What size cooler for gylcol reservoir

[homebrewdude76]

Not sure how I will cool the gylcol yet. Planning on a 14gal connical and want the option to add a second.

Plan on separate pump circuits for each.

Room is 75 in summer and want to go to 65 for ales.

What size cooler should I look for?

 

[homebrewdude76]

I see some people use a dehumidifier or AC unit. Seems AC unit is cheaper. So maybe I get that first and see how large the coil is?

 

[Drumminguy81]

I have an ac unit glycol cooler that I just finished building, I used a 54 quart stainless coleman cooler. Seems to work great but I haven't been able to test it in the heat of summer and I haven't bought glycol yet. But with just water I was able to get from 75 ambient down to 50 so im pretty confident it will work well.

 

[homebrewdude76]

How hard was it to convert the AC unit? Do you have a dedicated temp controller to run the AC to a set water temp?

 

[ajdelange]

There is no need to use a glycol solution unless it it is going to be chilled below freezing. If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing because if it did the air coils would frost up and the device wouldn't be very efficient. Low side design pressures in such devices are set for saturation temperatures of a little above freezing for this reason. Now if the low side pressure drops because of a refrigerant leak then you may see an evaporator at below freezing but then there is a system problem which needs to be fixed.

Systems designed with liquid to liquid exchangers (e.g. Cornelius soda dispenser systems) are a different matter. Their evaporators are designed to run at below freezing temperature and the load side fluid needs to be protected with glycol (preferably the less toxic propylene glycol).

As for the volumes of buffer tanks: this is really easy to compute as 1 BTU raises 1 pound of water (or beer - approximately) 1 °F. After 'mixing' two masses, m1 and m2

m1*T1 + m2*T2 = (m1 + m2)*T3

The masses are really thermal masses but we can consider water and beer close enough to being the same in specific heat and specific gravity that m1 and m2 can even be in gallons so that if we have m1 gallons of beer at T1 and coolant at T2 and want final temperature T3 we would need
m2 = m1*(T1 - T3)/(T3 - T2)
gallons of coolant. For example if we have beer at 85 that we want to cool to 60 and have coolant at 40 we need (85 - 60)/(60 - 40) = 25/20 = 1.25 times more coolant than beer.

That may be of interest for calculating how much coolant (and/or it's temperture) you need to get wort filled into the fermenter down to pitching temperature but the other part of the problem is determining how much you need to keep it cool. To find this out you need to fill your fermenter with a known amount of cold water and set it in the room you are going to use it in at the temperature you expect the room to be. Then you measure the temperature of water and the room over time (try to keep the room at constant temperature). Now multiply the mass of the water (pounds) by the degrees and make a plot against time. You will have a curve that starts steep (when the water is cold) and levels off as it approaches room temperature. What you are looking for is the slope of this line. It is in units of °-pounds/hr (BTU/hr). It is the rate at which beer in your fermenter absorbs heat from the air and, as the curve heels over it is clear that this depends on the temperature difference. You can calculate it at several points along the curve and then divide by the temperature difference (between room and water) at each point. This should give a fairly consistent set of numbers which represent the thermal conductivity of your fermenter (units BTU/hr•°). Once you have that you can calculate the BTU's per hour that you have to dispose of via the coolant system and then use the formula above to calculate the amount of coolant at what temperature it takes to get that.

 

[jddevinn]

I use a standard 48qt cooler and a 5000 btu A/C set at 20°C with a 7°C hysteresis. I can cold crash (3) 14 gallon conicals.

 

[augiedoggy]

There is no need to use a glycol solution unless it it is going to be chilled below freezing. If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing because if it did the air coils would frost up and the device wouldn't be very efficient. Low side design pressures in such devices are set for saturation temperatures of a little above freezing for this reason. Now if the low side pressure drops because of a refrigerant leak then you may see an evaporator at below freezing but then there is a system problem which needs to be fixed.

Systems designed with liquid to liquid exchangers (e.g. Cornelius soda dispenser systems) are a different matter. Their evaporators are designed to run at below freezing temperature and the load side fluid needs to be protected with glycol (preferably the less toxic propylene glycol).

As for the volumes of buffer tanks: this is really easy to compute as 1 BTU raises 1 pound of water (or beer - approximately) 1 °F. After 'mixing' two masses, m1 and m2

m1*T1 + m2*T2 = (m1 + m2)*T3

The masses are really thermal masses but we can consider water and beer close enough to being the same in specific heat and specific gravity that m1 and m2 can even be in gallons so that if we have m1 gallons of beer at T1 and coolant at T2 and want final temperature T3 we would need
m2 = m1*(T1 - T3)/(T3 - T2)
gallons of coolant. For example if we have beer at 85 that we want to cool to 60 and have coolant at 40 we need (85 - 60)/(60 - 40) = 25/20 = 1.25 times more coolant than beer.

That may be of interest for calculating how much coolant (and/or it's temperture) you need to get wort filled into the fermenter down to pitching temperature but the other part of the problem is determining how much you need to keep it cool. To find this out you need to fill your fermenter with a known amount of cold water and set it in the room you are going to use it in at the temperature you expect the room to be. Then you measure the temperature of water and the room over time (try to keep the room at constant temperature). Now multiply the mass of the water (pounds) by the degrees and make a plot against time. You will have a curve that starts steep (when the water is cold) and levels off as it approaches room temperature. What you are looking for is the slope of this line. It is in units of °-pounds/hr (BTU/hr). It is the rate at which beer in your fermenter absorbs heat from the air and, as the curve heels over it is clear that this depends on the temperature difference. You can calculate it at several points along the curve and then divide by the temperature difference (between room and water) at each point. This should give a fairly consistent set of numbers which represent the thermal conductivity of your fermenter (units BTU/hr•°). Once you have that you can calculate the BTU's per hour that you have to dispose of via the coolant system and then use the formula above to calculate the amount of coolant at what temperature it takes to get that.

I use a food grade glycol water mix for its ability to slow bacteria growth in the liquid.

 

[jddevinn]

There is no need to use a glycol solution unless it it is going to be chilled below freezing. If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing because if it did the air coils would frost up and the device wouldn't be very efficient.

This is wrong. The "air" coils are placed in the glycol solution. They will not frost as they are covered in glyco. ! See below image of my system in operation, obviously below water freezing temp.

Low side design pressures in such devices are set for saturation temperatures of a little above freezing for this reason. Now if the low side pressure drops because of a refrigerant leak then you may see an evaporator at below freezing but then there is a system problem which needs to be fixed.
t.

No. Breweries usually operate the glycol mixture in the 25-27°F range, not slightly above water freezing. See attached glycol guide from pro prewer.

View attachment GLYCOLGUIDE.pdf

m1*T1 + m2*T2 = (m1 + m2)*T3

The masses are really thermal masses but we can consider water and beer close enough to being the same in specific heat and specific gravity that m1 and m2 can even be in gallons so that if we have m1 gallons of beer at T1 and coolant at T2 and want final temperature T3 we would need
m2 = m1*(T1 - T3)/(T3 - T2)
gallons of coolant. For example if we have beer at 85 that we want to cool to 60 and have coolant at 40 we need (85 - 60)/(60 - 40) = 25/20 = 1.25 times more coolant than beer.

No. You are assuming that there is no heat removed from the coolant system. The purpose of the chiller is to remove this heat. You definitely do not need more coolant than beer. I run less than 48qt [12 gallon] of glycol mix and have no issues cold crashing 42 gallons of beer at the same time.

To cool 12 gallons of beer from 85°F to 60°F 2,508 BTUs need to be removed from the system. A window A/C at 5,000 BTU/hr can reject this heat in 30 minutes. The reserve simply acts as a thermal buffer.



The size of this thermal buffer would be dependent on the specific heat of the mixture. At a 30% by weight polyethylene glycol mixture and 20°F initial set point. (40°F is too high, see PDF) To raise the temperature of glycol to 60°F each gallon will have a thermal buffer of 320 BTU.

ie To change the temperature from 20°F to 60°F each gallon of glycol mix will absorb 320 BTU. You would need 7.8 gallons of glycol mix (2,580/320) to chill 12 gallons from 85°F to 60°F without the chiller turning on. The glycol chiller will be removing heat from the glycol bath at 5,000 BTU/hr.

