Dear ppquintal,

Theory:

In a bend in a pneumatic conveying system, there is complete separation of air and material (in this thread cement).

There is no interaction between the air and the cement.

The result of this, is a gas flow pressure drop over the bend, which is minimal.

Secondly, the cement is losing velocity, due to the friction against the outer wall.

After the bend, the cement is re-accelerated again at the cost of an extra acceleration pressure drop.

This extra, re-acceleration pressure drop is referred to as the “pressure drop of the bend”.

Thus, a material loses velocity in a bend and as a result, a pressure drop occurs in a pipe section after the bend.

The question of how much pressure drop is caused by the bend is reduced to the question:

How much velocity is lost in a short radius bend versus a long radius bend?

Mathematically:

At any moment, the particle-velocity and the covered distance as well as the covered angle can be calculated as follows:

For a horizontal bend:

= bend angle

vmaterial(n+1) = vmaterial(n)+a(n)*∂t

∆vmaterial = vmaterial(n+1) -vmaterial(n)=a(n)*∂t

a(n)=-[f(friction)*vmaterial^2/R]and∂t=∂L/vmaterial and∂L=R

gives: ∂t=(*R)/vmaterial

resulting in:

∆vmaterial=-[ffriction*vmaterial^2/R]*(*R)/vmaterial

∆vmaterial=-[ffriction ]**vmaterial

where

vmaterial is the entrance velocity.

The entrance velocity is determined by the previous pipe section, where the material flow is assumed fully developed (quasi stationary conditions).

Conclusion:

The reduction in velocity in a bend is irrespective of the radius.

In the approx. 20 years of experience in pneumatic cement conveying (building cement unloaders and maintaining cement unloaders), I encountered and used 95% 1.5D bends and 5% long radius (2.5D - 5D – 8D) bends and noticed no significant differences in behavior (Nor in rate or pressure drop, nor in wear)

Hopefully, this serves your question.

Have a nice day

Teus ■

Dear ppquintal,

The special elbows, which you mention, are designed to minimize or eliminate bend wear.

Velocity drop in the bend (= pressure drop after the bend) is not the issue for these bends.

The principle of these bends is that the material impacts on itself instead of the bend outer wall.

In this process, there will be more velocity lost (therefore some more pressure drop), compared to a smooth bend.

However, the difference in outgoing material velocity between a special elbow and a smooth elbow is taking place in the lower velocities.

As the resulting difference in pressure drop is proportional to velocity^2 (and the velocity is low), the resulting difference in pressure drop is also low.

Smooth elbows can be extended in life by fixing an outer box on the bend.

When the elbow is worn through, the material starts wearing at itself.

In the meantime, there is some bend material mixed with the conveyed product.

For food processing, the impact of bend wear and/or food retention in the elbow for a longer period of time on the quality of the product has to be considered.

Take care

Teus ■

I work at a high density polyethylene plant that specifies that 8D bends be used in pneumatic conveying piping. Why is this? ■

Dear mkkonen,

As explained earlier in this thread, the pressure drop is theoretically irrespective of the bend radius.

Short bends have high impact angles and therefore high velocity losses, resulting in a higher re-acceleration pressure drop after the bend.

Medium radius bends have lower impact angles and after impact, the particles stay against the outer wall of the bend. The combined velocity loss (impact and wall friction) is minimum, although the particle velocity and particle properties should be “tuned”.

Very long radius bends have small impact angles, however, this could result in higher pressure drops as the velocity loss in the bend increases due to multiple impacts. The very long radius is then a series of angled pipe sections.

The impact angle and the particle velocity are important to particle degradation, and a medium to long radius bend could minimize the particle wear.

Also the orientation (horizontal to vertical-up or vertical-down to horizontal) of a bend is of influence, due to the direction of the gravity.

Many articles point out that the impact and friction forces in a bend contribute to the forming of streamers in f.i. polyethylene conveying by heating up.

However, if the friction energy is calculated, the temperature increase is very little. Too little to cause weakening of the polyethylene.

Electrostatic phenomena are also mentioned in relation to angel hair formation.

Take care ■

Originally Posted by ppquintal

What is your experience using short radius bends (i.e. Hammertek type) instead of long radius bends for cement's pneumatic conveying, dilute phase? do they really decrease pressure drop?

Regards

When conveying abrasive materials it is well known that long radius bends, specially made from aluminum, wear out quite quickly. Steel or stainless steel bends last longer. But a long lasting solution is to use either wear-resistant bends or special bends such a Hammertek and Pellbows.

Various studies and tests have been done to compare the pressure drop in long radius, short radius, Tbends, Hammertek bends, etc. Book written by Dr. David Jones (Pneumatic Conveying Design Guide) gives a comparison of these pressure drops. Test data of Dr. Jones shows that the long radious bends have the lowest pressure drop and the Hammertek bends the highest.

My experience with these different bends is that in an typical operating plant with 5 or 6 bends, this increase is quite small. If the conveying system is running at its maximum pressure drop capability, this increase would need to be addressed.

Regards,

Amrit Agarwal

Pneumatic Conveying Consulting

email: polypcc@aol.com ■