Vertical Slurry Pipelines 6

[ECD40]

We are planning to pump solids in a slurry form from a deep underground mine. The slurry particles will have a top size of 4.5 mm and a bottom size of -300 mesh. A laboratory test will be restricted to using a 2 inch diameter vertical pipe and centrifugal pump with a maximum lift of 50 feet. The fluid will be saturated brine and the solids will be a mineral that would dissolve in water, but not in the brine. A full scale plant would be using an 8" or 10" diameter pipe to convey the slurry vertically. The vertical lift height for the production plant will be 1,000 meters using positive displacement pumps for the high pressures involved. The specific gravity of the slurry will be 1.64.
Does anyone have any knowledge of scaling up from the laboratory to a full scale plant in a vertical lift hydraulic application? I've looked for empirical relations for scaling-up, but find none. As far as I can tell, this pumping application has not been done before.

 

[LittleInch]

As far as I can tell, this pumping application has not been done before.

I'm not surprised.

I can only start to imagine what happens when this pipe stops for any length of time.

But anyway one key parameter will be velocity - i.e. whatever you've got in your test set up needs to be similar in the full scale version

1000m vertical lift in 1.64 SG fluid in one go sounds like too much to me. I was wondering if a gas lift type of operation would be better, but the gas velocities will be huge when you get to the top.

I think I would be looking for lifts of 100m at a time so that I could drain down more easily when it stopped.

What sort of PD pump were you thinking of - progressive cavity?

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

 

[ECD40]

Hello LittleInch - thanks for your comments. My responses are below.
It has to be one 1,000 meter lift as the pipes will be in boreholes with no intermediate access.
The PD pumps will be of the reciprocating diaphragm design.
The slurry velocity will be above the particle settling velocity. About 9 feet/second is expected.
A pipeline draining system will be incorporated at the mine bottom, in the even of a power failure or pipeline plug.
Question - have you ever come across any empirical formulae for scaling-up the lab results to the production results?
Thanks for your interest in this problem and solution.
ECD40

 

[HTURKAK]

You did not define the type of ore , if not compressible and not having porosity , IMO also , vertical transport of solids in a slurry from deep underground mine is a viable concept . I looked the pumping of fresh concrete for the Burj Khalifa tower ( the highest structure up to 2010 ) The fresh concrete ( having 2.5 SG ) has been pumped to 600 meters high.

The flow and behavior of slurry varies with ( pipe size , SG and size of the particles, slurry density, pipe position sloped , vertical etc..) and the lab tests will give a limited idea and full scale pipe flow still could have substantial discrepancy.


I looked to the web and found the data sheet of the concrete pump used at that project ( Pressure 250 bars )

There are different suggestions for Vc to be proportional to D**0.25 , D **0.33 depending on the particle size etc.. .

The following paragraph is copy and paste from the book SLURRY SYSTEMS HANDBOOK (E. ABULNAGA )

I will suggest you to look DESIGN OF SLURRY TRANSPORT SYSTEMS ( by B.A. JACOBS ) and SLURRY SYSTEMS HANDBOOK (E. ABULNAGA )

This is not a common thread. When you decide upon your final solution , please share your solution with us for record and keep us involved.

 

[TiCl4]

ECD40,

A couple of things.

1. You have roughly 2,300 psi of lift to raise a 1.6 s.g. liquid 1,000 meters. Additionally, 8" pipe at 9 ft/s is roughly 1,300 gpm, depending on schedule of pipe (1,289 gpm for schd 80). That puts total hydraulic power at something like 1,700 HP, not including frictional losses or pump loss. Actual motor horsepower is likely to be 2,500+ HP. I don't think any diaphragm pumps in existence can do that. To be quite honest, I'm not sure what type of pump is even available for both high pressure and flow rate like that.

2. There isn't really a need for lab-to-full size scaling for simple pressure drop calculations - pressure drop can usually be calculated directly from the fluid properties. Refer to sections 5 and 6 in Perry's Chemical Engineering Handbook for typical fluid flow mechanics and pressure drop calculations.

Along these lines, when doing lab tests, make sure to measure the rheology of the slurry. This will greatly affect the pressure drop calculations, especially if the fluid is non-Newtonian (as most slurries tend to be).

3. You need to be very deliberate in your drain-out design. I'm pretty sure that if you have a power failure and the material is allowed to settle, it has a high likelihood of NOT draining when you open the drain valve. Many slurries will tend to pack out if allowed to settle and then become almost like a single mass. This is something a lab test may be able to answer.

On the other hand, if you open a drain line at the bottom of the drain from a liquid full pipe, you have 2,300+psi on the other side of that valve. In the event that it DOES drain, exit velocity will be enormous and dangerous.