That may be of interest for calculating how much coolant (and/or it's temperture) you need to get wort filled into the fermenter down to pitching temperature but the other part of the problem is determining how much you need to keep it cool. To find this out you need to fill your fermenter with a known amount of cold water and set it in the room you are going to use it in at the temperature you expect the room to be. Then you measure the temperature of water and the room over time (try to keep the room at constant temperature). Now multiply the mass of the water (pounds) by the degrees and make a plot against time. You will have a curve that starts steep (when the water is cold) and levels off as it approaches room temperature. What you are looking for is the slope of this line. It is in units of °-pounds/hr (BTU/hr). It is the rate at which beer in your fermenter absorbs heat from the air and, as the curve heels over it is clear that this depends on the temperature difference. You can calculate it at several points along the curve and then divide by the temperature difference (between room and water) at each point. This should give a fairly consistent set of numbers which represent the thermal conductivity of your fermenter (units BTU/hr•°). Once you have that you can calculate the BTU's per hour that you have to dispose of via the coolant system and then use the formula above to calculate the amount of coolant at what temperature it takes to get that.

The holding requirements for heat transfer will be WAY below what is required to change temperatures. No need to calculate this at all unless you just want to for fun. Insulation will help hold temps but is not required, if you don't want to get condensate dripping during a cold crash it is required.

 

[iijakii]

Not sure how I will cool the gylcol yet. Planning on a 14gal connical and want the option to add a second.

Plan on separate pump circuits for each.

Room is 75 in summer and want to go to 65 for ales.

What size cooler should I look for?

I have an 8 gallon reservoir for my two conicals with a 5k btu AC unit. It is more than enough to cold crash etc. Not sure how much smaller you could get away with. Food grade glycol is something stupid like $30gal so keep that in mind. If you are using it for ales only and dont want to cold crash I'd test out water only first.

 

[jddevinn]

I have an 8 gallon reservoir for my two conicals with a 5k btu AC unit. It is more than enough to cold crash etc. Not sure how much smaller you could get away with. Food grade glycol is something stupid like $30gal so keep that in mind. If you are using it for ales only and don't want to cold crash I'd test out water only first.

As long as the evaporator is completely covered while in operation (with cooling coils/jacket/hose and piping filled) no more volume is needed.....the volume is just a buffer to reduce the cycling time of the chiller and reduce cooling times in some situations (ie remove more than 5,000btu/hr temporarily).

 

[ajdelange]

This is wrong. The "air" coils are placed in the glycol solution. They will not frost as they are covered in glyco. I confused

Yes you are. What I said was:

There is no need to use a glycol solution unless it it is going to be chilled below freezing. If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing because if it did the air coils would frost up and the device wouldn't be very efficient. Low side design pressures in such devices are set for saturation temperatures of a little above freezing for this reason.

Read this again more slowly this time.

See below image of my system in operation, obviously below water freezing temp.

And obviously with an evaporator operated below freezing. I also said:

Systems designed with liquid to liquid exchangers (e.g. Cornelius soda dispenser systems) are a different matter. Their evaporators are designed to run at below freezing temperature and the load side fluid needs to be protected with glycol (preferably the less toxic propylene glycol).

Read this.

No. Breweries usually operate the glycol mixture in the 25-27°F range, not slightly above water freezing. See attached glycol guide from pro prewer.

Read it again. "Systems designed with liquid to liquid exhangers..." would include those used in breweries (at least all the ones I've ever looked at does).

No. You are assuming that there is no heat removed from the coolant system.

If the guy is using a reservoir of cold glycol as his source of cold, there isn't.

[qu

The purpose of the chiller is to remove this heat.

What chiller? You should read his post as well as mine. He never said anything about active cooling. He said he didn't know how he would chill it. Lots of guys when they ask this question want to know how much glycol they should put in their chest freezer to do a particular cooling job.

You definitely do not need more coolant than beer. I run less than 48qt [12 gallon] of glycol mix and have no issues cold crashing 42 gallons of beer at the same time.

If in absorbing the required amount of heat from the beer, the volume of coolant you have will rise above or even within say, a couple of degrees of, the desired temperature, you need more coolant. In some cases, depending on volumes and temperatures, you will need more coolant. In an active system it is different. I cool two 2.5 bbl fermenters with a gallon of glycol but they are in an active loop.

To cool 12 gallons of beer from 85°F to 60°F 2,508 BTUs need to be removed from the system.

Approximately.

A window A/C at 5,000 BTU/hr can reject this heat in 30 minutes.

A device which can remove 5000 BTU/h can do that, yes but does your 5,000 BTU A/C do that? My brand new 5 ton heat pump delivers 4.5 tons of heating and that is operating liquid to liquid as it is designed to do. What an actual unit delivers depends on source and load temperatures.

The reserve simply acts as a thermal buffer.

Well, yes, In addressing this question I usually advise people to size the reservoir by the amount of cooling they need quickly as in feeding a wort chiller. If they do that there is usually more than enough to cover keeping a fermenter cool, especially if the fermenter is insulated. As I noted earlier, 1 gal buffer is plenty for my system.

The size of this thermal buffer would be dependent on the specific heat of the mixture.

Yep, but if the guy is asking questions at the level he is asking them at I think 1 BTU/°•lb is good enough.

 

[jddevinn]

You are obviously confused. Of course he is going to have active cooling in the glycol, not just a reserve "magically" at a low initial temperature.

Yes you are. What I said was:


There is no need to use a glycol solution unless it it is going to be chilled below freezing. If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing because if it did the air coils would frost up and the device wouldn't be very efficient. Low side design pressures in such devices are set for saturation temperatures of a little above freezing for this reason.

Read this again more slowly this time.

I don't need to read it any slower. It is still wrong. A window A/C with the evaporator submerged in glycol solution will not freeze the air coils, it will operate below the freezing point of water.

If you are kluging something using a dehumidifier or AC no part of the system is going to go below freezing

And obviously with an evaporator operated below freezing. I also said:

Which you just said I couldn't do??

Read it again. "Systems designed with liquid to liquid exhangers..." would include those used in breweries (at least all the ones I've ever looked at does).

A A/C unit with a submerged evaporat is a liquid to liquid heat exchanger.

If the guy is using a reservoir of cold glycol as his source of cold, there isn't.

What chiller? You should read his post as well as mine. He never said anything about active cooling. He said he didn't know how he would chill it. Lots of guys when they ask this question want to know how much glycol they should put in their chest freezer to do a particular cooling job.
If in absorbing the required amount of heat from the beer, the volume of coolant you have will rise above or even within say, a couple of degrees of, the desired temperature, you need more coolant. In some cases, depending on volumes and temperatures, you will need more coolant. In an active system it is different.

No idea how or why you determined this. He said he hadn't determined his cooling source yet but was looking at reservoir sizes first. He's obviously going to cool the glycol not just have a room temperature cooler full of it. If you aren't cooling the glycol you are just going to operate at room temp. If he's not cooling how are you thinking he gets the initial source of chilled glycol?!?

Plus he says that he is going to get a chiller.

I see some people use a dehumidifier or AC unit. Seems AC unit is cheaper. So maybe I get that first and see how large the coil is?

I cool two 2.5 bbl fermenters with a gallon of glycol but they are in an active loop.

The cooling jackets in a 2.5bbl fermenter is larger than a gallon, no?!?! And that not including the reserve, the heat exchanger and the piping?!?

Approximately.

Ummm.. I showed you the math here but sure whatever "approximately".

A device which can remove 5000 BTU/h can do that, yes but does your 5,000 BTU A/C do that? My brand new 5 ton heat pump delivers 4.5 tons of heating and that is operating liquid to liquid as it is designed to do. What an actual unit delivers depends on source and load temperatures.

Ummm, again no.

"Ton ratings" on A/C and heat pumps refer to how many tons of air the unit can move between standard temperatures in standard conditions in a certain about of time. BTU ratings are the thermal capacity of A/C units and heat pumps they are dependent on the properties vapor compression cycle of the unit, not what they are cooling. Yea an A/C unit operating in an air to evaporator and condenser operation may be restrained to transfer heat from the air due to air movement (and the unit will frost up, instead transferring that energy from water in the air); however, a evaporator submerged in circulating glycol will have no issues transferring this heat all from the glycol.

Well, yes, In addressing this question I usually advise people to size the reservoir by the amount of cooling they need quickly as in feeding a wort chiller. If they do that there is usually more than enough to cover keeping a fermenter cool, especially if the fermenter is insulated. As I noted earlier, 1 gal buffer is plenty for my system.
Yep, but if the guy is asking questions at the level he is asking them at I think 1 BTU/deg is good enough.

