Conical Cooling Conundrum (Spike CF10)

[mongoose33]

Is the krausen line you are talking about on the coil or on the side of the tank? I realize the coil is attached to the lid so it's being tilted so you can take the pic, but it seems to me if you had around 6 gal in the conical that only 3-4 of the coils were actually in the wort which would possibly lead to lack of cooling performance. Anyway you could put an identical amount of water that you had in the last batch into the conical and rerun the temp tests and then try it with 10-11 gals to see if there's a difference?

I'm pretty sure that the krausen line on the coil corresponds to the line on the side of the conical, but now you have me questioning that a bit. Later I'll double check that. If only 3 or 4 of the coils were in the beer, I agree--that's not a lot of cooling.

 

[btcost]

X2 on how many coils were in contact with the beer. You will need all of them in contact to cool efficiently.

If the water in the beer didn’t freeze @ 28. It shouldn’t @ 22. Below freezing is below freezing. And the beer itself will never get to 22

 

[mongoose33]

X2 on how many coils were in contact with the beer. You will need all of them in contact to cool efficiently.

If the water in the beer didn’t freeze @ 28. It shouldn’t @ 22. Below freezing is below freezing. And the beer itself will never get to 22

I really didn't think it would freeze at 28 degrees; not only is there alcohol in the beer which reduces the freezing point, but it was also under a 10 psi pressure which should also have limited freezing.

So you think that under similar circumstances, the water wouldn't freeze on the coils at 22?

 

[Auger]

My first thought was to slow down the flow of coolant through the coils, but if your beer is at 38 and your coolant exit temp is 35, realistically there's not much more energy that coolant is going to pull out of the beer through a SS coil. I think with that beer volume you're likely reaching the temperature equilibrium.

X2 on how many coils were in contact with the beer. You will need all of them in contact to cool efficiently.

If the water in the beer didn’t freeze @ 28. It shouldn’t @ 22. Below freezing is below freezing. And the beer itself will never get to 22

this would be true if the beer temp was uniform throughout the volume. The beer closest to the coils will cool down faster and if it's going to freeze, it will freeze there first. That's why in the pictures posted, there was ice only on the coils vs the whole volume of liquid suddenly flashing from liquid to solid uniformly.

 

[mongoose33]

X2 on how many coils were in contact with the beer. You will need all of them in contact to cool efficiently.

Is the krausen line you are talking about on the coil or on the side of the tank? I realize the coil is attached to the lid so it's being tilted so you can take the pic, but it seems to me if you had around 6 gal in the conical that only 3-4 of the coils were actually in the wort which would possibly lead to lack of cooling performance.

OK, home now, and I double-checked the depth of the coil into the fermenter: the krausen on the coil lines up with the krausen line around the fermenter. So most of the coils were submerged.

 

[auburntsts]

Hmmm. Seems the only variable left is the glycol cooling side of the system. As has been previously suggested, maybe your fridge setup isn’t up to the task. Perhaps we can get someone who has a CF10/TC-100 combo and has gone with a dedicated chiller (DIY, aquarium, etc) to post how low they can chill 5 and 10 gal batches to so we can compare numbers.

 

[mongoose33]

Hmmm. Seems the only variable left is the glycol cooling side of the system. As has been previously suggested, maybe your fridge setup isn’t up to the task. Perhaps we can get someone who has a CF10/TC-100 combo and has gone with a dedicated chiller (DIY, aquarium, etc) to post how low they can chill 5 and 10 gal batches to so we can compare numbers.

That has occurred to me too, except....the reservoir of the chiller holds at 28/29 degrees. I think the issue lies in either the flow rate, or that there's too much heat being absorbed by the fermenter.

Heard back from Spike today. Here's the conclusion: "We only recommend 38 as a cold crash standard, so you are just at the temp we recommend."

Well...if so--and I can't find anywhere that this "38 as a cold crash standard" is listed on their website....it appears I'm at the best I can do. I might try to find a better pump and try that, but it sure seems as if I should be able to take a 5-gallon batch lower than that.

It is what it is.

