Private Forums
Inflow testing with a shallow packer
24 March 2020

Colleagues,

I need help in either convincing my colleagues that I am right, or help understanding why I am wrong.  In question is what happens below a packer when doing an inflow test when the hydrostatic in the annulus below the packer exceeds the hydrostatic inside the drillstring.  Consider the example below (and shown in the attachment):

1.       The well contains 12.0ppg mud.

2.       Open-ended drillpipe is run to 2000mRT, with a packer high up in the string at 200mRT.

3.       The drillpipe is displaced to 2000m with 8.5ppg seawater with a surface U-tube pressure of 1194psi.

4.       The packer is set.

5.       The U-tube pressure on top the drillpipe is bled off and the drillpipe left open to the atmosphere.

6.       What happens next?

I maintain that the mud in the annulus below the packer will backflow up the drillpipe until equalized, leaving a void below the packer with a pressure of zero (gauge) below the packer, and on top of the mud. 

My colleagues, who are experienced engineers, maintain that the mud on the backside would remain stationary since there is nothing to fill the void below the packer that would be created if the mud U-tubed away.  The pressure below the packer would have to be negative, or I suppose zero (but without a void create by equalization).  Neither of those outcomes seem possible to me.

Their argument is that this is the way inflow tests have always been done.  But I maintain that this could only be done (without backflow) if the packer is set deep enough so that the hydrostatic in the annulus below the packer is less than the hydrostatic inside the drillpipe.

Any views? 

Also, could anyone help me to understand why the pressure inside the void below the packer would only go down to zero (gauge), and not create a vacuum (of negative gauge pressure)? 

Thanks for your help.

Paul Boudreau

paulboudreau@yahoo.com


Documents uploaded by user:
Inflow Test with Shallow Packer.pdf
13 answer(s)
Paul Boudreau
Deepwater SDE
Posco Daewoo
Total Posts: 3
Join Date: 11/02/16

Gents,

OK, sorry for not responding sooner, my day job kept getting in the way.  (But boy am I glad I have it!)

Anyway, thanks for everyone’s input.  I have attached my (the) solution to this particular scenario, which I am confident is correct.  I should have posited the question more directly as:  “Please confirm that, if a communication path exists or is created between the annulus and an inner string, the external and internal hydrostatic pressures at the point of communication will equalize once allowed to do so.”  The inflow test and packer scenario were provided as a practical (or in this case, an impractical) example.

This all seems pretty elementary, but to my surprise it generated quite a bit of debate in the office.  A couple of things that strengthen my view:

It is hard to find relevant examples in drilling texts.  But in “Well Control for Completions and Interventions” by Howard Crumpton (apparently a Scotsman, with perhaps more credibility than a Sepo like me), in his section on “Opening the Circulation Path”, he comments on p.248:

·        There is a risk of the wireline tools being “blown up the hole” if that differential is from annulus to tubing.

·        When punching a hole in the tubing, there is a risk of the wireline tools being “blown up the hole” if there is too great a pressure differential from annulus to tubing.

He also works an example very similar to one I present (“Calculate the fluid level (H) in the annulus if the tubing head pressure is bled to 0psi”) on p.258. 

My own experience was when testing a gas well with a packer and seal assembly, the string contracted to the point that the seals pulled out of the packer.  Of course, the mud in the annulus U-tubed up the tubing string immediately.

Anyway, to address some of the specific comments you have made:

Apologies to the brainier amongst you.  I have not considered the effects of temperature or compressibility, nor ballooning or reverse ballooning.  All real and appropriate for consideration, but not really critical in understanding this simple concept.

Simon, I am guessing that you make the simplifying assumption that the annulus hydrostatic below the packer at the point of communication with the inner string is less than the internal hydrostatic (thereby resulting in a positive pressure acting on the bottom of the packer).  That, I fully agree, is the norm and is the correct way to do inflow tests with a packer.  But that misses the premise of my question.

Allan, the equalization will occur even without breaking the string on surface.  But if you break the string and open up the heavy side to the atmosphere, it will happen more dramatically (read quickly).  [Will bring up whether to use buoyed or air weight in my next post.  ;-) ]

Wayne, my old friend, let’s not complicate it.  I’m trying to focus on “first principles”.

