Off the shelf Cooling Jackets/Wrap around HX for pipework 1

[ProjProcEng]

Hi all, first time poster,

I'm hoping someone out there can provide some direction for me, I need to procure a cooling jacket for a pipe.
can someone recommend a vendor or a model
I have consulted several vendors in my locality and they have not provided any solution.

Something like the heating jacket in the link below, but with a cooling medium running through it, instead of the heating element

https://www.saequip.com/item/8/Heat/EX-Pipe-Heating-Jackets.html

I cannot get a suitable production window to install an "in line" BPHE or other conventional HX on this line before the summer. That constraint is what is driving this modification toward a "wraparound" solution.

BRIEF
A purified water line is operating at a high temperature during summer months and triggering alarms, and also damaging filters due to the high temperature.
The water must be cooled upstream of the pump, filters and associated alarms.
The purified water loop in in constant production use, so a non invasive solution, one which does not require a shutdown production window, is preferred.
The solution as envisaged by the client is a wrap around or bolt on cooling jacket, which does not require line breaks into the purified water line.

SYSTEM PARTICULARS
Medium = purified water
Purified water flow rate = Maximum 800 litres / hour
Pipe type ID ¾” or 20mm ASME BPE 316 Stainless steel
Max length for contra wraparound o = 2200mm
Water alarm temperature = 30 C
Max water temperature = 35 C
Desired temperature less than = 20 C *NB it may be desirable to over design the system to ensure no 30 C alarms are ever triggered.

EST 10- 20kw cooling max
Area of pipe exposed = 2.2 *3.14*.01*2 = .14 m2
K or U value between pipe and Jacket = TBC

 

[LittleInch]

Try these guys maybe?

http://fabritekengineers.com/bolt-on-jackets.html

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LittleInch]

Or just spray water at it?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Latexman]

Spiral wind some SS (or copper if short lived) tubing around it and hook up the water. Then insulate around it.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[LittleInch]

Of course to get to your 20C you're going to be looking at a water temp of around 10C in a counterflow situation to get there.

And you're correct - this is about 13kW of heat rejection to get your temp down by 15C.

In 2.2 m of 20mm tubing that seems a lot.

Or try this stuff and blow cold air at it ( well colder than 20c), or fit a water hose and make a lot of holes so the water drips down past it.

https://www.amazon.com/FABTEK-CLR-2-Hydronic-Baseboard-Additional/dp/B00GHUSRSY

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ProjProcEng]

I appreciate all the answers guys,

I am based in Europe, but those vendors look like a good lead.

I may need to sacrifice some of the performance in order to get the bolt on jackets but I think I've over spec'd with 20C desired Tout.

My end user will be happy if the alarms at 30c are not activated on a regular basis ( last July they went off 7-8 times a day), that means I could get away with 27C as an output Temp.
Furthermore only the very hottest days on record are 35C in my region.

Which leads me to believe that 3-4C is going to be the typical load, and I think the "bolt on" will deliver that.

Thanks again I'm impressed with the efficacy speed of response on this forum, I will be back again!

 

[Compositepro]

Those jackets, whether heating or cooling, are not good heat exchangers. They are intended to maintain temperature and not for maximizing heat transfer in BTU/(minxft2). Also heating jackets will always work more effectively than cooling jackets because there is much less limitation in delta T possible.

Do you seriously expect to get 10-20 kW of cooling through 0.14 m2 of pipe surface area?

 

[LittleInch]

You almost certainly need to add some fins of some sort even with a jacket around then.

either clamp on or fillet weld on some fins

I quite like the fin clamp things, but you're going to need some pretty cold fluid going past at probably 2-3 times the volumetric flow of your water to get even a 3 C drop

If you back calculate your required U value, for a 5C drop its about 5KW.

So for a 0.14m2 pipe that's a U value of 2,500 W/m2 K assuming a 15C differential temperature from your water to the cooling water.

That's high.

See this for some typical U values

http://www.cheresources.com/content/articles/heat-transfer/u-in-heat-exchangers

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Rputvin]

My former employer makes several products that might be what you need. Tube-in-tube heat exchangers using stainless inner tube, along with water-traced jackets for continued thermal stability.

https://www.viscosity.com/sealer-and-adhesive-systems

 

[LittleInch]

I'm just not sure anything is going to work well given that you have about 3 seconds from water entering to the same molecule leaving your 2.2m long section.

Is there any way you could cold tap the pipe online to create several parallel branches to increase your surface area?

Could you get a window to replace this 20mm ID pipe with a much larger one?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ProjProcEng]

Hi all,

I have reviewed my parameters and have decided there is scope to scale down the HX.

The critical requirement to keep the temperature under 30 C

Most of the heat is generated in the recirculation, loop, as the water enters at approx. 8-12 C
and the system really only alarms at night, when the recirculation is in play.
Recirculation has a 3kw pump and a UV unit on it, along with a number of RO filters.

