What's new
  • Please note members who been with us for more than 10 years have been upgraded to "Veteran" status and will receive exclusive benefits. If you wish to find out more about this or support IcMag and get same benefits, check this thread here.
  • Important notice: ICMag's T.O.U. has been updated. Please review it here. For your convenience, it is also available in the main forum menu, under 'Quick Links"!

DIY -50C Recirculating Chiller

WaterFarmFan

Active member
Veteran
So, I thought that I would create a thread to document my experiments on a budget DIY recirculating chiller that uses a dry ice/ethanol/propylene glycol slurry with an external magnetic drive pump. My target temps will be in -50C to -60C range, but I have no idea how the pump will handle it over the short or long term. This is a couple hundred dollars compared to the lab versions at MANY thousand dollars, which most only get to -30C. Yes, it takes dry ice, but not that much to justify the difference.

Here is the pump that I am using:

https://www.amazon.com/CHUGGER-PUMP-CPSS-CI-1-Transferring-Compliant/dp/B01N9HERFS/

HOME BREWING PUMP - The CHUGGER PUMP can be used to ease transfers between kettles, cool wort via a counterflow chiller, or recirculate ice water for quicker chilling!
HIGH QUALITY MATERIALS - Stainless steel pump head, Magnetic Drive w/ base, Teflon thrust washer, Clear Silicone O-ring, Ryton/Teflon Impeller, 55" cord with plug.
PREMIUM PERFORMANCE - Featuring a max flow rate of 7 GPM and 18 Feet of head to handle back-pressure. Capable of handling liquids up to 250 degrees F.
The CHUGGER PUMP is the only home brew pump with ETL certification and USFDA food compliant materials tested for safety and performance standards.
3/4" MPT inlet and 1/2" MPT, allowing easy configuration for any application!


I choose 2x (6L and 15L) relatively thin stainless stock pots as the system reservoir. I used 8lb density 2-part Liquid Urethane foam kit (2 quarts total) to create a thermal barrier, which varies from 2 1/2 inches at bottom to ~ 1 1/2 inches at the top. The 8lb cured absolutely rock hard and was quite difficult to cut by hand. I used every drop of the 2 quarts for this outer shell. 8lb density (4lb actual weight kit) yields roughly 1/2 cubic feet, but varies depending on room temperature. I think mine had a little less expansion resulting in an even denser foam.

After the foam set (and cooled), I cut off the top to make smooth surface for lid. Using a set of cobalt and titanium step bits, drilled 2 separate holes for pump inlet and return line from condensers. The outer hole is just big enough to squeeze the outer washer for the bulkhead thru. The inner hole is slightly larger than the 1/2" NPT male bulkhead fittings. I had to use a drill bit to clean up the inside of the foam to make sure there was a flush surface for silicone o-rings.

Most of this equipment is for beer brewing, so I used lots of stainless 1/2" NPT to 3/8" barb adapters. Just as with my heated nitrogen backfill, I am using stainless/nylon braided ptfe lined tubing in -8AN size, which is slightly larger than 3/8", but secures very tightly when 3/8" barbs are sufficiently wrapped in ptfe tape and fastened with hose clamps.

I created a lid out of ~2" thick foam and notched the edges to drop into the inner pot. Foam bottom is covered with aluminum foil, and radiant barrier material was used to create a layered border insulation system. The outer pot lid is secured to foam and RB material with foil tape.

This has many applications depending on temp range, but I will be using this at these very low (-58F to -76F) temperatures for terpene extraction.

I am still waiting on a few things (stainless mash tun to filter dry ice from entering pump) and will post a final parts list once tested, but here a few pics of the work in progress:

picture.php

picture.php

picture.php

picture.php
 
Excited to see what you're doing here! I'm also exploring a lot of different options for subzero cooling. I'm considering all options, from immersion chillers plumbed to AC units, to dry ice, to cascaded peltier elements. All three of these should be able to be built for much much cheaper than commercial lab chillers ($500-3000/HP(or ~750W cooling power).
 

WaterFarmFan

Active member
Veteran
Excited to see what you're doing here! I'm also exploring a lot of different options for subzero cooling. I'm considering all options, from immersion chillers plumbed to AC units, to dry ice, to cascaded peltier elements. All three of these should be able to be built for much much cheaper than commercial lab chillers ($500-3000/HP(or ~750W cooling power).

