Q&A No.718 - Effect of Sample Type on Sand Retention Testing Results

0 votes
We are currently conduct Sand Retention Testing (SRT) for screen size selection.

In the absence of any core samples, we only have access to nearby analogous, cuttings and few side wall cores (SWC)

Relevant nearby analogous data (PSD and installed screens size) suggest 250um is sufficient for sand retention.

SRT on cutting samples indicates <200um screen is required (200um failed to retain with produced sand >0.12 lb/ft2).



My question is in regards to the effect of sample type on the SRT results.

While we are comparing PSD data of SWC and cutting samples, I am wondering if anyone had any experience with SRT on cuttings and SWC

What are the Pros/Cons of sampling types (Core vs Rotary/mechanical sidewall core vs Percussion sidewall core  vs Cuttings) ?

Any insights or experience would be most appreciated.


asked Dec 20, 2016 in COIN by Abdullah Eshtewi (240 points)

3 Answers

0 votes
We typically use conventional core samples. Would think cuttings PSD heavily biased on whatever sorting is going on in the mud system. Side wall cores should be okay...perhaps some damage from percussion samples and rotary best


Steve Hicking
answered Dec 21, 2016 by Steve Hicking (140 points)
0 votes

Hi Abdullah. 

I would be very cautious about using PSD from cuttings and also SWCs. It is no surprise that these are giving a finer PSD than analogues. In your case, I would go more by analogues, - particularly if you can show that the reservoir sands and depositional environment is similar to yours. 

When selecting sand screens and doing screen sizing, very much focus has traditionally been given to sand retention testing. Rule of thumbs have been used to determine critical sand production (like 0.12 lb/ft2) without focusing on what needs to be achieved. To make this simple, the screen should retain sand to prevent critical sand production to surface (or into the completion). At the same time, the screen must allow flow-back of mud and fines from the drilling phase. 

Critical sand rate varies form case to case, but in a normal oil producer, fairly large initial sand production is acceptable as long as the coarse fraction is retained. A simple conservative indication is to use the D10 value, but more aggressive testing indicates that D5 is more than sufficient to start retaining the sand and build up a natural sand pack. A larger slot opening may lead to higher initial sand production, but as soon as a natural sand pack starts building up on the sand screen, this sand production stops.
The other element which often tend to be forgotten is the minimum slot opening. 200um is in the low end and requires properly conditioned mud or a clean brine as completion fluid. Dissolving fluids should also be considered in this case. A larger screen size like 250 um will be more robust with respect to particle flow back. In the case of very high permeable formation or very high solids content in the mud, even a larger minimum slot opening could be considered. If the screen size is too small, a low permeable layer may build up on the screen resulting in a high differential pressure. This could result in hot spotting and screen erosion and is probably a more common reason for screen failure than too large screen sizing.
Best regards, - Lasse Hermansson and Terje Moen, Ridge AS
answered Dec 23, 2016 by Lasse Hermansson (180 points)
0 votes
Answer from Fraser Martin of Chevron - from 21/12/2016

Working with cuttings samples for sand retention work  will not give a representative result because the cuttings will be invaded by fluid loss material from the mud plus ground and fractured grains of formation. Even if the cutting make it to surface relatively intact this will still be the case. Taking a particle size distribution from the side wall cores will be significantly better than using cuttings. Whole core would be better than side wall cores.

Using cuttings to perform the PSD will almost certainly result in selecting a finer mesh size than you actually require. I am sure you are aware that the finer the mesh the greater the  risk of plugging and poor performance or hot spotting and early screen failure.

If you simply do not have good enough core to perform adequate PSD then your offset well data assuming it is for the same formations and relatively close might be a better bet. If you think that your PSD is smaller than reality because of the reasons mentioned above the you could choose to be quite aggressive with the mesh sizing and aim to capture only the largest 2 to 5% of the distribution. If the formation is actually fine you should still be able to build a natural pack on the surface of the screens though it may require a longer period of lower flow rates until this pack builds up. You will be producing solids for this period  so flux management will be key to preventing premature failure.
answered Jan 3, 2017 by Fiona Curley (1,480 points)
Response from Abdullah:

We are indeed moving towards using analogue data and ignoring the cutting PSD samples

My focus is now to make sure the key analogue we are using is actually a relevant and similar analogue

It is a developed field with plenty of data (core screen installed etc etc)

We will also be getting some cuttings samples from this analogue and conduct a SRT

Hopefully will demonstrate it is failing SRT contradicting the fact that we are not having any sand production issues in the field

Really appreciate your feedback