 

[btcost]

My first thought was to slow down the flow of coolant through the coils, but if your beer is at 38 and your coolant exit temp is 35, realistically there's not much more energy that coolant is going to pull out of the beer through a SS coil. I think with that beer volume you're likely reaching the temperature equilibrium.




this would be true if the beer temp was uniform throughout the volume. The beer closest to the coils will cool down faster and if it's going to freeze, it will freeze there first. That's why in the pictures posted, there was ice only on the coils vs the whole volume of liquid suddenly flashing from liquid to solid uniformly.

In theory, the thermodynamic action taking place inside the fermenter shouldn’t allow any beer to be in contact with the sub-freezing coil long enough for it to freeze.

There should still be enough heat energy in the beer to keep freezing from happening. As the warm beer moves towards the cold coils.

I am thinking that everything the original poster is doing is simply right on the “edge” of what’s possible with this set up.

One 32 degrees is right at freezing.

His glycol was running 28. And he achieved 10 degree delta (right where it belongs)

It was “only” a 6 gallon batch. So there were exposed coils. You can’t underestimate how much heat energy was lost to these exposed coils.

I’m curious if the system will work better @ 10 gallons. In fact I think the original poster said it worked perfectly in his water “dry run”. Which was with a full fermenter.

 

[mongoose33]

ANOTHER UPDATE:

I checked the flow rate on the pump this morning. I was shocked to find that in 30 seconds the pump produced.... a quart. A single quart. I also hooked up the copper coil in the loop, and the throughput was virtually the same--a quart in 30 seconds.

That's 1/2 gallon per minute, 30 gallons per hour. That strikes me as very slow. The pumps that are sold on the Penguin Chiller site are rated at 3.5 GPM.

I have a Jaded Hydra immersion chiller. The Jaded people note that for maximum cooling, you need to run the water through the chiller as fast as it will flow. Faster water = faster chilling. Should be the same with the fermenter chiller--faster is colder.

So at the moment, I'm thinking the pump is the problem. I understand about the delta, but this flow rate is so slow that it has to be related to that.

 

[btcost]

Yes. Chilled water is very sensative to flow rate.
Too slow, not enough heat exchange, too fast! Not enough heat exchange.

Fingers crossed getting the flow rate where it belongs fixes it.

 

[otbrewin]

ANOTHER UPDATE:

I checked the flow rate on the pump this morning. I was shocked to find that in 30 seconds the pump produced.... a quart. A single quart. I also hooked up the copper coil in the loop, and the throughput was virtually the same--a quart in 30 seconds.

That's 1/2 gallon per minute, 30 gallons per hour. That strikes me as very slow. The pumps that are sold on the Penguin Chiller site are rated at 3.5 GPM.

I have a Jaded Hydra immersion chiller. The Jaded people note that for maximum cooling, you need to run the water through the chiller as fast as it will flow. Faster water = faster chilling. Should be the same with the fermenter chiller--faster is colder.

So at the moment, I'm thinking the pump is the problem. I understand about the delta, but this flow rate is so slow that it has to be related to that.

Hi mongoose33, I just saw this thread and it sounds like an exam question from decades back when I used to know how to do thermodynamics and fluid dynamics calculations. They say it is important to solve complex problems to keep your mind sharp as you age, so here does nothing...

I’ve been trying to think of what could cause such a significant change in performance between your test run with starsan and your full batch and I think you are on the right track with the pump/coolant flow rate being the culprit. Higher flow is always better through a heat exchanger system like this (outside of the heat added by a bigger pump to increase the flow). If you go too slow, you stop mixing the coolant within the lines (laminar flow), which will dramatically reduce your ability to cool because it basically turns the coolant near the edges of the cooling coil into an insulation layer. This loss of mixing (switch from turbulent to laminar flow) happens very abruptly with minor changes in flow rates, so it is possible you had a high enough flow rate during your starsan test to get good performance, but the pump has degraded a little or there is a little more restriction in the lines (or you added more glycol to the coolant mix?). If you don't already have a second batch going, I’d suggest running another starsan test with the exact same configuration you currently have and see how it performs.

