PDA

View Full Version : How cold will it get?







Aerobyrd93
29-01-2015, 08:03 AM
Hi there,
I am an Engineering student working on a project which requires a cold trap. The plan, as it stands at the moment, is to hack open a freezer, or a refrigerator, or whatever I can get my mitts on dirt cheap, and try and bend the evaporator coils around, then shove them into an insulated tube. The idea being that water passed through will condense on the coils, and not go through my system. Thing is, I need these coils to reach a temperature of hopefully about -40F (the lower the better) Without going into cascade refrigeration, does anyone know how this can be done? If I just stuff normal coils out of a refrigerator or something, how cold will it get if left on continuously?

Brian_UK
29-01-2015, 10:18 AM
How cold also depends on the type of refrigerant used.

Thisisausername
30-01-2015, 02:38 AM
I do know that has something to do with it. The new standard, r410a has a boiling point of like -70, but other than that, does anyone know how I can go about solving this issue?

hyperion
30-01-2015, 02:36 PM
Attempting to reshape the reclaimed evaporator coil once you have managed to extract it from the potted foam would be fraught with problems of how to ensure that you bend all of the coils carefully without causing a fracture. The pipe work will have age hardened and is usually fairly thin walled as it is partially supported in place by the potted foam.
The overall length of the coils would need to be matched to the duty of the compressor. Finding a small R410A compressor and operating electrics might also be a challenge.
Most of the pipework may not be suitable for the off cycle pressures when using R410A.
The system would need to be high pressure tested to prove the integrity and with R410A this test pressure can be as high as 41.5bar.
Getting the system to work on a capillary tube would present yet another challenge or else you would need to carefully select a suitable expansion valve to work with R410A. There is a fairly limited choice.
Appreciate that all of the above is not much of a positive help, but it might prevent you have too many false starts on the project.

Thisisausername
30-01-2015, 09:11 PM
Thank you for the response Hyperion, ill keep those pointers in mind! So, now that I'm sufficiently terrified, I wonder if someone might point me to any resources to get me started in the right direction? So far I still have no idea even where to begin.

xxargs
31-01-2015, 03:48 PM
R410A is not suitable for -40 degree C = -40 degree F , it give to high head temperature and short life system and also need to handle real high pressure system.

Better to try system with R290 (propane) or if you can get, R404A (going to banned i EU-country), possible you can find suitable compressors in food or icecream-freeze display cabinet (glass door type) used in restaurants and stores, this have LPB-compressor running on R404A (old) or R290 (new) with better capacity than compressors used in household freeze (using R600a)


Is important to use LBP (low back pressure) compressor and 'cold' (high thermal capacity) refrigerant as R404A, R290 and R600a to avoid high head temperature or equipment wears out and break very quick.

I think you need building pipework from beginning with copper-pipe and braze together ie. try reuse pipe in cabinets is more or less not practicable, is very brittle and breaks easily on bending or reverse bending to straight and mostly build in steel - one thing you possible can reuse is capillary pipe, but dryer need to replace.

R290 and R404A refrigerant can use same compressor either for R290 or R404A with some adjust of capillary pipe length, R600a compressor can only use R600a as refrigerant, not R290, R404A or R134a (also going to banned).

And read lot about how this works and reading on Danfoss site and lot of technical page on this place can be very good idea.

Oh i see - you living in US, so then can be problem to use R600a and R290 depend on some stupid law directly forged from refrigerant company interest in US, if not cannibal some Tom&Jerry freezer box/glass cabinet with R290 as refrigerant - her in Europa is forced to adapt, learn and use HC in household and restaurant machine.

So you choice seems be R134a (old R12) if using house-hold equipment and R404A on commercial freezer.

Thisisausername
01-02-2015, 06:26 AM
Thanks for the response! If that is the case, then perhaps I am wrong in my thinking. Elsewhere I had read that a refrigerant would not go any colder than it's boiling point. none of the refrigerants you listed have ones much lower than -40, while r410a is something like -65 or thereabouts. Also, I've been trying to read up on this stuff, but to be honest, I'm having a tough time getting started, without even knowing where to look.

