Water-cooling calculations (check my math? Help?)
Sci
vacuum station. I've stumbled through the math so far, but am by no
means confident in it.
The pump's manual says the input water temperature should be no lower
than 10c, and output temperature no higher than 35c. And the minimum
water flow should be no less than 75l/h^-1, and at a maximum water
pressure of 5bar. (pump body operating temp is 200-250c. 400c at base.)
Looks like choice of pump is pretty wide; seeing a number of computer
cooling pumps that well exceed 75l/h. And using the advice listed here (
http://www.physicsforums.com/showthread.php?t=465007 ) I came up with a
figure of about 2.2Kw/7512BTU of heat needing to be radiated (not
counting the baffle and valve, for which there are no numbers, but
"colder than the pump" and "only indirectly heated" may help in
ballparking).
So do my numbers make sense so far, and any guesses at a fan-assisted
radiator of sufficient size?
PC-style water-cooling radiators seem out, car heater-box matrices are a
possible, air-conditioning radiators seem likely.
* Details prattling follows;
I'm imagining water passing through valve and baffle before reaching the
pump currently, as a secondary radiator is rather more complex.
I'm hoping we can get some inline temperature sensors to keep a check on
it, and have grabbed a flow meter that should watch for pump failures. A
variable-speed pump and fan should help keep things optimum, but it
needs to be planned for absolute limit ratings as the diffusion pump is
slow to start and stop. (again, planning to insert a stainless
temperature probe in the pump-fluid drain-plug to monitor temp better too)
~ Sci
Will Pearson
same as large normal radiator running on max, so if in a small
enclosed space for long periods of time you are likely to get
decreased cooling due to the temperature differential not being so
good.
I'll show my working and see if I get the same ballpark as you.
Flow of water in kg per second = 75/360 = 0.208 (kg/s)
Temperature in = 250
Temp out = 25 (room?)
Heat capacity 4.18 kJ / (kg. K)
Power = 4.18 * (250-25) *0.208 =195 kW.
Err. Something went wrong. Are you cooling down from 250? or 35. Even
from 35 I get 8 KW.... So mistakes are somewhere.
How many kW is the thing we are trying to cool actually rated for? I
suspect we are calculating the maximum amount of heat we can transfer
and what we are cooling isn't going to produce that much, so we should
a spec radiator based on that...
Will
On Feb 12, 3:50 pm, Sci <s...@sci-fi-fox.com> wrote:
> Been trying to work out the cooling system capacity required for the
> vacuum station. I've stumbled through the math so far, but am by no
> means confident in it.
>
> The pump's manual says the input water temperature should be no lower
> than 10c, and output temperature no higher than 35c. And the minimum
> water flow should be no less than 75l/h^-1, and at a maximum water
> pressure of 5bar. (pump body operating temp is 200-250c. 400c at base.)
>
> Looks like choice of pump is pretty wide; seeing a number of computer
Adrian Godwin
The best you might do is run a fairly high temperature gradient over a
bit of metal, keeping the water below 100 (sure, you can pressurise it
but that creates all sorts of other interesting problems).
-adrian
Will Pearson
hour Should be
Revised calc
Flow of water in kg per second = 75/3600 = 0.0208 (kg/s)
Temp out = 25 (room?)
Heat capacity 4.18 kJ / (kg. K)
Err. Something went wrong. Are you cooling down from 250? or 35. Even
Will
Nigel Worsley
The pump's datasheet says the heater power is 850W, if you have more heat than that to dissipate then
I suggest a quick trip to the patent office for an over unity free energy device.
Nigle
Adrian Godwin
> Been trying to work out the cooling system capacity required for the
> vacuum station. I've stumbled through the math so far, but am by no
> means confident in it.
>
> The pump's manual says the input water temperature should be no lower
> than 10c, and output temperature no higher than 35c. And the minimum
> water flow should be no less than 75l/h^-1, and at a maximum water
> pressure of 5bar. (pump body operating temp is 200-250c. 400c at base.)
>
> Looks like choice of pump is pretty wide; seeing a number of computer
> cooling pumps that well exceed 75l/h. And using the advice listed here (
> http://www.physicsforums.com/showthread.php?t=465007 ) I came up with a
> figure of about 2.2Kw/7512BTU of heat needing to be radiated (not
> counting the baffle and valve, for which there are no numbers, but
> "colder than the pump" and "only indirectly heated" may help in
> ballparking).
>
I think the problem here is that you've calculated the heat input (and
therefore the dissipation requirement) as the heat needed to take
75l/hr from 10C to 35C. This isn't actually the requirement : as Nigel
says, that can't be more than 850W.
The min/max figures are LIMITS. I don't know how they come up with
75l/hr but the 10 and 35 figures are not expected to both be reached
at once - you're supposed to feed water in at a temperature and rate
that means the output is below 35, and if the rate is so low that the
input would have to be below 10 to dissipate the power, you should use
a higher rate.
-adrian
Sci
So what I actually calculated was the heat input over an hour then, not
what needs dissipating?
I suppose I assumed there'd be some sort of rate calculation involved,
but yes it does make sense it'd only ever need to radiate as much as the
heater puts in.
So, should be looking for maybe a 900-1Kw radiator, to include keeping
the upper pump manifolds lower than 10c. Depending on use, the chamber
may generate some heat as well.
Looks like we're back into PC/heater-matrix territory for radiator
candidates.
Thanks! :)
~ Sci
Sci
> Where are you planning to run the radiator? That amount of heat is the
> same as large normal radiator running on max, so if in a small
> enclosed space for long periods of time you are likely to get
> decreased cooling due to the temperature differential not being so
> good.
Good point, we may have to duct the output like the laser-cutter.
- Sci