A Machine that turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog post: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge)

Apologies for my delayed response to latest posts on this topic.

The reasons why septage treatment plants fail are complex and vary from country to country. However, based on field studies in Indonesia and review of secondary information from the Philippines and other countries, I would say that the following are important:

1. Overestimation of demand, resulting in plants that are grossly oversized. This is a common problem in Asia and I suspect may also be an issue in Africa.
2. Poor design details, which make it difficult for operators to carry out routine operational tasks. The biggest problem here is usually desludging. This is a task that requires large effort at infrequent intervals and often gets delayed or neglected, with the consequence that facilities fail or underperform.
3. Selection of inappropriate technologies - in particular those that require regular inputs of chemicals, highly trained operators and, most crucially, a regular electricity supply.
4. Inadequate funding - septage treatment plants do not recover their operating costs and need to be subsidized from charges for septage collection and transport.
5. Inappropriate management systems - perhaps the most serious constraint on good operation and maintenance. This relates to the low priority that most municipal decision-makers give to wastewater and septage management. This, in turn impacts on the availability of funds and the ability to attract and retain appropriately trained operators. (Most operators are paid at the lowest rates and have no capacity to operate anything other than the most simple plants).

In my opinion, land availability is not a huge issue and land requirements can be reduced by careful design. There is something on this in my recent WEDC Conference paper but I will try to put together some notes for this forum.

When assessing the suitability of a technology such as the Omniprocessor, it will be important to consider (a) the likely loading and its relationship to the service area (b) the reliability of the electricity supply and (c) the ability to recruit, train and retain appropriately skilled workers and (d)ongoing operational costs in relation to likely income. I think that the onus is on those proposing new technologies to show how its use will be compatible with the likely loading, the availability of a reliable electricity supply, the finances available for operation and maintenance and the systems to recruit and retain effectively skilled workers. It is always good to remember Eric Dudley's two principles - 'how does this technology solve the problem?' and'what might go wrong?'

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
You need to login to reply
  • muench
  • muench's Avatar
  • Moderator
  • Freelance consultant (former roles: program manager, lecturer, process engineer)
  • Posts: 2106
  • Karma: 44
  • Likes received: 620

Re: Bill Gates blog post: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge)

Dear Andrew, Kevin and all,

I am following your questions and answers about the Janicki Omniprocessor with interest, please keep the discussion going.

I saw this image in the Guardian today - so a prototype of this machine is now in Dakar, Senegal:



Source:
www.theguardian.com/global-development-p...s-senegal-sanitation

(I have just mentioned this article also in another thread about use of mobile phone ICT for sanitation: forum.susana.org/forum/categories/197-mo...nitation-esani-msani )

Section from the article (nothing new to what we discussed above, I think):

Designed by Janicki Bioenergy under the backing of the Bill and Melinda Gates Foundation, the $1.5m omniprocessor dries faecal waste and burns it, to drive steam through turbines to produce electricity. Meanwhile, the steam produced when drying the sludge is filtered, condensed and treated to produce drinkable water.

The omniprocessor can produce a net power of 250KW of electricity – which could run 25,000 households – and around 80,000 litres of drinking water per day (enough for about 35,000 people) – which could be sold to the national grid or bottled like mineral water.

While Curtis welcomes this innovation in the much-neglected field of sanitation, she remains cautious about the complexity of safely collecting and transporting human waste from pit latrines to the plant.

Many of these pits are down very narrow alleys, inaccessible to emptying trucks,” says Curtis. “The waste processing plants are far away as nobody wants to live anywhere them – so you need a solid business model for waste collection.”


It will be interesting to know what comes out of the trials in Dakar!

Oh wait, I just discovered that the man himself who has done so much to raise public interest in the sanitation crisis already, Bill Gates, wrote about the omni processor on his blog. This is exciting! (OK, yes, he probably has staff members who wrote the article, but still).

See blog post here:
www.gatesnotes.com/Development/Omni-Processor-Update

Note the first paragraph:

You may recall that the JOP takes human waste and turns it into drinking water, electricity, and ash. (It is actually one of several Omni Processors being developed that treat human waste and produce something of value.)* It’s tempting to focus on the drinking water, for obvious reasons. But the goal is not to provide water. The goal is to dramatically improve sanitation for all the cities in poor countries.


