DOUBLE WALL HEAT TRANSFER (FOR TANK HEATING) 4

[RintoS]

Is there any application in industry where a "double wall" heating system is being used? I am referring to tank heating by means of double wall surrounding the tank and hot water circulating inside the walls.
The purpose where are looking for is getting a flush inside tank surface in such a way that cleaning will be easier (I mean easier than if heating coils inside were used)
- What is the most adequate speed for the heating medium inside the wall?
- What are the possible patterns for water circulating (we are thinking on a zigzag vertical routing).
- Medium to be heated would be a sort of fuel subjected only to the natural convection due to the heat transferred from the walls. These heat transfer coefficients are already known.

Thanks in advance for your attention

 

[bchoate]

bchoate
I have just a few comments regarding the use of 'jacketed' tanks. Although reasonable heat transfer coefficients are possible, limited heat transfer area is usually the case. Spiral baffles are usually used in the jacket annulus to improve heat transfer. It has been our experience that such jackets foul easily with water in the jacket. Jacketed tanks are more expensive. An alternative to jacketed tanks is to use exteriorly applied plate coils such as 'Panelcoil'. This are designed for turbulent flow on the water side. A good thermally conductive cement is necessary to insure maximum heat transfer (Thermon). For plate coils contact someone with 'Mueller'. If more information is needed about jacketed tanks, DeDietrich is a good source.
Bill Choate

 

[Montemayor]

Rinto:

We spent a lot of time developing thread391-71471, "Half-pipe coil overall coefficient". At first, you may think This is not the subject you are talking about; but it is. You'll note that all a "double-wall" vessel is, is nothing more than a jacketed vessel - be it a reactor, a process tank, etc.

I still am very partial to a simple, cylindrical, carbon steel jacket with a spiral baffle(s) between the jacket and the vessel. If you use hot water instead of steam, it will be less efficient. The velocity criteria to be applied is that which yields complete turbulent flow in the jacket. That's precisely the problem: it's very difficult to achieve turbulence because of the velocity needed. The baffles are certainly needed to obtain a decent velocity, but the spacing and quantity are the problem. Your specific application will reveal this.

You don't state size, fluid properties, configuration, controls, and other specific basic data. Therefore, I can only give a qualitative response. The best I can guess is that you'll have an overall "U" of 100 or less - depending on turbulence inside the vessel and in the jacket.


Art Montemayor
Spring, TX

 

[negrita]

This message is for Mr. Choate, or anyone who can help:

I would like to ask the following questions:

1. Where do I get DeDietrich a good source for jacketed tanks? Is this a book or technical reference for designing cooling jackets?

2. Do you know of a source for Heat Transfer Analysis for vessels mixed with pulse jet mixers and cooled with cooling water jackets? All the correlations available in the literature are for vessels agitatated with mechanical impellers and cooled with cooling coils. There are none for vessels agitated with pulse jet mixers. The fabricator of the pulse jet mixers does not have any heat transfer correlations. Pulse jet mixers do not have mechanical parts and are free of maintenance. This is ideal for mixing radioactive slurries, some of which are non-newtonian. Because they do not require maintenance, they are also ideal for vessels processing nuclear waste where no one can enter the plant for radiation safety reasons.

3. Do you know of a reference containing data of fouling factors for radioactive slurries. The thickness of the stagnant layer(scum), also called sidewall fouling factor that forms on the walls of the vessel seem to be a great obstacle to remove heat from the vessel. I wonder if perhaps you know a good reference for these parameters. While the heat transfer coefficients for the vessel wall thickness, jacket fouling factor, jacket film coefficient are easily obtained, I can not find heat transfer correlations for the vessel inside film coefficient, and the inside fouling factor. I have been unable to find any heat transfer correlations for vessels agitated with pulse jet mixers. As I mentioned before, the fabricator of the pulse jet mixers does not have any heat transfer correlations.

4. Do you know about a good technical reference for sizing air spargers whit no holes. The sparger consists o a tube with a nozzle at the end. The nozzle can be replaced by a pinching the tube at the end.

I will be anxiously looking for your responses.

 

[Montemayor]

Negrita:

I hope Bill Choate adds to this query with his helpful advice. I can say that DeDietrich is a reactor vessel fabricator - much as Pfaudler is - specializing in glass-lined products. They also fabricate other items. You can find them at their website, identified in a search engine such as Google.

Your questions are very important questions if you are an engineer involved in liquid radioactive waste slurry material or other similar non-newtonian fluids. You have not specifically stated that; but I suspect this from your questions and from other posts you have written in other forums. I strongly recommend, if your questions are (as I suspect) serious & professional in nature that you write an original question in a new thread within the Mechanical Engineering Heat Transfer Forum. I consider this subject to be so important and unique that it can best be handled on it's own merits as an important and practical heat transfer problem. Please be very specific and detailed in your basic data and explanation. The quality of your submittal will determine the quality of the professional responses that you will obtain.