On a separate note, I did manage to find a freezer available locally which is perhaps within my budget. I wonder if there is any way to know more about how cold I could get it. the spec sheet for it lists r134A (xxarg's suggested coolant) at 300 high, 140 low. Anyone have any ideas? 11953

hyperion
01-02-2015, 05:33 PM
Most domestic freezers are capable of achieving -20C after a period of time. Normally they are thermostatically controlled. If the thermostat was linked out, you might get a cabinet temperature of -25C to -30C at its best. The thermal losses through the relatively thin insulation would be a limiting factor.
From your original post you mention passing water through the insulated tube so that it condenses on the coils. With a coil temperature of -40C it is more likely to freeze onto the tube and coils and just become a block of ice.
Perhaps if you were to elaborate on the end use of the air that you are using in your process, the forum might be able to suggest a more economical method of stripping trhe moisture out of the air.

mad fridgie
01-02-2015, 10:00 PM
You are focusing on the wrong part to start with!
Explain what you are trying to cool (inlet temp and how much) and why!
You may find that you could just use some "dry ice" if this is a relatively short term range of experiments!

Thisisausername
01-02-2015, 10:40 PM
You are focusing on the wrong part to start with!
Explain what you are trying to cool (inlet temp and how much) and why!
You may find that you could just use some "dry ice" if this is a relatively short term range of experiments!

Sure.
I am building a freeze dryer. The part I am building now has a dual purpose inside it, but is primarily a cold trap. I intend to make a tube with the frozen coils inside it to pull moisture out of the air,, which comes from the drying food, so it is not going into my vacuum pump, which doesn't like moisture. Water at the low pressures I'm using freezes at lower than normal temps. I anticipate having a gallon or two condense on the coils, which will be defrosted after the cycle.
Further, I want to use this same refrigeration unit to deep freeze the food (faster wod be better) by circulating the cold air between the trap and the main chamber, which is a large insulated tAnk ( used to be a well- pressure tank)

The biggest constraint is budget (shoestring) so I intend to try and do this with craigslist salvage.

Brian_UK
01-02-2015, 11:05 PM
Also, the 140 and 300 figures are test pressures, not running pressures.

mad fridgie
02-02-2015, 07:40 AM
Sure.
I am building a freeze dryer. The part I am building now has a dual purpose inside it, but is primarily a cold trap. I intend to make a tube with the frozen coils inside it to pull moisture out of the air,, which comes from the drying food, so it is not going into my vacuum pump, which doesn't like moisture. Water at the low pressures I'm using freezes at lower than normal temps. I anticipate having a gallon or two condense on the coils, which will be defrosted after the cycle.
Further, I want to use this same refrigeration unit to deep freeze the food (faster wod be better) by circulating the cold air between the trap and the main chamber, which is a large insulated tAnk ( used to be a well- pressure tank)

The biggest constraint is budget (shoestring) so I intend to try and do this with craigslist salvage.

A freeze dryer on the cheap?
So what are you using to heat the product, and to ensure it stops below its eutectic point?
How are you going control the refrigeration capacity when load varies greatly, original flush to end of process.
The coil not to protect the vacuum pump, it is to dramatically y reduce the vacuum pump size. The coil acts as a high compression ratio device? (density of water vapour to density of ice is a single sublimation process.)
I very much doubt you could build a working freeze dryer, it likey to be just a vacuum drier. (where the products goes above its eutectic point)

xxargs
02-02-2015, 05:52 PM
You can make lower temperature than -43 degree C for example R290 - trick is pumping below atmospheric pressure on suction side and every freezer box with R600a as refrigerant in Europa do it.

LBP-compressors means have high cc suction volume per rev. and build for near atmospheric pressure (134a) and 0.5 bar lower than atmospheric pressure (R600a) on suction side in normal running for freezer box.


If normal freezer cabinet have hard to receive below -20 degree C, is mostly depend heating from outside trough insulation in box coming to balance of cooling capacity on compressor in actual temperature - if make better/thicker insulation, you can receive lower temperature without any 'fix' on compressor system.