One other thing to look out for is short circuiting from the return loop back into the pump inlet due to the combination of the low flow rate and placement of the return compared to the pump inlet. Make sure the return loop is as far away from the pump inlet within your reservoir as possible, and I would put the sensor for your coolant reservoir right in the path of the returning coolant. Again, this could explain the difference between your test run and your real batch if things shifted around a little between the two runs.

Keep us updated on how your next batch or test goes!

 

[mongoose33]

Hi mongoose33, I just saw this thread and it sounds like an exam question from decades back when I used to know how to do thermodynamics and fluid dynamics calculations. They say it is important to solve complex problems to keep your mind sharp as you age, so here does nothing...

I’ve been trying to think of what could cause such a significant change in performance between your test run with starsan and your full batch and I think you are on the right track with the pump/coolant flow rate being the culprit. Higher flow is always better through a heat exchanger system like this (outside of the heat added by a bigger pump to increase the flow). If you go too slow, you stop mixing the coolant within the lines (laminar flow), which will dramatically reduce your ability to cool because it basically turns the coolant near the edges of the cooling coil into an insulation layer. This loss of mixing (switch from turbulent to laminar flow) happens very abruptly with minor changes in flow rates, so it is possible you had a high enough flow rate during your starsan test to get good performance, but the pump has degraded a little or there is a little more restriction in the lines (or you added more glycol to the coolant mix?). If you don't already have a second batch going, I’d suggest running another starsan test with the exact same configuration you currently have and see how it performs.

I plan to do this--right now I have another batch going so can't do it, but I plan on it. There's one other clue here I'm going to document in the next post.

One other thing to look out for is short circuiting from the return loop back into the pump inlet due to the combination of the low flow rate and placement of the return compared to the pump inlet. Make sure the return loop is as far away from the pump inlet within your reservoir as possible, and I would put the sensor for your coolant reservoir right in the path of the returning coolant. Again, this could explain the difference between your test run and your real batch if things shifted around a little between the two runs.

There's about maximum distance between the return line (which comes out of the copper coil on one end of the reservoir), and the pump, which draws from the opposite side. Good idea though.

Keep us updated on how your next batch or test goes!

Oh, I will. Part of me is intrigued by the puzzle, the other part of me just wishes the chilling performed better.

 

[mongoose33]

Another clue, related to the exposed coils in the fermenter. Above the pic showing the krausen ring indicates about 1/3 of the chilling coil in the fermenter is not submerged.

I'm doing another batch and I capture the produced CO2 off the fermenter and run it through the soon-to-be receiving keg to purge it of air using the pure CO2 from fermentation. I connect it to the "OUT" post.

Because I like to monitor the progress of the fermentation, I connect another line to the "IN" post and run that into an airlock jar. This lets me determine when fermentation begins as well as monitor the vigor of the fermentation.

It's bubbling happily away, but i noticed something interesting. When the chilling pump turns on, directing 30-degree chilling fluid through the partially-submerged coils in the fermenter...the bubbling stops for a minute!

The only thing I can attribute this to is that the exposed coils are cooling the headspace gas, which then contracts a bit...stopping the bubbling. It then starts again after a bit.

The two closeups of the airlock jar below show, first, the drop tube in the jar when bubbling. I tried to catch a pic w/ a bubble rising but as a photographer I'm not, apparently, that good.

The second closeup shows that tube when I just turned on the pump. Note how the liquid in the tube has drawn back.

I didn't figure the exposed coils were all that important in the grand scheme of things, but they do seem to have an effect.

 

[mongoose33]

This is the never-ending story.

Spike replied, and suggested I test the pump without running it through the coils. Duh! I can't believe I didn't think of that.

Anyway, I removed the lines from the coil into the fermenter, turned on the pump--and proceeded to get my pants and the floor wet as it gushed out of the pump like a fire hose.

It's not the pump.

The only other alternative now is that there's a blockage or kink someplace. I blew through the return line, doesn't appear there's a blockage there, but hard to tell just blowing air through it. I'll do some more testing tonite. Harder to do because I have a batch working in there. I'd love to pull the coil and test it separately but I won't do that until this batch is done.

