Well Integrity Technical Section

All,

Well is completed with surface casing set at ~ 4,000', production casing set at ~ 12,000' & 7" Liner is set at ~ 13,200' in high pressure reservoir of ~ 6,500 psi & ~ 350 degF.

CBL/VDL/USIT logs were run for production casing & Liner section as well. All logs suggest no cement or very bad cement behind both casings. USIT log indicates macro channeling. 

From well integrity point of view, what corrective measures should be taken to mitigate the integrity issues while ascertaining the cement sheath bond. (Cement remedial job)?? If cement squeeze job is unsuccessful, what other materials can be used to avoid macro channeling?  

From industry best practices, What is the minimum thickness of cement patch is required behind production casing in order to avoid surface casing pressurized??

Dear Lajpat,

First I would ensure you know what's going on behind your casings:

• I guess with macrochanneling you mean mud channeling: does the CBL match the USIT? Compute the bond index from the CBL and add up the non-brown pixels of the USIT and compare the curves: if they match, the "liquid" USIT pixels are grouped together and they cover less than about half of the circumference (i.e. BI<=50%) then you may have a mud channel. If you do, you should be able to tell it from your cement job: density & rheology hierarchy, well deviation, caliper and centralisation and so on. Did you expect channeling? If not then it may not be a mud channel or you may have to increase your control on cement practices.
  • If the CBL doesn't match the USIT because it shows more bond (i.e. CBL BI>USIT BI), then it may be a case of lead-in-tail channeling so you don't have a problem. If the CBL shows less bond (i.e. CBL BI<USIT BI) and the USIT solid band is not continuous, then you may have a fluid-filled microannulus.
  • How long after the job did you acquire your logs? I take you logged the production casing after cementing the liner, so whatever was behind was likely set, but it's always good to be sure about it.
• No cement behind casing is rare, unless you didn't expect any: did you experience any losses? If so, where (study the cement job data and ask you neighbourhood geologist for weak/thief formations) ? Do you know what your hole volume should be (check the pressure at the end of displacement if your well is not terribly deviated) ?
  • If you put cement, but you don't see any then you may have debonding. If the USIT shows gas and the acoustic impedance is well calibrated (sensible ZMUD), then you have a gas-filled microannulus.

Second, I would map the possible leakage pathway:

• Is the channel extending from the liner shoe or the reservoir to the top of cement? Are there intermediate layers that would take a leak (lower pressure reservoirs, aquifers, weak formations...)?
  • If the log shows a continuous defect to the top of cement and the leak would end at the wellhead in the B annulus then check for sustained casing pressure: if there is evidence of a leak and the gas comes from the reservoir, then you definitely have a pathway.
  • If you should see a leak but you don't and things will not get worse with time (e.g. if you're naturally producing a gas reservoir)  then you may be safe. But watch out: you have to be sure you would see SCP. You may have downhole circulation that at best saps your recovery and at worst may quickly corrode your casing, charge intermediate horizon and even cause a safety & environmental disaster.
• If you have SCP I would clock the leak to have an independent measure of risk and figure out its causes. If the flow rate is of the order of 1-100 tons per year then it's most probably a microannulus; more than that may indicate gas migration while setting or a mud channel.

Only as a third step I would think about repair:

• If you have a mud channel (or a chimney from gas migration) and you have a leak and it's unsafe to keep the well in this state, then you have to fix it. If you have a microannulus I would definitely not squeeze unless you have overwhelming reasons to do it (again, safety).
• Squeezing a mud channel is not too different from primary cementing, except this time you should do it right: perforate and circulate with water (or a clear kill fluid); have two sets of perforations if you cannot establish circulation through the B annulus. Only after you clean up place your neat G slurry fairly slowly. If gas migration was the problem in the first place you may want to use finer cement and a version of hesitation squeeze to keep the slurry moving while setting.
  • You can also use thermally activated polymers or solutions, but these exotic solutions are most suited for microannuli than standard mud channels, for which Portland cement is good enough. Again, if you are going to squeeze a microannulus, you'd better make sure it's really necessary and steel yourself for a long and uncertain journey.

Finally you ask about the length of a cement barrier. You have two options, both of which require you to re-log:

• You can follow standard industry rules-of-thumb, such as those in SPE 12141 or in reference books such as Stolen's. Those require a vertical distance with BI above a certain threshold. The vertical interval is in the range of 1.5-5 m depending on casing size.
• Or you can use NORSOK D-010: casing cement is the well barrier element 22 (p.178), and requires around 30 m verified by logs. The requirements are a bit more complex, full of ifs and buts, but that's the reference vertical length. Mind that "verified by bonding logs" is left fairly vague: qualified personnel should do it, but there is no hard and fast value of a single curve.

To my knowledge there is no sound engineering basis for a minimum length of annular cement for pressure isolation: if cement is really competent and well bonded, one meter may be all you need. Most length requirements are there to deal with uncertainties in measurements and environment.

At the end of the day if you have SCP, and it is unacceptable, then you have to keep repairing until you get rid of it.

Good luck,

Matteo

All,

Well is completed with surface casing set at ~ 4,000', production casing set at ~ 12,000' & 7" Liner is set at ~ 13,200' in high pressure reservoir of ~ 6,500 psi & ~ 350 degF.

CBL/VDL/USIT logs were run for production casing & Liner section as well. All logs suggest no cement or very bad cement behind both casings. USIT log indicates macro channeling. 

