4 replies to this topic

[Lolly84]

Hello ChemEng Gurus,

I would greatly appreciate some points in the right direction, as I think I may be incredibly confused around this area of heat transfer.

I've got a batch reactor that needs to be heated from Tb1 to Tb2. It is heated via a jacket and HTF. The HTF is heated by steam in a heat exchanger.

I am trying to establish what the area of the heat exchanger needs to be in order for the batch duty to be achieved.

My thoughts, and where I've got to.. (I'm think I'm muddled because I'm getting confused between batch and continuous).

1) Steam flowrate determines the temperature of the HTF.
2) As the temperature differences on the batch are continually changing, it is not possible to use the log mean temperature difference to model the reactor as a heat exchanger.
3) As the batch temperature increases, the temperature gradient will decrease until a constant temperature is reached.
4) As the batch temp increases, there will be a temperature fall on the HTF side. However, over the entire process, the inlet temperature of HTF to the jacket SHOULD NOT(???) remain constant. It should in fact fall to a temperature close to the target batch temperature.#
5) Should point 4 be taken into consideration for heat duties, as it is perhaps just a result of tuning parameters?
6) Is the way forward to look at heat balances over the batch in (say 1 min) intervals, to determine what the inlet (and hence outlet) temp of the HTF to the jacket should be. This jacket temp should be the same as the outlet HTF temp from the steam heat exchanger, so we can work out the heat duty there as well. or am I going about this backwards?

However, Kern's models all seem to assume a constant HTF inlet temp.... so I'm wondering if my thoughts are wrong..

Basically, I think I'm quite confused! And I'm having trouble actually visualising what's going on.

I've been told that just looking at the overall temperature differences is wrong. I've also been told that I can't just look at the point of maximum heat duty, because the heat exchanger area will not be dependent soley on this. This is the point that I have trouble understanding most. Surely, the most heat duty occurs at maximum temp difference across the batch. This would obviously be the case in continuous processes - why should it be different in batch?

any help to clear my confuddled mind would be very very greatly appreciated.

Thanks in advance,
Lolly

[Art Montemayor]

Lolly:

You have failed to mention the magic word.

Heat transfer is a Unit Operation that is basically divided up into two sections:

1. Steady State Heat Transfer; and,
2. Unsteady State Heat Transfer.

You are stuck with the hardest version – the latter. If you are truly reading, studying, and working the examples in Don Kern’s Opus, Process Heat Transfer, then you are on the right track. To get a quick summary review of what Don Kern tells you in his 850-page volume, go to the following websites that I got in Google using “Heat Transfer, Unsteady State”:

http://www.unb.ca/ch...ab/unsteady.pdf

http://www.unb.ca/ch...ed/unsteady.pdf

http://www.nzifst.or...httrtheory4.htm

http://www.eng.fsu.e.../whatisuss.html

http://www.eng.fsu.e...usseducate.html

http://www.engin.umi...potato/heat.htm

I hope that you have realized that the bottleneck and the prime factor in your application is the reactor’s jacket. No jacketed vessel is optimized for heat transfer. It can’t be. All a jacket does is use as much of the available surface for heat transfer – it does not guarantee that the available surface is sufficient to do the desired job in the desired time. No decent engineer in his/her right mind is going to expect to rely on a jacket to give process heating requirements. It is a “token” heat transfer - at best - and is usually employed with the hopes of pre-heating the contents and keeping them hot during idle time. If you have a need to continuously (under steady state) heat up a vessel’s contents, then you are forced to use conventional heat transfer apparatus such as re-circulating external heat exchangers, internal coils, or internal heating panels – complete with an efficient agitator to ensure good convection heat transfer coefficients.

I recommend you read Kern’s section on Unsteady Heat Transfer and work out his derivations for the equations used, as well as all the examples he gives you. By doing this and studying well the algorithm, you will quickly grasp the basis of unsteady state and be able to fully “visualize” just what is going on. I think you will find unsteady state heat transfer much more challenging than steady state and very interesting.

Good Luck.

[JBradley]

All good stuff. But please, jacketed vessels can and are employed for many process heating applications (for example where gentle indirect heating is required).

regards

Grand master of the Worshipful Guild of Jacketed Vessel Manufacturers.

[Art Montemayor]

Mr. Bradley:

No one has claimed or stated that jacketed vessels are not employed for many process heating applications. A lo contrario. I believe it was stated that jackets are used for pre-heating the contents and keeping them hot during idle time.

Additionally, the stuff is not only good, but I would add the qualifier of “true”. A batch reactor vessel is never designed and fabricated to suit the heating duty of its jacket. This is basically reverse engineering and to apply that criteria would be to negate the basic scope of work of any economically justified batch reactor: to process out a given quantity of product in a given amount of time available. To follow this prescribed scope of work one never even contemplates the related heat up or endothermic requirements in determining the physical size of the vessel. The related heat transfer area requirements are always resolved after the vessel size is determined – using internal or external heat transfer equipment as mentioned before. The application of the jacket, then is done – more often than not – simply because it is there and available. The jacket’s heat transfer capability does not control the capacity of the vessel. If it does, something is wrong in the economic justification for using that vessel and procedure for the proposed reaction.

The above has been proven and applied for many years in the business not only by myself but by those that passed on the technology and taught it to me and others. It works that way not because text books say it should, but rather because it makes for common engineering sense and – more importantly – it is economic and makes for a profitable batch reactor operation. Time is of the essence in almost all batch operations. The shorter the batch time, the greater the number of batches per operating day. And that is the name of the game when one is trying to make a profit by operating a batch reactor.

I believe that it is very important for Chemical Engineering students to learn very early on about the strengths and weakness in the Unit Operations that they study. This is the sort of details that are often left out or merely mentioned in passing over equipment characteristics in University courses and clearly explains Lolly’s dilemma in not being able to visualize what are the main factors to consider in a batch reactor operation and what is the logical sequence of design steps to undertake in designing such a system. That a jacket is to be employed on the batch reactor is a given known, and accepted necessity. However, that the same jacket is of heat transfer priority in the operational requirements of the batch reactor is, in my opinion, a false and naïve expectation. Batch reactors are dependent on – and thermally controlled by – either internal coils/panels or external heat exchangers and recirculating pumps. This is done not only for the purposes of accurate and dependable heat transfer control(s), but more importantly because of safety control considerations. This is important stuff and often is never taught in university at the right and indicated moment for the student to not only understand, but to also appreciate and apply later in his studies and projects. I believer it is practically unrealistic to rely on a jacket heating system to accurately control the temperature within a batch reactor.

Respectfully,

The Ancient Aztec Gate Keeper of Safe and Controllable Batch Reactors.

[Lolly84]

Hi Art Montemayor,
Thank you so much for your help, and those webpages are incredibly useful.
I had suspected that it was unsteady state, but I seem to have a bit of a fear of putting a name to problems in case I'm wrong. Need to get over that really, because I guess that's one of the fundamentals in being able to apply theory to practice.

I am aware of the limitations of jacketed vessels for unit ops, but I would hope that some of these issues would be lessened with agitation.

I will continue with Kern then for the time being, and hopefully get it clear in my head soon. Your webpages have really helped me on the way though. Thank you very much for your time. I may have a question on humidification in the near future if I can't get to the bottom of it myself!

Thanks again,
Lolly