Hot Oil Efficiency

[Gottburg]

In our plant we have a hot oil heater running with natural gas.
Equipment capacity= 13.7 MMBTU/hr (absorbed heat)

Currently we are running about 14 MMBTU/hr of released heat (burning this amount) and we are getting efficiency about 65% measured as Heat absorbed by the oil/heat released by combustion. And Combustion efficiency (using stack temperature, and NG fed) about 83 to 84%

Do these values make sense for y’all? Is the efficiency to adsorbed heat too low?

Can somebody share his/her experience?

Fernando

 

[25362]

Apparently, too much (combustion or tramp) air.

 

[Gottburg]

Thanks 25362.

Air was tunned to 10% 4 week ago.
So for us does not seems to much.

 

[macmet]

Could your heater tubes need cleaning?

 

[Zoobie]

I don't think the discrepancy could be explained by fouled tubes as your stack temperature should rise if heat transfer to the tubes was impaired which would manifest itself in a lower efficiency (by the stack temperature, excess air method). At the same time, that doesn't mean that your tubes are OK.

Is your stack temperature accurate? It would have to be way out though (about 600F + using Fig.8-17 GPSA Manual, 10th Ed.) Looking at the figure, I would have a hard time explaining the discrepancy to a bad excess air measurement.

What about leakage? I am not at all an expert on this kind of equipment but could you be getting some air leakage that is throwing your measurements off.

Where are you calculating your process duty (heat absorbed by oil). Is this measured across the heater itself or is it indirectly measured (like using the downstream process load)?

Random thoughts.....back to work.

 

[25362]

Assuming 3% heat is lost through the walls you're still left with 15% unaccounted for.

If you aren't superheating steam in the convection section, it is clear you are transferring heat to other sinks beside the hot oil. Either tramp air or xs combustion air. Could it be that your estimation of xs air is in error ?

According to a quick estimation your stack gases should be around 330^oC for a heater efficiency of 83-84% and 10% xs air. Could it be that the stack gas temperature reading is wrong ? Please explain how did you calculate the heater efficiency of 83-84%.

Kindly refer to the comments above.

 

[Gottburg]

Thanks to all for your feedback, here there are some answers:

Combustion Efficiency: I was using fabricator manual: He suggests using the %air excess and the stack temperature go in a table (a table for NG and with 80F/60HR Air) and get “%H”, this value should be then subtract from 100%. Then other minor sustraction can be made: loss thru shell etc.
In our case: 10% and 340C. The “%H” from mentioned table is: 15.02%, so efficiency is 100-15.02-1(loss) = 83.4.

Absorbed heat is measured (calculated) using hot oil flow and IN and OUT temperatures. All these variables are stored in DCS.

Burner calibration was done for a “respectable” company. Should I suspect a miss-tuning?

 

[Zoobie]

What sort of oil are you using and how old is it? Are the density and heat capacity known fairly accurately?

 

[25362]

Fernando, preste atenci[ó]n al exceso de aire. If the stack temperature is OK, I still think that xs air is actually near 30%. How do you measure it ?

Long ago I've had my fair share of experience with portable Orsat analysers. I hope that you're aware that the typical analyser is inherently far from reliable, the sampling technique, the freshness of the reagents, the pH of the water, etc., are factors that influence the readings.
A 3% absolute variation in repeatability and reproducibility on the % O₂ is quite common.

For a natural gas fuel, such an absolute error in the oxygen measured on the stack gases could easily explain a 20% absolute error in xs air estimation.

Besides, air leaks (such as from open peep holes) must be found and sealed.

 

[Gottburg]

Gracias 25362....
The xs air 10 % was reported to me by the company that tuned the burner...I'll ask more detail.

Zoobie: I’m using Terminol 72, 3 year old. For Cp, SG I’m using manufacture correlations at T average thru the heater.

Thanks y’all so far

 

[Zoobie]

Are you reading the correlations off of a chart? How does the Cp and SG vary from inlet to outlet conditions? Is it linear in the operating range?

This may have nothing to do with the discrepancy that you are observing but if your Cp is 5% out then your duty will be as well. If SG is out as well then your error is compounded.

 

[Gottburg]

Zoobie,
For Cp and SG I'm using the equation presented on the product data sheet.

I agree with "compounded error".
I'll go thru a instrument calibration campaing to check on these.

 

[Zoobie]

What does the equation give you at inlet, outlet, and T average? Is this a linear equation?

 

[Gottburg]

Cp and SG are linear (type sg=a*T(C)+b also Cp is linear same general equation.
I realized that delta T thru our equipment is just 13C, so each 1C in T difference lead to up to 8% error in absorbed heat.
We are going to a instrument checking effort now.

Thank to everybody for the comments and help.