Category: Test Results

Particle Size Distribution Analysis for Ceramic Pot Water filter production

Maria del Mar Duocastella and Kai Morrill

Potters Without Borders, Enderby, British Columbia, Canada – September 2012

Abstract: To develop a standard Particle Distribution Analysis testing protocol for use in Ceramic Pot Water Filter factories.

Introduction: Ceramic Pot Water filters are generally manufactured from sources of raw clay that vary in their consistency, some factories have begun using particle distribution analysis to qualify clay batches, as well as for blending multiple clay sources in order to maintain a more homogeneous clay body. In order to promote common testing methods between factories, we have begun herein to develop testing protocols that utilize widely available apparatus and materials. It is desirable to develop an effective test that is easily accessible to individuals with limited laboratory experience. This test must be able to be performed in extremely rudimentary conditions with limited resources while presenting reliably accurate results. We hope that by establishing stabilized testing standards specific to filter production the test data will be useful in comparing clay bodies between all participating filter factories. We find that difficulties in ensuring that identical lab equipment is used (cylinder dimensions) may make it difficult to accurately compare results across different factories. Several standards already exist for soil classification; particles can be classified into categories of Clay, Silt or Sand. These categories are demarcated recognizing that suspended particle size is in direct relationship to settling time. For our purposes, we established a baseline for classification by comparing other standards and examining the results of our tests.

Although it is useful for general comparisons to define the samples by the three categories (Sand, Silt, Clay), for the purposes of detailed clay sample comparison, it is better to collect data from various particle sizes, thus developing a curve of particle size distribution. For this reason we tested samples at 13 different time intervals: 30 seconds, 1 minute, 2 minutes, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 6 hours, 8 hours, and 24 hours. Having this expanded range of sample data allows us to compare samples in greater detail. These times were also chosen in order to complete the test within an 8 hour work day. *Note 1: Samples in Appendix 2 (Raw Data) which fall outside the standard testing procedure (Those prepared 24 or 48 hours before testing) were excluded from the final averages as there was significant variation in their results. It would have been interesting to use the results gathered to compare particle distribution results to burnout mixture ratios used in the participating factories. This proprietary information did not receive specific approval prior to publication.

Download (PDF, 1.21MB)

Creative Commons License
Particle Distribution Analysis for Ceramic Pot Water Filter Production by Potters Without Borders is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
Based on a work at http://potterswithoutborders.com/?p=3499.

Criteria Report for Household Water Treatment Solutions

When are ceramic water filters appropriate?

Community Choices Tool for Water, Sanitation, and Hygiene Pacific Institute

Household Water Treatment Solutions Criteria Report
Criteria Used for Ranking Household Water Treatment Solutions
Below you can view information that clarifies how your individual answers to the questions impacted the Community Choices Tool’s
recommendations for technologies and approaches that are appropriate for your situation and needs.
NOTE: This demonstration prototype of the Community Choices Tool contains rankings for the few solutions we have in the database. We
envision that once fully developed, the Community Choices Tool will be able to evaluate and rank hundreds of technologies and approaches
for the entire WASH sector, and from those it will be able to create fully customized solutions for each user (rather than the static solutions it
has now).

Download (PDF, 32KB)

Pug Mills Increase the Strength of Filters

Pug mills really do increase the strength of our filters. At our facility in Yunnan China, we did an experiment where we made 12 blocks as we usually do (banging them on the floor) and each weighing 8.6kg (as usual).

Derek Chitwood wrote…

“Pug mills really do increase the strength of our filters. At our facility in Yunnan China, we did an experiment where we made 12 blocks as we usually do (banging them on the floor) and each weighing 8.6kg (as usual). In the next set of 12 we added 0.2kg to each block (8.8 kg) thinking that the extra material would cause extra compression – and work like a pug mill in effect. Our 3rd set of 12 had an extra 0.4kg in each block (9.0kg) – that was a lot of waste and frustrating for the technician. Then the 4th set of 12 had all the clay first go through a pug mill once and then was made into normal 8.6kg blocks (as usual – pounding them on the ground). The results where that the crack point of our filters increased about 9%. “  See the figure:
— in Kunming, Yunnan, China.

Data courtesy of Derek Chitwood

GTZ Yemen Ceramic Water Filter Flow Test Results

GTZ Yemen Ceramic Water Filter Flow Test Results

Summary

6 -#1 Mix Filters 2.33 liters/hour average

10 – #2 Mix filters all more then 3.5 liters per hour

9- # 3 Mix Filters- All more then 3.5 liters/hour

1-#4 Mix Filter- More then 3.5 liters per hour

No # 5 Mix filters survived the forming process –due to fragility

4 -#6filters- All more then 3.5 liters/hour

Total Filters Flow Tested-30

Filters formed- 48

One Hour Flow test-After forming and firing to a temperature of between 912-923C most filters were soaked in water for more than twenty four hours. The filters were then filled with water and allowed to stand for one hour. The amount of water filtered was measured using a pvc tee shaped gauge which was set on the filters. (See photos)

Mix #1

Formula Batch Wt.

