5 replies to this topic

[pmbvila]

Hello everybody

I am trying to calculate how much water I can chill from 20 to 10ºC (12 mbar-9 torr) using about 50 ton/h steam (6 bara; 200ºC) with an ejector system in a vacuum chiller.

The corresponding flash vapour in the drum is about 1.5% of water inlet flow. Is this vapour constantly removed through the ejector suction? Is the suction load the entire flash vapour quantity?

I think a two-stage intercondensing ejector system would work, but I need help calculating the required steam consumption. Given the suction load (flashed vap+air seepage) how do I calculate motive steam flow?

I usually use Perry's design curve for single stage ejectors but the expansion ratios I require are far outside the curves. The steam consumptions I have arrived at are far greater than those reported by suppliers.

Any sort of input would be welcome.

Thanks,

Pedro

[Art Montemayor]

Pedro:

What is your the type of design that you are after? Are you just studying the application? Are you seriously having to design the application? Or is this an academic problem? Please tell us all the basic data and the scope of work in order to help you out the most that we can. We are just as busy as you are and hate to waste a lot of time and effort needlessly trying to guess just exactly what it is that you need. We can't furnish specific answers when we have generalized questions.

For your basic information, I can tell you up front that you don't generate the motive steam rate for the vacuum jet. You specify the process conditions to a steam jet manufacturer and the manufacturer will advise you what the steam rate for the specific application will be.

Where are you getting your steam rates from? --and under what conditions?

If you have already "toured" the Internet, you probably already found the attached information. But maybe it can be of help in giving you an idea.
 Artisan_Vacuum_Water_Chiller.doc   175KB   134 downloads

[pmbvila]

Good day

Thanks for answering, I will try and sum up the whole story. I am just doing a preliminary study to implement a vacuum chiller in our petrochemical complex. We are expanding the complex and are playing around with ideas that will allow us to maximize LP steam consumption. It would be in our power plants interest to spend some 50 ton/h LP steam. If we can provide more chilled water (max. 10ºC) than the current and future needs it would be ok, because we are thinking a bit on energy integration terms, and it would also give us the possibility of maximizing steam or electricity consumption depending on the time of day. We would also become trigeneration. On the downside I have cooling water consumption increase and corresponding costs.

"For your basic information, I can tell you up front that you don't generate the motive steam rate for the vacuum jet. You specify the process conditions to a steam jet manufacturer and the manufacturer will advise you what the steam rate for the specific application will be."

But surely there is a way to calculate this just to check.

"Where are you getting your steam rates from? --and under what conditions?"

I am using both Perry's Chem. Eng. Handbook's design curves for single stage ejectors, which give you entrainment ratios for the intended compression and expansion ratios, and Walas's Chem. Process Equipment, wich provides steam consumptions for multistage ejector systems with or without intercondensing. These curves are for the standard 100 psig 10 degree SH steam and I use a correction factor of 1.2 to correct my 6 bar 200ºC steam. Our cooling water is about 22ºC (70ºF I think).

I also have several manufacturers brochures and reports. My problem is the results I get from literature and manufacturers are quite different, and I need to be sure of what I am doing.

Any more information you think is important let me know, I will answer the next day as we are in very different time zones.

Thanks for your time, please don't spend too much of it on this. No need to hurry.

Pedro

[pmbvila]

Art:

Just read the doc you sent me. Thanks. It seems what I need is similar to the last case on the table

gals/min: 125
ºF out: 50 (spot on)
ºF return: 68 (spot on)
cooling: 100 ton
tower sise: 30x50
steam: 70 psig (against my 70 psig 40ºC SH)
lbs/hr: 2,500
CW: 600
CW temp: 90 (ours is 70-80ºF throughout the year)

I would just like to ask you:

If the cooling is in US standard ton of refrigeration (3500W)? Because if I calculate the DT=10ºC heat loss for 125 gals/min water I don't get 100 tons refrigeration.

Is the suction load on this system about 150000 lb/hr? Just to see if I am thinking correctly.

Is the superheated steam especially unadvisable?

Can you recommend any books on the subject? I am curious about the calculations involved.

Answer whenever you have the possibility. You have already been a great help.

Pedro

[Art Montemayor]

Pedro:

It’s nice to find you receptive to further information on one of my more satisfying and interesting subjects: the production of vital, profitable process cooling using waste steam. This is a type of project that can literally make a big, big hero out of an engineer. When done correctly and with practical common sense, this project can put money in your pocket and more profitable product out the front gate. In other words, this is the type of situation that can quickly turn a process situation that is considered waste and full of expensive problems into a “win-win” situation. It is a potentially rewarding type of project – both professionally and economically. And it is very important from an ecological and efficiency point of view. You simply can’t lose – if you do it right. I’ve done this twice, and I hope things come out just as rewarding for you.

The answers to your specific questions are:

1. I believe the water cooling (refrigeration) effect shown on the table is in US standard ton of refrigeration (12,000 Btu/hr) and the figure of 100 tons refrigeration is correct. See the attached spreadsheet that I used to verify the published figures. Of course, the calculated figures were rounded out.
2. I don’t quite understand your question, “Is the suction load on this system about 150,000 lb/hr?”. As you can see I calculate approximately 62,550 lb/hr of water load on the refrigeration effect.
3. The book that explained and showed me how to apply this system is: Applied Process Design for Chemical and Petrochemical Plants by Ernest E. Ludwig, ISBN 0884156516. This is one of the all-time great process design books ever published and I highly recommend all 3 volumes of the series. If you don’t own the volumes, buy them. If you can’t afford them now, borrow them. You will never regret reading and digesting all the knowledge that Ernie Ludwig put into these books.

Now to comment on your prior response. Steam rates are empirical. They are dependent on the design and configuration of the jets. Just as in the case of steam turbines, you can calculate the thermodynamic ideal steam consumption, but the efficiency and the simplicity of the application deem that accurate estimates are best done empirically with published tables by the manufacturers. It’s what everyone does. I would recommend that you stick with the manufacturer’s estimates rather than the literature. Books like Perry’s and Walas are OK for rough figures and academic problems. Out in the plant, where you have to get as practically close to the real thing as you can, you are far better off going with the guys that “wrote the book” – the actual manufacturers. They, for all practical purposes, are the experts – not Perry nor Walas.

The basic strength and main thrust of employing this type of cooling water refrigeration is to employ waste steam – a source of waste and operating costs in a normal plant. The idea is to convert this waste to a more profitable use. Therefore, using superheated steam, in my opinion, is going off in the wrong direction and doesn’t do justice to this application. Superheated steam is produced primarily for two purposes:

• to ensure that condensation in steam distribution headers is minimized;
• to employ as the motive force for steam turbines.

I have used 30 psig turbine exhaust steam as the driving force for jet refrigeration with success in the past. I could never justify the use of superheated steam in this application. It is too expensive.

I hope this experience is of some help to you.
 Steam_Jet_Refrig.zip   209.68KB   93 downloads

[pmbvila]

1. As far as refrigeration is concerned I had an error in my excel sheet.

2. I meant the suction load removed from the drum by entrainment through the ejector suction. I calculate 1.5-2% flash vapour (most of load) + air from leaks (little bit) + air dissolved in water (also a little bit)

3. 6 bar steam with a 40ºC superheat is our exhaust steam from back-pressure turbines, we need to spend it to keep the turbines going, and its better than releasing it to the atmosfere.

I have a couple of good documents from GEA Jet Pumps to help me with the calculations. Did a preliminary analysis based on fuel, emission, electricity and cooling water cost and I think vacuum cooling will be worthwhile even without further integration. Thanks for all your help.