Thanks! It really depends on your lowest target temps and of course means of reliable throughput. I like this system because I can vary the ethanol/PG ratio to bring the temp up or down, and just keep enough dry ice (an inch or two) to maintain the slurry at its max temp. Dry ice floats to top and pump draws from bottom via a 150 micron tube screen.
 

Gray Wolf

A Posse ad Esse. From Possibility to realization.
Mentor
ICMag Donor
Veteran
So, I thought that I would create a thread to document my experiments on a budget DIY recirculating chiller that uses a dry ice/ethanol/propylene glycol slurry with an external magnetic drive pump. My target temps will be in -50C to -60C range, but I have no idea how the pump will handle it over the short or long term. This is a couple hundred dollars compared to the lab versions at MANY thousand dollars, which most only get to -30C. Yes, it takes dry ice, but not that much to justify the difference.

Here is the pump that I am using:

https://www.amazon.com/CHUGGER-PUMP-CPSS-CI-1-Transferring-Compliant/dp/B01N9HERFS/

HOME BREWING PUMP - The CHUGGER PUMP can be used to ease transfers between kettles, cool wort via a counterflow chiller, or recirculate ice water for quicker chilling!
HIGH QUALITY MATERIALS - Stainless steel pump head, Magnetic Drive w/ base, Teflon thrust washer, Clear Silicone O-ring, Ryton/Teflon Impeller, 55" cord with plug.
PREMIUM PERFORMANCE - Featuring a max flow rate of 7 GPM and 18 Feet of head to handle back-pressure. Capable of handling liquids up to 250 degrees F.
The CHUGGER PUMP is the only home brew pump with ETL certification and USFDA food compliant materials tested for safety and performance standards.
3/4" MPT inlet and 1/2" MPT, allowing easy configuration for any application!


I choose 2x (6L and 15L) relatively thin stainless stock pots as the system reservoir. I used 8lb density 2-part Liquid Urethane foam kit (2 quarts total) to create a thermal barrier, which varies from 2 1/2 inches at bottom to ~ 1 1/2 inches at the top. The 8lb cured absolutely rock hard and was quite difficult to cut by hand. I used every drop of the 2 quarts for this outer shell. 8lb density (4lb actual weight kit) yields roughly 1/2 cubic feet, but varies depending on room temperature. I think mine had a little less expansion resulting in an even denser foam.

After the foam set (and cooled), I cut off the top to make smooth surface for lid. Using a set of cobalt and titanium step bits, drilled 2 separate holes for pump inlet and return line from condensers. The outer hole is just big enough to squeeze the outer washer for the bulkhead thru. The inner hole is slightly larger than the 1/2" NPT male bulkhead fittings. I had to use a drill bit to clean up the inside of the foam to make sure there was a flush surface for silicone o-rings.

Most of this equipment is for beer brewing, so I used lots of stainless 1/2" NPT to 3/8" barb adapters. Just as with my heated nitrogen backfill, I am using stainless/nylon braided ptfe lined tubing in -8AN size, which is slightly larger than 3/8", but secures very tightly when 3/8" barbs are sufficiently wrapped in ptfe tape and fastened with hose clamps.

I created a lid out of ~2" thick foam and notched the edges to drop into the inner pot. Foam bottom is covered with aluminum foil, and radiant barrier material was used to create a layered border insulation system. The outer pot lid is secured to foam and RB material with foil tape.

This has many applications depending on temp range, but I will be using this at these very low (-58F to -76F) temperatures for terpene extraction.

I am still waiting on a few things (stainless mash tun to filter dry ice from entering pump) and will post a final parts list once tested, but here a few pics of the work in progress:

View Image
View Image
View Image
View Image[/quo


Good experiment! The proof will be in the testing.

Extreme thermal cycling can cause threaded fittings to loosen, so we ended up welding couplings through the tank walls, instead of using bulkhead fittings on the WolfWurx tricore heat exchangers, that used an alcohol dry ice bath.

I have the same questions regarding the suitability of the chugger pump at low temperatures as well, but they have to have some added tolerance to be able to rate them at -40C, sooo, ah wate with bait on mah breath!
 

Old Gold

Active member
Chugger pumps suck, I couldn't even get one to push liquid up a 25 ft coil (roughly 8-12" head) and it had a gallon of priming reservoir directly overhead. Based on its viscosity rating, it can handle ethanol all the way down to below -100°C

I had much better luck with this:
https://m.grainger.com/mobile/product/FINISH-THOMPSON-1-4-HP-Polypropylene-115V-5FZW1

I've run that thing for hours and hours on end at -40°C and seen it push as cold as -55°C alcohol with no flow issues. The only thing that kills it is ANY amount of dry ice being put directly into the "chiller fluid" stream instead of the external cooling bath (which I've tried just to cheat the time needed to cool 10 gallons). Even if there are no solid chunks of dry ice left in the alcohol, it shuts itself off for a healthy few hours.