So 1 gallon of reserve is enough for your ~60,000 BTU unit you are using on 5bll of beer but homebrewdude here needs 1.25 times the amount of beer that he is cooling???? Guys looking to put coolant in a freezer normally want to put it in a keg not a cooler! I assume he know that he has to cool the glycol, but maybe I'm off base. AJ, I'm not really interested in arguing with you as you don't have a good grasp on the technical design of these systems. (sorry, most of your posts that I see are on chemistry and you obviously have a great handle on that )

Homebrewdude, if you have any other questions about reservoir size, chilling unit (i'd suggest A/C as the most economical), controls or anything else here let me know.... I've built a couple of these

 

[jddevinn]

Photos

 

[ajdelange]

You are obviously confused. Of course he is going to have active cooling in the glycol, not just a reserve "magically" at a low initial temperature.

Here's what he said.

Not sure how I will cool the gylcol yet.

If that says 'Of course he is going to have active cooling in the glycol' then I can't help you.

I don't need to read it any slower. It is still wrong.

Clearly you don't know too much about HVAC. I don't either to be honest but I work on two systems that cools air: a normal air conditioner and an walk in cooler. The normal (central) air conditioner is probably more relevant to the discussion. It's low side saturation temperature is about 34° F. Kept at that level so the coil doesn't frost over. When I arrived here last summer the low side saturation temp was in the 20's. Ooops. An indication that the system was low on refrigerant (there was indeed a pinhole leak where the fuel oil guy dented the suction line with his bowser). Recharge it to the proper level and set the TXV for the proper superheat and the sat. temp is in the 30's - where it belongs. The walk-in system uses a different refrigerant (R-22) and has a receiver so charge isn't important. When the TXV is set for 20 ° superheat the saturation temperature is in the 30's.

A window A/C with the evaporator submerged in glycol solution will not freeze the air coils,

No. it won't because the evaporator is not below freezing.

it will operate below the freezing point of water.

Probably not. At least that's not what it is designed for. But I can't say that with complete authority because I have never put gauges on a window A/C with its coil immersed in liquid. What's the low side pressure? And what's the refrigerant?

Which you just said I couldn't do??

Sorry. I can't figure out what this refers to. AFAIK I never said you couldn't do anything.

A A/C unit with a submerged evaporat is a liquid to liquid heat exchanger.

Yes it is and if the evaporator is operated below freezing the coolant should be protected against freezing if circulation stops (or even probably if it doesn't). My comment was intended to reflect systems that are designed with low side pressures that give evaporator temperatures below freezing. Do they apply to air conditioners repurposed in a way they were never intended to operate? I really don't know. If you immerse the evaporator in a medium other than air at 65 °F then that can't be any worse than operating the unit in 65 °F air unless you are loading it with more heat in which case coil temp will go up. Low side pressure should stay about the same and sat temp, consequently about the same so the coil should stay above freezing.

If he's not cooling how are you thinking he gets the initial source of chilled glycol?!?

As I said in my last post most people who ask this question want to know how big a reservoir of glycol to put in a chest freezer to feed the jacket on a fermenter. Since he said he didn't know how he was going to cool it it is natural to assume that the pre-cooled reservoir is an option - and the easiest one to calculate for sure.

The cooling jackets in a 2.5bbl fermenter is larger than a gallon, no?!?! And that not including the reserve, the heat exchanger and the piping?!?

Don't know. When I replace the glycol seems to me I put in about a gallon. Maybe it's two. Does that change the intent of the comment?

Ummm.. I showed you the math here but sure whatever "approximately".

To cool 12 gallons of beer from 85°F to 60°F 2,508 BTUs

Don't see any math there but if you just said 2500 BTU then I'd have let it go as it is a nit. But if you put 2508 that implies that you have considered specific gravity of the beer and its specific heat where clearly you have not.

Ummm, again no.

"Ton ratings" on A/C and heat pumps refer to how many tons of air the unit can move between standard temperatures in standard conditions in a certain about of time.

Where did you ever get that idea? Tons are BTU/h divided by 12000. This is the number of BTU required to melt a ton of ice divided by 24 hours (which actually gives 11,917 but we round up to 12,000). Thus a ton of airconditioning is about the amount you would get in an ancient system that melted a ton of ice per day. It has nothing to do with air handling. The heat pump system that I referenced in an earlier post pumps hear from one fluid loop to another. No air is involved. Yet it is rated 5 tons.

BTU ratings are the thermal capacity of A/C units and heat pumps they are dependent on the properties vapor compression cycle of the unit, not what they are cooling.

BTU ratings measure the transfer of heat under specified conditions and they definitely depend on what is being cooled or rather on the source and sink temperatures. The Second Law demands this. The reason I see 4.5 ton performance out of my heat pump is because the source temperature is around 40. To get the full 5 tons the source water would have to be closer to 60.

Yea an A/C unit operating in an air to evaporator and condenser operation may be restrained to transfer heat from the air due to air movement (and the unit will frost up, instead transferring that energy from water in the air); however, a evaporator submerged in circulating glycol will have no issues transferring this heat all from the glycol.

No it shouldn't and that is why I am pretty sure temperatures within the evaporator would stay above freezing.

So 1 gallon of reserve is enough for your ~60,000 BTU unit you are using on 5bll of beer but homebrewdude here needs 1.25 times the amount of beer that he is cooling????

Huh? Where did this come from? I cool the two fermenters with a small, probably half ton (never looked) chiller. My house heating cooling unit is rated 5 tons (60,000 BTU/h - apparently you can do the BTU to ton conversion)

AJ, I'm not really interested in arguing with you as you don't have a good grasp on the technical design of these systems.

This from a man who thinks a ton of air conditioning relates to the weight of air and that a system designed to cool air would be designed with an evaporator temperature below freezing!

 

[jddevinn]

AJ, none of my posts are intending to be rude... although I know I come across that way sometimes in technical discussions.

If that says 'Of course he is going to have active cooling in the glycol' then I can't help you.

Second post in this thread.

I see some people use a dehumidifier or AC unit. Seems AC unit is cheaper.

Third post in this thread.

How hard was it to convert the AC unit? Do you have a dedicated temp controller to run the AC to a set water temp?

Seems to me that homebrewdude76 is going to have a cooling source.

Clearly you don't know too much about HVAC. I don't either to be honest but I work on two systems that cools air: a normal air conditioner and an walk in cooler. The normal (central) air conditioner is probably more relevant to the discussion. It's low side saturation temperature is about 34° F. Kept at that level so the coil doesn't frost over.

There is a big difference between a HVAC system that cools air and and a liquid cooler. You have to keep the temperature on a air cooler below above [edit: typo pointed out by AJ] the freezing point of water at the installed location to prevent water vapor in the air from condensing on the evaporator or any of the tubing. A submerged evaporator doesn't have the problem of water condensing. Yes you will need to keep the glycol mixture above its freezing point. A recirculation pump and a set temperature will prevent this freezing, and is already required for this project. I've posted in another thread wiring diagrams to do this with a BCS, and will get to posting on how to do it with multiple STC style controllers eventually.

Probably not. At least that's not what it is designed for. But I can't say that with complete authority because I have never put gauges on a window A/C with its coil immersed in liquid. What's the low side pressure? And what's the refrigerant?

It 100% will. I have shown images in this thread of a window a/c doing this and multiple other people have built similar systems. These units use R-410A. Low side pressure is not specified on any of these units, but several people (including me) have empirically proven that the pressure is high enough that 10°F glycol bath can be obtained and the R-410A will still change phase as required for the vapor compression cycle to operate.

Yes it is and if the evaporator is operated below freezing the coolant should be protected against freezing if circulation stops (or even probably if it doesn't). My comment was intended to reflect systems that are designed with low side pressures that give evaporator temperatures below freezing. Do they apply to air conditioners repurposed in a way they were never intended to operate? I really don't know. If you immerse the evaporator in a medium other than air at 65 °F then that can't be any worse than operating the unit in 65 °F air unless you are loading it with more heat in which case coil temp will go up. Low side pressure should stay about the same and sat temp, consequently about the same so the coil should stay above freezing.

This is overly unnecessary. The system does not need to be custom designed to match the low side pressure with the glycol concentration (freezing point). Commercial glycol chillers do not match these. In the earlier PDF it has a section on selecting glycol concentration, the same chiller would be used for all of these systems.

A pre built glycol chiller is essentially the SAME EXACT THING as a window A/C with a submerged evaporator.

Where did you ever get that idea? Tons are BTU/h divided by 12000. This is the number of BTU required to melt a ton of ice divided by 24 hours (which actually gives 11,917 but we round up to 12,000). Thus a ton of airconditioning is about the amount you would get in an ancient system that melted a ton of ice per day. It has nothing to do with air handling. The heat pump system that I referenced in an earlier post pumps hear from one fluid loop to another. No air is involved. Yet it is rated 5 tons.

I've never worked with Ton units before. I quickly Googled and apparently read a source with wrong information. I apologize.