Thisisausername
02-02-2015, 07:41 PM
A freeze dryer on the cheap?
So what are you using to heat the product, and to ensure it stops below its eutectic point?
How are you going control the refrigeration capacity when load varies greatly, original flush to end of process.
The coil not to protect the vacuum pump, it is to dramatically y reduce the vacuum pump size. The coil acts as a high compression ratio device? (density of water vapour to density of ice is a single sublimation process.)
I very much doubt you could build a working freeze dryer, it likey to be just a vacuum drier. (where the products goes above its eutectic point)

That pretty much is the idea, yeah.

To answer your other questions, I am using several coils of nichrome, insulated with kapton, to heat the product, which will be located inside the tank on several trays, controlled by a relay board. as for the refrigeration capacity, I would like to have the initial be able to handle a content with a water capacity of a gallon or two (more would be better, depends on what the refrigeration will handle).

As for the coil, it does indeed reduce the pump's size, My pump is about as small as they come, and so far has been able to keep up with a small load, and bring it down to about 30 pascals, measured by capacitance manometer. But the coil also serves to reduce moisture, so as to not crap up the pump by pulling moisture through it.

As for the eutectic point, I have an array of about 18 different temperature and humidity sensors placed in/around the machine, a capacitance manometer, several servo-controlled valves, and windmill gauge for vapor flow. If I can manage to create the conditions (particularly the ones regarding cold), I believe I should be able to control the process quite effectively. At the moment it is all being controlled by an arduino Mega via Labview.

Beyond that, I'd sincerely love to hear of anything I may be missing that would make such a production impossible. Good resources have proven hard to find on every facet of this project, and I've been working on it for quite some time.

Thisisausername
02-02-2015, 08:26 PM
You can make lower temperature than -43 degree C for example R290 - trick is pumping below atmospheric pressure on suction side and every freezer box with R600a as refrigerant in Europa do it.

LBP-compressors means have high cc suction volume per rev. and build for near atmospheric pressure (134a) and 0.5 bar lower than atmospheric pressure (R600a) on suction side in normal running for freezer box.


If normal freezer cabinet have hard to receive below -20 degree C, is mostly depend heating from outside trough insulation in box coming to balance of cooling capacity on compressor in actual temperature - if make better/thicker insulation, you can receive lower temperature without any 'fix' on compressor system.

Hmm, that does explain a few things. I'm certainly going to do my best in terms of insulating it. I guess what I'm currently at a loss on is the kind of device I need to be looking for. Do I need to be looking into making my own system from scratch? if not, is there any kind of small, non-commercial system that would be able to fulfill this sort of requirement? I haven't really found any freezers in the US with sub-atm pressures.

Example: I found a r134a fridge/freezer with a test pressure of 140-400. Is there a way to modify such a system to go colder?

BradC
03-02-2015, 06:26 AM
I just built a cold trap for a vacuum still. I used a Danfoss BD35 compressor and a hydrocarbon blend that works like R12. It only gets to -28C but that is what I designed it for. R134a would work similarly in this application but I can't get any. I didn't want to use straight propane as the compressor is not designed for those maximum pressures. I just wanted to trap any stray noxious vapor before it eats my vacuum pump, so it's not critical.

The key is the heat-load. My evaporator is a 1-3/8" tube, with a 3/16" coil inside it and a 1/2" tube down the middle creating an annular chamber with a 3/16" coil wound through it, The refrigerant enters at the bottom of the annular chamber and is sucked out the top, so it's a flooded evaporator. the 3/16" tube goes to the still, and the 1/2" tube goes to the vacuum pump. Vapor is condensed in the tube and falls into a bottle. The top of the bottle connects to the vacuum via the 1/2" tube. This way any vapor that makes it out of the bottle condenses on the wall of the 1/2" tube that runs up the middle of the evaporator. The evaoprator has 1" of Aeroflex insulation on it so it stays cold. The aeroflex took me from -18C to -28C. If I'm distilling anything with water in it I need to limit the evaporator to about 2C or things freeze up. Usually I'm distilling dichloromethane or relatively pure alcohols, so I don't get freeze up.

If you reduce the heat load, you can reduce the refrigerant flow (longer cap tube or smaller orifice) which will reduce your suction pressure and make the refrigerant evaporate colder. You are limited by how your compressor is cooled and what sort of compression ratios you can get out of it. This unit uses the longest length of the smallest cap tube my supplier had and it's still probably a bit big.