Mounting an explosion proof motor would be a good idea, even if only for ethanol recirculation.
 

WaterFarmFan

Active member
Veteran
Good experiment! The proof will be in the testing.

Extreme thermal cycling can cause threaded fittings to loosen, so we ended up welding couplings through the tank walls, instead of using bulkhead fittings on the WolfWurx tricore heat exchangers, that used an alcohol dry ice bath.

I have the same questions regarding the suitability of the chugger pump at low temperatures as well, but they have to have some added tolerance to be able to rate them at -40C, sooo, ah wate with bait on mah breath!

Thanks. How exactly did you weld the couplings?

EDIT - I thought about just filling my voids with some kind of resin to create a leak proof seal. Downside would be that I could not replace bulkhead nipple. Any recommendations on an epoxy (from jbweld or other) that when cured could withstand potentially -78C? I need to seal the top exposed foam with epoxy as well.
 

WaterFarmFan

Active member
Veteran
Chugger pumps suck, I couldn't even get one to push liquid up a 25 ft coil (roughly 8-12" head) and it had a gallon of priming reservoir directly overhead. Based on its viscosity rating, it can handle ethanol all the way down to below -100°C

I had much better luck with this:
https://m.grainger.com/mobile/product/FINISH-THOMPSON-1-4-HP-Polypropylene-115V-5FZW1

I've run that thing for hours and hours on end at -40°C and seen it push as cold as -55°C alcohol with no flow issues. The only thing that kills it is ANY amount of dry ice being put directly into the "chiller fluid" stream instead of the external cooling bath (which I've tried just to cheat the time needed to cool 10 gallons). Even if there are no solid chunks of dry ice left in the alcohol, it shuts itself off for a healthy few hours.

Mounting an explosion proof motor would be a good idea, even if only for ethanol recirculation.

Appreciate you sharing your experiences with the Chugger. I need mine to pump up about 3 feet to a pair of inverted 400mm Dimroth condensers. I am hoping that gravity from the coils helps this thing "chug" along. If it does not cut the mustard, I will return it and go with another option.
 

WaterFarmFan

Active member
Veteran
Mash tun filter arrived today. It is 150 mesh not 150 micron, but should work just fine. I need to get my ptfe hose cut, but should be ready to test tomorrow.

picture.php
 

Gray Wolf

A Posse ad Esse. From Possibility to realization.
Mentor
ICMag Donor
Veteran
Thanks. How exactly did you weld the couplings?

EDIT - I thought about just filling my voids with some kind of resin to create a leak proof seal. Downside would be that I could not replace bulkhead nipple. Any recommendations on an epoxy (from jbweld or other) that when cured could withstand potentially -78C? I need to seal the top exposed foam with epoxy as well.

I TIG welded the 304SS couplings in the 18-8 stainless pot using 316SS wire.

Solly, not familiar with low temperature epoxies.
 

SkyHighLer

Got me a stone bad Mana
ICMag Donor
Veteran
Thanks. How exactly did you weld the couplings?

EDIT - I thought about just filling my voids with some kind of resin to create a leak proof seal. Downside would be that I could not replace bulkhead nipple. Any recommendations on an epoxy (from jbweld or other) that when cured could withstand potentially -78C? I need to seal the top exposed foam with epoxy as well.

Is J-B Weld resistant to water and/or gasoline?
When fully cured, J-B Weld is completely resistant to water, gasoline, and about every other petroleum product or automotive chemical. For wet-surface or submerged water or gasoline repairs, try our SteelStik or WaterWeld.

Will J-B Weld hold up to extremely cold temperatures?
Original J-B Weld has proven to maintain its characteristics to temperatures as low as -67º F.

Just says "maintain it's characteristics," as long as it doesn't crack or loose adherence it may still be durable enough for your purposes.

https://www.jbweld.com/pages/faqs
 

WaterFarmFan

Active member
Veteran
I found this in a PDF:

What Happens to Epoxies at Cryogenic Temperatures?

Exposing epoxy to very low, even cryogenic temperatures, will not cause any degradation of the material, however, some performance properties may change.

Most EPO-TEK® datasheets list -55°C as a minimum operating temperature, but that is due to test equipment limitations, not epoxy constraints. Many epoxies typically will operate well below -55°C; including cryogenic temperatures ranging from -150°C to absolute zero (-273°C).

