set point question

[ndbisbrew]

If I'm trying to crash to 35 F what should my chiller setpoint be set at? It's currently set at 28 F and tanks will only get to 36F

Thanks
Ted Hoffman
Founder/Brewer
Buffalo Commons Brewing Co
Mandan Nd

[dfalken]

If I'm trying to crash to 35 F what should my chiller setpoint be set at? It's currently set at 28 F and tanks will only get to 36F.

Thanks
Ted Hoffman
Founder/Brewer
Buffalo Commons Brewing Co
Mandan Nd

Curious to hear what others have to say. My chiller is set at 25 but my tanks don't get below 36F. Outside temp is around 88F during the day and 75F at night.

[SouthHouseBrew]

35F is pretty cold. Unless you're making Eisbock, that 1F difference shouldnt be a big deal.

Have you checked the temp of your glycol? You may be set at 28F, but the chiller may not be cooling to 28F.

[BrewinLou]

Could be lots of things. More info is better. Is your piping set up with the first in last out flow design? Is everything insulated properly? How far is your chiller from said tank? What else is being cooled while this tank is crashing? Cheers

[dfalken]

There are some good points made above about system design and conditions. One other thing I'll throw in is flow rate. We run glycol at around 28 and have no problem crashing tanks to 30-32 degrees. One trick I've found useful is to leave an uninsulated metal valve on the inlet and the outlet of the tank glycol (or an uninsulated chunck of copper tubing works too). When the tank has been calling for a while, check to see what kind of frost is building up on the valves. If the incoming valve is thick with frost/ice and the outgoing valve is equally so, then your problem may be that the glycol is passing through the tank so quickly that it's not effectively transferring heat, so the outgoing glycol is still super cold. If you start slowly closing back the outgoing glycol valve until it builds up minimal frost then you'll know that you exchanged a good amount of heat in passage. Interestingly enough, I've found that the same trick works to prevent icing inside the tank from glycol that is too cold. There are obviously much more elaborate ways to tuen your coolant flow rates, but I've found this one to be simple, effective and practical, not to mention low budget.

Steve, I am no cooling expert but this is not logical. If you want your wort coming out of your heat exchanger to be colder do you speed up or slow down the speed of your cooling liquid? The faster the cooling liquid flows, the more heat you are able to remove. Having said that, what you said about frost on the outgoing valve is probably true as well but not for the reason you mention. If the coolant is flowing very quickly then each unit of coolant (call it liter, gallon, whatever) picks up less heat than if it was flowing slowly but in the same amount of time more heat is exchanged because the temperature differential between the coolant and the media you are cooling remains larger.

[dfalken]

Could be lots of things. More info is better. Is your piping set up with the first in last out flow design? Is everything insulated properly? How far is your chiller from said tank? What else is being cooled while this tank is crashing? Cheers

Joel,

My piping is set up first in last out. My chiller is about 50 feet from the tanks which was my first suspicion but I installed a thermometer on the return line right before the glycol goes back into the reservoir and it is only a degree or two warmer than the glycol in the tank. It doesn't matter whether I am trying to cool all the tanks in the brewery or only one tank, I still can't make any of them reach 32F even if I run the system for 48 hours non stop. Only thing I can imagine is that the tanks are picking up more heat from the environment than the glycol can remove but I can't seem to understand how or why as they are properly insulated as far as I know. I don't see any weeping on the tank other than at the manway or at the leg connection or the non-insulated domed top but all of these places are expected to weep. My tanks have a dimple jacket on the cone and a half-pipe coil on the body so the other thing I could imagine is that the coil doesn't have enough surface area to achieve the needed heat transfer but even when I've had the tanks filled only to the top of the cone (meaning I am only using the dimple jacket for cooling) I still can't get them to 32F.

Any ideas?