 

[jsharp4684]

@mongoose33 I'm working on cold crashing my first batch in my CF5. I'm using a DIY glycol chiller made with a 5k btu A/C unit. Like you, I am able to hold 28F reservoir temps, but can't get 6 gals of beer below 38F. I prefer a 32-34F cold crash, as that's what I've always done with my ferm chamber.

Have you considered that maybe we're asking the pump to push too much glycol vertically? My CF5 is on a stainless table and my chiller is on the floor. It's pumping about 5 feet vertically. I haven't checked my flow rate, though.

Have you made anymore progress?

 

[brewbama]

I can't comment on the system getting down to 32F as I've never tried. I've had my conical in a cold garage and inside my warm house; I've never had an issue. I'm going off of what multiple professional brewers have told me they cold crash at. You're try to crash below what is recommended which is what I suspect is causing issues.

Though I don’t use this system, as a reference point, I cold crash to 32*F. I do this because I asked several folks (pros and hobbyist) about it and they say that’s what they do.

When asked, Denny Conn said, ”3-7 days at 33F. Length depends on when I get around to kegging.”

Another said “limit my cold crash to 24 hrs at 30F.”

And another said, “lower the temp to 29-33 degrees rapidly.”

I guess my point is it’s not as unheard of as you make it seem.

 

[mongoose33]

@mongoose33 I'm working on cold crashing my first batch in my CF5. I'm using a DIY glycol chiller made with a 5k btu A/C unit. Like you, I am able to hold 28F reservoir temps, but can't get 6 gals of beer below 38F. I prefer a 32-34F cold crash, as that's what I've always done with my ferm chamber.

Have you considered that maybe we're asking the pump to push too much glycol vertically? My CF5 is on a stainless table and my chiller is on the floor. It's pumping about 5 feet vertically. I haven't checked my flow rate, though.

Have you made anymore progress?

Well, it makes me feel marginally better that I'm not the only one.

I'd been in communication w/ Spike about this--the suggestion was to check the flow coming out of the coil, and I got about 2.25 quarts in 30 seconds, or about 1.25 gallons a minute. But it slows from 2.25 out of the coil to about 1 quart coming out of the return line.

Unfortunately, I haven't been able to pursue this further. The last place I have to look is the return line, it's possibly kinked as it moves through the top of my refrigerator. I don't think it is, but until I check I won't know for sure. I had a brew in there while much of this was going on, and since then I just haven't had time to fool with it.

One question for you: are your coils completely submerged? Mine has about the top 1/3 exposed with a 5-gallon batch as it's a 10-gallon fermenter.

Something I checked with mine was the temp of the return glycol--it picked up about 5 degrees of heat. I still wonder if there's just too much exposed surface on the fermenter picking up ambient heat. I'm also pretty sure that if I could increase the flow rate it would do better. I've even thought of putting two of these pumps in series to see what that might do.

 

[jsharp4684]

I have two coils that aren't in beer. I think all the conicals use the same number of coils, with longer "legs" to the 4" TC flange. I'll try to measure the return temp. Something I have noticed: the 2" valve and piping from the bottom aren't as cold as the 1.5" accessories that are higher in the conical.

 

[jsharp4684]

I have a theory. The flow is too high. The stainless coil isn't an effective enough conductor to get the temperature down that low with that small of a temperature delta and a constant flow of glycol. I found that I can get the beer down below 38F by cycling the pump and allowing the glycol to sit in the coils for a bit. I'm sure there is some fancy math that would dictate the optimal length of time, then the pump could be "pulsed" to exchange all the glycol in the coil, then allow it to sit, and repeat. I'm sure there's a way to do this with a BrewPi or something. You might be able to accomplish the same thing by unplugging the heater and babysitting the temperature controller while stepping it down 1 degree at a time.

 

[mongoose33]

I have a theory. The flow is too high. The stainless coil isn't an effective enough conductor to get the temperature down that low with that small of a temperature delta and a constant flow of glycol. I found that I can get the beer down below 38F by cycling the pump and allowing the glycol to sit in the coils for a bit. I'm sure there is some fancy math that would dictate the optimal length of time, then the pump could be "pulsed" to exchange all the glycol in the coil, then allow it to sit, and repeat. I'm sure there's a way to do this with a BrewPi or something. You might be able to accomplish the same thing by unplugging the heater and babysitting the temperature controller while stepping it down 1 degree at a time.