From well integrity point of view, what corrective measures should be taken to mitigate the integrity issues while ascertaining the cement sheath bond. (Cement remedial job)?? If cement squeeze job is unsuccessful, what other materials can be used to avoid macro channeling?  

From industry best practices, What is the minimum thickness of cement patch is required behind production casing in order to avoid surface casing pressurized??

Thanks for sharing your opinions and ideas.

We had perforated  two intervals of 10' each in production casing with 4-1/2" 12 spf HSD gun. Interval was selected behind which USIT/CBL/VDL logs shows gas filled macro channeling.

Set highly viscous gel plug of ~ 500' below bottom perforations. Set packer between two perforations & performed injectivity test with brine. Observed no injectivity & circulation even pumping pressure was kept at ~2,000 psig. After which suicide cement squeeze job was performed - found no success even pumping pressure was kept ~ 2,500 psig.

During pumping operation, observed no returns at surface and no pressure in B annulus (production x surface casing) ( valve was kept open).

Observed B-annulus pressure build up when valve is closed. Fluid analysis suggest hydrocarbons. 

What other remedial job can be done to cater the integrity issue??

Regards,

Lajpat Rai

Lajpat,

This is the problem with gas-filled channels on ultrasonic logs: almost always they are thin microannuli, and you definitely cannot squeeze microannuli with cement. It should be easy to tell them apart from logs.

Then again, if you cannot inject water/brine, then you cannot squeeze the temperature-activated clear fluids that are better candidates for the job. I suspect your perforations are too far apart. You can follow Orville's suggestions and space them at 100'-200'; I think a bit of mechanical modelling can tell you how far (and how fast) you can pump for a given pressure: just remember that the highest the pressure at the lower perforations, the more you'll tend to balloon the casing and close the microannulus, so not all the pumping pressure will go into pushing the circulating fluid - and you'd better put the packer as close as possible to the perforations.

The issue is that you're not only dealing with casing ballooning vs. friction pressure drop. You also have to pump at a higher pressure than the migrating fluid and you have to avoid fracturing the formation and losing your repair fluid.

So in my opinion you should try and understand what's going on:

• What is the leakage pathway? That is (at least partly) a microannulus, but what is the origin, and how does it evolve going up? This is important: if it catches up with a chimney you can always try squeezing the upper defect and leave the microannulus alone...
• What is your caprock? Is it creeping? If so then you can split your problem: you can mitigate the leak (even far from the source) and the caprock will repair the microannulus in due course.
• Can you manage SCP by producing? If this well is related to your previous posts about a lot of CO2 and carbon steel casing, then you cannot allow a leak.
• By the way: if you have a lot of CO2 in the gas the situation may not be so bad: water will be produced by cement, slowing the leak, and under some conditions the pathway may even heal.

If you have SCP (which you seem to do) then you should by all means do a couple of build-up and production tests. Once you assess your top of cement behind the production casing with logs, do a constant pressure production tests measuring the multiphase rate/volume (as simple as counting your gas and allowing fluids to accumulate) and then let the pressure build up to equilibrium. That will allow you to identify the source formation and its pressure, characterise the leakage pathway, and have ideas about effects of creep and multiphase flow.

At this point, once you know what's going on, you may even be able to use simple mechanical methods to control the leak.

The alternative is the tried and tested "shoot first, ask questions later" approach: get hold of some hardening clear fluid, perforate 100' above the bottom one (or 30', or 50'...), establish circulation, pump the fluid, wait and then repeat until it's fixed. If you're lucky, you're looking at 2-3 attempts maximum. Or you can mill the casing & cement and repeat the primary cementing with a patch (or set a plug you'll drill and patch later)... The problem with these blind approaches, except for the non negligible chance that they will fail, is that if the microannulus appeared once it may appear again: if your casing pressure (installing a pump) or temperature (circulating cold fluids) drops, or your formation pressure increases (injection or stimulation), you may restart the leak.

Best regards,

Thanks for sharing your opinions and ideas.

We had perforated  two intervals of 10' each in production casing with 4-1/2" 12 spf HSD gun. Interval was selected behind which USIT/CBL/VDL logs shows gas filled macro channeling.

Set highly viscous gel plug of ~ 500' below bottom perforations. Set packer between two perforations & performed injectivity test with brine. Observed no injectivity & circulation even pumping pressure was kept at ~2,000 psig. After which suicide cement squeeze job was performed - found no success even pumping pressure was kept ~ 2,500 psig.

During pumping operation, observed no returns at surface and no pressure in B annulus (production x surface casing) ( valve was kept open).

Observed B-annulus pressure build up when valve is closed. Fluid analysis suggest hydrocarbons. 

What other remedial job can be done to cater the integrity issue??

Regards,

Lajpat Rai

Thanks for sharing your opinions and ideas.

We had perforated  two intervals of 10' each in production casing with 4-1/2" 12 spf HSD gun. Interval was selected behind which USIT/CBL/VDL logs shows gas filled macro channeling.

Set highly viscous gel plug of ~ 500' below bottom perforations. Set packer between two perforations & performed injectivity test with brine. Observed no injectivity & circulation even pumping pressure was kept at ~2,000 psig. After which suicide cement squeeze job was performed - found no success even pumping pressure was kept ~ 2,500 psig.

During pumping operation, observed no returns at surface and no pressure in B annulus (production x surface casing) ( valve was kept open).

Observed B-annulus pressure build up when valve is closed. Fluid analysis suggest hydrocarbons. 

What other remedial job can be done to cater the integrity issue??

Regards,

Lajpat Rai