Gas Turbing Inlet Cooling by Evaporation 1

[dfrobles]

Hello all,

I know the topic I will ask about has been already discussed in a few other threads but, after reading them and getting as much information as I could on the internet, I was unable to find the answer to my question.

We operate 2x 27MW gas turbines in our process, not to generate electricity but to boost our process gas. Since we are operating in dessert conditions, the GT capacity is very affected by the warm weahter we get over here. Temperatures above 40C (104F) (sometimes well above this value) are common for around 4 months/year. Since when this occurs, relativy humidity is very low (around 15%), it has been thought that the use of a inlet air cooling system based on evaporative cooling (either with a wetted media or fogging) would be a possibility to increase power outlet during theses warm months.

After reading a lot of information on the internet, I see this technology is well proven and, doing the calculations, could work fine with our conditions: from a 40ºC-104F dry bulb temperature we have around 20ºC-68F WB temperature. It means up to around 20ºC-36F gain. This is pretty interesting. I can also easily calculate the air consumption.

Now my question, and please forgive me if it is stupid, but I haven´t been able to find the answer and my heat and transfer knowledge in the psycometrics doesn´t allow me to answer. In order to saturate the inlet air with water, and get the temperature decrase above, what would be the temperature of the water that we will be, either spraying either circulating in the wetting media? Since this is a complex heat and mass transfer problem, involving latent heat as well as sensible heat, I wonder if it has to be lower than 20ºC (lowest air temperature) in order to the water droplets to vaporise and saturate the air or it can be higher than that.


The above is important since, in the climate, air is not the only thing that gets warm. Available potable water (that will have to be demineralised) is also hot, around 30ºC-86F. So if water has to be cooled down prior to be sprayed, an additional small refrigerating system will be required. This will add to a small demin system to get proper water and the skid itself (fogging) or the wetting media system.

I hope my explanation is clear and you can give me an answer or some reference to better understand the heat and mass transfer involed in the system.

Best regards
David

 

[MortenA]

A psychrometric chart should tell you what you need to know.

Best regards

Morten

 

[zdas04]

Think energy transfer from the water, not ΔT of the water. The big change in enthalpy happens with the phase change, so your refrigeration plant would actually be counter productive (because the cooler the water, the less evaporation). I did one of these where we used produced water that entered the process about 15C warmer than ambient and it worked very well.

Also, you don't need DI water for this. If I'm using water that has more TDS than potable, I just increase the water flow rate to get some dilution of the solids build-up.

If you have produced water available, using it for this sort of thing is very useful. In the conditions you described, I would be looking at an evaporative cooler for the air intake, but I'd also be looking at using a plate and frame heat exchanger (with produced water on the outside) on the combustion air out of the power turbine compressor. That can be a very large boost in available power, while adding considerable heat to the produced water (accelerating evaporation).

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[willard3]

The water/air temperature mixture will approach the wet-bulb temperature for the conditions you outline. Wet bulb can be found on a psychrometric chart.

 

[dfrobles]

Hello zdas04,

Thanks for your reply. From it, it is now clear that we could perform the turbine air inlet cooling with the availiable water at 30-35ºC. It is still not clear in my head the mechanism for this to happen but first, the urgent, and then, looking for more information, the important.

Concerning the water quality, you seem to be partially right. In the information I could gather, it is true that demineralised water is not compulsory in the evaporative cooling. However, if we finally use the "fogging" system, it looks like demin water is a must. But that is something we will discuss with the vendor once one option or the other is chosen. Some vendors even supply the small demin water plant together with the "fogging" skid.

I don´t think I fully understand your last paragraph, about using a plate and frame HX on the combustion air out of the power turbine compressor. Do you mean the combustion exhaust gas or the compressed air before entering the combustion chamber? Anyway, we do not have any produced water (we do not use steam) so I think, it is not applicable to our case.

Again, thanks a lot.

MortenA, if you do not have anything to add to the discussion, you may consider remaining quiet.