I appreciate the reply.

I think it's actually the opposite, though--faster is better. You're right, stainless steel isn't a particularly good thermal conductor, but the faster the glycol goes through the coils, the more it cools the inside of the coils--which then conduct heat from the wort/beer.

I still have a few things to try to nail some of this down, I'll keep reporting as I learn things.

 

[tdibratt]

Hi folks wondering if you all figured out the problem. I have my glycol at 26F and cannot get my CF10 below 44. Have 8 gallons in the conical doing a test run.. coils are not frozen and pump is running. Any ideas?

 

[mongoose33]

Hi folks wondering if you all figured out the problem. I have my glycol at 26F and cannot get my CF10 below 44. Have 8 gallons in the conical doing a test run.. coils are not frozen and pump is running. Any ideas?

What's your ambient temperature? What kind of glycol chiller are you using?

Do you have the neoprene jacket on it?

 

[jsharp4684]

I solved mine by following Spike’s suggestion. I swapped my input and output and now the system has no trouble cold crashing to 34F and keeping it there.

 

[tdibratt]

Ambient temps in basement are around 70 F. I have the complete T100 pkg from spike which includes the jacket and it is on. The chiller is made from a 5000 BTU AC. I have 2 gallons of Glycol and 6 gallons of water in an insulated cooler. It gets down to 28F no problem and I have it for 1 Deg variance. Morning after and Conical is 43.7F. Set for 38F. Pump is running gycol through. I had the input/output set and then realized I should have the input set on the length of SS that goes straight down to bottom then turns back up through spiral and then back to cooler. Still cannot get better that 43-44F. My solution works out to be 25% Glycol, but no signs of freezing or crystals in the mixture.. Further information, the chiller coil has frost on it above the water line, and glycol still moving through. The temp of the glycol is 28F as measured from a temp probe as well and the glycol returning to the cooler is about 29.1F.

 

[btcost]

Ambient temps in basement are around 70 F. I have the complete T100 pkg from spike which includes the jacket and it is on. The chiller is made from a 5000 BTU AC. I have 2 gallons of Glycol and 6 gallons of water in an insulated cooler. It gets down to 28F no problem and I have it for 1 Deg variance. Morning after and Conical is 43.7F. Set for 38F. Pump is running gycol through. I had the input/output set and then realized I should have the input set on the length of SS that goes straight down to bottom then turns back up through spiral and then back to cooler. Still cannot get better that 43-44F. My solution works out to be 25% Glycol, but no signs of freezing or crystals in the mixture.. Further information, the chiller coil has frost on it above the water line, and glycol still moving through. The temp of the glycol is 28F as measured from a temp probe as well and the glycol returning to the cooler is about 29.1F.

Can you try to slow down the glycol flow rate? If the chilled water (glycol) is moving too fast. You won’t exchange the heat.

Aim for 10 degree delta. So 28 in 38 out. You will get that slowing down the flow rate.

 

[mongoose33]

Can you try to slow down the glycol flow rate? If the chilled water (glycol) is moving too fast. You won’t exchange the heat.

Aim for 10 degree delta. So 28 in 38 out. You will get that slowing down the flow rate.

See, I don't get this. I have a Jaded Hydra immersion chiller. The thing is a beast. What Jaded says to do to get the best chilling performance is to run the water as fast as possible.

I don't understand how the glycol isn't going to chill the insides of the tubing worse if the glycol is moving fast than if it's moving slowly. All the stainless tubing knows is how cold is the material next to it, flowing fast or not.

I'm always willing to learn something new, and if I'm wrong on this, please correct, but I can't see how slowing it down improves the performance.

 

[SanPancho]

one problem with nice round coils like these is that you actually want a messy and turbulent flow inside the tubing. if you get laminar flow your cooling power goes to shitt. higher speeds contribute to that.

 

[mongoose33]

one problem with nice round coils like these is that you actually want a messy and turbulent flow inside the tubing. if you get laminar flow your cooling power goes to poopyt. higher speeds contribute to that.

