Smart questions
Smart answers
Smart people
Join Eng-Tips Forums

Member Login

Remember Me
Forgot Password?
Join Us!

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips now!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

Join Eng-Tips
*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Jobs from Indeed

Link To This Forum!

Partner Button
Add Stickiness To Your Site By Linking To This Professionally Managed Technical Forum.
Just copy and paste the
code below into your site.

Vig16 (Mechanical) (OP)
25 Mar 11 12:30
OK, so I've been working on a few different things lately and one that I've been fumbling around with is being able to heat up water to about 70 deg C, then back down to about 38 deg C...all in less than 2 minutes too lol.

I can't really think of a way to get the water down to almost half the temperature of what it was in that short of a timespan.  I'm sure by having a fan at one end of the tube of passage and having that moving quickly I could do it, but I haven't had much luck with that theory...

Any ideas would be GREATLY valued!

Thanks so much!!
MintJulep (Mechanical)
25 Mar 11 12:37
How about just not heating it up in the first place?
ione (Mechanical)
25 Mar 11 12:39
You've apparently forgotten to specify the amount of water to be cooled (not a trivial detail)
Vig16 (Mechanical) (OP)
25 Mar 11 12:49
Mint -  I heat it up to kill some contaminants in whatever I'm trying to heat up.

ione - I'm not trying to do it to a big volume of water, ~8 ounces.
amorrison (Mechanical)
25 Mar 11 13:24
I would look for  some type of solid copper spheres (the size of bb's or shotgun shot). Have one "cup" of hot spheres for the heating that are added to the water then the water is poured out into a cold "cup" filled with chilled copper spheres.

Calculations NOT to follow(:<)).

IRstuff (Aerospace)
25 Mar 11 13:57
TE coolers might do the job, with adequate heat sinking.  The standard polymerase chain reaction (PCR) system heats to ~98ºC and cools down to around 20ºC within seconds, albeit with rather tiny volumes.

Your nominal conditions will require about 253 W of cooling.  Can you physically move the container from the heater to a cooler in a aluminum container?  What kinds of container can you use, i.e., something with lots of fins in it?


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

ione (Mechanical)
25 Mar 11 14:08
3 kg of ice might do this job
Vig16 (Mechanical) (OP)
25 Mar 11 15:41
Thanks for all the replies again, guys!

IRstuff - are TE coolers what's used in PCR to get the temperature to drop so quickly?  I read about PCR, but didn't see anything relating to TE cooler.

And the container cannot be moved from under the nozzle to a cooler in an alumininum container.  I can use any container that would allow me to pass the 70 deg C water through for a short period of time (even just a pulse), then I'd have to introduce some way of cooling that initial "pulse" down to about 38dC.

ione - the ice would definitely do the trick, but I wanted to avoid ice to reduce the temp, sorry for not including that in my description.
IRstuff (Aerospace)
25 Mar 11 15:52
The TECs are built in to the hardware, which consists of a thermal plate with wells for the ampules, and underneath the plate is a TEC that does both the heating and cooling.  PCR is relatively mindless; you simply load up the ampule with the DNA strands to be replicated, add in the reagents, and run the temperature cycle about 20 times or so.  The high temp step breaks up the replicated strands and the low temp step starts the replication process over again.

The thermal conductivity of ice is quite low, relative to water, but, lots of really cold water, say 5ºC, with lots of flow and lots of contact area is certainly a possibility.  Hypothetically, a 32ºC reduction in 8 oz of water can be accomplished by raising 7 cups of 5ºC water to 10ºC.

How often do you need to do this, as this would scope the overall cooling capacity needed.  If it's one cup per day, that's quite different than one cup every 2 minutes.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
25 Mar 11 15:58
Well, I think it would probably be needed once every 2-3 hours.  I don't have a problem with getting the heat up to 70dC, I just need to find a way to get the end product down to 38dC.

Would a TEC do this for me?
vpl (Nuclear)
25 Mar 11 16:05
Are you thinking about something like this?

It looks like they keep the water hot.  If you want the final product cooler, you add a bit of cold water from the cold water dispenser.

(By the way, there is pdf of the parts manual on the link.  You might get some ideas from looking at what they use.)

