LMTD question
LMTD question
(OP)
I have a TEMA type BXU
shell side (glycerol)
Tin: 338 °F
Tout: 122 °F
Dew point: 241.1 °F
tube side (cooling water)
Tin: 95°F
Tout: 115°F
Correction factor : 0.99
T1 = 338-115 = 223
T2 = 122-95 = 27
this means LMTD corrected = 0.99 * [223 - 27]/LN [223/27] = 91.9 °F
According to the heat exchanger spec sheet they arrive at 97.73 °F .
Can someone help me out on this one?
Thanks
shell side (glycerol)
Tin: 338 °F
Tout: 122 °F
Dew point: 241.1 °F
tube side (cooling water)
Tin: 95°F
Tout: 115°F
Correction factor : 0.99
T1 = 338-115 = 223
T2 = 122-95 = 27
this means LMTD corrected = 0.99 * [223 - 27]/LN [223/27] = 91.9 °F
According to the heat exchanger spec sheet they arrive at 97.73 °F .
Can someone help me out on this one?
Thanks





RE: LMTD question
It is, apparently, a weighted MTD. Follow these steps:
1. Estimate the heat load Q1 from 338 to 241.1oF, and the heat load Q2 from 241.1 to 122oF.
2. Now estimate the water temperatures for these two heat loads to get the intermediate water temperature (against 241.1oF).
3. Calculate the LMTD's for both sections separately.
4. The weighted LMTD would be:
RE: LMTD question
There's nothing wrong in your calculations, except the fact that for your application you probably don't have to use LMTD.
I'd suggest to follow 25362's advise and use weighted MTD.
Take a look to the link below to get a more precise idea about special cases which ask for the weighted MTD approach.
ht
RE: LMTD question
The heat exchanger specification sheet has a higher MTD than the straight line MTD. Normally in a case like this where the dew point is significantly lower than the inlet temperature, it is the other way around. Is there a bubble point? If this is pure Glycerol, there should be.
RE: LMTD question
RE: LMTD question
RE: LMTD question
I agree with you. In my previous post dated 21st December, I suggested the OP to follow 25362's approach. In a broader view the classical method to calculate the LMTD (i.e. that which leads to LMTD = 91.8 °F) when dealing with complex HX doesn't work.