Gas quality variation 2

[ivymike]

I'm trying to find some data to support a product development decision at work, and I don't have the right contacts - I need to figure out how much gas quality (LHV and MN) can swing at sites supplied by wellhead gas gathered from a few wells. Consider remote power generation for oil production in the Middle East, where several gas gensets run on low MN gas derived from the oil well & processed to some extent. Anyone have any gas samples taken in a time sequence that show a swing in LHV and/or MN over the course of an hour/day/month/year?

 

[catserveng]

I did a bunch of data collection on this issue twice, all data was collected in North America (USA and Canada). Let me talk to the customer on that effort to see if they'll let me release any of that info.

In a nutshell, in both efforts, a wide variation in how gas quality changed over time. We looked at weather conditions, type of field, type of pretreatment, draw rates on fields, and time of day. Some fields vary stable, others had wide variations that didn't seem to have any good reason. We also had some data from three sites operating in India, but no real good indication at the time on how to predict variations in field gas quality, at least in all the info I saw.

We looked at this issue for gas fueled engines supplying oil field power of all kinds and for dual fuel applications primarily for diesel engine driven units for drill rig power.

If you'd like to discuss directly my contact info is here,

http://www.pactectric.com

MikeL.

 

[georgeverghese]

There are limits for fuel gas quality swings for both Wobbe Index ( gas LHV/sq rt(gas sg)) and its rate of change for gas turbines, and the same would apply for gas engines - engine manufacturers usually have these in their catalogues. Have you spoken to Caterpillar, Waukesha, Borsig etc?

 

[BigInch]

Over short terms they can be very stable. Over several months to a year there can be little to considerable variations. Concentrate on establishing and working with the lowest limits over longest times.

You should not even attempt to correlate any data across fields in North America, never mind NA to India. That's crazy. This is highly reservoir dependent. Could easily be well dependent. Sometimes you can lose a whole well (30MMSCFD - 300SCFD) in some very tight formations within a couple of months.

Reaction to change doesn't stop it

 

[zdas04]

BigInch,
What exactly are you basing the statement that short term data is stable on?

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[ivymike]

Hi Georgeverghese,

My question is about what the wells will put out, not what the engines can accept as input. The latter is easy to understand. Given that engines have limits on fuel quality, one has to determine whether the potential fuel supply will fall within those bounds & that's what I'm interested in learning about.

Regards

 

[LittleInch]

I think you need to really talk to the rock doctors aka reservoir engineers.

However any prediction they make is heavily influenced by performance so what they predict at the start of field life will change considerably after some time in production.

All this is so reservoir and well dependant there is, IMHO, no such basis for making any decision based on a different field. Like catserveng says "Some fields vary stable, others had wide variations"

" I need to figure out how much gas quality (LHV and MN) can swing at sites supplied by wellhead gas gathered from a few wells." Simple answer - A lot over 2-3 years or maybe very little over 8-10 years.

You would probably get a decent range able to be burnt for power over a 3 year min period. If your economics need a longer period then every month / year after that is increasing risk it won't supply the range of gas you want. Your risk.





Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ivymike]

Thanks - my biggest concern is around variation that happens over a duration of ~5-10 minutes, and secondarily the variation/drift that happens over a spark plug change interval, say 3-6 months. Variations slower than that could be accommodated with manual adjustments.

 

[georgeverghese]

Agreed, you'll have to talk to reservoir engineers about this - some people in your project team may have the reconstituted Corelab fluids composition for these reservoirs by now (?) , and there should be some process engineers who can set up a CVD ( constant volume depletion ) profile ( over time and or over depletion rate) on Hysis or Pro II for you for each of these reservoirs and mix them up as you wish to get the mix feed gas mole wt / LHV / Wobbe Index. A dynamic simulation may be required to enable short term swing info on these parameters for the operating scenario changes of interest.