 65 F.

This condition would require near infinite water flow. It ain't gonna happen.
Why isn't this question a Pro forum question?

[dannym]

No idea which forum to put it in. I've done by best to read common performance charts for AC but I'm unclear on how this works.

Water flow doesn't need to be "zero" delta-T across the laser. Maybe +5F (62.5F in,67.5F out) as an upper end, that calcs to 2.8 gpm, that- and much better- is quite achievable, so for general calcs I'm looking at that delta-T as "minimal".

[jayguy]

...However, the laser is not cooled much below room temp. It cannot go below the dewpoint as condensation would be a catastrophe...

I have seen a lot about hacking an AC window unit by pulling out the evap core and getting it dipped in a water tank without cutting the lines, which sounds great....

you have a laser, where condensation would be a catastrophe...why would you "hack" something together from a window shaker? engineer and build the proper unit.

...I know BTU output decreases with increased temp differentials. Reduced evap airflow produces very cold air but relatively few BTU. Higher outside temps (higher condenser temp) reduces BTU output for a fixed room temp. ....

this first statement is very vague as it is not true in all cases, however, i think i know what you are trying to say.

...So, this is a critical question since it affects the choice of AC size and sizing the electrical circuit accordingly: how does the BTU performance increase with a 75F condenser airflow and a 65F evap temp? If there's 2 lasers operating, that's 13,600 BTU. Question on my mind is, under these more favorable operating conditions (higher suction temp, lower condenser airflow temp) will, say, a "9000 BTU R410a window unit" pump 13,600 BTU? It would be good to settle this before buying units and spending many hours to modify and test.

you are going to be very disappointed in your "hack". just because you pick a unit that puts out 9k to 13k btuh doesn't mean that it will "stay on" the entire time providing 65F leaving water temperature...under various circumstances, it may put out more btuh...causing it to cycle. under another set of conditions (slightly different, but enough to cause changes), it will put out less btuh causing your leaving water temperature to be higher. you will need a larger unit (which will cause cycling but will get you to temperature every time) and then you will need controls and lots of water volume (relatively speaking) to help the unit maintain temperature while it is off. if you are asking for engineering help on designing a system, this isn't the place to get it. it would be best to hire an engineering firm to design/build a system for you.

as a side comment, just because you have a 75F room, doesn't mean that you will get a 75F refrigeration condensation point. keep in mind that just because you pick such a point, many compressors are not allowed to run below a certain condensation pressure. you will need to check your compressor documentation for details.

[jayguy]

...Why isn't this question a Pro forum question?

he is not a professional member so he can not post there.

[dannym]

Actually "hacks" of window AC units have been reported to be pretty effective water chillers for aquariums and grow ops. Seem superior to what's on the market for actual mfg chillers. The system is expected to cycle on and off to maintain a constant temp. There is a concern that the required off-time of the compressor will be too long and the water will heat up before it can restart, but with a few gallons of capacity this doesn't seem to be a real issue.

I can get an R134a Neslab CFT-75 chiller, it's rated 2100W (7170BTU/hr) @20C water temp. It consumes 9A@208V (1872W) by mfg info, and that's not starting amps, but I haven't measured it myself. Aside from being a massive, bulky, expensive thing, the 208v line is gonna be an extra difficulty, and, well, this power consumption looks really poor. That's an EER of 3.8, and it's only producing a 20C output. It does include the wattage of the water pump but obviously not wattage of an evap fan. Even if I'm slightly misreading the conditions that spec was taken under, it seems exceptionally poor performance.

When I "hack" something together, I do it right (usually). Part of that is doing everything I can to see that it works on paper before dropping money and time and space into it.

[dannym]

Actually "hacks" of window AC units have been reported to be pretty effective water chillers for aquariums and grow ops. Seem superior to what's on the market for actual mfg chillers. The system is expected to cycle on and off to maintain a constant temp. There is a concern that the required off-time of the compressor will be too long and the water will heat up before it can restart, but with a few gallons of capacity this doesn't seem to be a real issue.