Sana’a Clay- 95% 8550

Sawdust 5% 450

100% 9000gr.

Of 6- Mix #1 Filters

2.5liters

2.5 liters

2.5 liters

2 liters

2 liters

2.5 liters

Flow Test-2.33 liters/hour average

Mix 2– 5% More clay then 50/50 Proportion

Formula Batch Wt.

Sana’a Clay- 90% 8100

Sawdust 10 900

100% 9000gr.

Flow test

Of 10- Mix#2 filters- All filters were more then 3.5 liters per hour

Mix 3– 50/50 Proportion by Volume

Formula Batch Wt.

Sana’a Clay- 85.6% 8127

Sawdust 14.4 1372

100% 9499gr.

3.5 liters water (estimated)

Flow test–

Of 9- Mix #3 Filters- All filters were more than 3.5 liters/hour

Mix 4 – 5% Less clay then 50/50 Proportion

Formula Batch Wt.

Sana’a Clay- 80% 7136

Sawdust 20% 1784

  1. 8920

Flow Test

Only 1- Mix# 4 filter emerged from the mold intact, because of fragility-( One filter broke during flow testing) This filter was more than 3.5 liter/hour

Mix #5

Formula Batch Wt.

Sana’a Clay- 75% 6750

Sawdust 25% 2250

100% 9000gr

Flow Test-

No Mix#5 filters survived the forming process

Mix #6

Formula Batch Wt.

Sana’a Clay- 92.5% 8325

Sawdust 7.5% 675

100% 9000gr.

Flow test

Of 4 –Mix# 6 filters-all were more than 3.5 liters/hour

Download full document 34kb: yemen-ceramic-water-filter-flow-test-results.docyemen-ceramic-water-filter-flow-test-results.doc

April 28th 2007 filter firing Yemen GTZ

Time Description Door Bottom Door Top Chimney Side



Kiln On 4 corner burners






:30
60 93 122



1:00 Damper at 10 cm 69 93 128



1:22 6 burners (2 more near to the door)






1:30 Damper @ 15 cm 93 118 156



2:05
111 133 170



3:06
173 195 248



3:35 Smoke beginning to generate from filters 187 209 265



3:40 Heavy smoke






3:57
208 230 307



4:23 smoke reduced 228 245 337



4:26 8 Burners Damper@ 24 cm






4:38 Damper @ 27 cm 293 295 388



5:00 10 Burners 345 317 415



5:17
347 401 448



5:22 12 Burners (2 on low)






5:27
417 401 457



5:45
438 480 465



6:41 All burners on low 474 518 481



6:53
504 536 501



7:00 Increase two burners either side of the door 509 537 509



7:20
539 575 530



7:32 Increase ten more burners 554 588 543



8:00 Increase ten burners again slightly more 610 646 602



8:22 Damper @ 24 cm 650 684 639



8:40 Increase 8 burners on chimney side to full on 675 708 684



9:00 Increase 4 burners on door side to full 712 744 717



9:12 All burners on full 729 763 738



9:35
750 779 755



9:45 Increased pressure on gas manifold by kiln 784 813 784



9:58 Switched propane cylinder bank to +pressure 798 830 804



10:05 Cone 013 down 012 bending






10:10 Cone011 Bending 832 862 836



10:12 Cone 011 @ 3 o clock






10:14 Cone 011 down 844 874 845



10:16 Open damper to 22 cm 848 877




10:27
865 891 867



10:30
871 897 874



10:35 Begin Soaking period 874 899 878



10:40 All Cones except 09 down -reduce pressure 877 901 881



10:52 increase pressure slightly to maintain pressure 880 903 885




upper Cone 09 @ 3 o clock






11:00 ` 883 906 888



11:06
885 906 890



11:26
880 902 886



11:39 Kiln Off






Notes on the Sana’a, Yemen filter firing April 28th, 2007
The firing of the ceramic filters followed a firing of tanoor ovens by two days insuring that the kiln was well dried.
In firing the filters I was particularly concerned that all of the sawdust would be completely burned out so that trapped carbon would not cause a problem with the filters. This is necessary so that the finished filter be as strong as possible and so that flow rates not be effected by carbon in the filter body. Care must be taken in the stacking of the filters in the kiln to separate the filters using spacers so that heat can infiltrate throughout the setting. Filters should not stacked bottom to bottom without spacers as well. The filters staked on the floor of the kiln should be stacked on top of brick to elevate them. Another thermocouple was installed near the floor of the door to monitor the difference in temperature between the floor and the upper part of the kiln. It is recommended that monitoring of the temperature advance is done with a combination pyrometer, Orton temperature cones and draw trials.