This whole thing seems rather fraught with difficulties, dangers, and challenges. Have you or your colleagues ever seen anything similar in design in the mining industry? If not, you may want to ask yourself why that is.

Coming from a non-mining perspective, I'm with LittleInch on using multiple lift stations if you can. There are some fairly large hurdles to overcome with your current proposal.

 

[EdStainless]

There are many old papers that discuss long distance pumping of coal slurries.
This gives some general information on issues such as flow loss, stratification of particles, and plugging issues.
the only deep lift slurry work that I have ever seen was done with jet pumps.
Huge pumps are the surface that delivered a high-pressure fluid to the bottom of the bore and that was used to bring the slurry to the surface.
These were only about 250-300m lifts and at the limit of what they could do.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[TiCl4]

Thanks, Miningman, I hadn't looked to see OP's history with this. For my own edification, how are slurry lifts out of deep mines normally handled?

 

[miningman]

Well , in general we dont try to lift slurry , We attempt to dewater as much as we can which then gives relatively clean water to dispose of as well as a much smaller volume of high solids slurry , This can then be deliberately introduced into the run of mine ore or waste streams or it can be deliberately introduced into a mined void, known as a stope. Think about it , we are being paid to extract large volumes of rock.
This generates large empty voids. These are often backfilled with different types of fill material that can be co mingle with waste slurry.

I suspect the OP is trying to work with potash but that doesnt change some of the fundamental engineering or practical limits assoiated with underground work.

 

[MJCronin]

From what I have read, you have a situation at the extremes of slurry pumping engineering experience

IMHO, you should call in an industry consultant to evaluate this "cutting edge", unique and extreme problem.

I would not rely on the vague opinions and recollections of strangers ...

Pay the consultant's fee for an authoritative and comprehensive report ...

It will be money well spent ...

MJCronin
Sr. Process Engineer

 

[LittleInch]

Miningman, good memory - I looked back and remember some of the earlier threads.

ECD40 - I don't now how far you've got with this project but I agree with TiCL4 - the scaling is not the issue here, it's all about what this "slurry" look's like and how it behaves. Is it more like concrete or dirty water with a few stones in it? How slow can you pump it before it starts to separate? Does it create a smooth paste?

What happens in no flow situations?
What is the viscosity / shear thinning? Non-newtonian fluids do strange things.

1000m vertical is no joke and it seems from the discussion above you are running into some serious issues which cannot be ignored or glossed over which are fundamental to your project.

Scaling up creates issues in that things which don't happen in a 2" pipe could in an 8" or 10" pipe due to lower surface are per volume of the pipe wall. Also you're going to be lucky to get more than 20 or 30m vertical pipe in most lab settings.

But let us know how you get on for sure.

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

 

[ECD40]

Thank you all for some very good comments and observations. Let me add more information as a result of the above.
Firstly, this is a real mine that has reached its production capacity from conventional skip hoisting. Now, there is a requirement for a significant increase in mine production. A new shaft and hoisting plant will cost in the region of $1 billion and take about 4 years to construct. We are conducting a prefeasibility study on alternative means of increasing production at the lowest Capital and Operating cost possible.
We have decided that lab testing will be of little or no benefit at this time, because we have no good means of predicting the results of a 'scale-up' of capacity. Most likely (next year) we will go to a pilot plant using a 4" vertical pipe and 100 meters high. This will give better results than at a lab scale, but still leaves much uncertainty in a further 'scaling-up' to a production size plant.
The only case of vertical transportation of a slurry (ours will be 50% solids by mass) is a coal mine in Germany that operated for about 5 years and an uranium mine in northern Canada that has very high grade radioactive ore. Neither of these cases has sufficient relevance for us to use as a guide.
We have an evaporite type ore, which is water soluble, so we need to use saturated brine as the transport medium.
One of the biggest unknowns to us is how the solids will distribute themselves in the pipe cross section and what will be the relative velocities between the smallest and the largest particles in the pipeline. There is a term called 'herding', which is poorly understood and may have a significant influence on friction factors, pressure loss and possible pipeline plugging. 'Herding' does not occur in the horizontal transportation of slurry. Those issues may be quite different in a 2" diameter pipe compared to a 10" diameter pipe carrying the same distribution of particle sizes (PSD). The manufacturers of PD pumps assure that their pumps can meet the performance requirements, but our 'unknowns' will be beyond their control.
To summarize, this application will be a first of its kind, but it has a very high economic benefit. We have to recognize the risks and mitigate them before they happen.
I appreciate all of your comments and welcome hearing from you. Ask any question that you wish, but the mine name and location is confidential.
Thanks again,
ECD40

 

[miningman]

I had two posts here, one of which had earned three/ stars. I guess at least three posters found my opinions relevant. I have NEVER had a post deleted previously. I can make a real good guess as to who flagged my posts , simply because he did not like my content.

