Yuri,

Is there a reason that you want dense phase for this conveying system? These materials can be easily conveyed in a conventional dilute phase system.

Regards,

Amrit

Amrit Agarwal

Consulting Engineer

Pneumatic Conveying Consulting

Email: polypcc@aol.com

Ph and Fax: 304 346 5125 ■

Dear Yury,

From the data, which you have given, the following can be calculated:

particle size = 4 mm (4000 micron) and particle density = 910 kg/m3

The suspension velocity = approx 8 m/sec

12 tons/hr and SLR=20 results in approx 0.14 m3/sec of air flow

With a 102 mm pipeline the end velocity will be aprrox. 17 to 20 m/sec

Trying to run the program with an assumed product loss factor resulted in much lower capacities than 12 tons/hr at much higher pressures than 0.4 bar.

Even with the sedimentation detection switched off, the results did not improve.

Due to the lack of reliable field data for this product, I cannot generate meaningful figures in this case.

The program indicated however, a significant part of the pressure drops is used for air losses and keeping the particles in suspension.

(100 m of 4” pipeline at 20 m/sec can cause a pressure drop, which is large compared to 0.4 bar)

This time

Sorry

Teus ■

Teus,

Is it possible for you to run calculations for this conveying system in dense phase mode also?

I think that Yuri prefers dense phase due to the problem of fines and streamers in dilute phase conveying of PET. But for PP's this should not be a problem.

Amrit ■

Dear Amrit,

When the subject of dense- or dilute phase is raised, then first of all the definition of dense- and dilute phase conveying must be clear.

see also: Dense phase- or dilute phase pneumatic conveying:

https://news.bulk-online.com/?p=238

The “agreed” definition is based on the Zenz diagram, in which the pressure drop is plotted against the gas velocity for a given installation and a given capacity.

Pneumatic conveying at gas velocities lower than the gas velocity at which the lowest pressure drop occur is considered dense phase conveying.

Pneumatic conveying at gas velocities higher than the gas velocity at which the lowest pressure drop occur is considered dilute phase conveying.

In this definition, SLR is not a meaningful parameter. In long conveying lines, the minimum pressure drop occurs at higher gas velocities as in shorter conveying lines and at longer conveying lines the maximum SLR is lower.

My computer program calculates the whole range from dense phase to dilute phase.

see: Pneumatic conveying, Performance and Calculations:

https://news.bulk-online.com/?p=65

The calculation algorithm assumes that if the wall velocity drops below a suspension velocity related value, sedimentation takes place, until the remaining and conveying cross-section generates enough (new) wall velocity to continue further dilute conveying.

The sedimented mass in a horizontal line is considered not moving and carried by the bottom of the pipeline.

The sedimented mass in a vertical line is considered not moving and carried by a pressure drop.

This approach nicely results in the expected Zenz diagram.

As an example see thread:

https://forum.bulk-online.com/showth...t=16863&page=4

For the question in this thread (conveying PET particles of 4 mm size), the calculation program generates output, which is not the problem.

The output however, based on the initially, by Yury, assumed input values of rate, air flow, pipe size and pressure drop is not consistant with the assumed input.

The calculated output shows low capacity at far higher pressures and sedimentation, caused by too low gas velocities.

Moreover, I do not have reliable field data of existing installations on which I can base a product loss factor.

The calculation of a Zenz diagram, in order to find the lowest pressure drop, therefore, is in this case a shot in the dark.

Best regards

Teus ■

Dear Yuri,

PET can be conveyed in dense or dilute phase, this choice depends upon the economics.

For dilute phase you will need internally roughened pipe line and equipment for removal of fines and streamers.

For dense phase you can use internally smooth pipe line and will not need this additional equipment.

But the capital and operating cost of dense phase may be more because of the higher pressure equipment.

Regards,

Amrit ■

Dear Yury,

The velocity is NOT constant during all conveying.

The pressure at the beginning of the pipeline is high and drops along the pipeline to atmospheric at the end of the pipeline.

This means that the specific volume of the air increases from the beginning of the pipeline towards the end of the pipeline.

The air velocity is proportional to the specific volume of the air and therefore the air velocity increases also from the beginning of the pipeline towards the end.

The density of air, which you refer to, of 1.3 kg/m3 is related to pressure and temperature.

Assuming air of 40 degrC, the pressure is then 1.153 bar.