Chris, the packer was being considered to prevent putting a negative pressure (top to bottom) across the BOP rams/annular.  Since then, that plan has been abandoned. 

Justin, on that point, I’m not sure if a void would see 0psig or 0psia.  I’m guessing closer to 0psig, as presumably there’d be enough entrained gas/air to come out of solution to occupy the void. 

Thanks again to all commenters.

Paul

Documents uploaded by user:
U-tube issue 2.pdf
Toriati.pw
Well Control Instructor and HPHT/MPD Coach
SPREAD Associates
Total Posts: 1
Join Date: 31/08/13
Paul, - Paul Howlett is correct. You have not conducted an inflow test to a draw-down of 1194psi. You have to include the set packer in your calculations. It looks like Pete Aird has done the calcs for you. The main reason for setting the packer deeper would be to achieve the required draw down. Allowing a drop in the annulus would make it very difficult to spot a leaking liner lap or shoe. And you are right to question this procedure.

cmcandrew
SPREAD Associates
Total Posts: 37
Join Date: 16/08/10
Peter, fine until:
"The annular hydrostatic pressure change at the end of the bleed down that resulted is a change of -1193psi at 2000m and -119.3psi at 200m and 0psi change at surface. "
Leave out hydrostatic from the above , (and also delete bit about -119.3 psi at 200m - it is above packer.). When set packer , existing hydrostatic is retained and isolated.
There is only one BHP.  Hydrostatic of a column of mud is still the same ( if totally incompressible ):  therefore only way to balance pressure down drill pipe and in annulus is that annulus drops/utubes, leaving a void or vacuum above: as in the diagram on right.
IwanGunawan
Sir Drilling Specialist
KUFPEC
Total Posts: 7
Join Date: 27/03/17
We usually do this inflow test using a packer, circulating valve and a tester valve. Circulating valve and ball valve used to spot lighter fluid inside DP/tubing, after packer is set at predetermined depth (we suspect the leak is lower than packer setting depth. Thus when the test/ball valve opened, observing tubing pressure ( no worries on annulus as it's still on kill fluid ) If tubing is coming in, then the well is alive with a leak below packer,somewhere. This inflow test is typically done on hpht wells where you want to do an underbalance annulus DST, to make sure that casing holds differential pressure outside in. Trying to displace to lighter fluid without packer is very risky..over displace and you may end in a big mess, like in Macondo. In my opinion, Macondo is a failed, uncontrolled inflow test. You bring the dragon out, unprepared.
Companyrep
Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 441
Join Date: 10/01/05
-1193psi at 2000m in annuls not -119.3psi at 200m after bled down with packer set.
Companyrep
Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 441
Join Date: 10/01/05
Paul,

Nice one, got me thinking.

Experiential norm answer is that I know this, i.e. I expect a barrel or so back when I bleed down the pressure and that's it.

To explain why?, I first went back to refresh my fundamental understanding of pressure. Source for this was (Schlumberger Completions 'Hydraulics calculations handbook) a great technical aide by the way. 

Pressure?
1.  Applied (no pressure is applied from surface in this case)
2.  Hydrostatic (fluid inside outside and below.)
12ppg annulus (4261psi) 
8.5ppg string (2897psi)
3. Differential  ( drillstring annulus 1193psi)

So this is all about hydrostatic and differential pressure.  
Plus factor so
- barriers  
the drillstring surface barrier after displacement to retain a differential
packer provided barrier to flow to bled down and inflow test (where the differential pressure after displacement is assumed to be applied)
- wells ability to flow or not (is also central and key) 

The evident process and pressure changes now analysed as follows. 

String is displaced where the hydrostatic pressure in the string changes from surface to 2000m. (Atm-4261psi to  Atm-2897psi) 
Annular hydrostatic have not changed. (4261psi at 2000m)
A differential pressure of 1193psi now exists and is now applied throughout the well. (annulus from 2000m to surface has been hydrostatically reduced by -1193psi at 200m to 0psi at surface)

If we were to open the drillstring barrier? the 1193psi differential pressure would be lost and annular level would drop to hydrostatically balance drillstring and annulus. 

Well bottom would also loose the 1193psi differential pressure applied. Mud level would drop 360m in annulus. Pressure at 2000m now  3494psi (767psi less)

We however maintaining the 1193 differential pressure, we assume we close the annulus (set the packer) and then bleed down.