If I went to R290 (propane) and a compressor more suited to that task, I could get it colder. I've done ~-40C with propane and a bigger compressor on another unit. It just depends on what you want to do (and more critically what your heat load is). Most fridge/freezers I've seen run either R134a or R600a and static cooling. Getting either of those much colder than -20C will require such a low mass flow that you'll probably cook the compressor.

With the R290 unit, I used a surplus R404a condensing unit with fan cooling of the compressor, so I can run the suction in almost a vacuum and it just keeps going. The condensing unit cost me about $250 on E-bay brand new. Add a custom evaporator, and capilliary tube and you have a very cheap unit.

Thisisausername
03-02-2015, 06:41 PM
I just built a cold trap for a vacuum still. I used a Danfoss BD35 compressor and a hydrocarbon blend that works like R12. It only gets to -28C but that is what I designed it for. R134a would work similarly in this application but I can't get any. I didn't want to use straight propane as the compressor is not designed for those maximum pressures. I just wanted to trap any stray noxious vapor before it eats my vacuum pump, so it's not critical.

The key is the heat-load. My evaporator is a 1-3/8" tube, with a 3/16" coil inside it and a 1/2" tube down the middle creating an annular chamber with a 3/16" coil wound through it, The refrigerant enters at the bottom of the annular chamber and is sucked out the top, so it's a flooded evaporator. the 3/16" tube goes to the still, and the 1/2" tube goes to the vacuum pump. Vapor is condensed in the tube and falls into a bottle. The top of the bottle connects to the vacuum via the 1/2" tube. This way any vapor that makes it out of the bottle condenses on the wall of the 1/2" tube that runs up the middle of the evaporator. The evaoprator has 1" of Aeroflex insulation on it so it stays cold. The aeroflex took me from -18C to -28C. If I'm distilling anything with water in it I need to limit the evaporator to about 2C or things freeze up. Usually I'm distilling dichloromethane or relatively pure alcohols, so I don't get freeze up.

If you reduce the heat load, you can reduce the refrigerant flow (longer cap tube or smaller orifice) which will reduce your suction pressure and make the refrigerant evaporate colder. You are limited by how your compressor is cooled and what sort of compression ratios you can get out of it. This unit uses the longest length of the smallest cap tube my supplier had and it's still probably a bit big.

If I went to R290 (propane) and a compressor more suited to that task, I could get it colder. I've done ~-40C with propane and a bigger compressor on another unit. It just depends on what you want to do (and more critically what your heat load is). Most fridge/freezers I've seen run either R134a or R600a and static cooling. Getting either of those much colder than -20C will require such a low mass flow that you'll probably cook the compressor.

With the R290 unit, I used a surplus R404a condensing unit with fan cooling of the compressor, so I can run the suction in almost a vacuum and it just keeps going. The condensing unit cost me about $250 on E-bay brand new. Add a custom evaporator, and capilliary tube and you have a very cheap unit.

That's very interesting, Brad. I've been considering the possibility of changing a system to a propane coolant. From what limited knowledge I've been able to glean, it seem happier to go to very cold temps than some others ( I think). Unless I'm mistaken, to get the whole system down to -40 will probably require going down even colder on the coils.

Please someone correct me if I'm wrong, but at the moment I'm envisioning finding the largest compressor I can (like 1/2hp or 1hp or something if I can find it) and simply adding a normal ball valve between the two sides, which could be adjusted as required. This valve, as it got progressively further closed, should make the colder side colder (again, correct me if I'm wrong).

As for filling it, and the amount I should put in to an unknown system, is there a way of calculating this? I would think it would be a simple pv=nrt, but part of the refrigerant is in liquid form, so it seems that would not work. How do I know how much to put in?

BradC
03-02-2015, 11:22 PM
Please someone correct me if I'm wrong, but at the moment I'm envisioning finding the largest compressor I can (like 1/2hp or 1hp or something if I can find it) and simply adding a normal ball valve between the two sides, which could be adjusted as required. This valve, as it got progressively further closed, should make the colder side colder (again, correct me if I'm wrong).