The conveying pressure at the beginning of the pipeline is higher and at the end of the pipeline lower. You must have calculated the velocity somewhere in between.

To perform a more or less reliable calculation, it is necessary to know at least 2 variables.

1)the suspension velocity of the particles

From the particle density and the particle size with an assumed drag factor (which is shape depending), the suspension velocity can be approximated. In this case, this approximation results in 8 m/sec

Based on this suspension velocity, an end air velocity is chosen.

The chosen air end velocity of 17 – 20 m/sec is causing sedimentation in the pipeline and therefore considered too low.

2)the product loss factor

This factor determines the pressure drop related to particle collision- and friction losses.

The influence of this parameter increases with the conveying length and in this case of 100 m length, a product loss factor value obtained from an existing installation is necessary.

If I make a calculation, based on the before described criteria, supplemented with assumed parameters, the result will be mathematically right but (again) a wild guess.

When dense conveying is here referred to as a kind of plug conveying, whereby a series of plugs is moved by a pressure drop over each plug, then another calculation algorithm is required.

Best wishes

Teus ■

Dear Yury,

The discussion about dense phase or dilute phase is now becoming a problem-obscuring subject.

According to the definition (derived in this forum), it is possible to design a low pressure system or a high pressure system in both conveying regimes.

There is always a lowest pressure drop possible.

In this case, we face the problem that we do not know the pneumatic conveying properties of the two products.

Especially, choosing the conveying velocities is very tricky in relation to the formation of streamers.

To avoid deceptions, the best way to go, is to ask a quotation from well know manufacturers of pneumatic conveying installations with proven experience for PET and PP.

An installation, using a rotary lock as feeder is possible in combination with a low-pressure pneumatic conveying design.

The compressor volume has to be increased, to compensate for the rotary lock leakage.

A venturi is another design, for which a manufacturer is needed.

A blow tank system is also an oportunity, however, possibly more expensive.

Self-designing a critical installation without the proper theoretical background and experience is strongly advised against.

We need more input from forum members with experience in these products.

best regards

Teus ■

Dear Yuri,

If you convey PET and PP in the same conveying system, most likely you will have cross-contamination between tho very different materials. After conveying one of the materials you will have to 100% clean the entire system to remove any residues of the conveyed material. This clean-up is not easy. I hope you have considered this potential problem when deciding on using the same conveying system for these 2 entirely different materials.

From technology viewpoint both of these materials can be conveyed in dense or dilute phase.

If you select dense phase, conveying velocities must be below the velocity at which fines and streamers are generated. My experience with these materials is that the maximum velocity in the conveying line should be about 15 ft/sec. We should then work backwards to calculate the velocity at the beginning of the conveying line, or vice-versa. In dense phase mode. lower the velocity, higher is the conveying pressure. We have to keep both in mind. System clean-up is easier in dense phase because the pipe line is smooth.

Best regards,

Amrit Agarwal

Consulting Engineer

Pneumatic Conveying Consulting

Email: polypcc@aol.com

Ph and Fax: 304 346 5125 ■

Dear Yury,

I made the calculation, which you asked for.

The results are attached.

I also checked, whether the design is dense or dilute phase conveying.

The conclusion is that the design is barely dilute phase conveying and that a small reduction in air volume results in sedimentation, starting at the beginning of the pipeline.

However, the air velocities are ranging from 22 m/sec to 40 m/sec and the particle velocities are ranging from 12 to 20 m/sec.

These velocities may be high in respect to the formation of streamers and angle hairs.

A stepped pipeline causes a reduction of the velocities and thereby the risk of particle deterioration.

The calculation, I made is NOT based on verified data and must therefore be regarded as INDICATIVE.

Your choice of the “product friction factor” and the “velocity factor” are also (educated?) guesses

Remark: Why do you cool the air down to -18 degrC?

In addition, the now calculated velocity of 40 m/sec is very much higher than the velocity of 4.57 m/sec (15 ft/sec) as mentioned by Mr Agarwal.

An air velocity of 4.57 m/sec compared to a suspension velocity of approx. 8 m/sec is far too low to keep the particles in suspension, which means that there will be no pneumatic conveying.

To support a vertical column of 15 m requires then 15* 0.54/10 = 0.81 bar.

Overcoming the friction of the particles against the wall (like the storage of bulk material in a small diameter and high silo) requires substantial extra pressure.

