Dear RJB324,

Your question does not contain the necessary information to formulate an answer that can make sense.

Obviously, you are overfeeding the pneumatic vacuum system.

A partially opened knife gate valve is most likely not a proper dosing device.

Overfeeding causes a too high vacuum and thereby the airflow reduces too much, causing the suction intake choke immediately.

Whether the system is designed and built properly requires more information.

Material and size of the bead.

Pipesize

Air displacement of vacuumpump

Pipe routing.

Straight pipe section length before first bend.

Interesting to know the cause of your problem and the reason why it is designed as it is.

BR

Teus ■

Dear Rob,

From English units to METRIC units:

3300 cfm = 93 m3/min = 1.557 m3/sec

9.5” Hg = 0.3266 bar vacuum

4’ = 1.2 m

10”pipe = 0.26 m inner diameter

Plastic 400 micron bead suspension velocity based on a guessed material density of 1200 kg/m3 (please confirm) is approx. 3.2 m/sec

Intake air velocity = 1.557 * (1-.3266) / (3.141593/4 * 0.26^2) = 19.6 m/sec

Air velocity/suspension velocity = approx. 6 (Rather high ratio, but not alarming)

Distance to the first T-bend of 1.2m should be sufficient for particles to be accelerated sufficiently.

I still suspect an over feeding problem. (knife gate valve)

Can you describe the intake feeding arrangement?

If you can supply piperouting including the designed performance then I can try to calculate the system.

In addition, the material density and the bulk density of the plastic beads are required.

Take care

Teus ■

Solid flow rate is still missing from the data. Assuming that your stated air flow is at exhauster intake and the system runs at 300 mbar vac. the pick up velocity is approx 20 m/s, which for a 10” line is marginal. At these velocities the accelerations lengths to achieve min. slip velocity are in the order of 4 -9 m. But on site there never is that much distance from the pick up point to the first bend and system still do work !!!

You have not mentioned the orientation of the bend is it horizontal to horizontal or vertical up. This will make a lot of difference if it is a long vertical up that cause problems. Blind tee’s are very commonly used for plastic pellets to avoid angel hair generation even though they give slightly higher pressure drops. You should replace the first bend with a long radius one.

As mentioned above make sure you are not over feeding the system. Start the system with material feed valve fully closed then open it in little increments and let the system stabilize before opening the valve further. If the system fails to run at reasonable rates then you will need to put the starting 4` section and the first bend to 8” this will increase the pick up velocity and hopefully will make the system run without blockages. ■

Dear ,

A rather complicated pipeline.

I am not certain, whether I interpreted the sketches correctly.

On SK1 I see new piping with 10 TEE-bends and 1 long radius bend.

At the end of that pipeline I see an arrow, obviously pointing against the material flow.

The pipeline end (with the arrow) is connected to the pipeline on SK2, where the code PT”A” is written.

From there the pipeline encounters another 5 bends.

The mentioned knife gate is at the end of the pipeline on SK1 and just before the pipeline on SK2.

That means that the knife gate is in the middle of the suction pipeline and when partially open it creates an orifice and increase the vacuum.

The vacuum can become too high, reducing the air mass flow and a blockage is eminent.

Feeder control of a vacuum system must always be executed at the beginning of the pipeline and never in the air/material flow.

However, before I jump into conclusions, please verify whether I am on the right track.

Interesting case

Have a nice day

Teus ■

The knife gate and the adjacent piping is 6".

Is that correct?

The pipe, being 6", will cause an extra pressure dop.

regards

Teus ■

Dear RJB324,

As the pneumatic conveying installation consists of 44 elements, it will take some time time to model it.

The 6"pipe section can cause air velocities up to approx. 56 m/sec.

Therefore this section is very important.

The rotary lock size is also important to know (volume and rpm) as this component creates a leakage to the system. (lowering the convey air mass flow)

Stay foot

Teus ■

Then, I am still a bit puzzled.

I set up the pipeline in sketch and found a conveying length of 109 m.

If the 6"pipeline is the plastic bead feeding, controlled by the 6"knife gate, where is the material mixed with the convey air and in what direction is the material flowing?

Is point TP/A perhaps connected to PT"A" and flows the material in that direction?

We will get this right somehow.

Teus ■

Ok, that is clear now.

Leaves me with one last question.

Where is the convey air inlet and where is the convey air mixed with the material?