I see your technician qualifications. I'm a mech engineer with years experience designing and manufacturing some niche equipment requiring intensive thermodynamic specialization per application...... I may misuse an HVAC term but I definitely understand a simple glycol cooling unit.

The big disconnect between us here seems to be cooling method. If homebrewdude76 was to use a reservoir in a fridge then your math is correct...... but it's way off if he is using active cooling, which by the posts in this thread tell me he is.

 

[ajdelange]

Seems to me that homebrewdude76 is going to have a cooling source.

He either is or isn't. That's up to him. Not me or thee. The basic principle that I was trying to put across is that 1 BTU changes the temperature of 1 pound of water or beer by approximately 1 °F. With little more than that one can figure out what is going to happen with or without active cooling.

There is a big difference between a HVAC system that cools air and and a liquid cooler.

Not in principle. The PE curves are the same for my air to air systems and liquid to liquid systems (they both use R410A). The curves for the air to air systems in my cars are the same as the air to liquid system I use to cool my fermenters (both use R134a). Why would there be a difference? As long as the same thermal mass passes the non-refrigerant side of an evaporator or condenser in a given unit of time what difference does it make if it is water or air? All these systems like about 10 ° superheat and about the same amount of subcooling. They can all pump about a ton per horsepower.

You have to keep the temperature on a air cooler below the freezing point of water at the installed location to prevent water vapor in the air from condensing on the evaporator or any of the tubing.

This says you have never seen an air cooling system and I'm sure this is not the case so you just aren't thinking. In an air cooling system the evaporator is kept above freezing so that the condensate on it does not freeze. If it does air flow to the evaporator is blocked and the unit does not cool. In cases (like a freezer) where the evaporator must be below freezing some sort of defrosting is required like that provided by the defroster in a refrigerator or the reversing valve in an air to air heat pump. But in a room A/C, walk in cooler (not freezer), automobile A/C, central A/C, or dehumidifier the coil is kept at above freezing. This is easily verified by, for example, looking up info on how to service one of these units. A quick search this afternoon turned up a question as to what are good high and low side pressures for an auto air conditioner. The given answer was 30 pisg. That's 3 atm absolute and if you look up the R134a table you will find that 3 bar absolute corresponds to a saturated vapor pressure of just a bit above freezing. And yes, the suction line often sweats (depending on superheat, relative humidity...). This is one of the reasons it is insulated (the other being that there is no point in adding heat to the system from the outside air if you can avoid it with a couple of bucks worth of insulation).

It 100% will.

Will what?

I have shown images in this thread of a window a/c doing this and multiple other people have built similar systems.

Still not sure what 'this' is but I suppose it means using a window A/C to serve as a chiller. I have no doubt that it can be done. As I said in a previous post cooling water to 65 wouldn't be, in terms of lo side pressure much different from cooling air to that temperature if the flow rate is such that the heat absorbed is the same.

These units use R-410A. Low side pressure is not specified on any of these units,

You have never put a gauge set on one?
Used as intended low side pressure is going to be between say 107 and 118 for a saturation temperature of 35 to 40 °F (my air to air heat pump runs in that range).

...but several people (including me) have empirically proven that the pressure is high enough that 10°F glycol bath can be obtained and the R-410A will still change phase as required for the vapor compression cycle to operate.

How would you know this if you never attached a gauge set? Besides that you need a pressure low enough to allow 10 °. It's just like water which boils at lower temperature in Denver than it does in Death Valley.

To make things easy lets say you have a superheat of 0 and that a 10 °F bath is attainable with a 10 °F saturation temp. That would require low side pressure to be 62 psig. Supposing your compressor were a typical one with a 3:1 ratio you would then have 186 psig high side pressure for a dew point of 65 °F. Thus, with R410A, if you are holding the condenser at or below 65 °F (is it in the room?) I might believe that you could have a 10 °F bath.

The system does not need to be custom designed to match the low side pressure with the glycol concentration (freezing point).

I don't recall seeing anywhere in this thread a suggestion that it should be. Evaporator low side pressure is chosen to give the desired evaporator operating temperature and that is determined from the other system requirements and trades.

A pre built glycol chiller is essentially the SAME EXACT THING as a window A/C with a submerged evaporator.

Sure. And a double bassoon is essentially the SAME EXACT THING as an oboe.

I've never worked with Ton units before. I quickly Googled and apparently read a source with wrong information. I apologize.

No need to apologize. Just remember that Abraham Lincoln said that the trouble with information garnered from the internet is that it is difficult to verify that it is correct.

I see your technician qualifications.

Don't take those too seriously. They are not particularly hard to get but no one can will sell you CFC's unless you have that first one. They do not mean I have the years of experience required of a good HVAC tech. Of course the real reason I got them was that the guys the HVAC companies kept sending around didn't either.

but I definitely understand a simple glycol cooling unit.

You have made several statements that might cause someone to question that. Perhaps you were just being hasty. For example, you insist that air cooling applications for loads above freezing require an evaporator below freezing. This defies common sense and anyone who knows anything at all about air conditioners knows that it is not true. The picture below shows a reading from an air conditioning heat pump.

The big disconnect between us here seems to be cooling method.

Other than the tons thing, the evaporator temperature thing and your apparently reversed understanding of the relationship between saturated vapor pressure and temperature there is no disconnect. This whole dialogue is because you called many, if not most, of the statements I made in my first post incorrect based on your misunderstanding of the context.

 

[Calichusetts]

 

[ajdelange]

Yes, that looks like me all right but be aware that by responding to these posts I gain insight. This morning I was able to sit down with the R410A PE diagram and sketch out the design for a 1 ton chiller (air or liquid) which I don't think I would have been able to do earlier. I also realized that my suggestion to OP is the answer to a long standing problem I've had in my brewery which is how to get the wort from my wort chiller in the high 60's down to pitching temperature. The answer is to fill one of my fermenters with water and run the chiller on it over night so that it is at 35 or so on brew day and then run that water through a second wort chiller.

 

[jddevinn]

He either is or isn't. That's up to him. Not me or thee. The basic principle that I was trying to put across is that 1 BTU changes the temperature of 1 pound of water or beer by approximately 1 °F. With little more than that one can figure out what is going to happen with or without active cooling.

Except your answer to his question is wrong if he has active cooling. He does not need to have a reserve bigger than his wort.

Not in principle. The PE curves are the same for my air to air systems and liquid to liquid systems (they both use R410A). The curves for the air to air systems in my cars are the same as the air to liquid system I use to cool my fermenters (both use R134a). Why would there be a difference? As long as the same thermal mass passes the non-refrigerant side of an evaporator or condenser in a given unit of time what difference does it make if it is water or air? All these systems like about 10 ° superheat and about the same amount of subcooling. They can all pump about a ton per horsepower.

Yes the Pressure Enthalpy relationship of the R410A is the same; however, the transfer from the evaporator to what you are cooling is way different.

If I disabled the thermostat on an A/C unit and allowed it to run with air on both hx coils the evaporator would ice over and lose the ability to transfer the heat from outside to the R410A. Even though the "goal" is to cool air as much as possible the evaporator must remain above the freezing point of water to prevent icing (or have a defrost cycle.) The evaporator can get MUCH colder than the air it is cooling due to the heat exchange rate between the air and the evaporator; requiring that the evaporator temperature itself be monitored.

With the coils submerged in a recirculated glycol bath the evaporator must only remain above the freezing point of the glycol bath. With the high heat exchange rate between the glycol and the evaporator the evaporator will never get to a temperature below what the reservoir bath is. Therefore a temperature set point of the glycol bath can be used to prevent the evaporator from icing. The set point simply has to be one that is achievable at the condenser temperature or the unit will hit the point that the evaporator is the same temp as the glycol.

This says you have never seen an air cooling system and I'm sure this is not the case so you just aren't thinking. In an air cooling system the evaporator is kept above freezing so that the condensate on it does not freeze. If it does air flow to the evaporator is blocked and the unit does not cool. In cases (like a freezer) where the evaporator must be below freezing some sort of defrosting is required like that provided by the defroster in a refrigerator or the reversing valve in an air to air heat pump. But in a room A/C, walk in cooler (not freezer), automobile A/C, central A/C, or dehumidifier the coil is kept at above freezing. This is easily verified by, for example, looking up info on how to service one of these units. A quick search this afternoon turned up a question as to what are good high and low side pressures for an auto air conditioner. The given answer was 30 pisg. That's 3 atm absolute and if you look up the R134a table you will find that 3 bar absolute corresponds to a saturated vapor pressure of just a bit above freezing. And yes, the suction line often sweats (depending on superheat, relative humidity...). This is one of the reasons it is insulated (the other being that there is no point in adding heat to the system from the outside air if you can avoid it with a couple of bucks worth of insulation).