The only flow capable to result as we bleed down is the compressional effect off the 1193psi differential pressure as applied to the well system i.e. a barrel or so depending on PVT of wells system)

At the end of the bleed down the total wells system at the top of the string to the bottom of the string and to wells TVD  is now hydrostatically underbalanced by (1193psi - compressional volume of mud that was bled back into the string).
The annular hydrostatic pressure change at the end of the bleed down that resulted is a change of -1193psi at 2000m and -119.3psi at 200m and 0psi change at surface.  

And why the wells annulus does not u tube and why annular fluid does not flow into the well. 

Should we immediately open the packer at this point the annular column of mud would U tube and flow to hydrostatically balance the well.   


Both the string and annulus would equalise and change pressure to equal each other. Hydrostatic pressure would be 3494psi. Annular level would drop by 360m.

Is my evident reckoning of this. 

If I am mistaken, I am sure someone can guide me to correct my errors. 


cmcandrew
SPREAD Associates
Total Posts: 37
Join Date: 16/08/10
Yes, you are right, ( as on diagram on right)- for theory, can  keep it simple and treat as incompressible fluids.
I will just add, that apart from adding complexity of this backflow, the main reason to set test packer deep enough is for safety, on circulating the influx out above the packer. 
If given choice between either your shallow packer "plan" or just circulate entire hole to lower MW without any packer,( and watching pits) I would choose the latter !
simonleiper
Product Champion WBCU
Reactive Downhole Tools Ltd
Total Posts: 13
Join Date: 17/04/09

Hi Paul,

My masters thesis last summer was on the behavior of fluids during inflow tests and devising a compter based model to predict the flow rates as a result of this.  Without writing another 10,000 words on the subject with citations and reference, here is my simplest explanation…..

In an ideal world without the effects of thermal expansion of fluids, compressibility of fluids and ballooning effect on drill pipe, the pressure would bleed to zero and there would be no flow (as per your left diagram).

 

Considering ballooning, there will be a dP across the string down to the water / mud interface.  This will cause the pipe to squeeze and resulting in a small flow.  My investigations found this to be minimal.

 

Now consider compressibility of the fluid.  In this case water (and I assume WBM) is generally not compressible so will result in minimal flow.  In the case of base oil and oil based mud the effects is more exaggerated as the base oil component is much more compressible, so you have to determine the compressibility of the base oil and also the compressibility of the oil, water and solid phase of the mud (generally assuming the solid phase to be incompressible except under ultra-HPHT conditions).

 

Next consider the expansion of the fluids due to temperature and also compressibility as well.  As we go deeper in the well pressure and temperature increases.  Temperature causes thermal expansion (also reducing density) while the increase pressure works oppositely compressing the fluid and increasing density.  These effects are far more profound in oil based fluids, and it is known that temperature is the dominant effect.

 

The next step is where it gets interesting (yawn)….. before performing the inflow test it is common the circulate the well to get a homogenous mud, which tends to cause the temperature gradient of the fluid in the well to be fairly consistent from top to bottom.  As soon as we stop pumping, the well begins to move back toward the geothermal gradient.  As a rule of thumb, the top 1/3 of the annulus will cool and the bottom 2/3 will heat and if plotted will form a straight line with a pivot point.  As we commence pumping the cold, dense underbalancing fluid, we are displacing hotter fluid up the annulus.  The cold dense fluid will try to gain temperature equilibrium drawing heat through the pipe and in turn from the annulus.  Initially the top 1/3 will be colder than the annulus, and begin to heat.  Once at equilibrium with the annulus it will cool together to geothermal gradient (faster above the mud line).  The bottom 2/3 will heat until equilibrium is achieved.

 

It is this heating and cooling and the effects of expansion and contraction that we witness at surface when we see flow during the inflow test.  This tends to manifest itself as an ebbing and flowing again far more profound when using OBM.  It is also predictable if you have good data to work with.

I attached some slides on my project.

Thanks

learning.lifewayne@g
D&C Project Coordinator / Decommissioning
Society of Petroleum Engineers
Total Posts: 32
Join Date: 25/01/16
Interesting dynamic Paul.  Best to track you volumes through each sequence.  When you run the tubing what is the pit gain at surface and when on bottom does the well sit stable before setting the packer.   After displacing the seawater are the volume in and volume out equal and does the final displacement pressure and hydrostatic balance theoretically relative to volumes.