I can get an R134a Neslab CFT-75 chiller, it's rated 2100W (7170BTU/hr) @20C water temp. It consumes 9A@208V (1872W) by mfg info, and that's not starting amps, but I haven't measured it myself. Aside from being a massive, bulky, expensive thing, the 208v line is gonna be an extra difficulty, and, well, this power consumption looks really poor. That's an EER of 3.8, and it's only producing a 20C output, with room-temp air on the condenser. It does include the wattage of the water pump but obviously not wattage of an evap fan. Even if I'm slightly misreading the conditions that spec was taken under, it seems exceptionally poor performance.

When I "hack" something together, I do it right (usually). Part of that is doing everything I can to see that it works on paper before dropping money and time and space into it.

[wolfdog]

As someone who has worked on many laser chillers for at least 5 different brands; you have not provided enough information to make any kind of reasonable answer.

[jayguy]

As someone who has worked on many laser chillers for at least 5 different brands; you have not provided enough information to make any kind of reasonable answer.

i guess my assumption was that this was an important piece of equipment and reliability was important.

[dannym]

Coherent Diamond G100: actually spec is 2.5KW cooling load, not 2KW, although the max power consumption is 48v@50A=2.4KW, and they're about 20% efficient, so 1.92KW is realistically the max waste heat in the system.

Coherent Gem 100L: basically the same

Had a guy say he was running a 100W (output) laser at 50% power (900W cooling load) and his 7 gal capacity dropped 6 deg F in 5 min with a 5000BTU/hr window unit modified into a chiller. If I calc that right that's 7400BTU/hr, performing at 148% of rating. No, I don't take that anecdotal report as gospel, I didn't measure it. I did find some info about more exact R410a performance which says a chiller with room-temp condenser air and somewhat higher evap side temp should perform at 122% of rating.

[Poodle Head Mikey]

To avoid the cycling-too-quick problem just use two units and a (help me here - just that quick I forgot the name! <g>)

What's that thing called which alternates starts to even equipment wear?

But anyway: use two units and start a different one for each cycle.

You will have to substantially 'agitate' the tank water with a force pump. An air coil (especially a comfort cooling coil) immersed in water will act more like a solid cold block than it will like a heat exchanger. The actual surface area will be nowhere near to being active unless you 'blow' water flow through the fins.

PHM
--------

Actually "hacks" of window AC units have been reported to be pretty effective water chillers for aquariums and grow ops. Seem superior to what's on the market for actual mfg chillers. The system is expected to cycle on and off to maintain a constant temp. There is a concern that the required off-time of the compressor will be too long and the water will heat up before it can restart, but with a few gallons of capacity this doesn't seem to be a real issue.

I can get an R134a Neslab CFT-75 chiller, it's rated 2100W (7170BTU/hr) @20C water temp. It consumes 9A@208V (1872W) by mfg info, and that's not starting amps, but I haven't measured it myself. Aside from being a massive, bulky, expensive thing, the 208v line is gonna be an extra difficulty, and, well, this power consumption looks really poor. That's an EER of 3.8, and it's only producing a 20C output. It does include the wattage of the water pump but obviously not wattage of an evap fan. Even if I'm slightly misreading the conditions that spec was taken under, it seems exceptionally poor performance.

When I "hack" something together, I do it right (usually). Part of that is doing everything I can to see that it works on paper before dropping money and time and space into it.

[dannym]

Probably "load leveling" as a term. Yeah, I noticed that in calcs. A window AC commonly uses a 3 min lockout period before restart. So, regardless of size, it's presumed the unit will have more than enough capacity to exceed the load. If there's 2x 2KW loads expected to drop <2F in 3min, the water mass would need to be close to 50 gal. Either 2 units, or one other thought would be to route condenser heat into the water to reduce performance as-needed so it won't ever go below the temp lowpoint and require cycling.