I am still skeptical about our figures and advise NOT to decide to build an installation, based on these figures.

Take care

Teus

Attachments

petcalculation (PDF)

■

Dear Yury,

quote

In this calculation "15* 0.54/10 = 0.81 bar." 0.54 and 10 what is this? “

unquote

If there is NO pneumatic conveying, because the air velocity is below the particle suspension velocity, then there is a column of particles in the vertical section of your installation of 15m.

With a bulk density of 540 kg/m this column can be replaced by a water column of

15* 540/1000 = 15 * 0.54 = 8.1 mWC = 8.1/10 = 0.81 bar (10 mWC = 1 bar)

Keep in mind that SLR is NOT a decision factor.

SLR is a calculation factor, defined as material mass flow/gas mass flow.

The SLR follows from your installation design and installation settings s.a. controlled feeding.

The installation with 154 mm diameter pipe and 29.8 m3/min results in a lower pressure drop but also in lower velocities and because of the lower velocities, the risk of sedimentation in the pipeline is eminent.

Increasing the airflow to 0.631 m3/min the air velocity is increased beyond the risk of sedimentation.

The risk of the formation of streamers still exists.

Using the higher airflow, the 12 tons/hr are conveyed at a calculated pressure of 3917 mmWC.

Bigger pipe diameters result in oversized installations, which are then underused and therefore not the optimum choice.

Regarding the air intake temperature, it is better to calculate the highest occurring temperature as the air mass flow then will be the lowest and the resulting SLR the highest.

have a nice day

Teus ■

Dear Yury,

Sedimentation in a pneumatic conveying pipeline occurs when the air velocity is becoming too low for keeping the particles in suspension.

This condition occurs primarily along the pipe wall.

The air velocity along the pipe wall is, due to wall friction, lower than the average air velocity.

SLR is nothing else than a calculation factor.

I used an Aerzen blower, type GM35S at 3800 rpm for my calculations and this blower delivers 0.631 m3/min at 1 bar.

When running at a lower pressure, the lower internal leakage is accounted for in the calculation

The same remark goes for the rotary lock losses.

Calculating a 154 mm bore pipe with a blower displacement of 29 m3/min at 0.4 bar pressure result in 20 m/sec to 28 m/sec and the calculated wall velocities in horizontal sections appear to be too low.

Sedimentation is then the case.

A local cooler manufacturer can tell you the best option for coolers.

Water to air or air to air. Pressure drop only a few mmWC, if you wish.

take care

Teus ■

Dear Yury,

Here the requested calculation.

Success

Teus

Attachments

petcalculation2 (PDF)

■

Dear Yury,

The figures 1.0 – 1.51 are the ratio of the local wall-air-velocity divided by the local-suspension-velocity. (not m/sec)

If the calculated ratio is too small, the respective cell is marked red.

Then the sedimentation is changed, whereby the remaining cross section is decreased until the required ratio is restored.

The value is increased by increasing the wall-air-velocity and that is reached by increasing the blower volume.

best regards

Teus ■

Dear Yury,

The calculation for a pipeline with a bore of 114 mm is already done.

See “CalculationPET.pdf”

The air volume for a pipeline with a bore of 154 mm and no sedimentation is 0.648 m3/sec.

The capacity is then 12 tons/hr at (calculated) 3916 mmWC.

The condition for sedimentation is taken as:

horizontal: vwall = 1.2 * vsuspension

vertical: vwall = 1.5 * vsuspension

Scaling pneumatic conveying installations is (almost) impossible. The number and complexity of scaling factors is very, very high.

F.i. are the conveying properties of the PP the same as the PET, we are referring to and if not in what respect are the properties different and what is the effect of those differences.

Best regards

Teus ■

Dear Yury,

First I repeat:

The calculation, I made is NOT based on verified data and must therefore be regarded as INDICATIVE.

Self-designing a critical installation without the proper theoretical background and experience is strongly advised against.

Further:

The motor size of the blower of 45 kW should be a little bit higher (approx.55 kW).

Then, there is enough motor power to increase the pressure to 0.6 bar. In case the design is wrong, there is always some pressure reserve. In my calculations, I used an Aerzen blower.

The volumetric efficiency of your blower choice should not be lower than the Aerzen blower, in order to comply with the calculated airflows.

Are the pneumatic conveying properties of the PP the same as those of the PET?