A warm up period with a target temperature of 300 Celsius in the first 4 hours proved adequate to insure both moisture and carbon was not trapped.
After the warm up, an advance of 100 C per hour was desired up until a target temperature of 900 C. In practice we took longer to advance in temperature due to the amount of smoke which was generated by the burning out of the sawdust . At target temperature the firing was soaked for one hour. I had intended to soak for a longer period as per advice by other filter technicians, but after discussion with Bernd Pfannkuchen the kiln designer, and through evaluation of the draw trials* at the target temperature I decided that one hour soaking was adequate. An examination of the finished filters showed that there was no trapped carbon. The original recommendation of a longer soaking period is based on the use of the Mani kiln which is using solid fuels (wood) whereas the Sana’a catenary kiln fires with Propane under oxidizing conditions. Having the designer of the kiln to assist in the filter firing was a real advantage. Bernd gave advice and the kiln design proved very versatile. It is noteworthy that the Orton cones and the pyrometers showed less then one cone differential in temperature throughout the kiln.
*The draw trials were made from the same mix of clay and sawdust and formed into rings. These trials were positioned in the kiln on two levels near the floor and withdrawn through the ports in the door. These rings were then examined for residual carbon as well as for strength.

Pororsity Tests

Pororsity Tests*-150 to 200 gram Sections of fired filters were cut out of the lip, weighed, soaked for 24 hours and then re-weighed to evaluate.

Ceramic Water Filter Test Batch No. 1
Thursday January 18, 2007
Pororsity Test*- 38.8%
******

Ceramic Water Filter Test Batch No. 2
Saturday January 20th, 2007
Pororsity Test*- 32.2%
Mix Proportion represents 10% less sawdust than mix No. 1

******

Ceramic Water Filter Test Batch No. 3
January 22, 2007
Pororsity Test*- 34.5%
Mix Proportion represents 20% less sawdust than mix No. 1

******

Ceramic Water Filter Test Batch No. 4
January 24, 2007
Pororsity Test*- 45.6%
Mix Proportion represents 10% more sawdust than mix No. 1
******
Ceramic Water Filter Test Batch No. 5
January 25, 2007
Pororsity Test*- 49.7%
Mix Proportion represents 20% more sawdust than mix No. 1

Technorati Tags : ,

Ceramic Water Filter Series Batch Tests Firing

Feb 8th, 2007
20 filters in firing – Target temperature 890 C.
Gas firing

Kiln On  — 1 Burner
1:00 187 C 2 Burners
2:06 277 C 4 Burners
3:20 523 C 6 Burners
4:00 544 C
4:58 560 C
5:40 682 C
6:06 707 C
6:38 805 C
7:12 850 C
7:21 861 C
7:40 890 C Begin Soak
8:00 893C
8:21 897C
8:40 894 C
9:00 Kiln Off

Firing Descriptions:

Of 20 filters 17 cracked during the firing. These cracks appeared to be dunts due apparently to rapid firing. The firing also revealed carbon retained in the filters mostly in the bottom of the filters. Of the three filters which were not cracked, two had carbon in the bottoms. The retained carbon appears to be caused either from insufficient soaking and or the method of stacking the ware in the kiln.

Ceramic Water Filter Series Batch Tests

Ceramic Water Filter Test Batch No. 1
Thursday January 18, 2007

Mix Proportions @ approximately 50-50 by volume

Plainsman Red Earth Clay 44lbs. (20 KG) 86.3%
Screened Sawdust 7lbs. (3.18 KG) 13.7%
Water 21 lbs 11.5 oz.

(62.31 lbs total mix used for 4 filters or 15.58 lbs per filter)

Clay and Sawdust were placed in mixer and mixed for 10 minutes.
Water was added and mixed a further 10 minutes.
20lbs. 3 oz. of the resulting mix was used as a charge for each filter
Four filters were made from this mix and were marked with Iron oxide with the date. *********************************************************

Ceramic Water Filter Test Batch No. 2
Saturday January 20th, 2007

Plainsman Red Earth Clay 44lbs. (20 kg) 87.5%
Screened Sawdust 6lbs.4.8oz (2.86 KG) 12.5%
Water 21 lbs

Mix Proportion represents 10% less sawdust then mix No. 1

********************************************************

Ceramic Water Filter Test Batch No. 3
January 22, 2007
Plainsman Red Earth Clay 44lbs. (20 kg) 88.7%
Screened Sawdust 5lbs. 9.6oz (2.55 KG) 11.3%
Water 20lbs
Mix Proportion represents 20% less sawdust then mix No. 1

******************************************************

Ceramic Water Filter Test Batch No. 4
January 24, 2007

Plainsman Red Earth Clay 44lbs. (20 kg) 85.1%
Screened Sawdust 7lbs. 11.2oz (3.5 KG) 14.9%
Water 21 lbs-9oz (9.8 KG)

Mix Proportion represents 10% more sawdust then mix No. 1

******************************************************

Ceramic Water Filter Test Batch No. 5
January 25, 2007

Plainsman Red Earth Clay 44lbs. (20 kg) 84%
Screened Sawdust 8lbs. 6.4oz (3.82KG) 16%
Water 26lbs (11.82 kg)

Mix Proportion represents 20% more sawdust then mix No. 1