As temperature decreases, the modulus will increase, and cold epoxies are more brittle than they are at room temperature. Epoxies that perform well at low temperatures tend to be those with lower moduli. When the modulus increase is minimal, it induces less stress on the bonded components as a result of the changes in temperature.
 

WaterFarmFan

Active member
Veteran
Here are some more good details on mechanical and thermal properties:

Understanding Mechanical Properties of Epoxies For Modeling, Finite Element Analysis (FEA)

The unique molecular structure of epoxy allows for a large variety of mechanical properties through varying formulations. Epoxies can range from very soft and flexible to very hard and rigid. Softer materials can relieve stress while harder products are used for their high strength and acoustic properties. The thermosetting nature of epoxies cause them to behave differently than thermoplastics and other engineered materials when stresses are applied. Understanding these properties can help in the selection of the best possible product and aid in the modeling of systems such as finite element analysis (FEA).

Strength

In the majority of applications, the largest stress on an epoxy is shear, not tensile. For this reason we perform two types of shear testing to determine epoxy strength: lap shear and die shear. Lap shear is performed by bonding two overlapping aluminum coupons and applying a shear force by pulling the coupons in opposite directions until they fail in a shear mode. The lap shear strength provides a good approximation for strength in structural applications and is reported in units of pounds per square inch (psi). Die shear strength is typically perfomed using a probe to push a 2mm x 2mm gold die, adhered to a gold plated kovar substrate. Die shear strength is reported as a force in kilograms (kg) and as a stress in units of pressure (psi). For the bonding of small components such as electronics, this is the preferred test method.

Modulus

Modulus is an excellent property for judging how rigid or flexible a material may be and is very important in modeling. Epoxies highly cross-linked polymer structure exhibit mainly elastic response to loadings, however there is also a viscous response which causes some plastic deformation. The combination of responses is known as viscoelasticity, which is measured using Dynamic Mechanical Analysis (DMA) and provides a more accurate characterization of the material.

DMA uses a cyclic Three Point Bend Test which allows for the viscoelastic response to be seen as a phase shift in the response of the material. The phase shift, reported as tan(δ), can also be expressed as the ratio between the loss modulus and the storage modulus. The loss modulus represents the complex or viscous component, while the storage modulus represents the real or elastic response. This allows the storage modulus to act as a good approximation of the Young’s Modulus for an epoxy.

Temperature

Epoxies are thermosetting polymers, which causes them to behave quite differently from thermoplastic materials at high temperatures. Unlike thermoplastics, epoxies will not flow or melt when brought above their glass transition temperature (Tg). Above its Tg, an epoxy becomes softer as evidenced by a decrease in the modulus. However, the cross-linking of the polymer chains causes the material to maintain its shape and adhesion through this transition. Many epoxies are commonly used well above their Tg with excellent results. Above the Tg, there is an increase in the free volume within the structure which allows for more movement of the polymer chains. The increase in chain movement accounts for the decrease in modulus and an increase in the coefficient of thermal expansion (CTE).

Many times, instinct dictates that a lower CTE value will lead to better resistance of thermal stresses, this is not always the case. While low CTE values indicate less of an expansion in the dimensions of the epoxy with an increase in temperature, it can be impossible to exactly match the CTE of the substrates being bonded. Low CTE materials also tend to be very rigid, so any thermal stresses transferred to the bond line can often lead to de-bonding of the epoxy. Instead of having a low CTE, the best materials for overcoming thermal stresses usually have a lower modulus. A lower modulus allows the epoxy to absorb stresses caused by temperature changes, even if the epoxy has a high CTE value. This is especially important for larger parts where the forces caused by thermal expansion are proportionately greater.

The above information provides a general guideline for epoxy selection in finite element analysis and can aid in predicting the performance of an adhesive for specific applications. Many times, adhesive performance can be optimized by adjusting cure schedules, application method and surface preparation.

Hardness

In addition to strength and modulus, another important property is the hardness of an epoxy. Hardness is a useful approximation of the rigidity of an epoxy. Along with the modulus, hardness can provide additional data regarding the compliance of an epoxy. Hardness is measured using a Shore® Durometer. Higher hardness values indicate a more rigid material, while a lower values, a softer one. There are two scales that are used for measuring hardness; Shore D for more rigid materials, while Shore A is used for softer products. Each hardness reading can vary from sample to sample and even within the same sample if measured by different operators. This can lead to a variation of approximately ±5 for most products.

