Understanding a #natgas reciprocating compressor package (part IV)

It has been over 4 months that I’ve been working on this live inquiry from one of our prestigious clients. The tender document on a whole seems to be very simple but if one focuses on the technical aspect of it, and then the complications arise. Like I said in my previous post, gas composition is one very important aspect when it comes to designing a compressor. To keep it simple, more the heavier hydrocarbons in the gas, more complex the process gets. We have to understand in detail the behaviour of gas at different pressures and temperatures. This would help us to decide the compressor metallurgy for that particular application. If it is a sour gas that is gas with high amount of hydro-sulphide or water with high amount of carbon dioxide then the metallurgy of the compressor will change to stainless steel. The piston rod, valves and cylinders may change to stainless steel. To be honest, a compressor can handle any type of gas provided it is informed clearly to the compressor venders.

A reciprocating compressor package consists of many other items such as pressure vessels, driver and a heat exchanger. The driver can be a motor or a gas engine depending on the client’s requirement. The heat exchanger can be shell and tube type or an air-cooled heat exchanger. The exact gas composition in mole% is required while designing the air-cooled heat exchanger. The metallurgy of the vessel also plays a vital role while conceptualizing the package. Generally, upstream gas is requires a good process study. On the other hand, gases contain C1 + C2 + C3>90% makes life easy for the compressor guy as the other items can be finalized easily and accurately. Once the gas is analyzed, selecting the compressor is as easy as pie.

This ends our discussion on selection of reciprocating gas compressors and hope it’s helpful. Many electrical factors also affect the compressor performance which we’ll see later as it involves understanding the compressor package on a whole. Even selecting right instrumentation like valves, gas detectors etc is important which we will discuss later.

Since, we are interested in compressor selection, so stay tuned for the next post where, we will study in detail on selecting a screw compressor for natural gas applications.

Understanding a #natgas reciprocating compressor package (part III)

In my previous post, we studied in detail on how to select cylinders for a particular frame for a given application and how different parameters affect the estimated power calculated by the compressor selection performance software. Once the compressor frame, compressor cylinders, primary driver i: e motor or engine is selected with necessary input parameters; we shall run the compressor to understand the output. Firstly, we need to see if it shows any major errors like exceeding rod loads, cylinders of lower RDP are selected, high discharge temperatures, improper degree reversal values and unacceptable suction or discharge volumetric efficiencies. To solve these issues, let us have a look at each one of them individually as this will enhance our problem solving ability:

  1. Exceeding rod loads: This is the most frequently occurring problem when we select cylinders having large-bore diameters to achieve the required capacity. Rod load is defined as the amount of weight exerted on the connecting rod by the gas and the by the inertia of the moving parts. In this case, we either reduce the cylinder sizes and compromise with the flow or else we can select frames of higher BkW. Rod loads also have to be check for safety relief case where discharge pressure is 1.1*given discharge pressure. The most advisable value should be 80% of maximum allowable rod loads.
  2. Cylinders of lower RDP are selected: This is an error when we select cylinders which cannot operate at that particular stage as the pressure at that stage is higher than the rated discharge pressure of those cylinders. This error ultimately leads to high rod loads.
  3. High discharge temperature: This generally happens when the cylinder are previous stages are either too big for the frame to handle or the cylinder in the latter stages are bigger than the previous stages. This also indicates that the compression ratios are higher in the latter stages. This may also lead to high gas loads in any of the stages.
  4. Degree reversal: As far my experience and observation goes, this happens when the bore diameters of cylinders in the initial stages are very big as compared to the bore diameters of the latter stages. In order to maintain compression ratios for a balanced machine, we cannot use improper cylinder size combinations.
  5. Unacceptable volumetric efficiencies: The ideal volumetric efficiencies should be about 75% to 80%. If this has to be achieved then cylinder selections need to have decreasing compressor ratios and discharge temperatures.

To understand the above, one has to sit with the compressor performance software and try every combination. Once results flash for all different combinations then it will result in better understanding of the working of the compressor. The points discussed are the major errors which we have to tackle in order to achieve an efficient compressor sizing. There are other minute things and also, many concepts related to the gas composition. The working of the compressor highly depends on the gas composition. If the gas contains heavy hydrocarbons then we have to consider many other possibilities. So, tune up for the next post where we will look into details of the compressor relation and the gas used.

Understanding a #natgas reciprocating compressor package (Part II)

In my previous post, I had spoken about the estimated power shown by the compressor performance software which is dependent of a few major factors.

