November 23, 2007

Understanding Compressors Curves

Compressor Curves are generally misunderstood in day-today working of process engineers and many confusions arises out of discussion among cross functional teams. This becomes more confusing in case of Recycle loop systems for example in Ammonia synthesis loop OR Ethylene oxide Synthesis Loop. In view of this, I decided to write this article based on my recent experience.


In case of Compressors, first thing is to understand the Head Vs Flow Curve. Please do not get confused between Head and discharge pressure which are significantly different than pumps where usually fluid specific gravity is around 1.0.

Now Let us first note down the requirement, as below

  1. Curve Readings - Head & Flow from Design Curve.
  2. Speed at Which the Curve is applicable.
  3. Design Suction P & T.
  4. Design Molecular Weight of the Gas
These are the minimum requirements to simulate your compressor for any modification study.

Now if you need to consider any speed change it can be (As shown in attached file), otherwise new operating conditions can be specified. So the first step is to consider the new flow & head curve based on revised speed.

Then the program in attached sheet calculates actual suction volume at design conditions and then calculates developed head at new speed at specified flow. This head can be converted to new conditions in terms of discharge pressure.

Or if desired head is given new flow can be calculated by iterations from the given curve at revised speed.

In this way when we convert actual condition to design basis and consider converted suction volume the original curve is still applicable. This helps in avoiding any confusion due to change in suction condition in actual plant operation compared to design. The fundamental is simple - if you convert gas volume at design condition that will be understood by the compressor as if it is operating on the curve.

OR if compressor handles Actual M3 of flow at design condition then the developed head point will lie on the original curve or revised curve if speed is changed.

The assumptions in this sheet are
  1. No speed change beyond 10% of design value because errors are larger due to change in internal flow pattern.
  2. No change in efficiency. Though I have considered efficiency correction factor if you wish to apply it manually.
  3. No thermodynamic calculation of discharge temperature is done.
If you have any query related to any compressor problem kindly let me know. I have also developed a BC++ program for any high pressure real gas mixture compression system which uses equation of states for the calculation of thermodynamic properties of the gaseous mixture and its efficiency, which is tested for highly non ideal system also upto 220 bar pressure.

To Download the file , just click .... Compressor Evaluation

No of Downloads:-

Readymade pack is available for Hydrogen, Air, Nitrogen, Oxygen, Synthesis Gas of Ammonia plant, Recycle gas for EO plant, CO2. For any other gaseous mixture it can be easily modified.

Get Free Updates:
*Please click on the confirmation link sent in your Spam folder of Email*

Continue to read this post...

November 08, 2007

Oil Tanks - Tonnage Measurement - New Cost Effective Method

Recently Zaki had posted a write up on his blog related to this issue, whereas I commented about inaccuracy of this method & my experience. I had been doing it in my earlier days but I was never satisfied with the results due to large inaccuracies. So I thought to share it with you which will help you in taking care of these aspects.

One may not think or consider an inaccuracy of 0.5-2.5 % seriously but the view can be made clear from a simple example in case of tanker of etiher diesel or fuel oil. The tankers usually do not have more than 2 meter in height. If you dip tape & measure where an error of <1">) than you have an error of 0.5% (10mm / 2000mm which is tank height)to 1.2%.

The Method
In the conventional method a metallic dip tape is dipped from the top of the tanker or storage system, which goes in due to its weight and thus we measure the level.

The Problem
The problem arises due to three factors i.e. Density of Oil, Viscosity of Oil & Surface Tension (to some extent). As said, tape goes inside due to its weight so a buoyancy force also act on it which increases its floating nature.

Secondly due to viscosity & floating tendency (as described above) the tape goes inclined or slacks inside the tank. This goes unnoticed due to invisibility.

