7 replies to this topic

[Amit1]

Hello

I am working on a revamp project. I have rated datasheets for the existing exchanagers. Without spending too much time on rigorous rating like HTRI or rigorous simulation in Hysys or PRO II, I want to check adequacy of the heat exchangers for the revamped flow rate.

Is there any reliable pro-ration correlation between service U (overall dirty U on rated datasheet) vs shell & tube side flow rate, that I can use for a quick adequacy check? My intention is to short-list candidates for further detailed check (by rigorous simulation in PRO II). So I need to be quick as well as little bit conservative in my approach.

Thanks in advance.
-Amit

Edited by Amit1, 17 October 2011 - 11:13 AM.

[AlertO]

hi Amit

this depends on the information that you have. if you have the data of hi, ho, fouling and film coefficient, you can do the approximate calculation by updating only hi and ho as the following relation:
flow rate >> velocity >> Re>> Nu >> hi (the calculation formula is illustrated in general heat transfer book)
then, you can estimate the new U as you want.
in case of no infomation enough, you have to do the new calculation that performing HTRI simulation may spend the time shorter.

Hope this can help you

[ramlalithravi]

Heat transfer co-efficient for revamped condition is calculated as same as usual practice, since the revised condition will have change off properties like flow, pressure & temperature, which will impact on Convective heat transfer co-efficient of the fluid, is dependent properties for the estimation of Nusselt number and Prandtl number used in Convective heat tranfer co-efficient correlations.
Thermal resistance of the material will remain constant in the ravamped condition, will not be reflected to the Modified Overall heat transfer co-efficient. HYSYS or other simulation softwares will estimate the Co-efficient based on the process condition and duty of the Heat Exchanger.

[Amit1]

Thanks for your input

[srfish]

I don't know how you can closely determine the new overall heat transfer without knowing the h_i and h_o of the existing exchanger. For sensible heat services you could use the flow factor to the 0.8 power on h_i and the 0.6 power on h_o.. The factor would be the ratio of the newflow/oldflow. This calculation would be multiplied by the existing film heat transfer coefficients.

[Amit1]

srfish

As rightly pointed out by you, I tried to prorate using the factor 0.8 and o.6 to flow ratio. It indeed works reasonably well for sensible heat transfer. But for condensing/vaporizing duty, it doesn't work. I used below correlation.

U(new) = (1/3 + 1/3 * (shell side flow ratio)^0.8 + 1/3 * (tube side flow ratio)^0.6)* U(old)

I really don't know the source of this equation. That's what they did in a previous projects.

Thanks

Amit

[GS81Process]

You could likely quick screen the exchangers for flow velocity, pressure drop, rho v squared and a few key parameters. You could also check the new required duty versus the % excess area listed on the datasheet.

I would still recommend rating the exchangers in a rigorous program such as HTRI. There are other things to check which cannot be easily checked by simple screening such as vibration, baffle spacings, etc. A program like HTRI will also allow you to check thermal resistances and estimated % excess area available.

Edited by GS81Process, 21 October 2011 - 11:17 AM.

[kkala]

Trying to find suitable exchangers for desalter brine cooling, we searched more than 50 exchangers available in the refinery. Prescrening rejected all of them without proceeding to check by software; just by looking at the process data sheets of the exchanger, after having defined process conditions and approximate area required for the new exchanger. Type of existing exchanger, too low design pressures, too high or too small size, insufficient flow velocity in tubes, were reasons for rejection.
Obviously you rerate exchangers for same or similar service, so above prescreening has not rejected a lot of them.

Edited by kkala, 21 October 2011 - 05:10 AM.