I've seen a youtube hacker run a compressor off a VFD but I don't think that's an option for a single-phase 120v. Actually an inverter-driven mini-split WOULD do the job, but man that's extra pricey. Although... the evap does come as a separate piece, and that might be much easier to make fit into a water heat exchanger before evac-and-charge.

Yeah there'll be plenty of water circulation. You're right, without circulation, water is an exceedingly poor thermal conductor.

[Poodle Head Mikey]

Who are you?
What are your capabilities?
What technical / mechanical abilities do you have / are experienced with?
What do you do for a living now?

PHM
-------

Probably "load leveling" as a term. Yeah, I noticed that in calcs. A window AC commonly uses a 3 min lockout period before restart. So, regardless of size, it's presumed the unit will have more than enough capacity to exceed the load. If there's 2x 2KW loads expected to drop <2F in 3min, the water mass would need to be close to 50 gal. Either 2 units, or one other thought would be to route condenser heat into the water to reduce performance as-needed so it won't ever go below the temp lowpoint and require cycling.

I've seen a youtube hacker run a compressor off a VFD but I don't think that's an option for a single-phase 120v. Actually an inverter-driven mini-split WOULD do the job, but man that's extra pricey. Although... the evap does come as a separate piece, and that might be much easier to make fit into a water heat exchanger before evac-and-charge.

Yeah there'll be plenty of water circulation. You're right, without circulation, water is an exceedingly poor thermal conductor.

[dannym]

Oh I'm an electrical engineer with decent machinist and CNC/laser/fabrication skills. I'm an electronics/software engineer with a Texas semiconductor startup.

I'm always obsessive about doing any measurements and performance calcs I can, so BTU/hr, heat load, measuring the cycling off-time of a window unit and calculating the delta-temp/time, I'm gonna do that. But I also seek to reasonably question the accuracy and real-world applicability of measurements and calcs. You can indeed trick yourself into a wrong conclusion in either the "good idea" or "bad idea" direction. Yet, math remains critical, because wanting something, having good intentions, and having a vision alone isn't what makes technical projects work.

I help run a local makerspace (membership-driven community shop).

[Poodle Head Mikey]

I asked because I started to write more in-depth but then remembered that I am not allowed to help any non-refrigeration-skilled person do anything which may hurt them.

See if you can convince the moderators here to let you have professional-member status. You may be able to work it out as their standards must be very low. Just consider that they allow me to join - and even more startling; to remain. <g>

PHM
-------

Oh I'm an electrical engineer with decent machinist and CNC/laser/fabrication skills. I'm an electronics/software engineer with a Texas semiconductor startup.

I'm always obsessive about doing any measurements and performance calcs I can, so BTU/hr, heat load, measuring the cycling off-time of a window unit and calculating the delta-temp/time, I'm gonna do that. But I also seek to reasonably question the accuracy and real-world applicability of measurements and calcs. You can indeed trick yourself into a wrong conclusion in either the "good idea" or "bad idea" direction. Yet, math remains critical, because wanting something, having good intentions, and having a vision alone isn't what makes technical projects work.

I help run a local makerspace (membership-driven community shop).

[dannym]

I did consider the option of replacing an evap core with a different copper coil in water, but opening up, evac, and recharge is not my current skill set and hard to imagine being worth it. There are people who have done it for wort chillers (brewing). Rather, most people just carefully manipulate the existing evap core on the existing short copper lines and achieve immersion without ever opening the system. Bending the copper line too far and having it buckle and pinch closed would be a concern.

A mini-split system with a pre-filled R410a reservoir that just needs to be evac'ed out and the reservoir opened into the lines, I think that may be practical. Following evac instructions is not that difficult and the equipment is not hard to get. I've done car AC before. Currently not the plan I'm thinking of, and if I did go that way, I'd do more research and planning for the specific task.