Patricia Lougheed


Please see FAQ731-376: Forum Policies: Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

Vig16 (Mechanical) (OP)
25 Mar 11 16:24
vpl - that works too, but I'm looking to have it all done at once, so the final product will have 38dC water when you take the cup/bottle/container away.
IRstuff (Aerospace)
25 Mar 11 16:36
OK, so two of these guys: face-to-face, with serious heatsinks on the outsides, and a thermally conductive labyrinth between the faces for the water to flow through.  Assuming you can throttle the flow down to 0.066 oz/s, there should be sufficient cooling capacity to get the temperature down 32ºC.  That works out to be about 38 drops/s winky smile


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
25 Mar 11 16:55
So when you say face-to-face, should the hot side of one TEC face the cold side of the other TEC while the water tube runs through a "maze" of some sort and just have the TEC's programmed to have the hot side that contacts the water tube activated to heat the initial flow of water that's going through, then have that TEC shut off and activate the cold side of the other TEC as the water is passing through?
IRstuff (Aerospace)
25 Mar 11 17:36
No, both cold faces facing each other, since you need a minimum of 250W of cooling, and the TECs cited only provide about 200 W each at 30ºC deltas.  Now that I relook at that it might be better to put both of them in series rather than face-to-face, since you're not going to get much cooling at the start of the maze.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

macmet (Materials)
27 Mar 11 12:36
I've used TEC's and they work fairly well. I was using them in an application where size was critical and I found I had trouble cooling the hot side.  Depending on your system, this may not be an issue for you.

Have you thought about adding dry ice?  I used to use that to kick start the process.
Vig16 (Mechanical) (OP)
28 Mar 11 8:03
IR - Thanks for your advice!  I totally understand what you mean about running them in series as there won't be much cooling at the start of the maze.  I have enough wiggle room to try both configurations, but I'll give the series configuration a shot first...thanks again for all your help!!

macmet - I thought about dry ice, but I'd like to refrain from using ice to do this as I want as little operator assistance during use.  Thanks for the input though!!
Vig16 (Mechanical) (OP)
28 Mar 11 8:32
IR - I had one question about your amount of energy it takes to cool the water 32 degrees C.  You said it takes 253 W to do so; however, when I do the calculations, I only get ~134 Watts.

Since 4.186 J is needed to drop water by 1dC, I did 4.186*32=133.952.  And since 1 watt=1 Joule/1 sec, I don't see where you got the 253.

I probably calculated it wrong, but if so, where did I go wrong in my calculations?
ione (Mechanical)
28 Mar 11 9:21
Mass = 8 ounces = 0.223 kg (approx)

Specific heat of water = 4187 J/kgK

Temperature decrease = 32 °C

Time = 2 mins = 60 seconds

Power = 4187 * 0,223 * 32 /60 = 498 W (approx). Obviously if time available for cooling process decreases power will increase
Vig16 (Mechanical) (OP)
28 Mar 11 9:47
ione - Shouldn't it be ~249 W since it's going to be over 2 minutes?
IRstuff (Aerospace)
28 Mar 11 10:16
yah, 2 min is a bit more than 60 s

Shouldn't it be ~249 W since it's going to be over 2 minutes
> That sounds familiar! winky smile  You need to have remember that specific heat is in joules, and the 2 minute time element is what gets you watts.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
28 Mar 11 10:51
LOL, so did I forget to convert something in the way I arrived to 249 W?
IRstuff (Aerospace)
28 Mar 11 15:16
Ione's calculation, using 120 s, instead of 60 s, is the correct calculation.  You didn't divide by time, at all, even though you mentioned watt = joule/s


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

IRstuff (Aerospace)
28 Mar 11 15:19
Ohh, and YMMV, i.e., your mileage may vary.  Since the TEC does not do much pumping when deltaT is either zero, or max, the 250W is a ballpark estimate.  Ione's 500W might be closer to what might be actually needed, particularly in light of the fact that my TEC selection was two 400 W,max TECs.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
28 Mar 11 15:19
Oh, OK...thanks!
Vig16 (Mechanical) (OP)
28 Mar 11 16:27
IR - am I able to take the TEC's apart for do my own configurations for the labyrinth?  Or is there a different way I should do that?
Compositepro (Chemical)
28 Mar 11 16:52
Cooling could be done simply by running your hot water through a copper tube coiled in a water bath. 38C is far above room temp. so the water bath will cool naturally.
You could even use a tube-in-tube heat exchanger where cold water flows through a copper tube into a heated metal block, gets hot, and then flows out of the block through a tube that surrounds the the copper tube with the water coming-in (a counter-flow heat exchanger).  
IRstuff (Aerospace)
28 Mar 11 18:43
No, TECs must be treated as blackboxes that should never see the light of day.  I would think that a serpentine would be adequate.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

macmet (Materials)
28 Mar 11 19:02
Composite pro is on to something.  But I would suggest that you consider cooling the cold fluid to a low temperature (i.e. below 0).  When I did it I used propylene glycol so I could cool it with minimal concern of ice.