Yes, that was a below/above typo. I fixed it and credited you for noticing.

I cannot find published pressures on any of these A/C units. However, some searching gives me a couple sources that say for R410A low side is typically around (ish)

Low side (evaporator) is typically set for ambient -~20°F or ~40°F or so. (right above water vapor freezing) This corresponds to ~120psi.

High side (condenser) is typically set for ambient +~20°F or ~100°F or so. This corresponds to ~335psi.

However, this is just the design pressures and temperatures. While it is in the service manual that does not mean that is how the system will operate in other conditions, with the same compression ratio. In operation as a glycol cooler the condenser should be much cooler than 100°F.

Still not sure what 'this' is but I suppose it means using a window A/C to serve as a chiller. I have no doubt that it can be done. As I said in a previous post cooling water to 65 wouldn't be, in terms of lo side pressure much different from cooling air to that temperature if the flow rate is such that the heat absorbed is the same.

You have never put a gauge set on one?
Used as intended low side pressure is going to be between say 107 and 118 for a saturation temperature of 35 to 40 °F (my air to air heat pump runs in that range).


How would you know this if you never attached a gauge set? Besides that you need a pressure low enough to allow 10 °. It's just like water which boils at lower temperature in Denver than it does in Death Valley.

To make things easy lets say you have a superheat of 0 and that a 10 °F bath is attainable with a 10 °F saturation temp. That would require low side pressure to be 62 psig. Supposing your compressor were a typical one with a 3:1 ratio you would then have 186 psig high side pressure for a dew point of 65 °F. Thus, with R410A, if you are holding the condenser at or below 65 °F (is it in the room?) I might believe that you could have a 10 °F bath.

How would I know? Well the design pressures on these units aren't published. Instead of buying a unit, getting gauges, measuring pressures, calculating what it can do.... I had to buy the unit to do all that measuring anyway, so I just put it in service. I know that it can hold a 10°F bath with 70°F + ambient because I have run the system to test in these conditions.

I don't recall seeing anywhere in this thread a suggestion that it should be. Evaporator low side pressure is chosen to give the desired evaporator operating temperature and that is determined from the other system requirements and trades.

Sure. And a double bassoon is essentially the SAME EXACT THING as an oboe.

Right, but you are telling me I can't use a A/C for a glycol chiller.

From what I can tell glycol chillers use 134a or 410a in a similar compression ratio to an A/C. Condenser and evaporator coil design LOOK similar. I can't find anything on their compression ratios but am GUESSING they are similar. Yes they are purpose built but they are 10-20x the cost.

In regards to homebrewdude76 questions:

Not sure how I will cool the gylcol yet. Planning on a 14gal connical and want the option to add a second.

Plan on separate pump circuits for each.

Room is 75 in summer and want to go to 65 for ales.

What size cooler should I look for?

Assuming you are using a cooling source a standard sized 48qt cooler is more than sufficient. The larger the reserve you have the more heat you can move to the glycol at a time before the chiller turns on. However, the heat still has to be moved at some point. The reserve is basically a buffer that allows you to temporarily cool at a higher exchange rate than the chilling source can support.

I see some people use a dehumidifier or AC unit. Seems AC unit is cheaper. So maybe I get that first and see how large the coil is?

This is what I would suggest that you do. 5,000 btu window A/C condenser coils fit into the 48qt cooler very well.

How hard was it to convert the AC unit? Do you have a dedicated temp controller to run the AC to a set water temp?

The A/C was very easy to convert. If you buy one and have issues, post the wiring diagram on the interior of the unit and we will help you with the wiring. A manual controlled unit is easier than a digital one.

Yes, a temp controller needs to control the A/C and you will need to circulate the glycol. You can use an individual pump for this or valves. I used valves.

 

[jddevinn]

Yes, that looks like me all right but be aware that by responding to these posts I gain insight.

Yea, I don't mind a bit of internet arguing... gets me to think about a couple things I may not have before.

This morning I was able to sit down with the R410A PE diagram and sketch out the design for a 1 ton chiller (air or liquid) which I don't think I would have been able to do earlier. I also realized that my suggestion to OP is the answer to a long standing problem I've had in my brewery which is how to get the wort from my wort chiller in the high 60's down to pitching temperature. The answer is to fill one of my fermenters with water and run the chiller on it over night so that it is at 35 or so on brew day and then run that water through a second wort chiller.

[/quote]

This is a great idea!!! I've been using tap water, then a little ice water then transferring to the conical to get the last 10°F or so.

Making a couple assumptions. If I cool 12 gallons of wort to 85°F with tap water, I should be able to get down to 60°F with 12 gallons water at 35°F. Hopefully with a single pass goes from the conical to the drain.

Only concern is that the water may freeze around the conical coils with 20°F glycol. JP with jackets you wouldn't have this potential issue.

I thought about expanding my glycol reservoir for cooling, but once I calculated the volume required the glycol itself is too expensive compared to the benefit.

 

[WPStrassburg]

The condensing unit that I have puts out just shy of 4000 btu(1/3 ton) when I'm running it at a 15* suction for crash cooling. Normally when it's running at 45* suction I have over 7000 btu available. With a 15 gallon cooler reservoir maintaining a 68* ferm temp for a 15 gallon batch in an 80* garage the condensing unit was running about 15 minutes every 3 hours. So it has plenty of capacity to maintain ferm temp. For crash cooling once it was down to temp it runs about 20 minutes every 2 hours.

 

[ajdelange]

Yes the Pressure Enthalpy relationship of the R410A is the same; however, the transfer from the evaporator to what you are cooling is way different.

That's why I qualified what I said with:

As long as the same thermal mass passes the non-refrigerant side of an evaporator or condenser in a given unit of time what difference does it make if it is water or air?

You really need to read the posts to see what people are saying.

If I disabled the thermostat ....

Yes. Not sure what the point is here.

requiring that the evaporator temperature itself be monitored.

That's what a TXV is for.

With the high heat exchange rate between the glycol and the evaporator the evaporator will never get to a temperature below what the reservoir bath is.

Not so. For example my R410A liquid to liquid heat pump runs a super heat of 10 -12 °F. See the chart below. In case you aren't familiar with the term 'super heat' it is the difference between the saturation temperature of the gas leaving the exchanger and the actual temperature of the gas leaving the exchanger. As any real world exchanger has a finite approach the 'bath' temperature is appreciably higher than the saturation temperature. In this case, the coil is 30 ° colder than the 'bath'.

Therefore a temperature set point of the glycol bath can be used to prevent the evaporator from icing. The set point simply has to be one that is achievable at the condenser temperature or the unit will hit the point that the evaporator is the same temp as the glycol.

In a properly designed system the aquastat is set in accordance with the requirements of the load. The evaporator will probably, as is the case in the example, run much colder than that with the actual temperature determined by the refrigerant and available sink temperature (which determines hi side pressure which determines low side pressure which determines low side saturation temperature).

Yes, that was a below/above typo. I fixed it and credited you for noticing.

Whatever it was I am glad you are finally clear on how these things work.

I cannot find published pressures on any of these A/C units.

I wasn't looking for verification of what typical R410A pressure would be. I have plenty of data on that. The readings I posted yesterday are typical of an air to air system. I chose the data plotted here as it is for a liquid to liquid system. Note that the hi side pressure is only 210 for a sat temp of a little over 70 °F with, consequently, a lo side pressure of only 80 psig and, consequently, a saturation temperature of around 20 °F as discussed.

The reason I asked was because I am always interested in this sort of stuff and wondered what the actual performance numbers for your system would be. I am very curious about one thing. Apparently you have disconnected the evaporator from a home A/C unit and replumbed it into a cooler. This implies that you have a recovery machine, vacuum pump, scales etc. but you don't have a gauge set? If you don't have the other stuff either how did you do the work?

In operation as a glycol cooler the condenser should be much cooler than 100°F.

It doesn't have to be but if you can get it its nice as Second Law says the colder the condenser the more efficient the machine is going to be. This is why I asked about where the condenser was located in my previous post. I looked at the pictures and apparently it is in the room with the other gear. If that room is air conditioned otherwise then you can have the condenser at whatever temperature that room is being held. I do have to interject here the story of a facilities guy at a company that I worked at who put a window air conditioner in a small windowless (just hung it from the ceiling) room (to get rid of the heat from a STU II) and wondered why the room got hotter when he turned it on.

I know that it can hold a 10°F bath with 70°F + ambient because I have run the system to test in these conditions.

I gave you the calculations that show that this should be the case in the last post and wanted to know if the condenser was at 70 °F which would be required for such a low evaporator temperature. Apparently that is the case.