In theory I think I could test this. I have a Penguin chiller and two pumps; I can hook them in series, one pushing, one pulling, and I increase throughput about 50 percent when I do that.

The problem is that the temp of the mix coming out should be lower than if at a slower speed. But it's not clear what that would indicate--either that it's not picking up as much heat....or that it's actually chilling the coils to a lower temp than would otherwise be the case.

The only good test is the temp of the wort--and that I can do. I don't know if I'm brewing this weekend or not, but when I do, I'll try to do something with this.

BTW, yours is the first reasonable theory I've read about why faster would be less efficient. Not sure it's correct, but it's a nice approach grounded in some physics. BTW, what happens with the Jaded Chiller would seem to refute it, though w/ the jaded chiller it's copper, not SS, and it's water, not a water/glycol mix.

 

[SanPancho]

Jaded also has ridiculous amount of surface area vs your coil.

Surface area
Copper vs stainless
Contact time
Design efficiency
Temp differential

Lots of variables you can adjust. Jaded would be high conductivity high surface area which could make up for short contact time.

 

[btcost]

See, I don't get this. I have a Jaded Hydra immersion chiller. The thing is a beast. What Jaded says to do to get the best chilling performance is to run the water as fast as possible.

I don't understand how the glycol isn't going to chill the insides of the tubing worse if the glycol is moving fast than if it's moving slowly. All the stainless tubing knows is how cold is the material next to it, flowing fast or not.

I'm always willing to learn something new, and if I'm wrong on this, please correct, but I can't see how slowing it down improves the performance.[/QUOTE

All of my experience in this area have nothing to to with beer or home brewing but with commercial and industrial HVAC.

I am an air & hydronic systems balancer by trade. It’s my job to set the air and water flows to their correct numbers in order to achieve the engineers design. I live in Boston, so I work in a lot of hospitals and bio-labs. And of course all the schools (MIT, harvard, etc.)

Long story short I set up chilled water coils on large air handling units almost daily. And if the flow is not “right” you will not achieve design discharge air temp.

In your case, assuming you do indeed have 28 degree water (I trust you!)
The only other variable is flow rate. You are going too fast or too slow.

 

[augiedoggy]

See attached pic.

I have a 33% Glycol-67% Water mixture in a reservoir in my ferm chamber refrigerator. The system does well at moving temps down to the low 40s, but now, I can't complete a crash to 32 degrees.

Since last night, when the temp was 39.1, the temp this morning was 38.7. It simply isn't cooling, and I don't know why.

Five Gallon Batch (maybe 6 gallons in the fermenter).
Ambient in the garage: 64 degrees.
Conical wrapped in moving blanket down to floor.
Two towels on top cover the portion I can't reach w/ the moving blanket.
Condensation on both supply and return lines; they're cold.
Tilt Hydrometer in Conical confirms temp read on temp control.

Morrey does something similar with his SSBrewtech unitanks. He feeds his with a Penguin chiller which is much more responsive to temp than mine, but he keeps his at 28 degrees.

Mine is one degree lower--and the reservoir temp is more than low enough to take this down to 32 degrees.


Why? Is it possible I'm forming ice on the coils and that's limiting cooling? @Morrey suggested raising the temp of the coolant to 30 degrees. The only other thing I can think is that there's so much heat gain into the fermenter that it's offsetting the cooling. If so, how do any of the rest of you manage this?

To reiterate: it's not the temp of the reservoir. It's plenty cold, and recovers well.

View attachment 570130

from what I read from spike.. when using the stainless tubes you do NOT want to take the chiller down below 30-32 degrees since doing so causes ice to form on the outside of the cooling coils insulting it from cooling the wort. I know you show that theres no ice at the time you pulled the coil after draining so I would experiment with flow at this point.
Edit sorry just saw this is a old thread.

 

[tdibratt]

I suppose a valve after the pump and before the conical to slow down the rate would work

 

[Tartan1]

Found an old link with regards to temperature inversion. Interesting read.
https://cdn.shopify.com/s/files/1/0..._Product_Guide_FINAL.pdf?14033054371399362451

 

[laura2863]

Is it possible this is occurring? (From Spike's TC-100 pdf)