I used TEC's to cool the propylene glycol.
Vig16 (Mechanical) (OP)
29 Mar 11 7:51
IR - Since I can't open the TEC, how would I go about getting the serpentine path to touch the cold face of the TEC?

Compositepro and macmet - I like those ideas too, so I'm going to give yours and IR's a shot to see which is best for my application!  Thanks!
IRstuff (Aerospace)
29 Mar 11 10:21
My starting point might be a copper block attached to the TEC with high thermal conductivity adhesive.  A copper pipe serpentine would be soldered, brazed, or whatever, to the copper block.  The thermal mass of the copper pipe and block would help with the thermal efficiency, assuming you're allowed to precool the serpentine.  

The precool temperature should be as low as practical, since thermal transfer is proportional deltaT.  The colder the serpentine, the faster it cools, provided you don't get freezing or other annoying side effects of cold objects.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Compositepro (Chemical)
29 Mar 11 18:34
Those who brew beer at home must cool 5 gallons of boiling wort to less than 100F for fermentation. This is done in a few minutes by siphoning the hot wort through a copper tube coiled in a sink full of water. The top layer of water in the sink will become quite warm while the bottom stays cool. Keep it simple.
bithkits (Mechanical)
30 Mar 11 8:17
Be careful with TEC's. How will you cool the hot sides? Something to think about. If the hot side is not adequately cooled the TEC will actually *HEAT* up the cold side and put more heat into the system.

Stacking/staging doesn't work for larger heat loads - each layer has to remove the heat from the previous TEC as well as the electrical power supplied to that previous TEC.

This is important -  a 200W TEC puts out way more than 200W of heat on the hot side and only removes 200W of heat on the cold side for large delta T.

People also tend to underestimate the electrical power supply for TEC's - they can really eat up the amps.

On the other hand, reverse the polarity and the TEC's become heaters so you could use the same TEC's to heat and then cool the fluid.

I don't quite follow why you would want to open/dismantle the TEC's? Are we talking about the same thing here - thermo-electric coolers, also known as Peltier coolers...

See this site for more info (albeit for computer cooling):

I am a Mechatronics Engineer from South Africa.

Vig16 (Mechanical) (OP)
30 Mar 11 8:24
Compositepro - I like that idea, but how long does it take to cool the water for that volume?

bithkits - I heard the same thing from one of the other engineers here yesterday, which kind of gave me second thoughts about that.  Also, it seems to be a little more expensive than what we were looking to spend.
IRstuff (Aerospace)
30 Mar 11 10:05
We're not talking about stacking though, just talking about making the cooled area longer than it is wide.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
30 Mar 11 12:19
I was just thinking of something else with cooling the water.  What if I used air, such as a fan, to be blown through the water?  What calculations do you think would be necessary to see how much air I'd need to get the to cool the water?
Vig16 (Mechanical) (OP)
6 Apr 11 13:37
Sorry to bring up this old thread again, but I just came up with another idea.  What if I used a Phase Change Material (PCM)?  I found this material ( which would put me right in the 35-38°C range and used 1.25g of the PCM (since 1g would dissipate ~200 J and I need to dissipate ~250 J).

My plan for this is:

1.  Have the 70°C water running through the Silicone tubing that I have

2.  Insulate the Silicone tubing with the PCM which would be encapsulated with a copper tube.

Does this look like something that would work or am I misunderstanding PCM's?
Compositepro (Chemical)
6 Apr 11 14:01
Great idea! One of the more common PCM's is called "ice".
IRstuff (Aerospace)
6 Apr 11 14:41
The issue is rarely the material, the question is how to get access to the actual properties of the material in the timeframe alloted, without any ancillary side-effects.  

You'd need to come up with some sort of flow-through sieve-like structure that won't contaminate your samples with itty-bitty balls of PCM.  Then you need to bring the PCM back down below its melting point, which requires some sort a cooling system.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Vig16 (Mechanical) (OP)
6 Apr 11 14:48
Would I actually have to have the PCM's "in" the liquid?  Couldn't the PCM just be outside of the thin-walled Silicone tubing that the hot water is running through?
IRstuff (Aerospace)
6 Apr 11 14:55
You could, if you want to take 10 or 20 or whatever minutes it's going to take.  Silicone is considered to be thermal insulator.

You really need do the analysis, and your unfamiliarity with how the cooling is being accomplished is severely hampering your path to a solution.  You should read up on heat transfer; there are gobs of resources, including Wikipedia, a free HT textbook: all on the web.


FAQ731-376: Forum Policies
Chinese prisoner wins Nobel Peace Prize

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!

Back To Forum

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close