This mix-location is the start of the pneumatic conveying calculation

BR

Teus ■

Dear RJB324,

I modelled the pipeline and performed a preliminary computer calculation with an assumed product loss factor for plastic powder.

The preliminary results are given in the attachment.

Choking problems, because of insufficient material velocity, are very unlikely.

To improve the accuracy of the calculation, it helps if you have an existing installation with the same product that is operating.

Using the installation data and performance data of that installation it is possible to determine the product loss factor of the plastic beads and use this factor for other installations.

I cannot locate the make and type of the 4B3 blower.

If I have the blower data, I can use these as input for the pneumatic conveying calculation.

Can you supply those blower data?

Also, check for the air leakage under the cyclone. This air leakage has to be subtracted from the blower volume to get the conveying air volume. This effect increases with increasing vacuum

I am still thinking of overfeeding, now in combination with cyclone leakage.

All for now

Teus

Attachments

plasticbeads (PDF)

■

Dear RJB324,

Thanks for the information.

The blower specification shows that you are using a centrifugal fan as a vacuum pump.

This type of air mover is not a preferred machine.

The reason for this is the characteristic of rapidly decreasing airflow at a very narrow range of pressure change.

Imaging the following:

The fan is operating at a SLR=0 at low vacuum and high airflow.

Feeding material increases the SLR and thereby the vacuum. At the same time, the SLR increases again, due to the decreasing air mass flow.

Because of the increased SLR, the vacuum increases again and the air mass flow decreases again.

Whether the end situation is stable or not, is very difficult to estimate.

With low SLRs it might be possible. However, with high SLRs, the operating point moves to the flat section in the curve, where the air mass flow decreases very sharp with a small increase in vacuum and choking is eminent.

Therefore a positive displacement pump (characteristic) is always preferred in pneumatic conveying installations.

Further remarks are:

-The calculation I made earlier must be regarded as indicative, as not all parameters are sufficiently accurate known.

-The material/air mixing device is not clear.

-Pneumatic conveying data of working installations for the same material would be very welcome for reference purposes

-Any feeder control should be based on pressure (vacuum) rather than Amps

-Perform a test from closed knife gate to open knife gate in small increments and measure capacity and pressures (vacuums) and check the functioning of the cyclone (leakage).

-A calculation at 11.8 inch HG and 1600 cfm failed.

Before making any drastic measures to correct the installation must be preceded by a thorough investigation of the whole system.

Take care

Teus ■

Dear RJB324,

The application of a fan blower versus a PD blower is an interesting issue.

Therefore, I calculated your system for both cases.

I converted the fan curve into metric units and calculated the capacity for a number of points on the curve.

The calculation showed that at a vacuum of approx 3500 mmWC (11 “Hg) the pneumatic conveying becomes critical and a little bit more feeding makes the conveying to stop immediately.

Then a PD blower was chosen with a comparable displacement of the fan and the calculations were executed again.

These calculations show that the performance curves do not show any critical points in the working range and maintain a stable pneumatic conveying.

In the attachment, you can see the calculation results with the explanatory graphs.

Best regards

Teus

PS, did you already make progress in your problem finding?

Attachments

comparison fan blower versus pd blower in a vacuum (PDF)

■

Dear RJB324,

The first remark is that you only need a vacuum transmitter at the end of the system, which is the intake of the fan.

The vacuum setting should be approx. 9.5 “Hg, corresponding with a (calculated) capacity of maximum 33 tons/hr.

If the vacuum increases above the 9.5 “Hg, the feeder gate must close.

The Amps of the fan motor are in no way proportional to the vacuum of the fan.

Probably, the Amps are the lowest, when the fan airflow is the lowest with the pipeline choked.

The big drop off in capacity occurs when the fan-air-flow reduces to a value whereby the air velocity in the pipe becomes too low to carry the material anymore.

The pipe starts filling up with material, thereby reducing the cross section and maintaining in this way the minimum settling velocity, until the pipeline is fully choked.

Keep in mind that the air velocity in the pipeline is not only decreasing with increasing vacuum, but also because of the decreasing air volume displacement of the fan with increasing vacuum.

(Actually it is better to focus on the air mass flow changes)

see also extended attachment.

The first step, I would do, is improving the control of the feeder and then test again manually.

Before other drastic measures are considered.

Also, keep in mind that my calculations are based on an assumed product loss factor.

As soon as field data of the plastic beads are available, a more accurate calculation can be made.

Take care

Teus

Attachments

comparison fan blower versus pd blower in a vacuum_1 (PDF)

■