[TRNSYS-users] Tank loss coefficient

[Jeff Thornton]

> I have a question concerning the tank loss coeff. of Type4.
>
> Is the tank loss coefficient in TRNSYS also the u-value of the 
> insulation??

Using just the U-value of the insulation will under-predict the u-value 
of the tank.  You're not accounting for the convective resistance of the 
water-side heat transfer, the resistance of the tank wall, and the 
combined radiative and convective air-side resistance.  But all of that 
assumes the tank is ideally insulated and that's really never the case.  
There are quite a few "thermal shorts" that will reduce the effective 
u-value of a tank from that calculated by simply using the insulation.  
There are algorithms to calculate the u-value of a tank if you know the 
energy factor or have access to heat loss data.

> I switched the unit of the tank loss coefficient to W/m^2K. It´s the
> same unit as the u-value. I knew knew the heat conductivity of my
> insulation (lambda = 0.04W(mK)) and its thickness (100mm), i was able
> to calculate the u-value by using u = 1/(0.1m/(0.04(W/mK))) = 0.4.

That formula only applies to 1-d conduction and does not hold for 
cylindrical tanks - you'll have to use a radial calculation for 
cylindrical tanks.

Jeff


---
Jeff Thornton
President - TESS LLC
22 N. Carroll Street, Madison WI USA 53703
Office: (608) 274-2577  Fax: (608) 278-1475
www.tess-inc.com
E-Mail: thornton at tess-inc.com