In theory, this will work. Why not do that with a small compressor first to get a feel for how well it works? Hand valve expansion devices work as a demo, but they're not a great way to get results long term.



As for filling it, and the amount I should put in to an unknown system, is there a way of calculating this? I would think it would be a simple pv=nrt, but part of the refrigerant is in liquid form, so it seems that would not work. How do I know how much to put in?

Volume is a good start. You want the evaporator mostly full, the liquid line completely full and enough liquid backed up into the condenser to give you some subcooling. Calculate the volume your tubing holds for each portion and work out the volume, then just calculate for the liquid density of your refrigerant. Start with less than that and tune the charge from there based on your desired operating parameters.

Most small compressors specify a maximum refrigerant charge for the system. Make sure your refrigerant comes in under that (generally not hard).

Thisisausername
04-02-2015, 07:33 AM
In theory, this will work. Why not do that with a small compressor first to get a feel for how well it works? Hand valve expansion devices work as a demo, but they're not a great way to get results long term.



Volume is a good start. You want the evaporator mostly full, the liquid line completely full and enough liquid backed up into the condenser to give you some subcooling. Calculate the volume your tubing holds for each portion and work out the volume, then just calculate for the liquid density of your refrigerant. Start with less than that and tune the charge from there based on your desired operating parameters.

Most small compressors specify a maximum refrigerant charge for the system. Make sure your refrigerant comes in under that (generally not hard).

Okay, a few more questions for you guys :o . What is the downside to the hand valve which renders it impractical? I've heard this elsewhere. Also, as you decrease orifice size, pressure difference grows, system gets colder. I realize this increases strain on the system, does this decrease in size also lower overall cooling power?



As I continue to maintain my scavenge for systems, I wanted to ask. I've been assuming that the larger the system is, the more cooling power will be built into it. Does a freezer have a much larger compressor than a similarly sized fridge?

Also, is it possible to modify a capillary tube by, say, crimping it, in order to Increase deltaP? or would that be unsafe.

My continuing gratitude to all of you, this is a great help on what I consider to be a very important project :)

xxargs
05-02-2015, 06:42 AM
Okay, a few more questions for you guys :o . What is the downside to the hand valve which renders it impractical?



Not stabile over time and temperature of liquid. very sensitive for any oil drops or differense in oil-fraktion solved i refrigerant (coming from compressor), debris and moisture - if system is not very dry and using HC - you need hold moisture lever clear very below 25 ppm (~10 ppm) in HC or is make ice-plug in valve after short time... ie. dry - filter dryer is must to use and compressor is very well vacuumed (below 0.2 mBar steady vakum), if possible in 80 degree C temperature if compressor stored with open pipe to air some time. (don't try running compressor with electrical power to heating up under vacuum pumping, can starting electrical discharge and plasma inside between electrical part and connector and compressor destroys)


Hand valve or orifice is useable in system if liquid refrigerant flow can count in kg/minute as car AC etc.




I've heard this elsewhere. Also, as you decrease orifice size, pressure difference grows, system gets colder. I realize this increases strain on the system, does this decrease in size also lower overall cooling power?



If suction pressure going down depend of lower temperature - also sucked mass of refrigerant going down per compressor rev - also cooling power going down - if you looking compressor data sheet you can mostly see table of cooling power in different temperature and for -40 degree C is many case low as 5-20 Watt compare to -5 degree C is capacity around 200 Watt for fridge or freezer compressor - so to not make overflow of refigerant in evaporator you need take down flow of liquid refrigerant same scale you going down in temperature and why capillary line taken from fridge or freezer is way to short ie. have to much liquid capacity compare needed on temperature around -40 degree C and in critical filled machine, mostly of compressor capacity and working effort going to make flash-gas and completely non use in cooling process.

Flash gas (hissing in capillarity depend of compressed gas from compressor going trough capillary) is completely waste of compressor work and take down compressors potential cooling capacity

Ordinary fridge/freeze you hearing this hiss in periodic intevall and is designed for this (to simple solution of metric device to sacrifice some of compressors potential capacity ) - but if try going to -40 degree C - it's out of working range and hiss is almost constant and small periodic break of silence is time for squirt of some liquid refrigerant going trough capillary and mostly of compressor cooling power in -40 degree C range is going to waste, to only circulate compressed gas.




