Deposits in pipes 4

[SFTB]

Hi,
I have a deposit problem in a 1 inch pipe. The actual velocity is 1 m/s (3.2 ft/s). I would like to increase the velocity to maintain conditions where it would be impossible to have deposits. I would be easy for me to go up to 16 m/s (52 ft/s) with minimum modifications. How can I choose a sufficient velocity? Is there a "rule-of-thumb" that prescribes a velocity value for self-cleaning applications?
If so do you have a specific reference (Machinery handbook,etc.)?

Thanks,

 

[aspearin1]

In my chemical engineering text book, they refer to water velocities above 1 m/s as being used to reduce fouling inside heat exchangers. The book is Transport Properties and Unit Operations by Christie Geankopolis.
Like many undesireables, reduction is the best you can hope for. Complete elimination is not usually feasible. You'll need to balance the costs of running your pumps at higher speeds as well as the pressure stresses you may be placing on your system against the benefit of reduced fouling. You may find the right balance that gives you the best cost savings. But you'll probably have to do the math.

ChemE, M.E. EIT
"The only constant in life is change." -Bruce Lee

 

[makeup]

With any kind of laminar flow process you will always have the chance of deposit formation. No-one is exempt, hydraulic flushing of the system should be a regular maintenance procedure. If you need any further help with this let me know.

 

[CRG]

Increasing velocity may create other problems. Corrosion and erosion may accelerate with the higher velocities. The pipe may be protected from corrosion by a passivating film. Higher velocity may erode the passivating film resulting in accelerated corrosion. Also, damage may result from the erosion alone created by the higher velocity.
What is the pipe material, the fluid type, temperature, and lastly if this is a heat exchanger what is fluid film temperature. With this information, someone might be able to point you to a resource.

 

[friartuck]

It would help if you told us the process and what materials are being deposited.

If its simply scale, you might want to try a 'conditioner device. These are used on water pipework and believe it or not 'beer delivery pipes'. They work by producing a modulating electromagnetic wave in the fluid under flow. A signal is induced in the fluid and changes the characteristics of the deposits. In the case of water, the normal calcium deposits are alterd to argonite which cannot easily attach themselves to the walls of the pipe. The debris is simply flushed away and becomes a sludge i the hot water cylinder etc.

Success rates appear to be good, but whether this will work with your scenario is any bodies guess.

manufacturers are 'Waveform', Liff, etc. and they can be expensive.

Drapes

 

[SFTB]

Thanks for all your replies,

Here are my fluid characteristics:

Temperature : 20 to 30°C (68 to 86°F)
Vicosity : (similar to water) 1.12E-3 N.s/m2 (2.34E-5 lb.s/ft2)
Density : (also similar to water) 999kg/m3 (62.4 lb/ft3)

The flow is in a SS316 pipe and the solid is very soft, not abrasive **Hardness 3 HM on Mohs scale**

The solid in my liquid is calcite (2g/L) and it tends to deposit and form a thin coating everywhere because it is very sticky. Don't worry about all the calculations...(pump selection, pressure loss, pipe diameter) I'm OK with that. What I need to know is : what would be my target velocity to be in a "self-cleaning" mode.

For example, would 4 m/s (13 ft/s) would be enough. Is there a simple rule of thumb like "self cleaning properties are achieved when velocity over X m/s are found in a pipe".

 

[SFTB]

I guess I' m looking for an answer similar to the one aspearin1 gave me (2nd post)earlier. The thing is I know that 1 m/s isn't enough in my situation! (This fluid is not in a heat exchanger, my case is worst than the traditional amount of particuls you could have in a HX liquid.)

 

[Zapster]

Why are you concerned with the minimum velocity required to ensure that no deposits will form? Why would the deposits create a problem if this is not a heat exchanger? Just pump the volume of fluid required for your application with a positive displacement pump. If deposits form, the velocity will increase until you reach the minimum velocity where no additional deposits will form. You will know the opening vs. velocity is in equilibrium when the pressure to maintain flow levels off.

 

[SFTB]

Zapster,

I agree with you BUT, with a positive displacement pump, the pressure could increase until something bursts or until the pressure switch cuts off the pump relay to prevent mechanical damage. I guess my best shot would be to design the system with a nominal velocity that prevents deposits...

