Central bore cooling 1

[corus]

Water cooling through a central bore hole can be improved by inserting a rod through the bore to reduce the area and increase the water velocity. This rod would effectively provide an annular region through which the water flows to provide cooling to the outer radius of the bore hole.

What parameters would optimise the diameter of the rod in order to improve the cooling rate? My guess is that for the same mass flow rate then the smaller the annular region then the faster the water flow and higher heat transfer coeficcient.

corus

 

[BigInch]

To minimize pumping costs, you want to minimize velocity, therefore you should find out what the maximum temperature that you can except for exiting water is and slow the water down to a flowrate that will allow the water to heat to your max acceptable exit temp.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[Yorkman]

I'm wondering if you have the theory of this right, If the mass flow is going to be the same, with an increase in velocity because of the reduced cross sectional area, I think the leaving coolant temperature may rise, but I'm not sure the temperature of material being bored will be reduced. Your still moving the same amount of heat based on mass flow.
I agree with Big Inch that there will be a trade off at the pump due to the increase in H.P. required, brought on by the increase in pressure drop.
Please correct me if my thinking is wrong, and steer me in the right direction.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[IRstuff]

How long is the bore? If it's rather long, then some sort of swizzle stick with arms might be enough to stir the water, which is I think you're wanting?

The only reason an annular channel with the same mass flow increases thermal transport is if you think the flow is so laminar that the center part of the flow is not contributing to thermal transport.

How big is the bore? Too narrow a bore and the channel resistance will be dictated by the bore surface roughness and you'll need to run at a much higher pressure to get the same mass flow.


TTFN

 

[BigInch]

Yorkman,

There's still a chance it could work, if Corus is in the laminar flow regime now and (s)he increases velocity enough that he gets into turbulent flow. With turbulent flow, there's a greater heat transfer coefficient available in the film layer next to the pipe wall. Under those conditions Corus can increase overall heat transfer, but that's about the only way I see that it could give any net gain.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[Yorkman]

Thanks, Big Inch that clears up the muddy waters a bit.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[BigInch]

Is it too late to change "except" to "accept"?

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[insult2injury]

if you're just looking to make the flow turbulant, there are and more efficient easier ways of doing it without large increases in pumping requirements. A vortex generator could be used, the surface roughness could be altered or more commonly and easily accomplished, a twisted tape through the bore to produce the swirling effect. Regardless of what method is used, only minor thermal improvements will be shown so don't expect fireworks.

If you actually are looking to quantify the increase in heat transfer that the rod method will produce, there are existing nussult correlations for annular flow that are widely accepted. Also with the rod, if it is mounted eccentrically, I believe there is also another minor augment to the heat transfer coefficient.

 

[corus]

Thanks to all.
FYI the original bore size is 50mm dia. with a water flow rate of 60ltr/min.
As Yorkman implies, I was expecting that any increase in water coolant temperature is offset by the reduced temperaure of the bore surface, caused by the increase in heat transfer. Hence the advantage of the system is in reducing the bore surface temperature, with generally no significat increase in exit water temperatures. Thermally I can see no optimum annular size, other than to make it as small as possible. The increase in pressure is something I'd not considered and is worth more thought.
Ta

corus

 

[corus]

Putting aside pressure considerations, another aspect of the size of annular region is the effect of corrosion at the bore and rod surfaces. If the annular region were too small then the channel may block with corrosion after a while. Does the possibility of corrosion, preventing the water flow, limit the size of annular region? Are there any recommendations as to the minimum difference in the dimaeters of the annulus in order that corrosion will not ultimately cause a blockage?

corus

 

[monaco8774]

For turbulent heat transfer coefficients the heat transfer coefficient is proportional to Reynolds No raised to the power of 0.8

You need to use equivalent diameter to take into account the effect of the bore, this is calculated by 4 x flow area / wetted perimeter

Reynolds is calculated by diameter x density x velocity / viscosity

For pressure drop the losses are proportional to 1 / equivalent diameter to the fifth power

hope that helps

 

[moltenmetal]

At 60 L/min water flowing in an open 50 mm diameter round channel, you're already well into the turbulent flow regime. Assuming that the liquid film coefficient remains the controlling heat transfer resistance, your optimal rod diameter will be the one which gives the maximum pressure drop you can provide. At a certain point, though, the thermal conductivity of the thing with the 50 mm hole in it will come into play and you'll get diminishing returns.

If you want to do the math, here are the equations:

Nu = h D(hydraulic)/k
where h is the liquid film coefficient, a heat transfer CONDUCTANCE (i.e. bigger h means less heat transfer resistance). k is the thermal conductivity of water.

Nu = 0.023 Re^0.8 Pr ^0.3

D(hydraulic) = cross sectional area / wetted perimeter for your annulus, ie. (D^2-d^2)/(D+d) where D is 50 mm and d is the diameter of the rod

Re = D(hydraulic) V mu/k, where mu and k are fluid properties (i.e. constant for your situation) and V is the velocity of the flow

Pr is a function of fluid properties only (i.e. it's constant for your situation).

From this you could derive the dependence of h on D (50mm) and d (the diameter of the inner rod). All that matters for you, though, is that there is no optimum annulus- the smaller the annulus, the greater the heat transfer at a given mass flow- and the (vastly) greater the pressure drop too.

Practical considerations, such as how much pressure you can reasonably supply to force 60 L/min through this hole, and how much this pressure will increase with scaling, and how much erosion/corrosion you get at high velocities, will determine what the REAL optimum annular space would be.

As to using turbulators, static mixers or enhanced roughness rather than a rod in the ID, I doubt they'd make that much more effective use of available pressure drop for beneficial radial mixing in your situation than would be afforded by a simple rod, since your flow conditions are fully developed turbulent even without the rod in place. They'd make a world of difference if your flow was laminar, though!

 

[Yorkman]

A few other considerations might be, if the rod is too large would the debris from your boring process become lodged in the reduced area? What size is the material coming of the boring tool? Also will this debris act as an abrasive enlarging the bore diameter as it passes through the reduced space? How critcal is the bore diameter, and are the acceptable tolerances going to be affected by any errossion caused by this action?

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[IRstuff]

If you're allowed to put a rod down the middle, then you might consider putting fins on the inside surface, like an inverted heatsink. The objective should be to increase the effective contact area between the water and the pipe. Two fins in a cross across the entire inner diameter of the pipe will more than double the contact area of the inner surface. If you make them something like 0.08" thick, you would get about 12:1 aspect and decent fin efficiency.

TTFN