 

[25362]

Reading

http://www.muellerenvironmental.com/documents/100-020-88-GT-41.pdf

may be of help.

 

[racookpe1978]

Use as water as clean as possible: The mass of evaporated solids will quickly build up and "plate off" on the compressor blades - hurts both compressive flow paths (the geometry) and erodes the blades' material by impact and abrasion. DI water IS better!

Check your energy balance calc's again: Compare what you're doing now (104 inlet temp's, no water at all) with "cool" water (add in the energy requirement to chill the water!) and then with "room temperature" water.

 

[dfrobles]

25362,

Thanks for the paper, it is the first one I find where calculations are explained. It looks like that in evaporative coolers with recirculation the water temperature has no influence, or very little, and air outlet temperature can be calculate following the isoenthalpy line from the psycometric diagram: outlet air temperature = wet bulb temperature, if the system is perfect.

One final question: in case with use a "fogging" system instead of an evaporative cooler, can we consider, in terms of temperature calculation, as a system without recirculation and then, the sensible heat to cool the water from ambient temperrature to the wet bulb temperature must be considered, as explained in the paper?

Thanks a lot for all you have done up to now, everything is more clear.

Regards,
David

 

[EdStainless]

Have you talked with Mee?
With lesser quality water you would use an evap cooling system, if you can get good water then direct inlet fog.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[dfrobles]

EdStainless,
No, I haven´t talked to any suppliers yet. The idea is to present the different possibilites to the management, see if they are finally interested and then, if they are, go with a technology. Most likely, evaporative cooling/fogging, since chillers are expensive and not worth it for our small application. At that moment, I will talk to a supplier. Thanks for your comment, water quality is clear for me.

Regards,
David

 

[zdas04]

By "produced water" I was talking about the water that is produced from gas wells that must be disposed of. The Oil & Gas industry treats this water as a major pain and huge cost driver. I try to look for places where it can be an asset. If your gas comes from somewhere far downstream of the wells than this may not be a problem/opportunity.

When I talked about a plate & frame HX, I was taking about cooling the air into the combustion chamber. Dropping that temperature gives you 2-3 times more benefit than cooling the suction air, but it costs a lot more.

As to water quality in an evaporative cooler, it really does not have to be very good. I've used 10,000 mg/L water in this application. Typically, I adjust the water flow rate (with 1,000 mg/L water) such that 20% of the inlet water will evaporate. With 10,000 mg/L I would try to flow fast enough that 2-5% of the water would evaporate on an average day. A properly designed evaporative cooler captures all of the solids from the water either in the effluent or in the evaporation pads. It doesn't get on the pipe surfaces or on the turbine blades.

The evaporation vs. fogging discussion is really a field vs. process discussion. I have no problem (as a field engineer) designing an evaporative cooler. If I wanted to fog into a multi-million dollar turbine I would hand that off to a Process Engineer and expect him to simulate the living hell out of it to make sure that we fully understood required/allowable/maximum flow rates. I definitely would not design it by rules of thumb.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[rmw]

David,

I'd love to know more about the P&F concept. Is turbine compressor discharge hot air on the side opposite the produced water? After your last post, I went back and read the first one and that is what I make of it.

rmw

 

[zdas04]

Yep, you are circulating waste water through the HX to cool the air going into combustion. You can pick up a bunch of hp with a 10-15C increase in produced water temp. If you are pulling out of an evaporation pond those BTU's really help evaporation.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[dfrobles]

zdas04,
Thank you for all the useful information. The water quality is now clear, that was a major concern. This is one advantage for evaporative cooler since for fogging, demin water is compulsory, so a small reverse osmosis unit would be required. Apart from that, I see that fogging is also a well establised technology and seems to be in operation in many turbines.

For you other comment, we do not have a problem with produced water since all of it is reintroduced in the reservoir in order to help keep reservoir pressure. So, it does not seem too interesting for us. In addition, here it is better to keep things simple.

rmw, I have not understood anything in your post. Please clarify and I will try to answer your question.

Thank you all for this information.