 

[saxon]

SFTB, For sizing lines I have used the Reynold's Number to set pipe size to keep materials from settling out. As in any Fluid dynamics text, anything below 2000 is in the laminar flow regime, 2000-4000 the transitional flow regime, >4000 turbulent flow regime. The objective is to get the flow velocities into the transitional/turbulent flow regime by adjusting pipe diameter.

You muat be aware, that in some cases, there are materials that will plate out on surfaces over time due to the charge polarity of the piping and the process medium. I do believe that calcite is one such material.

Hope this helps.
saxon

 

[SFTB]

Thanks for your useful tips Saxon. I would like to know where could I check if calcite and my pipe material have electrostatic affinity ?

For the velocity, I will make sure that my Reynolds number is turbulent.

 

[crogan]

Coming from a mineral processing background, we generally have a rough rule of thumb of over 2 m/s line velocity for settling slurries. If there is a lot of coarse and/or low density (ie less hindered settling) and we know there will be a problem with sanding, then 4-5 m/s is typical. There is rarely a need for over 6 m/s as this just produces excess fricion losses and also increases pipe wear.

Generally any reasonably competent slurry pump manufacturer (Metso, Warman, Krebs, GIW) can also provide assistance with slurry transport and generally have their own calculation methods.

If you want to go through some calcs yourself, an Australian mineral processing guru, David Menne, once had a website going (i can no longer find it though). It had some formulae for such things. Luckily I saved it a few years ago and quoting directly from it:

---------------------------------------------------------

2. Solids volume fraction

Solids volume fraction is a most important slurry parameter; and many inferences can be made from values as compared to the so- called maximum solids volume fraction Cvmax.

Cvmax is generally meant to refer to the loosely packed condition, eg after pouring.

By tapping, vibrating etc, a higher value can be obtained, about 1.18x the loosely packed value.

3. Slope of a launder to ensure that it is self-scouring :

tan Th > 0.7 Fv [S-1]0.1 Dp0.2

where Th = launder slope
Fv = volume fraction solids Cv for Cv <0.34
= 0.34 for volume fraction solids Cv > 0.34
S = specific gravity of fastest settling solids
Dp = diameter of fastest settling solids, mm

Setting Fv = 0.34 gives a good conservative slope for rapidly hosing clean a spill surface or potentially boggable launder or pipe etc.

Using lower slopes, hosing time and water consumption increase almost in the ratio of cot Th.

4. Self-scouring throughflow velocity for horizontal pipe :

V > 3.3(1-Cv)0.13[gDp(S-1)]0.5

where g = 9810 mm/second2; g = 9.81 m/second2

 

[vpl]

I don't have my Crane (Technical Paper 410 - Flow of Fluids)Manual handy, but I seem to remember something that the fluid flow should be limited to around 10 ft/sec (3 m/sec) for water in pipe and that, generally, anything above 3 ft/sec (1 ft/sec)would keep silting deposition to a minimum. I don't think you can ever make it so it's impossible to have deposits.

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[d23]

SFTB

You may want to design from a recommended Q max. API RP 14E calculates erosional velocity to be:

Ve = C/(fluid density in Feet^3)^0.5.

Where:
Ve = Fluid Erosional Velocity in Feet per Second
C = Empirical Constant = 100
Fluid Density = Gas / Liquid Density at flowing pressure and temperature.

From experience I believe this calculation to be very conservative, but I do not deal with a lot of fittings like 90 degree elbows etc. I think you will find that erosional flow is in the 12 to 14 FPS range. At this velocity you will be in a turbulent flow, but should avoid erosional velocity.

Just remember in a turbulent flow if there are any existing system problems you will accelerate the failure rate.

The university of Berkeley posted a reasonably good paper about turbulent flow. You can find it at:

http://dnclab.berkeley.edu/~mschmit/html/E190_Final.pdf

Good Luck

 

[bulkhandling]

I made a calc using the Durand formula, for 1 inch pipe, 1.5 m/s should be more than enough for the normal calcite particles. You normally should not go beyond 2 m/s for 1 inch pipe because of abrasion, corossion, excessive friction and vibrating problem.
The problem is , if the solids is too sticky, it may tend to scale onto the pipe wall (like lime as I know). In that case, you may have to clean the pipe periodically. The method can be acid solution purge or by consulting the pipe cleaning companies.