Right, but you are telling me I can't use a A/C for a glycol chiller.

This is about the third time you are telling me that I told you that. I never in fact said that. It is a kluge, yes, but kluges often serve when a proper solution is not available (and I always note that the word means 'clever' in German). I always prefer something designed for the application over a kluge. If I needed a glycol chiller I'd go get an old Cornelius glycol chiller (which is in fact exactly what I did). But I have been the creator of many a kluge both in my professional and hobby lives. If I wanted to build something for the fun/challenge of building it I'd get one of the Hearts wort chillers which is really a refrigeration heat exchanger and plumb that into the output side of the compressor.

From what I can tell glycol chillers use 134a or 410a in a similar compression ratio to an A/C.

A compressor will do what a compressor can do and that's usually in the 3:1 ballpark.

Condenser and evaporator coil design LOOK similar.

In an air to liquid system the condenser is going to be the same as the condenser used in an air conditioner or heat pump. Its job is to transfer heat from refrigerant to air and it doesn't care how the refrigerant got hot. In a liquid to liquid system the exchanger is obviously going to be a refrigerant to water exchanger. As I said above the Hearts wort chiller (http://heartshomebrew.com/index.php?main_page=product_info&cPath=17_213&products_id=878) is in fact such an exchanger. On the evaporator side one can use something like the Hearts feading a reservoir or a refrigerant line (finned or not) submerged in a reservoir. I can't really see into my Cornelius but it looks like nothing more than copper tubing weaving back and forth in the tank. I've never looked into the Remcor as the reservoir is all sealed up in insulation.

I can't find anything on their compression ratios but am GUESSING they are similar.

About 3:1

 

[ajdelange]

The condensing unit that I have puts out just shy of 4000 btu(1/3 ton) when I'm running it at a 15* suction for crash cooling. Normally when it's running at 45* suction I have over 7000 btu available. With a 15 gallon cooler reservoir maintaining a 68* ferm temp for a 15 gallon batch in an 80* garage the condensing unit was running about 15 minutes every 3 hours. So it has plenty of capacity to maintain ferm temp. For crash cooling once it was down to temp it runs about 20 minutes every 2 hours.

How are you controlling suction temp? If it's by cranking in the TXV (or starving the evaporator some other way) then naturally the capacity is going to go down (when you need it the most). Try blowing an external fan through the condenser. This might allow you to get a lower temperature with lower super heat (and more refrigerant to the evaporator).

 

[helterscelter]

This is awesome information

 

[jddevinn]

That's why I qualified what I said with:

As long as the same thermal mass passes the non-refrigerant side of an evaporator or condenser in a given unit of time what difference does it make if it is water or air?

You really need to read the posts to see what people are saying.

Not so. For example my R410A liquid to liquid heat pump runs a super heat of 10 -12 °F. See the chart below. In case you aren't familiar with the term 'super heat' it is the difference between the saturation temperature of the gas leaving the exchanger and the actual temperature of the gas leaving the exchanger. As any real world exchanger has a finite approach the 'bath' temperature is appreciably higher than the saturation temperature. In this case, the coil is 30 ° colder than the 'bath'.

OK, lets think about this for a second, using an A/C as a glycol coolant. The difference between water and air is heat transfer rate, and in the context that we are discussing the actual temperture of the liquid.

Using your R410A example readings from the other day Your low side pressure is 119psig (w 14.7psia atmospheric) your vapor saturation temperature ~134psia at is 40.1°F and your coolant temp is 52.7°F so a superheat of 12.6°F. I assume these are somewhere close to the design parameters and info that you see in the manuals.

The evaporator transfers energy from what it is cooling by first absorbing the latent heat needed for the phase change that it is forced to take by the pressure differential, then by sensible heat caused by the superheating. The sensible heat transfer is only because the coolant is at a lower temperature than what it is cooling.

ok.

Let's now put the same evaporator in a glycol bath. For easy numbers we are going to set this bath set temperature at 30°F and a starting temp of 60°F. ie the system will run until the bath temperature is 30°F then it will shut off.

When the system starts it will operate as you are expecting (assuming the evaporator is within operating temp), 134 psia 52.7°F coolant.

However, once the glycol reserve temperature reaches the coolant (supersaturated) temperature of 52.7°F the second law of thermodynamics tells us that no more heat can flow from the glycol to the coolant. As the temperature of the glycol continues to decrease in temperature the coolant will superheat to a lesser extent.

When (if it can) the glycol solution reaches the setpoint of 30°F the low side pressure will no longer be the 119psig design and the discharge will no longer be superheated without violating the second law of thermo.

In a properly designed system the aquastat is set in accordance with the requirements of the load. The evaporator will probably, as is the case in the example, run much colder than that with the actual temperature determined by the refrigerant and available sink temperature (which determines hi side pressure which determines low side pressure which determines low side saturation temperature).

Yes, this will push the unit off design conditions (more than normal operation in air would). As long as the condenser is in a relatively "normal" temperature the system pressures are now influenced by the evaporator.

The liquid-liquid graph you show at the bottom of your last post is a differently designed system than the air-air we were previously discussing, at (what I assume without axis labels) 80psig subcooling the saturation temp is right below 0°F (assuming units again) and the ~15°F superheat puts the condenser at a lower temperature than a bath set at 30°F. Setting the temperature lower than the superheat would cause the same evaporator = bath condition as the above A/C Example.

I wasn't looking for verification of what typical R410A pressure would be. I have plenty of data on that. The readings I posted yesterday are typical of an air to air system. I chose the data plotted here as it is for a liquid to liquid system. Not that the hi side pressure is only 210 for a sat temp of a little over 70 °F with, consequently, a lo side pressure of only 80 psig and, consequently, a saturation temperature of around 20 °F as discussed.

The reason I asked was because I am always interested in this sort of stuff and wondered what the actual performance numbers for your system would be.

Yep, and this is where we get to what I think we are disagreeing/discussing about and what I honestly hadn't thought about in detail before. At the low cost of the A/C unit my math was "buy it and see if it works"

Yes a properly designed glycol cooler looks like it would have the same 3:1 compression ratio of a A/C; however both low and high side design pressures are lower than the A/C. As such the system can cool to a lower temperature, but requires a lower condenser side temp.

Seeing this math laid out makes me realize that someone using an A/C system or a purpose built glycol system in an unconditioned area may have issues with low temperature operation. It also makes me realize that the efficiency loss of my system at low set temperatures (were I keep it) is much greater than I was estimating (I knew I was shifting the "curve" but didn't realize that I got rid of all the superheat and was operating so far below the design Low pressure point).

When I get some time at some point I may dive into this deeper as many people are doing the repurposed a/c and dehumidifiers.

I am very curious about one thing. Apparently you have disconnected the evaporator from a home A/C unit and re plumbed it into a cooler. This implies that you have a recovery machine, vacuum pump, scales etc. but you don't have a gauge set? If you don't have the other stuff either how did you do the work?

Nope, no HVAC tools. I just bent the pliable copper lines to fit where I wanted without evacuating the system.

It doesn't have to be but if you can get it its nice as Second Law says the colder the condenser the more efficient the machine is going to be. This is why I asked about where the condenser was located in my previous post. I looked at the pictures and apparently it is in the room with the other gear. If that room is air conditioned otherwise then you can have the condenser at whatever temperature that room is being held. I do have to interject here the story of a facilities guy at a company that I worked at who put a window air conditioner in a small windowless (just hung it from the ceiling) room (to get rid of the heat from a STU II) and wondered why the room got hotter when he turned it on.

Yes my brew area is insulated and definitely has a heat input from the unit running.

This is about the third time you are telling me that I told you that. I never in fact said that. It is a kluge, yes, but kluges often serve when a proper solution is not available (and I always note that the word means 'clever' in German). I always prefer something designed for the application over a kluge. If I needed a glycol chiller I'd go get an old Cornelius glycol chiller (which is in fact exactly what I did). But I have been the creator of many a kluge both in my professional and hobby lives. If I wanted to build something for the fun/challenge of building it I'd get one of the Hearts wort chillers which is really a refrigeration heat exchanger and plumb that into the output side of the compressor.

Purpose built is great.... unless I can do the same thing for 10x less!! A 5,000 BTU chiller is overkill for (3) 12 gallon batches, and apparently so is whatever BTU my A/C is actually working at!

 

[jddevinn]

How are you controlling suction temp? If it's by cranking in the TXV (or starving the evaporator some other way) then naturally the capacity is going to go down (when you need it the most). Try blowing an external fan through the condenser. This might allow you to get a lower temperature with lower super heat (and more refrigerant to the evaporator).

