Chiller questions

Hey guys, I’m fairly new to extractions first of all. I’m looking for a way to boost or change my current setup. There’s 3 PolyScience AD-15s: 1 for the columns, 1 for the solvent tank jacket, and one for the inner coil on the solvent tank.

Now these units are rated and set to -40C, room is 19-21C…Column gets to -18, Tank jacket gets to -10, and Coil gets to -20 overnight or -22 after chilling all weekend.

Is there a way to get any colder without changing everything out? I’d like to get at least below -25 if not closer to -40. Budget isn’t a huge deal but yeah, cheaper while still reliable is always better!

Any ideas?

1 Like

How much gap do you have between them, how much space from behind to the wall? It is possible that you’ve crowed them and lowered their efficency. How many amps are you pulling and what is your panel rated for. Is each unit in a dedicated circuit or on its own breaker? @indofab said that they may be competing for power at peak draw.

1 Like

I have a theory about the power, but to start:
What coolant are you using?
When did you last replace the coolant?
Did they ever achieve -40?


Good question… could be a chronic issue …

All chillers will lose cooling power in an exponential (non-linear) manner as the temp drops. This simply means that it likely doesn’t have the power to combat the heat gained by the ambient atmosphere. In short try insulating everything better and see if that helps. The heat exchange from chiller to atmosphere could also be improved so better air flow, lower room temp, etc


I noticed in chiller manual it says
“Cooling Capacity @ -20°C (W) 265”
which to me says the lower you go, the less wattage is left for cooling (this makes sense). Does this translate to 'once you hit -20 on this 1000W chiller you are consuming 735 Watts? Leaving you 26.5% of power for a substantial amount of cooling (an additional -20c).

thank you (anyone) for the edification.

For wattage I’m actually not sure if they mean actual wattage consumed by the unit (they could I’m just not sure). I believe it means the amount of heat removed by the machine at that temp.
In my mind the chiller would actually use more power at a lower temp because the compressor would cycle more, it would have to work harder, etc.


You are correct.

I just ordered this exact chiller the other day so ill let you know my experience with it.

Whats your your coolant mix?
Whats your electrical situation?

Heat transfer fluid definitely affects how much heat can be transferred. If you’re aiming at -40C a potassium formate based fluid has a higher heat capacity than a glycol mix formulated for that temp.

When using a glycol mix, using the least amount of glycol possible to avoid slush is the sweet spot.

1 Like

@Soxhlet - There is about 4" between each unit and 12" behind it to the wall. There are 3 units and the tank on a 6’ metal table and they are spread out as much as can be. I might be able to get one chiller on the shelf underneath and increase the distance between the remaining ones up top.
As for the power questions, I’m not entirely sure. They are on separate breakers, they would trip if more than one was on a breaker.
@sidco - They are running straight glycol and the last flush was about 3 months ago. Haven’t noticed any change in temps, they just never got down that low. I’ve never seen any of them get below -24.7

Pure ethylene glycol freezes at about −12 °C (10.4 °F) but, when mixed with water, the mixture freezes at a lower temperature. For example, a mixture of 60% ethylene glycol and 40% water freezes at −45 °C (−49 °F)

I have a feeling a little bit of water in there would go a long way

Source:Ethylene glycol - Wikipedia


Your chillers should have minimum 3 feet from the wall, you also may want to spread them several feet apart.


Watts are simply an expression of work. Horsepower can be expressed in watts and as a matter of fact nearly every unit of energy of work can be expressed as watts. A watt is a unit of power defined as the rate of energy transferred per second.

What those wattage figures are trying to tell you is independent of what the serial number plate is telling you. The serial number plate usually near the inlet power tells you what to expect the machine to pull in either amps or watts as a unit running at full capacity in continuous operation. The plate never tells you peak amp draw nor does it tell you what mostly only electricians or engineers have heard of called “power factor”. In essence the data plate tells you how big your circuit breaker needs to be to keep from tripping and that is about it.

The chillers that publish their cooling capacity in watts/temps are telling you something very different. They are explaining that the ability to hold the base line temperature, expressed in watts, diminishes the colder you get. It gives the engineer a number to work from when designing that the data plate cannot. It means that if you want to hold the chiller at any given temp it will consume power just to hold the temp without anything external carrying heat into the coolant. In essence the wattage expressed at a temp is communicating to you how much heat energy the unit is capable of transfering and still maintain the set point temperature of the chiller itself.

Let’s say that your -40C rated chiller is set up correctly next to my -85C chiller. The rating tells us the maximum cooling (work) that the unit is capable of after the chiller itself has cooled to temp. Assume that we both have exactly identical set ups and we both want to run at -35C. At -35C your chiller will likely show very little wattage left over to actually cool the thing that is hot that needs cooling. Let’s say you have 10 watts of cooling power left at -35C. This means you can pump the equivelent of a ten watt light bulb of heat into the system every second and the chiller would hold temp but if you start pumping say 15 watts of heat through the unit then it will not hold temp.

My -85C chiller set at -35C might have 200 watts of cooling power left so I could pump the heat of a 200 watt light bulb through the system and still be able to hold -35C in this scenario with no problem. If more heat is applied than the unit can handle in watts for a given temp the unit simply does not get as cold. In fact, assuming your unit is working correctly, you could even predict how cold the unit will keep things by first knowing how much heat (work) will be run through the unit. Then an engineer can use the chart in reverse so to speak and predict the temperature the unit can hold and size a unit correctly for it.

The wattage on the data plate near the cord tells you how much amp draw on average to expect when the machine is drawing full current in normal operation. The wattage expressed in relation to the cooling potential tells you how much work will be left over to cool your process after the machine itself is cooled to the set point. Absolutely everything that allows heat to transfer into your process (like ambient air) adds heat and therefor takes away from the wattage you have left to chill the process.


Why spread them apart? Heat transfer from one to another?

They need a certain amount of airflow to exchange heat. Look at the manual, I am sure it lists a spec of wall distance.




Huh? :sunglasses:

@Killbill needs to know this shit! I know I do.

1 Like