Lets us understand each one of them separately:

  1. First and foremost, the design suction pressure specified by the purchaser for which the package is to be designed. If we observe, we will realize that as the suction pressure increases, the estimated power reduces with other parameters kept constant.
  2. The discharge pressure specified by the purchaser for which the package is to be designed. If we observe, we will realize that as the discharge pressure reduces, the estimated power reduces with other parameters kept constant.
  3. The input design capacity is always is 1.03 times the flow specified by the purchaser and the estimated power reduces with reduction in capacity with other parameters kept constant.
  4. The suction temperature plays a vital role in design of the air-cooled heat exchanger but as far as flow is concerned it affects it exponentially. The trend is as the suction temperature reduces the power reduces too with other parameters kept constant.
  5. One of the important factors is deciding on the number of stages which is a trade of between estimated power and cost of the air-cooled heat exchanger. As we increase the number of stages the estimated power of the main driver reduces with other parameters kept constant but the cost of the ACHE (Air-cooled Heat Exchanger) increases as the size of the ACHE increase to accommodate cooling of the gas passing through every stage.
  6. There is always a loss of pressure till it arrives at the battery limit i: e inlet of the package; so we have to consider it and it is called skid edge suction pressure loss. Logically, with increase in loss the estimated power reduces with other parameters constant.

Above parameters are the inlet conditions on which the estimated power variations are discussed. There are certain other functions like pressure drops per stage and discharge temperature after the discharge KOD on the estimated power fluctuates but that will seen in the output shown by the software. We will get into deep understanding of the output later. The actual power is always more the estimated power as estimated power is only a guideline to select a compressor frame. Once we select the frame which can be a 2-throw or a 4-throw machine depending primarily on the required capacity, it is time to select the compressor cylinders. While selecting compressor cylinders; a major point to keep in mind is the rated discharge pressure of that cylinder. Every stage has a unique discharge pressure depending on the number of stages we use to design the compressor. The RDP (Rated Discharge Pressure) of every cylinder should be at least 1.5 times more than the inter-stage discharge pressure. So we can keep increasing the sizes of the cylinders till the RDP of the cylinder higher that the inter-stage discharge pressure. However, keeping in mind that the cylinder sizes should be decreasing per stage as this would result in decreasing pressure ratios per stage. When the pressure ratios are decreasing the machine is said to be balanced. Due to this the discharge temperatures also reduce per stage and the thermal stresses are handled well by the machine. The driver to be selected depends on the requirement given by the purchaser. It can either be engine driven or motor driven.

Once all of the above is done, we can move ahead the check the performance and understand the results it shows. Based on the results, we will have to make a few changes which we will look in detail in the next post. Till then stay tuned to be a reciprocating compressor sizing expert.

Understanding a #natgas reciprocating compressor package (Part I)

Now for any amateur, understanding and conceptualizing a reciprocating compressed natural gas distribution package isn’t a cake walk. I have been working in this sector for almost 14 months now. First thing that comes to our minds is what does the customer exactly want? Frankly, this field is so complex, that at times even the customer has no idea what he is looking for. The sales people in this sector always follow the rule of “customer in” and not “product out”. We need to share all the information with our internal customers and external customers for designing an efficient machine.

So let’s get on with the designing aspects of a reciprocating package. It all starts with the parameters given to us and recording it in the compressor performance software. The most important ones are design suction pressure, suction pressure range, suction temperature, required flow, ambient temperature, gas composition and discharge pressure. This compressor performance software has many minute things which a rookie application person generally misses out. Nevertheless, any software you use as in of any company i: e Ariel or Cameron or Dresser-Rand, all having the same functions. It is the compressor that has limitations but not the software. Different companies include various assumptions which are calculated by the software internally which remain hidden. Estimated Power vs. Flow is something all the sales engineers are worried about. I would like to elaborate on the flow/capacity. This is the most essential piece of information given by the purchaser. The highlighting point is that the design flow should always be considered at 1.03 times of actual specified flow given by the purchaser. Most of the performance softwares do consider it but a few do not. To confirm the best person would be the operator at the site and the application engineer sizing the compressor. Sales person is the link between these engineers. The best advice would be to always consider 3% more flow than the required capacity. This extra consideration in the flow is also known as No Negative Tolerance (NNT). Now, let us move on to the power aspect. The performance software shows an estimated power on the basis of various input parameters. We’ll get into the details of these parameters later. The power flashed by the respective software is actually the frame BkW. The frame which we select shall be having a higher BkW than the estimated figure shown. The frame rating should not be very high as compared to the estimated power as that would result in under-utilization of the compressor frame.

Though, we are using a software to size our reciprocating compressor package, there are various input parameters like pressure drops and discharge cooler temperatures which the software calculates. However, these are figures which are calculated on basis of various formulae but when practically observed, these values differ. Thus we can say, compressor sizing is a perfect blend of theory and practical knowledge. Also, the skid suction pressure loss is to be taken as 1% of the suction pressure. The after-cooler discharge temperatures should be 6 deg C more than the considered ambient temperature.

I am sure the above information will be helpful in understanding the initial stages of compressor design. For more information, tune in for the next post where we get into cylinder selection and performance analysis.