Sounds to me he is getting the temperatures he wants.

I hate to assume But I would guess at 15°F he is not changing the vapor-compression cycle directly but simply operating off design enough that he has dropped BTU rating the same way I am.

 

[WPStrassburg]

Not controlling the suction directly, just changing the control temp for the liquid line solenoid. I've got plenty of capacity for what I need even when crashing so based on calc'd numbers and real life performance the 4-6000 btu ac units that people are using seem to be a good fit for one or more conicals. Certainly could go smaller based on the op's desired ferm temps and load, but ac units are pretty similar in price in this size range.

 

[ajdelange]

The difference between water and air is heat transfer rate,

The difference between air and water is that the specific heat of water is much greater than that of air. The amount of heat transferred depends on the thermal mass which passes the exchanger per unit time, the temperature difference between the refrigerant and the water or air, and the approach (conductivity, length) of the heat exchanger. To transfer the same amount of heat per unit time, ceteris paribus, requires a larger volume of air both because of its lower Cp and because it may, in addition to the thermal heat contain water vapor (latent heat) which also has to be removed.

..and in the context that we are discussing the actual temperature of the liquid.

It is the temperature difference which drives the transfer.

Using your R410A example readings from the other day Your low side pressure is 119psig (w 14.7psia atmospheric) your vapor saturation temperature ~134psia at is 40.1°F and your coolant temp is 52.7°F so a superheat of 12.6°F.

Well the refrigerant boils at 40.1, the gas leaving the evaporator is at 52.7 for the 12.6° superheat and air temperature in the plenum was probably 75 °F or so coming in and a couple of degrees lower than that coming out.

I assume these are somewhere close to the design parameters and info that you see in the manuals.

Well those are pretty typical numbers. These days techs are taught (I think) to charge to a specified weight of refrigerant and the only way to do that is recover it all, weigh it, and then put it back plus whatever you come up short. How many do you think are willing to do that and are you, the customer willing to pay for that? Consequently they usually just charge to 10 -12 ° superheat. Some systems have tables (and their are calculators) that tell you what superheat to set given outdoor dry bulb and indoor dry and wet bulb readings. So yes, these numbers are consistent with current practice (AFAIK). I'm not a refrigeration tech or refrigeration engineer.

The evaporator transfers energy from what it is cooling by first absorbing the latent heat needed for the phase change that it is forced to take by the pressure differential, then by sensible heat caused by the superheating. The sensible heat transfer is only because the coolant is at a lower temperature than what it is cooling.

If you put a skillet on the stove and let it get nice and hot and then drop a teaspoonful of water on it the water will boil instantly, but not flash unless the pan is really hot. In so doing, it absorbs the enthaloy of vaporization for water and in so doing cools the pan. If the vapor then passes over other hot parts of the pan it, being at 212 °F (all at sea level here) which haven't been cooled below 212 °F, the vapor will continue to warm. This is the super heat.

In a refrigeration system you squirt warm (but below its boiling point at the high side pressure) refrigerant into a tube at a lower pressure, low enough that it is well above the boiling point. So it boils. Again it does not flash (or shouldn't in a proper design) and again it absorbs the latent heat of vaporization from the other side of the heat exchanger. If, in so doing, it cools the walls of the heat exchanger below the boiling point of the liquid it will stop boiling and the evaporator will flood. This happens if there is insufficient cooling load (e.g. you are pumping a half ton's worth of refrigerant into an evaporator in a cold room that leaks 1/16 ton of air from outside. The obvious solution in such a case is the admit less refrigerant but clearly if you do that you will absorb less heat which is fine in this case because there is only 1/16th of a ton of heat to dispose of.

If the refrigerant is throttled back it will all boil in the first part of the evaporator and vapor, at the boiling point of the fluid, will pass through the rest of the evaporator continuing to warm as the air in the chamber is typically quite a bit warmer than the boiling point of the refrigerant. This is the superheat. It is clear that if you have too much superheat the system doesn't produce as much cooling as it could and if you have too little you are also produce less (as the enthalpy of vaporization is larger than the heat required to warm the gas). Thus controlling superheat is very important in a system that is properly designed and modern systems contain TXVs to maintain superheat within a desired range. The system continues to operate efficiently despite variations in cooling load and you won't be slugging your compressor with liquid refrigerant - or you wouldn't be if the techs understood this.

ok.

OK - that's Refrigeration 101.

Let's now put the same evaporator in a glycol bath. For easy numbers we are going to set this bath set temperature at 30°F and a starting temp of 60°F. ie the system will run until the bath temperature is 30°F then it will shut off.

What's going to happen is mostly driven by the condenser. Assuming that you have 75 ° air to blow over it and that you can blow enough to get a 5 ° approach then the liquid would leave it at 80 °F and with 10 ° subcooling (a typical number) you would have a saturation temperature of about 90 °F corresponding to 275 psig high side pressure which is 290 psia. Assuming the usual 3:1 compression ratio low side pressure would be 97 psia or 82 psig corresponding to 22 °F low side saturation. With 10 ° superheat and 5° approach you would never get to 30 °F with a nominal load. But if you reduce the load e.g. by slowing the circulation of the bath water you probably could get down that low. If you system has a TXV it might thwart you though. As superheat gets high it will try to fix that by admitting more refrigerant.

When the system starts it will operate as you are expecting (assuming the evaporator is within operating temp), 134 psia 52.7°F coolant.

Just to be clear the 52.7° is the temperature of the gas leaving the evaporator. As I said above the temperature of the air leaving the evaporator was probably more like 70 and the 'coolant', i.e. the air flowing over the condenser was in the 80's (it's hard to test this stuff in Canada).

However, once the glycol reserve temperature reaches the coolant (supersaturated) temperature of 52.7°F

That system had a condenser temperature in the 80's. It would never, normally operated, reach 30 °F. But you could make it do so by removing refrigerant (antoher way to starve the evaporator). It would be terribly inefficient (so that you would have to lighten the load) but I don't think anyone cares about that here. It would bother me to operate it well outside its design parameters but if the attitude is 'run it until it goes TU' then why not?

..the second law of thermodynamics tells us that no more heat can flow from the glycol to the coolant.

You are going to think me terribly pedantic but it is the Zeroth law that is at play here. The Second Law implies that it is harder (takes more work) to pump a joule of heat from 20 ° to 70 ° than from 40 ° to 70 °.

As the temperature of the glycol continues to decrease in temperature the coolant will superheat to a lesser extent.

As you starve the evaporator, which you will have to do if you can't get the condenser cooled enough, the superheat will actually go up. But that is a good thing. As superheat gets low you start to run the risk of slugging the compressor. This is why it is, IMO, important that you get a gauge set and see what is actually going on in your system(s?). It will cost you about as much as a decent pH meter ($150 or so). All this speculation on your part and mine is really meaningless when stacked up against real data.

When (if it can) the glycol solution reaches the setpoint of 30°F the low side pressure will no longer be the 119psig design and the discharge will no longer be superheated without violating the second law of thermo.

Have a look at the heat pump data again. There we have evaporator out gas at about 30 °F which could, with a good heat exchanger, get us close to 30 °F on the liquid side. In the case of the actual heat pump the reservoir is at 50 °F and were getting that by being able to cool the condenser output liquid to around 60 °F. Again, I think you want the zeroth law. The second law speaks to the efficiency of a system.

The liquid-liquid graph you show at the bottom of your last post is a differently designed system than the air-air we were previously discussing, at (what I assume without axis labels) 80psig subcooling the saturation temp is right below 0°F (assuming units again) and the ~15°F superheat puts the condenser at a lower temperature than a bath set at 30°F.

Yes, sorry about not labeling the axes. I do enough of these that I don't have to and I meant to put the labels on this one but forgot. As you probably figured out the left axis is pressures (gauge) and the right temperatures in °F.

For starters, the unit is cycling. There are two on periods and one off which is clearly demarcated by the hi and low pressures coming together.

No one would design a system to operate at 80° subcooling. That is indicated on the right hand scale and is about 10 ° as is the superheat. The low side boiling point is about 20 °F.

Setting the temperature lower than the superheat would cause the same evaporator = bath condition as the above A/C Example.

I'm not sure what you mean by this. A super heat of 10 - 12 is pretty common and you could have, in a walk in freezer, for example, a low side temperature of less than this. To reiterate, the superheat is about 12, the low side temperature about 32 and the low side saturation temperature about 20.

Yep, and this is where we get to what I think we are disagreeing/discussing

It is, for the most part, discussion. You have/had some misconceptions but they are, I think, pretty much cleared up.