I assume you want to verify the pore pressure of perforations below, or is it open hole? Wouldn't an underbalance to the formation flow as soon as you set the packer.  Simultaneously mud will tend to drop out the annulus depending on there being no leaks across packer or with the casing.  The final effect will be the the water and producted fluid will migrate up the annulus and the mud will fall.  However if there is no production, flow out the tubing should stop.  Measure the flows and the vacuum under the packer should stabilize the annulus but don't discount light fluid migrating up the annulus although that should not be detectable as you monitor drill pipe pressures.  So what was your formation permeability and fluid content / pressure.  Eventually the annulus will swap out gravitation ally with the produced fluid.  Something like that but calculating volume and displacements should give some indication of what is happening where.
gibsona7
Head of Well Design and Completions
MOL Group
Total Posts: 9
Join Date: 06/11/11
I agree with you Paul H and Paul B answer is also good in my humble opinion. Another similar scenario to help convince your colleagues is slugging drillpipe. After you pump a slug the level drops in the drilpipe and a vacuum is created which you can feel when you break the drillpipe connection after pumping a slug. Good luck with your discussions and I strongly recommend you don't open a different subject - buoyancy with these colleagues or you'll be forever doomed :-)

PaulHowlett
CEO
Sudelac
Total Posts: 111
Join Date: 10/04/08
Paul, You are not doing a full 1194psi inflow test with a shallow packer, if that worked you could just close the BOPs and not bother running a packer at all. The hydrostatic in the annulus has not been refuced by much with that method, you bled off the u-tube back pressure and some mud will have moved into the drill pipe but only until the drill pipe pressure is gone and mud cannot move anymore. The hydrostatic pressure in the annulus is still mostly there, 0psi at 200m and 3600psi at 2000m, your effective inflow is only 120psi, the difference between the mud outside and water inside at 200m. If this is not true why would anyone run the packer deep? Why run a packer at all just close the BOP after displacing the drill pipe.
jhardy
Completion Engineer
CNR International
Total Posts: 2
Join Date: 13/05/13
Paul,

I am of the  same opinion as you. The pressure will drop to zero below the packer if the hydrostatic at the packer setting depth is less than the inflow test value. It will not drop below zero as this is impossible. Strictly speaking when working out the heights of fluids in this scenario air pressure should be taken into account (14.7psi) which is why the thumb over a straw trick works. This obviously looks like the opposite to what you are saying but it isn't it is just the hydrostatic in the string is less than 14.7psi. So if the string was over ~34ft long (with fresh water @ sea level) a void would be formed at the top of the straw.

There are may examples of similar situations in the industry e.g. cementing, how do people manage to cement a well with full string of cement in the drill pipe - they don't it is the annular hydrostatic that dictates the pressure while pumping. So a void is formed at the top of the pipe. (You would still see the hydrostatic of the cement from the cement unit to the highest point in the cement manifold as a pump pressure). 

The same is true when gravel packing in particular with high density gel packs, and people normally run gauges that will prove this. i.e. drill pipe hydrostatic always remains ~ constant despite filling the string with a fluid that might be 4ppg heavier than the original fluid.

Obviously the above is only true in low friction environments as if the friction exceeds the hydrostatic difference no void will be formed.

I too have also had this discussion with other people and had to agree to disagree, so would be interested to see if someone can show any physics that disproves your theory as i think it completely backs it up.
harshmind
Drilling Adviser
SPREAD Associates
Total Posts: 8
Join Date: 14/11/15
Hi Paul, I hope it's going well.  While I have neither much experience with inflow tests nor a complete understanding of fluid behaviour at reduced pressures, I am interested in your question.

Firstly, is the mud water-based?
If it is, the fluid is relatively incompressible and by the same logic does not really expand when subject to lower pressures.
This suggests the fluid below the packer, if it expands, will U-tube by only a negligible volume.
What happens immediately below the packer?
The pressure decreases, certainly, but until it reaches boiling point for the local temperature there will be minimal void to fill.

I would ask why the packer is being run so high in the test string, though.

Best regards,
Chris

Jump to top of the page