Hot oil system design

[lotus7312]

Hi, all,
We plan to build a hot oil boiler to heat all the reactors in our plant. Due to the different temperature requirement of individual reactor, someone in our company brough up the following design:
1. Feed a medium(say kind of oil) in the reactor jacket
2. Install a sort of heating/cooling package for each reactor to control the temperature.The heater is heated by hot oil from boiler. The cooler is cooled by a cooling medium(say, cooling water).
3. Circulate the medium between the heating/cooling package and the reactor jacket by a pump. Therefore the material in the reactor can be heated or cooled.
To me, this design sounds a bit complicated and the heat transfer efficient seems low. Does anyone have better idea on this design?
Any suggestion or input is highly appreciated.

 

[unclesyd]

We do exactly what you are referring to on 9 large vessels(rotary dryers) using Therminol as a heating and cooling medium. We circulate the hot oil for 2 hrs then switch the oil through a cooler to cool the product. It works very well with very few problems.
We start with vaporized Therminol and take the condensate from this system for the hot oil and work from there. The system is balanced using two heat exchangers. 2 tanks and 3 pumps.

 

[lotus7312]

Hi, unclesyd,
Thank you for your reply. Is it possible for you to post a process flow chart on your system? Do you have individual heating/cooling package for each of your rotary dryer? If it is true, do you install an expansion tank in each system?
Thanks

Thanks

 

[unclesyd]

I don't have access to the PFD at the present time but will try to get it or a little better information.
All the users are inside a building and all the heating, cooling, pumping apparatus is outside. The three largest consumers each had a separate cooling system due to different temperature requirements with some crossover with the 6 remaining dryers.

I just found out that the three large dryers are now on single system as the new heating and cooling requirements are the now the same for all three. The new system was just described to me over the phone and I’m going to have to some clarification of what was said, so bear with me.

There are no expansion tanks other than small accumulators for inerts in the system.

We use Therminol VP-1 due our other requirements. For heating with liquid only you would probably use a different material.

http://www.therminol.com

 

[MJCronin]

lotus,

Is the use of steam out of the question due to temperature ?

To what limits must you heat and cool the process ?

It is my understanding that, if the demands are not too great, either a steam/water based system may work or perhaps a glycol based system may be considered.

Hot oil systems have advantages and disadvantages....

MJC

 

[lotus7312]

MJCronin,
It is not due to the temperature, it is due to the pressure. We want to heat the reactor at a high temp(say 250 degC), but then the reactor jacket should be able to stand high pressure if steam is used. A reactor with high pressure rating is really costy. Plus we have some experience with using hot oil boiler. So that we are thinking of using hot oil.

However, the problem we have not is to how to set up the hot oil system between reactor to have a better temperature control, simple piping, and good energy usage.

Unclesyd said they only have a cooler for their system. I think this make the piping easier than ours. I guess in Unclesyd's system, the hot oil is directly introduced into the reactor jacket to heat the material. And then the hot oil is switch to a cooler to cool the material. But
in our design, we have an individual heater and a cooler for each reactor. The hot oil is introduced into the heater to firstly heat up a medium that stored inside the reactor jacket. Then the heated medium will heat the mateiral inside the reactor. Same to the cooler. The medium inside the jacket is cooled by the cooler first and then the cooled medium is used to cool the material inside the reactor. To me, the good point of this design is that it may achieve good temperature control. But there are some bad points:
1. Energy waste due to the two time heat transfer: hot oil to medium, then medium to material
2. Complicated piping
3. Too many heat exchangers involved

Another problem is that we have 6 different reactors, the temperature requirement will be different. We may have one reactor at 250degC, while another one only ask for 50degC. But the hot oil outlet only has one temperature setpoin(say 300 degC). We are think of step down system: add one or more coolers in the outlet of the hotoil to reduce the temperature, which makes the system more complicated. I am sure this complicated system will cause operation confusion.

Is there anyway that I can simplify this system to meet our different temperature requirement?
Thanks

 

[jmw]

In the pharmaceutical industry they need to heat with steam to get the reaction running and then they quench with ethylene glycol/water at -40degC.
The main problems here are to detect the interface between steam/condensate and the ethylene glycol and obtain proper valve synchronisation. Condensate in the ethylene glycol can cause the lines to freeze because it changes the concentration and hence the freezing point.
The only other problem is valve maintenance which, due to the thermal shock, is a routine occurence.
SInce they use a lot of small reactors the problem is a maintenance headache.

 

[ixchawla]

Lotus,

You may review the following features of design and see if it is applicable for your case:
1. Split the hot oil boiler feed to partly pass through the roiler and balance to bypass (cold).
2. Consider a series cum parallel arrangement of hot oil users such that the desired temperature levels at users can be met. The cold bypass from hot oil boiler can be used to fine tune the temperature requirements.

You have used this scheme where temperature level for one user was restricted by degradation, another by controllability while other users temperature requirement was higher.

Regards

 

[abcmex]

You may get additional design details from the DIN 4754 and some safety information regarding this kind of installation.

http://www.standards.com.au/catalogue/script/Provider.asp?Db=DIN

There are two kind suppliers that can give you advise in this area
(a) Thermal oil suppliers

http://basic.bayerchemicals.de/index.php?page_id=132

http://www.marlotherm.de/

http://www.dow.com/heattrans/

(b) System suppliers
HTT

http://www.htt.de/

perhaps useful

http://www.process-heating.com/CDA/...ures/BNP__Features__Item/0,3156,20422,00.html

RGS

 

[25362]

The options mentioned seem to confront simplicity of operation (one heater, one cooler, and mixing of hot oil streams) vs the investment cost, energy savings and complexity of having individual heating and cooling packages for each reactor. The optimum option appears to be site dependent, that only lotus7312 will be able to decide upon.