Re: Accuracy Of Screw Conveyor As A Feeder
Posted on December 08, 1999 at 10:25:59:
In Reply to: Accuracy of Screw conveyor as a feeder posted by Utkarsha Joshi on November 17, 1999 at 09:35:31:
The accuracy of a screw feeder is dependent upon the flow reliability of the hopper and how this 'conditions' the density of the material filling the volume of the screw. High consistency can be secured with materials that settle rapidly to a stable density and good results obtained with product that do vary in density, if the feed hopper and screw are of good design. Powders are strong respecters of detail. Some care is required to secure within 1% bounds of variation, even with easy to handle products and +/- 2 or 3% is good with fine powders. Remember, it is the bulk material that varies, not the screw, so attention has to be directed to the averall flow system based upon measured values for predictable performance.
-------------------------------------------------------------------------------- ■
Re:
Hello Utkarsha Joshi,
It seems you are discharging the calibrated quantity of material from the hopper / bin. This quantity is fed into screw conveyor. The screw conveyor is supposed to transfer all this quantity into subsequent equipment. The accuracy of discharge at the hopper is +/- 0.5 but you are finding the accuracy after screw conveyor as +/- 1%.
This clearly implies that all the material going into screw conveyor is not reaching final point all the time and so the discrepancy arises which could be sometimes on minus side and sometimes on plus side. This means that the material is not lost from the screw conveyor but stuck within the screw conveyor only.
In my opinion screw conveyor is not so precise conveying equipment, sometimes some material will get stuck somewhere when quantity involved is very small. You may think about the use of totally enclosed narrow belt width belt conveyor moving at a slow speed. If material is not super fine (powder type) then practically all material will get discharged at the pulley and it will not have the type of problem you are mentioning. This suggestion is subject to verification / examination for your specific situation which is not very clear.
Possibly, Mr. Lyn Bates has replied considering that there is a screw feeder. But looking at the question, to me it appears to be a screw conveyor in situation as above. So, there is a difference in reply / understanding.
Regards,
Ishwar G Mulani.
Author of Book : Engineering Science and Application Design for Belt Conveyors.
Email : parimul@pn2.vsnl.net.in
Tel.: 0091 (0)20 25882916 ■
Re:
1.) A downstream, starve-fed HORIZONTAL transfer conveyor should have approximately the same accuracy as your gravimetric feed device, assuming it is appropriately applied and has no material loss or spillage. However if the transfer conveyor is on an incline, many variables now enter into the puzzle. Screw conveyors, once inclined, will lose efficiency due to material fallback. The constantly changing material variables such as flowability and density will never have repeatable fallback characteristics.
2.) It has been our experience with proprietary gravimetric controllers is that they tend to be “overly optimistic” in their accuracy calculations. Are you basing your feeder accuracy on the indicated value, or are you actually grabbing timed catch samples? Our controllers include a SPC screen that can use either the feeder scale, or an auxiliary catch scale, to determine true system accuracy.
Regards, Delmar Schmidt
Melfi Technologies Houston
Phone 281-298-8398
www.melfitechnologies.com
Attachments
■
Accuracy Of Screw Feeders
Mr Mulani and Schmidt rightly point out that there are various alternative feed device to screw feeders. Each of these tend to have their own benefits and drawbacks. The enquirer queried the accuracy of discharge from a screw conveyor, which I took to imply that a screw was in use to control the feed rate.
Expanding on my initial contribution any volumtric feeding device is only as good as the consistency of density that is attained in the supply channel. Whilst some products are far more stable in density than other it is usually possible to achieve relatively good stability with a well designed hopper and controlled operating dicipline. However good the control system, it cannot compensate for erratic fluctuations of product condition.
The way in which material leaves the end of a screw feeder can affect the short term accuracy of feed rate in different ways: - cyclic, cohesive collapse and avalanching. Details of these are given in a paper available from Ajax equipment on the feeding of solids. see www.ajax.co.uk
Regarding the measurement of accuracy I would commend the British Materials Handling Board publication - 'Guidelines for specifying Weighfeeder Systems' for evaluation of both gravimetric and volumetric metering devices. The period over which measurements are taken is a key parameter in this activity, and the significance of this is generally application dependent. Where two or more components are to be mixed, the final composition of the product is obviously affected by both the individual accuracies and the relationship of variations. A broad statement of 'plus or minus one percent' has very little meaning unless stringently qualified. It cannot be emphasised enough that where feed accuracy is a feature of contract conditions or detremines the product quality it is highly important to enclose the boundary conditions and closely specify the crucial parameters. ■
Accuracy Of Feeders
As I understand your product is a fine powder, oderwise you would not be considering aereation.