 

[rmw]

DFRobles, if your name is David, sorry. My post was directed to David of ZASD04 handle.

David of ZASD04, what size engines or turbines are you describing when you mention passing air through P&F's?

rmw

 

[zdas04]

The work I've done on it was recips. I did feasibility work on two sites: (1) the first was 4 Cat 3516 LE engines driving recip compressors; and (2) 3 Waukesha 7042 LE driving recip compressors. Both sites had adjacent evaporation ponds and the plan was to run the inter coolers, after coolers, turbo after cooler, and engine oil through the P&F HX and dump the heat back into the evaporation ponds.

In the first site we were going to save $5k/day 5 months of the year for DI water that they were spraying on the fin fan coolers to keep the machines running. The project added about 20% more hp (for the life of me I can't remember the hp rating of the engines at 6,000 ft elevation and it's too late at night to look it up) which was going to allow them to skip a multi million dollar station expansion to say nothing of not having to shut wells in because of lack of pond space. No reason was ever given for killing the project, it just died. The company engineer had zero ownership of the idea and kind of resented it being shoved down his throat by a contractor (his boss had hired me and the boss was fired for other reasons while the project was under review). The thermodynamics were amazing and the economics were phenomenal (payout of a half million dollar project in 18 days). If I had been the company engineer I would have strangled managers until I found one who was willing to find the money.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[dfrobles]

Hello zdas04,

I thought I had found all I could get from this thread but it seems I could get one more info from you. You seem to be aware of some prices involves with the installation of this kind of evaporative coolers in gas turbines and it could be really helpful for me in order to have a ballpark figure to present the management and before going ahead contacting suppliers.

In some presentation, I have read that the installed cost of such a kind of equipment would be around 20-60$ per additional kW. This is for fogging (Cadnwell) but I have also read that costs are pretty similar for both technologies. Phsycometrices tell me that, with my 38000hp turbine, in summer conditions I could get an extra power of around 3,2MW max. With this price range, I would get a total installed cost of 64k$- 192k$. Since our turbine is relative small, I would guess the cost would be closer to the upper limit than the low one. Do you think 200k$ is a good estimatite for the evarporative cooler? In some other presentation I also read a 26k$ per MW which is consistent with the other one.

I was pretty sure of this values but your last post has created doubts since your figures don´t seem to add up. Both turbines you talk about are ridiciusly small (in the range of 1MW for what I could find), then you talk about a half a million project which seems gigantic for 1MW. And then you also talke about huge savings, 5k$/day and since the payout of the 500k$ project was 18 days, it would mean around 22k$/day, huge.

So, if you can confirm my values, it would be great. A chiller system I have found to be in the range 300-500k$/MW, give a total cost of 1-1,5M$, reason why this option is not good for us, compared to the savings we can get.

Thanks a lot.
David

 

[zdas04]

"Ridiculously small?". In my world that is the upper end of the hp used in a single frame (BTW, it wasn't a turbine it was a reciprocating engine). Raw field gas is far too variable for big hp onshore. It sounds like if I were to dig up the actual numbers and was able to release them (which are still covered by an NDA) they wouldn't be germane to your "important" hp.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[dfrobles]

Hi zdas04,

I am sorry if I bothered you with my comment, but be assure it was far from being my intention. I was just shocked when I read your numbers, since making the calculations, my project would cost between 10-12 times more as compared with the ballpark figures release by some manufactures. That is the only reason of my comment.

It is for sure I am not interested in knowing the details of that project. My only question was if the figures I am using, according to your experience, seem reasonable or not. That´s all, if you have experience in having seen vendors proposals you may know, if not, you don´t. That is just to confirm my values I calculate from 3 sources I found in the internet.

Again, i am sorry if I bothered you.

Regards,
David

 

[dcasto]

if you don't mind here is another paper where they even use refrigeration to cool the air and get more HP out than the refrigeration used to cool the combustion air.

http://scholar.googleusercontent.co...t620M-QJ:scholar.google.com/&hl=en&as_sdt=0,6