Poisson’s Ratio

Another useful mechanical property for modeling systems is Poisson’s Ratio (ѵ). The Poisson’s ratio describes the relationship of the change in dimensions of a material in both the axial and transverse direction when a stress is applied. Epoxy Technology does not measure this property, but most epoxies exhibit a value of approximately 0.3-0.35. For modeling purposes, 0.3 is most commonly used.
 

WaterFarmFan

Active member
Veteran
Will J-B Weld hold up to extremely cold temperatures?
Original J-B Weld has proven to maintain its characteristics to temperatures as low as -67º F.

Thanks Sky. This confirms what the EPO-TEK pdf stated about testing as -55C = -67F, and it should work for even lower. I am just going to use the JB Weld for my cutouts, as I don't need too much.
 

WaterFarmFan

Active member
Veteran
By the way, I found some low modulus epoxy online, but it was a few hundred dollars for a very small amount. Way overkill for this project...
 

WaterFarmFan

Active member
Veteran
-40 °C can be achieved with a 1 to 0.8 ratio by weight of calcium chloride hexahydrate to ice.

Road salt = calcium chloride.

Cooling Bath

Hey troutman. I am not sure about the viscosity, but that should work with this setup, and is a good idea for slightly higher temps. I use lots of dry ice in most parts of my processes, so it is not too much of a pain keeping it topped off every few hours.

My current project requires temps as cold as possible. If the pump runs strong at -50C, I will push it as close to -70C as I reliably can. I just increase the amount of ethanol in the PG bath to lower temps. -50C will be roughly 50:50 ethanol to glycol and -70C would be 75:25.
 

troutman

Seed Whore
Fill up a bathtub or similar large container with the chilling mixture I mention and run your line coiled in it for maximum cooling.

Also regular ice and road salt is cheaper and easier to get than dry ice. :tiphat:
 

WaterFarmFan

Active member
Veteran
I had much better luck with this:
https://m.grainger.com/mobile/product/FINISH-THOMPSON-1-4-HP-Polypropylene-115V-5FZW1

I've run that thing for hours and hours on end at -40°C and seen it push as cold as -55°C alcohol with no flow issues. The only thing that kills it is ANY amount of dry ice being put directly into the "chiller fluid" stream instead of the external cooling bath (which I've tried just to cheat the time needed to cool 10 gallons). Even if there are no solid chunks of dry ice left in the alcohol, it shuts itself off for a healthy few hours.

Mounting an explosion proof motor would be a good idea, even if only for ethanol recirculation.

What are your thoughts on Polypropylene vs Stainless steel for housing and components to run at very low temperatures for extended periods of time? How important is HP for the pump if pushing up 4' of 3/8" coil? FT makes a 1/8HP model, like yours, as well, and there are some used ones, of various HP and components, on ebay.

Do you see any inherent flaws on this pump?

https://www.blichmannengineering.com/products/riptide-brewing-pump
 

WaterFarmFan

Active member
Veteran
Ok, I finally did some initial testing, and the chugger is a BEAST! With only water, for my first test using ~15 feet of 3/8 silicone tubing, I placed the pump and reservoir on floor and created a closed loop with hose. Pump was primed by gravity, and I turned it on and lifted hose to a head of ~7 feet. A stream was instantly shooting across my res and hitting the across wall on a near horizontal plane. I then rigged a coiled condenser on the end of hose, and with the same head of ~7 feet, water came shooting through in a flash.

Bottom line - This thing has some serious power. I have no idea how it will handle cold temps with extended run time, but I will soon find out. Next phase will be to run it with dry ice, ethanol and PG at -50C with condensers in a loop for 4-6 hours to torture test it. If that goes well, I will do another test using 100% ethanol dry ice with temps slightly above -78C for 2-3 hours, before putting it into production with a live run.
 

WaterFarmFan

Active member
Veteran
Chugger pumps suck, I couldn't even get one to push liquid up a 25 ft coil (roughly 8-12" head) and it had a gallon of priming reservoir directly overhead. Based on its viscosity rating, it can handle ethanol all the way down to below -100°C.

Just curious, when did own/use a chugger? Was yours inline or center input? I read center has better flow than inline. I have seen two versions used on ebay, and I wonder if you had an earlier model? No idea if it will last or replicate results at higher viscosity.

This is the first external fluid pump that I have ever owned, but I have gone through more a dozen of the submersible ones for hydroponic applications. This thing puts all of those to shame.
 
Top