Yes a properly designed glycol cooler looks like it would have the same 3:1 compression ratio of a A/C;

In the larger central systems the scroll compressors are now very popular because of their efficiency, low sound/vibration levels, longevity and resistance to slugging. But they don't seem to provide in practice more than about 2.4:1 compression ratios.

...however both low and high side design pressures are lower than the A/C.

The design pressures depend on the application. A liquid/liquid refrigeration machine is not designed to cool air. It is designed to cool liquid. Now the cooled liquid may be used to cool air or beer or whatever.

As such the system can cool to a lower temperature, but requires a lower condenser side temp.

You can make an air to air system cool to a tempreature as low as a glycol system. The designs are essentially the same (an oboe and a bassoon are really the same thing: double reeded wood winds) but the details differ. To go to lower temperature you need a colder condenser or a better compressor or both.

Seeing this math laid out makes me realize that someone using an A/C system or a purpose built glycol system in an unconditioned area may have issues with low temperature operation.

With compressors of limited capability you must do (if you insist on having a 20 ° reservoir) what a cryogenic designer does: use a multi stage system in which the condenser of the low temperature loop is cooled by the evaporator in the high temperature loop. In this case, of course, the high temperature 'loop' is your room or central air conditioner but the principle is the same.

It also makes me realize that the efficiency loss of my system at low set temperatures (were I keep it) is much greater than I was estimating (I knew I was shifting the "curve" but didn't realize that I got rid of all the superheat and was operating so far below the design Low pressure point).

This is where the Second Law comes in. The COP for a refrigerator is Tc/(Th-Tc) thus the closer the evaporator and condenser temperatures can be the higher the efficiency.

Nope, no HVAC tools. I just bent the pliable copper lines to fit where I wanted without evacuating the system.

Amazing! That's pretty cooperative copper, I'd say. I still think a gauge set would be an awfully good investment. My speculations taken from PE charts and tables are fine for elucidating the principles involved but there is nothing like solid numbers.

 

[ajdelange]

Not controlling the suction directly, just changing the control temp for the liquid line solenoid.

Got it.

I well remember the first time I saw a system (walk in) controlled that way and spent hours looking for the hidden wire between the thermostat and the compressor/condenser (outdoor) unit. When I finally figured out how it works I thought that was the biggest kluge I'd ever seen.

Blowing air over the condenser still might get you a little COP boost.

 

[jddevinn]

teaspoonful of water on it the water will boil instantly, but not flash

In a refrigeration system you squirt warm (but below its boiling point at the high side pressure) refrigerant into a tube at a lower pressure, low enough that it is well above the boiling point. So it boils. Again it does not flash (or shouldn't in a proper design) and again it absorbs the latent heat of vaporization from the other side of the heat exchanger.

Flash is typically the term used when the phase change of a liquid is caused by lowering its partial pressure. While boil is the term used when heat is added to cause the phase change. In this system the phase change is forced by an enlarging diameter and lower pressure, so technically it would be a flash.... unless I'm missing something.

properly designed and modern systems contain TXVs to maintain superheat within a desired range.

Correct me if I'm wrong but I thing small window A/C units in the 5,000 BTU/hr range still have fixed expansion not TXVs. I can't find any information to directly confirm or disprove but it doesn't look like the smaller glycol chiller units have TXVs, maybe the several Ton units do.

Just to be clear the 52.7° is the temperature of the gas leaving the evaporator.

When I say coolant I mean the refrigerant in the cooling system. If I'm talking outside of the system I say air or glycol.

As you starve the evaporator, which you will have to do if you can't get the condenser cooled enough

Ambient in my brewery during the summer seems to be cool enough... But an important consideration for people putting these in non controlled environments.

You are going to think me terribly pedantic but it is the Zeroth law that is at play here.

Zeroth law helps define the concept of temperature

Second Law expressed in:

Carnot's principle
the efficiency of a quasi-static or reversible Carnot cycle depends only on the temperatures of the two heat reservoirs, and is the same, whatever the working substance. A Carnot engine operated in this way is the most efficient possible heat engine using those two temperatures

Clausius statement
Heat can never pass from a colder to a warmer body without some other change, connected therewith, occurring at the same time

Kelvin statement
It is impossible, by means of inanimate material agency, to derive mechanical effect from any portion of matter by cooling it below the temperature of the coldest of the surrounding objects.

It will cost you about as much as a decent pH meter ($150 or so). All this speculation on your part and mine is really meaningless when stacked up against real data.

That's twice the cost of the unit though!!!

Yes, sorry about not labeling the axes.

Everyone remembers to check if OTHER peoples charts have labels.

No one would design a system to operate at 80° subcooling. That is indicated on the right hand scale and is about 10 ° as is the superheat. The low side boiling point is about 20 °F.

Said 80 psi discharge not 80°F... I didn't measure too well and got the ~15°F superheat.


The next time that I have all the fermenters empty I'll pump the glycol into buckets to weigh it, put it back in the system and let it warm to ~65°F or so and then cool to 20°F, log the data and back calculate the actual cooling capacity of the unit in 5 or 10°F intervals.

 

[ajdelange]

The usual statement of the zeroth law is that if A is in thermal equilibrium with B and C is in thermal equilibrium with B then C is in thermal equilibrium with A i.e. they are all at the same temperature and no heat exchange will occur between them. Yes, this allows us to design thermometers but the obvious implication here is that if material on one side of a heat exchanger is at the same temperature as the material on the other there will be no passage of heat between them.

With respect to the Second Law what we are interested in is the Carnot's statement of it: There is no engine operating between two reservoirs as efficient as a Carnot engine. As the efficiency of a Carnot engine is easy to calculate (1 - Tc/Th) we can bound the performance of a cyclic machine. As a refrigeration machine uses a Carnot cycle run backwards we easily get the maximum COP of a refrigeration machine as Tc/(Th-Tc) = 1/(Th/Tc - 1). Thus the closer Th to Tc the more heat you can transfer per unit of work. This is relevant to discussions of where to set evaporator temperature relative to your "hot reservoir" (air) temperature. Obviously the closer you can get it the higher the machine's COP, SEER etc and the happier the tree huggers are. Obviously (Zeroth Law) you can't have it higher than the lowest temperature you want to cool to.

 

[jddevinn]

Ok from original conversation.

I've been thinking about the cooling reserve to rapidly cool the wort after the boil is complete. I think I've come up with something I'm going to try next time I brew.

In my 3 vessel system, when the HLT is empty during the sparge I'm going to:

1. Fill HLT with "cold" tap water
2. Connect one of the lines from my glycol reserve to the HERMS coil of the HLT
3. Cool the HLT down to ~40°F, assuming I have 60% of the rated cooling of the A/C this will take about an hour. So it should be done when the boil is nearing complete.
4. Cool the wort down to about 100°F using cool tap water, this is normally quick.
5. A single pass of the 40°F chilled water should knock my 12 gallon batch to around 65°F

Hopefully I can cut the chilling down to 5 or 10 minutes!! If successful I'll likely start a seperate thread and look at getting a rain barrel or similar to use as a dedicated secondary "coolant" reserve.

 

[Drumminguy81]

Ill be doing something similar, but instead of cooling a vessel of water im doing it on the fly. I have 2 plate chillers, one is hooked up with the glycol as the cooling source, with my tap water hooked to the wort in side which should chill my water down quite a bit, it then goes to the second plate chiller as the chilling water with the hope I can cool 20 gallons from 200 down to 68 in a single pass. This is all speculation though.. ill report back after I try it.

 

[homebrewdude76]

I bought a used AC unit and started to open it up.
I dont really need a large cooler for this thing.

I am thinking of cutting the bottom steel in order to bring the cooling portion directly into the cooler.

Need to think how to wire this thing. I would prefer to bypass all controlls and use a control unit.

View attachment 1477239826352.jpg

View attachment 1477239835142.jpg

 

[Drumminguy81]

You're on the right track. Cut the bottom, carefully bend the coil to fit in the cooler with a small notch so the lid will close. Bypass the thermostat and controls and use a temperature controller like an inkbird to switch it on and off and get the glycol to your desired temperature.

 

[homebrewdude76]

Cut more metal to make more room for cooler

 

[homebrewdude76]

Photo

 

[homebrewdude76]

The andoid app always has issues...

View attachment 1479137714527.jpg

 

[homebrewdude76]

This is what I came up with. Just need to figure out what pump to push through the conical cooling coil

View attachment 1483574067446.jpg

 

[jddevinn]

This is what I came up with. Just need to figure out what pump to push through the conical cooling coil

FYI, I use this pump https://www.amazon.com/gp/product/B0009X8O2E/?tag=skimlinks_replacement-20.

Anything similar will work. The higher flow rate that you get the faster you can transfer heat (to a point).