Building the Ultimate chiller – AKA The GOAT (pics and data)

[setmash]

What a difference 24 hours makes. I put my thinking cap on last night and tried to figure out in my pea sized brain what was going on. I had the notion that perhaps the coolant and wort flow rates needs to be at a certain level before efficient heat transfer takes place. That would explain why the 22/10 performs so much worse than 15/10. I then thought maybe the 15/10 coolant flow rate of 3LPM wasnt optimum either. I was right...

My mains water comes up from a bore hole in the garden and is pumped in to the attic where its held in a big storage tank. I though feck this I'll connect the chugger up to the tap and see if i can force more coolant through the 15/10. I connected it up and got the flow rate up to 4.42 LPM from 3 LPM, doesnt sound much but i though bloody hell its worth a punt. I wasnt disappointed.

15/10 8M Goat cooling 25 Liters of wort from 99 degrees to 26.5 degrees .... 6 mins 37 secs using just 26 Liters of coolant exit temp 66 degrees (input temp 10.5 degrees). GET THE F*CK IN

Increasing the coolant flow by just 1.42 LPM had a massive effect on chill time. Im completely shocked. Imagine having a decent mains pressure. This 8M Goat is performing on par with the 12M hybrid i built. Im super chuffed....just wish id tested this bloody configuration at the start lol would of saved me a fair few quid in fittings and lecky bill.

I then got thinking about Aleman saying things would go bad when the summer comes and the coolant temp rises. I thought I will test this by using the HLT to raise the coolant temp to 21 degrees (i dont see it that high even in the height of summer). Because I have to raise the temp i used the HLT which sits just above the chugger (less head than my tank in the attic) - this dropped the flow rate through the goat to only 3.6 LPM...i though hmmmm this isnt going to go well....I was wrong

15/10 coolant flow 3.6LPM 21 degree summer input temp, time to chill 27.5 degrees 10.72 mins 40.6 Liters of coolant used coolant exit temp 60 degrees. With my 4.42 LPM Im sure to improve the chill time but im not sure Id want to, Either way the 8M 15/10 Goat is a superb summer or winter chiller with a meagre appetite for coolant.

Its clear that if you drop the flow of coolant or wort down to a certain point in these chillers there is a huge drop in efficiency.


Great results , Ill try write more detail next week. No more tests for a couple of weeks as Im away. A lot of hard work done this week but Im happy.

[setmash]

I made a video of yesterdays 8M 15/10 test..... hope its of use



https://www.youtube.com/watch?v=kFOzmva ... e=youtu.be

[lord groan]

Thats incredible Setmash, fantastic job! How have you sealed the 10mm tube inside the 'Tees'. Is it a 10-15-15 tee with the tube feeding through the tee and down inside the 15mm tube or have you built up the 10mm tube to 15mm where it enters the tee
cheers
lg

[boingy]

For me it's an interesting solution but way too much of a sprawling mess. I'll stick with the plate chiller and recirculated pond water for the coolant (really!).

I know what you are saying about not really knowing what is in a plate chiller but with careful rinsing and storage I'm pretty chilled (bad pun intended) about the tiny risk of infection from it. Beer only needs a moderately obsessive attention to infection control rather than a full-on OCD thing imho.

All that stuff to assemble, dismantle and clean each brew day is not for me.

[setmash]

Hi thanks all, LG if you look closely I use a 15mm equal Tee. I use a plastic push fit for my R&D model but I’ll move to copper and solder for my final version. You could use 15-15-10 but I prefer to use 15-15-15 and use a rubber bung with a drilled hole in it for the 10mm pipe. This makes it easier to construct but more importantly easier to de-construct . I can take it apart and play with stuff. The bung may not be effective at very high coolant flow and pressure rates – you'd have to test it. Or you could use 15-15-10 or 15-15-15 with a copper reducer to 10mm.

Boingy, the version you see on the floor is my R&D version. My final version will be tidied up and mounted on the back of my stand - nothing to assemble. It will be almost invisible and a very neat solution for my application. You don’t need such long ‘legs’ you could make them any length you require – just means more fittings. This is a way neater solution than an immersion chiller and fully cleanable...Plate chillers are great but you cant 'see' and you cant be 'sure' thats why the big boys use the ones that can be taken apart and cleaned. To be honest the Goat has such a massive bore that you wont need to use a brush, just recirculate your cleaner and flush as part of your process. job done.

[JamesF]

Very interesting project and I applaud your dedication to experimentation

I suspect there is no "ideal" arrangement of pipe diameters and lengths, but perhaps there are ones that are more suited to given flow rates of wort and water. Presumably you want the coolant coming out to be as close to the wort temperature as possible, but only just (ie it shouldn't reach that temperature until it's at the end of the chiller) for maximum efficiency and if you can't achieve that with useful flow rates of wort and coolant then you have to start playing with the chiller length or the pipe diameters.

It's certainly something I shall look at though, as this sort of design would fit quite neatly out of the way under a worktop or even mounted on a wall or the ceiling.

James

[lord groan]

Setmash,
Are you set up to be able to read the coolant outlet temperature alongside the current wort temperature. I imagine that maximum efficiency would occur where the cooling water is coming out at fractionally below the wort temp.

[JamesF]

I think achieving the highest possible coolant output temperature might be the key to minimising the loss of efficiency when the initial coolant temperature is higher, too.

For example, if in winter the coolant starts at 5C and the output temperature is 85C then obviously the temperature has been raised by 80C. If the coolant is 15C initially in the summer and leaves at 85C, you've lost 10C or 1/8th (12.5%) of its cooling capacity. On the other hand, if the output temperature is, say, 65C with 5C and 15C initial temperatures then you've lost 1/6th (just over 16.5%) of the cooling capacity with the higher initial temperature.

James

[setmash]

Raising the coolant output temperature is key to efficiency. The closer you can get the coolant to the input wort temp the better. The coolant temp increases with goat length and slower coolant flow. HOWEVER it appears from my tests if you slow the coolant flow down too far then the wort temp output rises dramatically. I need to do more tests but it may be to do with the pressure inside the HEX. There seems to be an optimum flow rate for both coolant and wort.

[setmash]

Latest data table with all tests. Added Chiller Z data from their tables for comparison.



Video now embedded

[setmash]

My final build. I added turbulation to the 10mm inner tubes by replacing the cutting blade in a pipe cutter with a 5p. The new chiller is mounted to the back of my brew stand - I choose to make it so that each arm of the goat is below the other so it is as flush to the back of the stand as possible and that it drains well. In hindsight I may modify this slightly so that it finishes a bit higher off the floor to make it easier to connect the drain pipe.

Adding the turbulation reduces the wort flow rate to about 3.6 LPM. The coolant flow rate is 4 LPM. It cooled 44 Liters of wort to 20 degrees in 12 mins and used about 55 Liters of coolant (9 degrees input temp). It also seems to be a lot more aggressive at chilling to lower temperatures, reducing the wort flow just slightly results in plunging wort temperatures - I could get the wort sub 12 degrees no bother, bodes well for summer cooling & lager. I circulated PBW as part of the cleaning cycle integrated in to my brew day so cleaning added no time to the brew schedule. ...and its spotless on the inside afterwards, no hop would ever lurk in here and if it did....out comes the Goat brush to flush the fecker out

Building the Goat - you can see the general idea:



Adding the turbulation to the 10mm pipe



Turbulation on the inner 10mm pipe



Pipe work at the back of the stand - you can see the coolant output feed



Coolant drain at bottom of stand - 10mm pipe is the hot wort input



10mm ends that carry wort



Rear View of rig with chiller mounted on stand:





Front view of rig with chiller mounted at the rear