A separate design calculation is strongly advised, in order to prevent future problems.

Also keep in mind the remarks of Mr Agarwal.

Cross contamination of material.

Different choice of velocities. Here we face some theoretical and practical contradictions, which are not yet solved. (We might be of track and therefore wrong)

More input of Mr Agarwal would be very welcome.

However, with the gained insight in pneumatic conveying, you can ask some experienced suppliers for a quotation and issue the order to one of them.

(Do not undertake this project yourself, regarding the uncertainties.)

This thread was a nice opportunity to investigate the physics of pneumatic conveying.

Success

Teus ■

Dear Yury,

I have no experience with this calculation method.

Only Mr Agarwal is able to verify the used figures.

In the mean time, we have now three mentioned velocities:

- 15 ft/sec = 4.57 m/sec (Mr Agarwal’s reply)

- 20 – 40 m/sec (My calculations based on a suspension velocity of 8 m/sec and the wall velocity criterium)

- 19.81 m/sec (Your PP calculation)

In addition, you are now referring to PP instead of PET.

From here on, professional assistance is needed.

Success

Teus ■

Dear Yury,

Is this mentioned velocity related to a particle size, particle density or suspension velocity?

(27.9 m/sec is the fourth velocity that emerges)

best regards

Teus ■

Dear Yury,

There is no direct relatiobship between bulkdensity and air velocity.

However, there is a relationship between air velocity, turbulence (wall velocity) and suspension velocity.

The suspension velocity is related to particle size, particle density and particle shape.

This is the background of my question.

best regards

Teus ■

Dear Teus and Yuri,

I have always used "saltation velocity" to determine solids conveying velocity. We get into dense phase regime when the solids velocity is less than saltation and are in dilute phase when solids velocity is above saltation. For calculating saltation velocity I use Rizk or Matsumoto co-relations and methods.

I guess your definition of suspension velocity is the same as that for saltation velocity. Is it true?

Regards,

Amrit

Amrit Agarwal

Consulting Engineer

Pneumatic Conveying Consulting

Email: polypcc@aol.com

Ph and Fax: 304 346 5125 ■

Dear Amrit,

We had discussions about the definition of dense- and dilute phase conveying before.

See the threads:

https://forum.bulk-online.com/showth...t=dense+dilute

https://forum.bulk-online.com/showth...+dilute&page=2

https://forum.bulk-online.com/showth...t=dense+dilute

I recall, that it was more or less agreed that the Zenz diagram is used to determine whether a pneumatic conveying system is dense- or dilute phase.

Left of the lowest pressure drop is dense phase and right of the lowest pressure drop is dilute phase.

The definition of the suspension velocity, which I use is just the free falling velocity of a particle under the local gas conditions.

The standard parameter “suspension velocity” is under atmospheric conditions at 0 degrC.

In the pneumatic conveying calculations, this value is corrected for the local conditions.

In the calculations, a check is executed, whether the wall velocity is high enough to carry a particle and if not, sedimentation is assumed.

The smaller remaining diameter causes an increase in velocity, which is then high enough to carry the product again.

I find the Rizk or Matsumoto co-relations rather complex and not easy to understand what they describe.

I also wrote an article in the Bulk-blog about this subject.

Dense phase- or dilute phase pneumatic conveying:

https://news.bulk-online.com/?p=238

It happens to be in most cases, that the saltation or forming of sediment occurs in the vicinity of the lowest point of the Zenz diagram and therefore close to dense phase conveying.

The absolute value of the SLR, however, is not a reliable indicator of dense- or dilute conveying, because the SLR is conveying length depending.

The above mentioned approach results in a smooth Zenze curve from dense phase conveying to dilute phase conveying, when calculated.

Have a nice day

Teus ■

Dear Yury,

As already explained before, there is no direct relationship between bulk density and air velocity.

Take f.i. the list for cement with a bulk density of 1200 kg/m3.

The air velocity, according to the table is then 2438 m/min # 40 m/sec

We all know that for cement, an air velocity of 10 m/sec to 12 m/sec is sufficient.

