Well Integrity Technical Section

This message has been cross posted to the following Discussions: Reservoir and Well Integrity .
-------------------------------------------

Dear colleagues,

Has anyone information or experience concerning effective permeability of installed well cement sheath?

We try to model potential vertical leakage of a gas well and are searching for reasonable input data.

What values can be assumed for good and defected cement? We do run regularly CBL- or USIT-surveys in our wells. Has anyone knowledge of correlations between the results of these logs and cement permeability?

Another question addresses the skin factor that is established by the mud-cake of a bentonite drilling fluid against permeable formations. We found in literature a figure of 60. Is this reasonable?

We are well aware, that it is very difficult to get reliable figures on these subjects. Nevertheless, every information would be helpful.

------------------------------
Josef Hoellwart
Head Subsurface Technology Reservoirs
E.ON Gas Storage
Essen
------------------------------

Are you experiencing an aquifer with total loss zone characteristics ?

Regards,

Othman Alyafei


Sent from my iPhone

------Original Message------

This message has been cross posted to the following Discussions: Reservoir and Well Integrity .
-------------------------------------------

Dear colleagues,

Has anyone information or experience concerning effective permeability of installed well cement sheath?

We try to model potential vertical leakage of a gas well and are searching for reasonable input data.

What values can be assumed for good and defected cement? We do run regularly CBL- or USIT-surveys in our wells. Has anyone knowledge of correlations between the results of these logs and cement permeability?

Another question addresses the skin factor that is established by the mud-cake of a bentonite drilling fluid against permeable formations. We found in literature a figure of 60. Is this reasonable?

We are well aware, that it is very difficult to get reliable figures on these subjects. Nevertheless, every information would be helpful.

------------------------------
Josef Hoellwart
Head Subsurface Technology Reservoirs
E.ON Gas Storage
Essen
------------------------------

Josef, let me attempt two answers to your questions. First the search for cement (equivalent) permeability has been conducted for some time by a few researchers in carbon storage, you can check papers by Sarah Gasda and Michael Celia.
However the basic problem is whether cement leakage pathways can be modelled through equivalent permeability, i.e. a linear relation between pressure drop and flow velocity.
There a reasonably limited number of possible defects to a cement sheath, and the most common nowadays is a microannulus. This is a very thin, thus elastic, gap at the interface between cement and casing or (less often) between cement and formation. Since elasticity is highly coupled to fluid flow, just like in natural fractures, velocity is proportional to the cube of pressure drop. So if you use equivalent permeability you need to have a pressure-dependent permeability, which is not ideal. For this reason it's probably better not to use the permeability analog to study flow in microannuli.
Now to the logs: in general they tell you something about the cement - its elastic properties and bonding - but not much directly about the shape of the pathways through it. And very little about their connectivity. To characterise leakage pathways and therefore get their hydraulic resistance (their equivalent permeability if you wish) you need to identify and measure them. This can be done reasonably effectively with the USIT, especially when coupled with the CBL, but unfortunately there is no direct simple measurement you can use, but rather an inversion method using all available measures.

If you're interested we can discuss this topic further offline.

Good luck,

------------------------------
Matteo Loizzo
Wellbore integrity consultant
Berlin
------------------------------

Josef,

I work with a number of Operators that have sustained casing pressure issues.  Most of them have 'decent' cement jobs but these have some micro-annuli that allows fluids/pressure to migrate.  Rather than use Linear Darcy permeability, the best 'mental model' I have found to explain these flow paths and migration is the concept of  'capillary flow'.  Capillary flow can exist when the flow path is very narrow and the flow path length is many times longer than the width.  This describes a microannulus/small channel pretty well with slow laminar flow rates.  Even very small flow paths, a few microns in width, will allow fluid to pass slowly and pressure build up over time. See wikipedia  Hagen Poiseuille Capillary Flow for a general explanation.

Once you have this mental model then the 'way forward' is pretty clear.  Without going into details, I  assess each well on 'capillary flow path' and other criteria.   If the flow rate is high and the channel is big then remediation may be a possibility.  If the flow rate is low, and the micro-annulus is small then remediation is much (!) more difficult. 

To prevent (?) flow in the cement annulus you have to either eliminate the pressure source or seal the annulus completely. In a normal well isolation of the source might require a new well redesign (cement slurry mix, stage cementing, liner) or remedial squeeze to isolate the pressure source from the annulus.  This might be an option for your gas storage projects.  Alternatively is a change in new well designs to improve the isolation mechanics between the surface and pressure source.  This may mean a change in well design (e.g. liner with mechanical sealing packer) and/or the use of multiple swell packers or similar technology to provide an alternative seal that is not dependent on cement quality.  

With a mental model of "Capillary flow" the Operator's focus can change to risk assessment and risk management.  We develop monitoring plans to assess if the flow rate/path is worsening and we calculate some maximum flow rates that could occur.  If remediation is appropriate we look at remediation methods and probability of success.

I hope this gives you some ideas on another methodology you might be able to assess your well issues.  Let me know if you have any questions.

PS: Drilling mud can provide a high skin but the skin is very pressure/rate dependent.  Mud 'lift off' pressure testing shows the clean up of mud in a few places at first and over a larger area as flow rate increases.  Lift off pressures are typically very low. So I would expect a 'high' initial skin but then a zone cleaning up over time...which is what we see in most wells.  You can ask your drilling mud company for 'lift off' pressures testing on your mud or similar mud formulations. 

I have found a Radial Darcy 'high skin' as a result of micro-annulus restrictions that are limiting the rate of flow which, I believe, have nothing to do with mud/filter cake.  Besifes cement micro-annulus above another explanation is barite settlement due to mud degradation.  You can test this to see how pressure/rate dependent your annulus flow rate by comparing various bleed off events over time.

------------------------------
Dan Gibson
TheWellDoctor
Completion / Well Integrity Advisor
Gibson Petroleum Consultants
FulshearTX
------------------------------