I am of course the only poster here who has operated underground at 1000 metres below surface and have 40 years experience doing this. I have to state that IMO, such behaviour is extremely unprofessional.

For those who are interested , the OP is working with potash. There are only two areas in Canada where potash is mined , both around 1000 metres below surface. Sussex in New Brunswick shut down a few years ago. All the other 1000 metre shafts are in Southern Saskatchewan.......and I spent 5 years at one of them. Lets see if the aggrieved poster somehow manages to get this post deleted...... its all factual!!!!

 

[LittleInch]

Miningman - I greatly respect your views on these issues and was one who gave you one of the LPSs and hence could see nothing wrong in your post.

ECD40 - if you re flagged MMs post then you appear rather thin skinned for this game - I saw nothing untoward in MMs honestly given opinion.

My points on your system remain unchanged - doing this in one single 1000m vertical segment is not really feasible IMHO.

one area you maybe could look at in a test or trial situation is what we do in oil wells and use gas lift. Now you could use air instead as its not flammable, but the volume increase over 1000m vertical is too high. 100m might be better and you may need to taper your pipe to allow enough of the slurry to move witht he expanding air bubble.

Other solutions I've seen include for stiff slurrys to inject a ring of water with multiple nozzles around the circumference to reduce frictional losses, but you key thing is still probably hydrostatic head and what I can understand about change of the slurry as you pump it vertically (herding)

Or have you considered a very long internal screw? the greeks knew a bit about that, but again you're almost certainly looking at staging as you go up

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

 

[HTURKAK]

I disagree with this approach . I have slurry line experience and IMO, lab tests is a MUST for slurry projects and the base data for the design.With lab tests,

- The optimum size ,
- The optimum density of the slurry,
- The interior lining ..
Can be decided.. I will strongly recommend the lab tests for the start point which the cost is nothing for the total cost of the project.

Consider my opinion only ..

This approach reminds me the proverb ; (For want of a nail the shoe was lost. For want of a shoe the horse was lost. For want of a horse the rider was lost. For want of a rider the message was lost. For want of a message the battle was lost. For want of a battle the kingdom was lost. And all for the want of a horseshoe nail. by B.Franklin or Tinghis Khan in my zone)

 

[1503-44]

Miningman, I tried, but I could only give you 1 star. There. I did it again.
Thanks for your opinion. I still remember it.

A black swan to a turkey is a white swan to the butcher.

 

[pierreick]

Hi,
You may want to consider this link and you may drop an email to the website owner.

http://www.svlele.com/piping/slurry_piping.htm

Good luck
Pierre

 

[TiCl4]

ECD,

First, I do agree with MM - the timeline of your posts didn’t give much confidence that this venture is moving anywhere any time soon. However, you say that it is, so here’s another’s thought below.

Other than the (huge) challenges mentioned already, and regarding “herding”: I assume you mean the various particle sizes will have varying rise velocities, and thus the larger particles may tend to accumulate, changing the local viscosity and other fluid properties due to the change in local PSD. I also understand that turbulent vortices generated at a given Re tend to have a specific length, so radial mixing from turbulence at a constant Re can become worse at higher pipe diameters. These two effects would combine to make scale-up difficult, as the game changes with each increase in pipe diameter.

Given your stated value of this project, I would go find the world’s expert in CFD slurry modeling and try to suss out some of these concerns in a rigorous model. When dealing with an investment this large, you don’t want to rely on scale-up experiments alone, and CFD is tremendously useful for studying and understanding multi-phase, non-Newtonian flow. It would certainly be more useful than asking a bunch if internet strangers.

Edit: One last thought. If viscosity starts being an issue for either pressure or mixing, you may want to consider grinding some of the particles to a smaller PSD so total PSD is bimodal. Bimodal PSDs with a ratio of ~3:1 (average particle size of the larger peak is approximately 3x the lower peak) tend to give much lower viscosities at the same %solids than a unimodal distribution.

 

[miningman]

Thanks for the vote of support guys. I have followed Piereick's advice and reached out to website's owner to see if the previous deletions can be reversed

 

[Artisi]

The pumping of a slurry
from 1000 metres underground with a single underground installed pump station is pie-in-the-sky. As others have said it will need to be a staged system, therefore my suggestion is to investigate a staged ejector slurry handling system, even this is method will be a major challenge in terms of operation and maintenance - although it will allow surface clean water pumps operating at much lower pressure to be used as the motive power source. For further discussion, it could even be air driven ejectors.

I would also suggest, a project of this nature will require a little more than a few posts coming from Eng-Tips members.

Good luck, but miracles do happen (sometimes).


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[pierreick]

Hi ,
Consider this handbook about pumping slurry.
Pierre