[SouthHouseBrew]

Steve, I am no cooling expert but this is not logical. If you want your wort coming out of your heat exchanger to be colder do you speed up or slow down the speed of your cooling liquid? The faster the cooling liquid flows, the more heat you are able to remove. Having said that, what you said about frost on the outgoing valve is probably true as well but not for the reason you mention. If the coolant is flowing very quickly then each unit of coolant (call it liter, gallon, whatever) picks up less heat than if it was flowing slowly but in the same amount of time more heat is exchanged because the temperature differential between the coolant and the media you are cooling remains larger.

It does seem counter-intuitive at first, but cooling media flowing across a heat exchange surface at too high a velocity will not transfer heat efficiently. It takes sufficient residence time (cooling volume / flow) to transfer heat efficiently.

In your wort chiller example, the plate HEX has an enormous amount of cooling volume, so excessively high fluid flow rates are unlikely.

We have this issue at my facility when the HEX cooling valve is opened very quickly. The product outlet temp drops dramatically after its manually closed 50%.

I like the idea of the uninsulated valve. With our HEX, we know we're not cooling effectively by observing condensation on both the inlet and the outlet of the cooling lines. Same idea.

Ted - you can try reducing the flow by cracking an outlet valve. More info would be helpful.

[St Rippy]

You are right, none of us are refrigeration experts. Do yourself a favor and call one. Somebody that knows what they're talking about will be well worth your time. You may have some other piping, pump, or air bubble issues that you don't know about.
All of our tanks have a restricting circuit setting valves at the outlet of the glycol valves. Basically a globe valve for glycol. Some of our tanks have one jacket, some have two, and some have three. Regardless of the number of jackets on the tank, we have one restricting circuit setting valve tied together for all of the outlets. During our installation I questioned my refrigeration expert about this setup. I thought it was a waste of a valve. He replied with the same response that Steve stated. When your glycol is moving so fast, it won't do a heat exchange as efficiently. When it gets a little longer to linger in the jacket, it has more time to exchange the heat. It's very counter intuitive, but it's the way it works.

[BrewinLou]

What tonage is your chiller, what size pipe is your glycol header and what does it drop down to going to the tanks? Is there a bypass valve at the end of your header? If you are going out at 28 and returning at 29 you may have the high flow problem was stated here...too high a flow rate for temp exchange. See if you can throttle a valve on one of your tanks back about halfway to see if it makes a difference. If so throttle them all or put some head pressure on you chiller send pump by closing a valve on the send line after the pump a little bit. Let me know how it goes.

Joel,

My piping is set up first in last out. My chiller is about 50 feet from the tanks which was my first suspicion but I installed a thermometer on the return line right before the glycol goes back into the reservoir and it is only a degree or two warmer than the glycol in the tank. It doesn't matter whether I am trying to cool all the tanks in the brewery or only one tank, I still can't make any of them reach 32F even if I run the system for 48 hours non stop. Only thing I can imagine is that the tanks are picking up more heat from the environment than the glycol can remove but I can't seem to understand how or why as they are properly insulated as far as I know. I don't see any weeping on the tank other than at the manway or at the leg connection or the non-insulated domed top but all of these places are expected to weep. My tanks have a dimple jacket on the cone and a half-pipe coil on the body so the other thing I could imagine is that the coil doesn't have enough surface area to achieve the needed heat transfer but even when I've had the tanks filled only to the top of the cone (meaning I am only using the dimple jacket for cooling) I still can't get them to 32F.

Any ideas?

[dfalken]

What tonage is your chiller, what size pipe is your glycol header and what does it drop down to going to the tanks? Is there a bypass valve at the end of your header? If you are going out at 28 and returning at 29 you may have the high flow problem was stated here...too high a flow rate for temp exchange. See if you can throttle a valve on one of your tanks back about halfway to see if it makes a difference. If so throttle them all or put some head pressure on you chiller send pump by closing a valve on the send line after the pump a little bit. Let me know how it goes.