As I continue to maintain my scavenge for systems, I wanted to ask. I've been assuming that the larger the system is, the more cooling power will be built into it. Does a freezer have a much larger compressor than a similarly sized fridge?




Yes - fridge compressor is designed for around -10 degree C evaporator temperature and not so much heat load.

Freezer have bigger capacity depend of more heat load and lower working temperature - I guess around -25 degre C in evaporator pipe under working to hold -18 degree C in stored food.





Also, is it possible to modify a capillary tube by, say, crimping it, in order to Increase deltaP? or would that be unsafe.




This a possible but if not make uniform shrinking whole way and only of some point of length, going to higher sensitivity of temperature on liquid, moist, oil and debris same way as hand tuned valve and is only one way adjusting ie. you cannot open up again if tightening is going to close

mad fridgie
05-02-2015, 09:33 AM
with sublimation point being below -30C you are going to need lower than -40 SST especially when you have on the evaporator.
To start with I would look at total wastage system first to get the rest of the process going first, the perhaps look for a second hand LT medical cabinet (2 stage cascade), and look at using that gear.

xxargs
05-02-2015, 10:09 AM
Sublimation can already start with -1 degree C of evaporator coil but for good transfer you need very low residue of air in coil area and heat inlet to item under freezer-drying or item cool down to same temperature as evaporator coil and moisture transport stops.

mad fridgie
05-02-2015, 11:17 AM
Sublimation can already start with -1 degree C of evaporator coil but for good transfer you need very low residue of air in coil area and heat inlet to item under freezer-drying or item cool down to same temperature as evaporator coil and moisture transport stops.
at 30 pa, sublimation point (condense) is below -30C, so it will not form on the evaporator at -1C, if heat is added to the product, then the product will sublime (boil) "freeze drying" is not about the process, but keeping the product frozen during the process (cell wall structure in place)
The purpose of the evap in a freeze drier is to convert a high volume into a very small volume. The vacuum pump is to remove non condensables from the chamber, not the water vapour!

Lloydy
05-02-2015, 11:36 AM
Just a quick point, as you are located in the USA, do you have EPA certification to be working with refrigerants.

If not you could be breaking the law and face a $37,500.00/day fine. Please see the link below, sorry for being the scare monger.

http://www.epa.gov/ozone/title6/downloads/Section_608_FactSheet2010.pdf

Thisisausername
05-02-2015, 09:00 PM
at 30 pa, sublimation point (condense) is below -30C, so it will not form on the evaporator at -1C, if heat is added to the product, then the product will sublime (boil) "freeze drying" is not about the process, but keeping the product frozen during the process (cell wall structure in place)
The purpose of the evap in a freeze drier is to convert a high volume into a very small volume. The vacuum pump is to remove non condensables from the chamber, not the water vapour!

You speak the truth, sir. I should mention though, that for the exact reasons you mentioned (given that at 30 pascals, the ice interface temperature is something like -60) The pump will not be running continuously, but will bring the chamber down to something like 200 pascals, where the interface temp is around 0F, and then the isolation valve will close and the pump will shut off. According to the literature, you want a coil temp 20 degrees below the interface temperature to achieve deposition, and have it freeze on the coils. I want it colder than this, however, because the chamber itself needs to be very cold indeed to start with, so I'm shooting for -40 chamber temp. I actually have looked at having a cascade refrigeration unit, and just buying one used, but it was far too expensive even for a scrappy one.

Thisisausername
05-02-2015, 09:09 PM
Just a quick point, as you are located in the USA, do you have EPA certification to be working with refrigerants.

If not you could be breaking the law and face a $37,500.00/day fine. Please see the link below, sorry for being the scare monger.

http://www.epa.gov/ozone/title6/downloads/Section_608_FactSheet2010.pdf

hmm. I do not. I may know people who would however, I'm not sure. At the moment, I believe I intend to go with a propane based system, which does not require an epa license, though the legalities will have to be looked into if I need to vent another system. I guess we'll see; thanks for the heads-up, I'll definitely be looking into that!