Check your special device feeder. and see if it really gives you the accuracy you were told.
As it was mentioned before, the special device could not be up to expectations or its calibration might have degraded.
If the screw feeder is not accumulating any material inside, it should not change the feeder responce , unless it is pumping air with the solids and affecting the feeder calibration. Highly unlikely unless you are running it at high speed or forming pluggs, normally your screw should run less than half full. I suppose you are not runiing a gas tight connection between the screw and the dosifying device.
Regards
Marco ■
Accuracy ?
I would like to draw the attention away from the equipment and flow details to something rather simple but nevertheless extremely important: statistics.
Utkarsha Joshi reported that his feeder is claimed to be accurate to +/-0,5% but he finds more than 1% deviation. Now, this could be a matter of comparing apples and oranges. To determine te accuracy of a feeder one typically does some measurements by catching product for 30s or 1min and doing some statistics on the collected sample masses. Obviously 1min sampling tends to give less scatter than shorter sampling times. Also you can consider the largest deviation of any one sample from the setpoint as measure of the accuracy or one can do some confidence interval statistics based of standard deviations to calculate accuracy.
The feeder manufacturer most certainly reported an expected accuracy based on short term standard deviation since this will give the lower number. What did Utkarsha Joshi do? ■
Accuracy
Of course you are right, but the batch size will depend on what is your delivery batch size, If you are loading laundry soap packages of one kg or 5 kg tortilla corn flour your sample size has to be as small as your package , now there can be many parts between your sample point and the feeder , and fine dust can accumulate in these parts , and then fall , kiling your accuracy.
If there are goverment regulations involved you might end filling some bags with less product , and some with more, compensating por the deviation could mean loosing valuable production. Or un reacted components left over, if this feeds a reactor.
Now the feeder supplier should have provided with mean and standard deviation specs, and also a reccomended sampling procedure. This should be reasonable considering the delivery package size, or batch.
The screw conveyor should not have spoiled the accuracy , unless the product is fuidising and flushing , some pet coke instalations have presicely this arrangenment and the screw conveyor (going up in a slope) sometimes back flows into the feeder.
I would also like to see a diagram or a picture of his instalation.
Just to better visualize the problem.
Regards
Marco ■
Re:
Various respondents have given good analysis and suggestions. But there is no proper clarity on the existing system i.e. without flow sketch, batch size, frequency of operation etc. So, the issue remains without suggestions in specific direction.
Regards,
Ishwar G Mulani.
Author of Book : Engineering Science and Application Design for Belt Conveyors.
Advisor / Consultant for Bulk Material Handling System & Issues.
Email : parimul@pn2.vsnl.net.in
Tel.: 0091 (0)20 25882916 ■
Screw Conveyor As Transfer Feeder From Metering Device
Others can comment more authoritavely than I on screw feeders, but considering the problem of accurate transfer of metered powder feeds, it is worth noting that aero-mechanical conveyors have been used very successfully for this duty. The reason they work on this application is clear. Given the questioner's duty of 250 kg per min and a typical aeromechanical conveyor velocity of 3 mps, a 6 meter conveyor (say) will contain in-flight material of less that 9 kg. - and that quantity would be contained within 30 separated 'packets' of only 300 grms.
One would, of course, want to know precise delails of the application before making specific recommendations.
Transfer of weighed batches may be similarly considered.
BILL ANGUS ■
Material Handling Standards
Posted on November 17, 1999 at 09:35:31:
We currently use a silo/hopper with a specisl device to fed a powder at around 250 kg/min. The discharge accuracy is claimed to be + or - 0.5% of the setpoint.
The powder is then transfered to a screw conveyor for discharging into a mixer. The variation in the final product weight can be significantly greater than + or - 1% (componded by the variabilityof another feed material in the mixer).
Am I right in thinking that, because thescrew conveyor is a volumetric device and can introduce aeration, that it is possible to 'undo' the accuracy of the precision feeder and therefore suffer the higher variability in the final product ?
What is the general accuracy of discharge from a screw conveyor claimed by various manufacturers ? Can + or - 0.5% be achieved in a screw conveyor ? ■