The Impact
The impact of this can be realized as below (Tanker example is already given above), Now we will consider an storage tank of ~5000 KL which is an usual size for FO storage & an usual height shall be ~10-15 meters. So let us consider 12 meter as general average height. If you go & measure the level in the tank which shall be somewhere ~70% of the total height than total H becomes ~8500 mm. Now in this case slack may go upto 2" (inches) which comes ~0.7%. practically it goes to as high as 1.5% however if I consider only 0.5% error than can you imagine the impact?

The impact come from your usage e.g. in a fertilizer plant the average daily consumption goes ~200-250 KL / day. Now 0.5% of it become ~1 KL / day. This will soon result in your inevntory error with in a month (30 KL difference) which is practically two tanker loads. So you might land up in great errors.

The Solution
There are two ways to solve this problem in practical life.
  1. Use some standard level gauge measurement system.

  2. There are many devices claimed to be accurate but the question is how much they are accuarte & what is the cost of accuracy you are paying for?

    Automatic gauging is the term coined for this solution but let us look at practical example & its cost.

    Equipment Cost / Tank - $3100 Average
    Installation - $500
    Monitor - $2200
    Cables etc. - $1600

    Minimum Total Cost - $7400 / Tank

    Accuracy Claimed - 0.1%
    Actually achieved - 0.3 - 0.5%
    RESULT - Practically no Gain, system remains same only adjustment frequency is changed.


  3. Use a simple manual method as described below.

  4. Based on my experience this is the best method so far I have found & is practically very low cost, in fact only ~1500$/year. Just install a simple tube of any transparent material if it is flexible than good, otherwise also, no problem e.g. Plastic tube or glass tube.

    Connect it to the bottom of the tank at any drain point or spare point. It doesn't matter where this bottom point is as long as your expected liquid level is above this point. Now there are two possibilities - 1. if you are using flexible tube & 2. If you are using straight (non flexible) tube.

    In case of flexible tube after connecting it to the bottom point (Drain should be closed) take it up to the expected level height manually. Open the bottom drain slowly & hold the tube from the maximum level; oil has risen in it. Touch it to the tank wall & straight the flexible tube portion in your reach on the tank wall. See if any drop or rise is there in the tube. Hold it straight on the wall of the tank & mark the level in the tube on tank wall using a pencil. This is your most accurate level point in the tank.

    In case of straight / nonflexible tube, it can be connected at the bottom of the same drain & top of the tube can be connected to the vapor space with a isolation valve so that vapor lock can be released whenever level is measured. It will look like a level gauge tube.

    However the disadvantage of this method is that it is not recommendable for volatile liquids e.g. Naphtha etc. rather they are not installed on big size tanks to avoid any mishap during leakages. So don't forget to install extra isolation valves on extended metal pipes required for connecting this tube. This will help in complete isolationof tube from tanks.

The Gain
In this case the error will not be more than 2-3 mm (Or say Max 5 mm) if done properly which will result in 0.05% max error. This will cause an error of 0.1 KL/Day for the same example of fertilizer plant. So the same error will occur after a year instead of a month.

Also it is very inexpensive compared to ~$7000 investment /tank but you need a person to do it which will cost ~$1500 / Year (Labour cost).

If you have nay question kindly let me know on the forum or on the comments section of this post.

Get Free Updates:
*Please click on the confirmation link sent in your Spam folder of Email*

Continue to read this post...

November 02, 2007

Site Updates

This post is a part of an ongoing updates on this site to keep you aware about developments & future planning.

I am adding DOWNLOADS section, which will provide useful downloads for all of you. Kindly feel free to browse it here OR select it from tab listed at the top of this page. You will find only those which I have used and found them useful for all.

Next I have removed Stores page, which shall be modified & relaunched by next month or so.

Next based on site stats I am finding that many of my readers are searching for cooling tower related queries. So I would request them to post their any kind of query related to Cooling Towers on my FORUM & I will answer them one by one including evaluations of performance, tower capacity related questions or anything else.

I am also including one GuestBook on downloads page & request readers to leave their feedback so that I can further improve this site.



Get Free Updates:
*Please click on the confirmation link sent in your Spam folder of Email*

Continue to read this post...