I should not need more than 3 kW cooling, and I will have to do the maths to ascertain if the to achieve that with a wound copper coil on the outside of the pipe.

The standing jacket at the end of the recirc loop is not cooled ether, so a water cooled jacket might work there too.

Thanks to all for input!

 

[LittleInch]

OK, that's better.

Only you can see your system but it strikes me that maybe there are things you can to reduce the problem at source, i.e. modifications to your re-circ loop. 3kW for such a relatively small flow looks quite high - if you describe / draw/ give us some data there may be something you can do to reduce the heating up part of your equation.





Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ProjProcEng]

Hi All,


Have scrapped wrapping the pipe in favour of wrapping a standing tank on the line,

it is awkward in that I can only wrap 3 sides,

NEW INFO

heat gain is approx. 0.6 kw, from the examination of trends, would like to get higher to be safe

I have a tank made of HDPE with about .6 m2 of area which can be wrapped

I am leaning towards using this product

https://www.powerblanket.com/solutions/industrial-cooling/

chiller available in -20c model, and 0-10 model, obviously I'm plumping for the latter.

My big remaining question is what kind of U value I can expect, the Vendor has not enlightened me yet

A U value of 581 W -m2 -k should do it, is that a realistic expectation

the tank is HDPE, the coolant is glycol, I do not know what the blanket is made of.

 

[LittleInch]

How thick is the HDPE? PE is quite a decent insulation material so that sort of U valve might be a little bit ambitious still, but what are you using for the delta T?

Now you have a vessel have you looked at "beer barrel cooling jackets"? They come as a complete package with chiller and cooler unit. do a search for someone local.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Compositepro]

Again jackets or blankets are intended to maintain temperature and are incapable of transferring much heat because of the poor heat transfer. You can get better transfer with a blower and cold air. An HDPE tank is unlikely to have any pressure rating. Can you not draw water out of the tank, chill it and return it? How about just air-conitioning the room? you can rent or buy portable units.

 

[ProjProcEng]

@composite pro

it will be an exercise in maintaining temperature, at the beginning of the recirculation the temperature is in range and rising to 30C over the course of about 2 hours.

I will attempt to model the blower and fins, (I assume you are implying the use of fins with the blower).

I would arrange the fins over the 2200 mm of pipe.

The orientation of the pipework means it will be difficult to deliver much air to the fins. I would need to run the air through a header.

@ Little inch

I could use up to 25 for delta T chiller
I have contacted the supplier to determine thickness, and confirm material type
but based on 2 mm thickness, at 25 delta T, and the thermal conductivity of HDPE being .48 W/mk
That's giving me 7 kw ????


The local beer enthusiasts may offer a saving as against the industrial optiohn

 

[LittleInch]

I would go for about 0.3W/mK and 2mm sounds very thin for any sort of tank.

So using 3mm that's about 2.5 KW with a 25 C difference.

I only mention the beer cooling one because you can buy or even rent the complete package quite cheaply and it is quite flexible in fitting.

Actually renting one for a week will tell you a lot about whether it actually makes any appreciable difference even if you set the jacket temp to 2-5 C.

The bigger the area and slower the internal fluid the better.

Water is an amazing substance and has a very high heat capacity so don't expect too much if you're really limited in external surface area to cool.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ProjProcEng]

My boss has gone off on the jacket, he has started looking at a shell and tube heat exchanger again.
I'm quite disappointed.

I am going to try to rent a unit as an experiment. Hopefully I can release the funds without having committed to the nature of the project

It will need to demonstrate it will work under the summer conditions, delta T will be reduced if we test the jacket now, but if I can demonstrate at least 1kW cooling with this solution it may convince him.


His big objection is that the standing tank will not facilitate sufficient cooling as there is no agitation.

the flow runs through the tank at 0.2 l/s
the tank holds 115 litres,
and as said before has 500x 400mm dims

Is the lack of agitation a legitimate concern?
I would have thought not, that there would be adequate heat transfer through the water itself?

 

[LittleInch]

I don't think you gave us dimensions before, but never mind, this is a bit more like it.

Residence time of nearly 10 minutes is a lot more like it.

Lack of turbulence is not ideal, but certainly isn't a deal breaker. If the outer annulus cools more than the inner section you will end up with currents in the vessel due to thermal gradients and it will eventually mix with the wall fluids.

However a lot depends on orientation and where the inlet and outlet is. What does this vessel look like in terms of horizontal or vertical and where does it come in and out? - can you post a sketch?

~e.g. if the outlet is low down at one end compared to the inlet then it might preferentially take cooler water from near the wall of the vessel. but other orientations it might end up with warmer water.

nothing like a good practical test to shut everyone up.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ProjProcEng]

http://www.astrosrl.it/files_cms/2-...o-quadro-per-liquidi-con-telaio-metallico.jpg

that is the exact tank as far as I can tell.

point to note*

I can only cool 3 sides as there is a PLC board mounted on the rear of the tank.

just makes it more difficult to mount the jacket