I have many suggestions and questions myself on this subject, but I prefer to not clutter this thread with them. I hope you respond in a new and independent thread as I suggest.

Art Montemayor
Spring, TX

 

[bchoate]

bchoate
The original post is a very good example of 'selected scope revelation'. I'm not trying to be unkind but the initial post was about 'jacketed tanks' and heat transfer. A 'fuel' was to be heated by natural convection at the warm wall (no mixing involved). Now we learn that not only is the tank to be mixed (particle suspension) but also the mixing modality is a not so common 'Pulse Jet Mixer'. In asking for fouling resistances, transuranic (radioactive) slurries are mentioned. Although I may have a few comments on the problem as expanded, it would seem that scope clarification is sorely needed.

Questions that come to mind include the following:
1) are you truly working with transuranic salts and sediments
2) if so, is this a pilot scale or grassroots project. There is a lot of literature about re-slurrying transuranic sediments in tanks at ORNL and Hanford and other places. It seems that DOE has this work in progress.
3) if possible to disclose, what are you working on

Let me comment on and perhaps anwswer some of your questions.
De Dietrich, as Art told you, is a manufacturer of chemical process equipment including jacketed tanks. The typical product is a CSTR with a mechanical agitator. They can fab the vessel for whatever design is required. Patterson Ltd in Cleveland, Oh and Addison Fabricators in Addison, Al also fab jacketed vessels.

Heat transfer with Pulse Jet Mixers (PJM) is about 2/3 that of a CSTR (mechanical). I'm sure you are familiar with the PJM design but let me go over it just to be sure we are talking about the same thing. Consider two eductors essentially in parallel. Compressed air is connected to the converging side nozzle of each. The diverging side nozzle of one is connected to a small tank; the other vents to the atmosphere. Aspirating nozzles are connected. A pipe connects the small tank to a nozzle(s) in the tank to be mixed. The cycle starts by feeding air to the eductor open to the atmosphere. This pulls a vacuum on the small tank and sucks fluid into the small tank. At the target level air is switched to the other eductor and the small tank is pressured to a target pressure. The contents on the small tank are forced back into the mix tank by pressure through a small nozzle. When the small tank is empty, excess pressure is vented and the cycle starts again.

The number of mix nozzles and mix pressure is a function of the volume of the mix tank, the viscosity of the liquid phase, the settling velocity of the dispersed particles, and the cycle time. Typically PJM's are designed for the fastest possible cycle time at a given pressure (about 90 s)
Longer cycle times can be achieved with more mix nozzles. It would seem that compressed air consumption could be high with PJM's. If there are hazardous materials in the air, it becomes more of a problem.

http://www.wmsym.org/abstracts/2001/31A/31A.8.pdf

describes PJM tests conducted with materials to simulate transuranic wastes. They homogenized a vessel with 17% solids.

Heat transfer with slurries is impacted by the particles. The particles increase the effective thermal conductivity of the fluid in the reactor. Hydrodynamically, heat transfer is affected by the particle influence on boundary layer thickness. Heat transfer to a slurry is enhanced over transfer to the liquid phase.

A simple model of this system could be achieved using a CSTR containing the liquid phase only and whatever heating medium desired in the jacket. A PJM mixed, jacketed CSTR will achieve about 2/3 the heat transfer of the model. This system needs to be modeled. A pilot scale model could answer some of the questions for which there is no data.
De Dietrich can run CFD on the CSTR. AEA can perhaps provide info about the mixing capability and design of the PJM.

Thermal conductivities for transuranic salts are in the range of 7-10 W/mK @ 400 K. Vessel wall fouling may be addressed if the application is grassroots or pilot plant. In a batch operation, a rinse cycle (dilute nitric acid) could be run whenever heat transfer begins to drop.

I am not able to assist you with detailed analysis of this vessel. I think we have gravitated to an exotic system.

An estimate for the mixture thermal conductivity can be estimated from the following:
k(mix)/k(liq) = 2*(1/(1-p^.3333)) where p is estimated from
p = (rho(solid)-rho(mix))/(rho(solid)-rho(liquid)).

I thank you for the opportunity to participate in your problem.
Bill Choate

 

[GenB]

So much to read.
well i could not find the desired temp of the system or product, if the heating of the product is anything above 160 deg F, steam will be the desired source of heat in a double jacket system (as a kettle), it is most efficient and simple to build or aquire.
ER