Success

Teus ■

Dear Yury,

Summary of properties of cement

- Average particle size 34 micron

- Particle size range 15 - 90 micron

- Average suspension velocity product 1.8 m/sec

- Average suspension velocity particle. 1 - 1.2 m/sec

- Critical can-velocity 1.4 m/sec (Vupwards between filter elements)

- Product density 3100 kg/m^3

- Poured density ( fluidized ) 1000 - 1200 kg/m^3

- Stored density ( after settling ) 1600 kg/m^3

- Angle of repose 25° - 30°

- Friction coefficient .35 - .6

- alpha = pvert/phor ( not fluidized ) 0.45

- alpha = pvert/phor ( fluidized ) 0.70 - 1.0

- alphamin ( fluidized ) 0.3°

- volume ratio at turbulance point int vol./total vol. 0.564

- vturbulance point air-velocity 1.31*10^-3 m/sec

- etadyn. Viscosity of fludised bed 0.1 Ns/m^2

- Hygroscopic

- Abrasive

- Chemically pure half-fabricate

Pneumatic conveying velocity = 6-7 * v-suspension = 10.8 m/sec – 12,6 m/sec

When you have a specific product to convey, the first thing that has to be done is to test a sample of that product.

Wikipedia is a great source in this respect

Have nice day

Teus ■

Dear Yury,

Here is your table.

No guarantee

Succes

Teus

Attachments

materialtable (ZIP)

■

[QUOTE=yury.bogdanov;56397]Hallo,

I am designing a system of pnemuatic conveying for PET and PP. And the data I have are as follows:

Material flow rate: 12000kg/h

PET:

dimension: 4000micro m

bulk density: 540kg/m3

Partucle density: 910kg/m3

The line:

horizontal: 80m

vertical: 20m

5 elbows

This conveying sistem is dense phase. I think to use rotary valve with venturi. And the feeder of aire is a positive pressure blower (pressure 1 bar max). And has the flex conection in the midle to conect the diferent destiny. The distance above is the longer way.

And I think to use 102mm of pipeline bore.

Now I need to calculate SLR, engine power and velocity of aire in this conditions. I think that for this product SLR is any 20 in dense phase. And supose that air mass flow reate is above 0.1kg/s. And the pressure with rotary valve is above 0.4bar.

Dear Teus, can you to calculate this system with you fantastic program, please.

regards[/QUOTE

Dear yury.bogdanov ,

Can you tell me how to calculate the pressure drop in 1.5m Horizontal pipe in which air carry 3mm particles.

I mean which kind of equation or theory you used to calculate the pressure drop.

If you can give me any suggestion regarding that then it would be helpful to me !

Regards,

Pritesh Patel

+49 176 64634215 ■

I viewed your spreadsheet.

The input is:

pressure in =1 bar

pressure out = 1.8 bar

Thus, pressure drop = 1.8 – 1 = 0.8 bar

The calculated pressure drop = 0.307 + 0.082 = 0.389 bar

The calculated pressure drop must be equal to the set pressure drop, which is not the case here.

That is the reason that an iteration process is needed

I also checked the unit dimensions of the calculated capacity in cell L28.

The unit dimension is ton/(min*m2).

The capacity must also be Air mass flow * SLR = 17.7 * 1.3 * 12 /1000 = 16.55 tons/hr

The volume change of the conveying air is not accounted for in the pressure drop calculation.

This is the reason that a numerical integration algorithm is necessary.

Calculating a pneumatic conveying system is not as simple as your spreadsheet assumes.

Reading the books of the many specialists, who had access to test installations and vast knowledge of physics and mathematics, support this.

Success

Teus ■

Based on the agrawal´s paper, I modified it to some extent and the spreadsheet which is attached herewith matches with the existing fly ash conveying systems that i know.

Attachments

pneumatic conveying calculation (ZIP)

■

for fly ash I use these two sheets, they agree with the observed data very well, both for dense phase and dilute vacuum phase

Attachments

conveying (ZIP)

■

Dear tathadream,

In the attachment, how I recalculated your installation.

In metric units.

BR

Teus

Attachments

recalculationflyash (PDF)

■

Dear Teus,

>But I think you have taken longer pipelines, the calculation I attached has very small pipelength ■

Dear Dear tathadream,

You are quite right. I made an error in the decimal points.

Attached the correct recalculation.

Have a nice day

Teus

Attachments

recalculationflyash2 (PDF)

■

hello teus,

I think my calculation and your program giving a similar result if I was able to comprehend your software properly ■

Could you please also review the other two excel files which I have attached one for dilute phase vacuum conveying and the other for dense phase conveying ■