Thanks you guys. The high speed = less efficient cooling is certainly not intuitive but it very well be might problem. My glycol system is definitely big enough for my needs. I have 10 tons of cooling for 3 x 40HL FV and 1 x40HL BBT, in the future I will add 3 x 120 FVs and 1 x 120 BBT and I should still have enough cooling power. My main line is 2" and my flexible hose into the tanks is 1" so I don't think the problem lies there. However, our pump does seem to be oversized and everytime it turns out it shakes the lines significantly (which has also been a concern of mine). We are also having problems with the normally closed diaphragm Burkett valves allowing glycol through when closed. (I asked for ball valves but my manufacturer screwed up). I thought it might be particles from the installation getting caught and in some cases it was. But in other cases we have opened up the valves, cleaned them out completely and we still have pass-thru. We do have a bypass valve between the in leg and the out leg and I think I might have made my problem worse trying to fix it. Initially I had that valve cracked but once I noticed my tanks weren't cooling properly I ended up closing it more and more each time until it is now fully closed. So if higher speed is less efficient, then I have made my problem worse.

I am going to crack that valve open significantly which will help with the knocking on the pipes when the pump comes on, should help with the pass-thru of the solenoids and will hopefully help with our cooling problem.

I'll report back if this solves our problem.

[dfalken]

I am trying to find more info online regarding this...

I found this which seems to dissprove the idea that faster flow of the cooling medium makes heat transfer inneficient:

The rate of heat transfer, Q, is determined from the equation:

Q = WC DT + WDH

where

W = flow rate of fluid (lb/hr)

C = specific heat of fluid (Btu/lb/degrees F)

Those of you stating that faster flow leads to less efficiency could you please present some supporting evidence or explain how you know this or came to this conclusion?

thanks in advance for your help.

[BrewinLou]

I sent Jim at Pro Refer a message to see if he would chime in on this one.

[jimvgjr]

Sorry to have been late to the discussion, this is very interesting.

I've always believed more flow is better, but this is assuming you have adequate surface area on the cooling jacket to exchange the heat. On a Unitank jacket the heat exchange is very passive, and inefficient, due to a minimal percentage of product in contact with the cooling jacket surface versus total overall product volume. In contrast, your Wort HX is very efficient because we not only have the high surface area but also the velocity and turbulence that is caused by pumping the wort through the HX.

I am not sure how too much flow would hinder cooling, unless, as Steve mentioned, it causes the cooling jacket to ice up- ice is an incredible insulator. Slowing the flow down would increase the dwell time, eliminating the risk of freezing the jacket and in this case, would definitely increase cooling ability. I think I'd give this a shot.

Another issue could be from insufficient surface area of the cooling jacket. I recently had a customer with a new unitank that cooled very slow (compared with existing vessels) and rarely pulled down to the setpoint temperature, eventually they had additional surface area added to the cooling jacket and this resolved their issue.

Not sure if I have helped, but hope this issue can get resolved quickly.

Best Regards,

Jim

Pro Chiller Systems
jimvgjr@prorefrigeration.com

[dfalken]

Thanks for your contribution Jim, I think the ice theory makes a lot of sense and goes well with the empirical evidence presented that the tanks didn't cool as well when the exiting glycol was below freezing and could freeze the out valve. I have opened my bypass valve a bit and also closed the valve infront of my pump a bit to try to slow down my glycol. I have also closed slightly the lower jacket valve on the tank I am testing right now. I'll keep playing around with this and if necessary will put more backpressure on the pump and/or open the bypass more. I'll report back if I am able to get the tanks to temp.

[OneMoreBrewer]

I am of the opinion that your problem may be stratification. There are some other posts on that issue, but the signs are a good beginning drop in temp that stalls out before reaching final target. To correct this, some of us switch the valves on our jackets to break the stratification layer. I would look into that scenario as I highly doubt you are icing your beer without dropping your thermo-probe below 35^F. That being said, I have seen the insulating ice inside tanks with a similar effect. Good luck.