This short video linked from the Bill Gates blog is good and worth watching ("Omni Processor in Senegal"):



You may or may not agree with the omni processor being a signification piece of the puzzle. But just because the name of Bill Gates is connected to this R&D work, the video already got 40,000 hits within 2 days of being uploaded! Our "normal" videos are lucky to get 2000 hits within six months... Whatever increases public interest in sanitation around the world, has got to be a good thing (only danger might be that unrealistic expectations could be created that the solution is really simple and that you only need the right technology to fix everything...)

Regards,
Elisabeth

* I used to think that this machine is "the omni processor" but I have since learnt that it is only one example of "an omni processor". There are other examples, too. See this Wikipedia page which explains it:
en.wikipedia.org/wiki/Omni_Processor

As expected, check out the spike in click rates of this Wikipedia article now that it's in the news again: stats.grok.se/en/latest/Omni_Processor
I didn't create the page but I made substantial improvements to it, see history page here:
en.wikipedia.org/w/index.php?title=Omni_...essor&action=history

Dr. Elisabeth von Muench
Independent consultant
Community manager of this forum via SEI
(see: www.susana.org/en/resources/projects/details/127 )
Wikipedian, co-founder of WikiProject Sanitation: en.wikipedia.org/wiki/Wikipedia:WikiProject_Sanitation

Location: Frankfurt, Germany
This email address is being protected from spambots. You need JavaScript enabled to view it.
Twitter: @EvMuench
Attachments:
You need to login to reply
  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Thanks to Elisabeth for the various links, giving further information on the Janucki omniprocessor. The links do provide information that helps to build up a better idea of what is happening and the objectives of the omniprocessor initiative.

Here are some points that may help the analysis. First, the population of Dakar in 2011 was 2.574 million From Tremolet Consulting report - tremolet.com/sites/default/files/downloa...FinalReport_WP4_.pdf . About 60% of the population was served by septic tanks or pits so the total potential population served by on-site sanitation facilities was just over 1.5 million. The total estimated septage production from this population was 1,017,842m3 per year, which works out at about 660 litres per person per year. (To put this into perspective, per-capita sludge accumulation rates in pit latrines in Africa are generally in the range 25 – 75 litres per person per year and the corresponding figure for Indonesia is less than 25 litres per person per year). The explanation for the high figure in Dakar appears to lie in the fact that much of the 'septage' removed from tanks is in fact water. (Presumably because of the high water table) The images in the video contained in the Gates blogspot referenced by Elizabeth confirm that the septage is indeed very watery. I would expect the moisture content of this septage to be up around 98% or even 99%.

According to the Janucki website ( janickibioenergy.com/s100.html ), the omniprocessor used in Dakar is of the S100 type. The website states that this has a capacity of 12.3 cubic metres per day and requires a moisture content not exceeding 84%. The larger and more recent S200 has a capacity of 92.3 cubic metres per day and the Janucki claim that it works with sludge water contents up to 100%. However, it requires a minimum dry solids weight of 9 – 11 tonnes per day. Assuming a generous 2% solids content in the Dakar sludge, 92 cubic metres of sludge would contain about 1.84 tonnes of dry solids per day and this will be the constraining factor unless another fuel such as solid waste is added to the omniprocessor. In theory, the S100 should not work with the moisture contents found in Dakar so it would be interesting to know (a) whether this was in fact a problem and (b) if so, how it was overcome.

To further complicate things, ONAS, the agency responsible for faecal sludge management in Dakar gives different information on its website ( www.pseau.org/outils/ouvrages/onas_boues_mag_n_3_en_2014.pdf ) This states that the omniprocessor deployed in Dakar can handle up to 450 cubic metres of raw sludge per day, with an average solids content of 1.5%. The ONAS site gives a similar figure for power output to that quoted by Janucki, which tends to confirm that it is indeed and S100 type, However, the information on the throughput of sludge is very different. This has important implications for planning - I would assume that the manufacturer's figure is correct.

This brief review of some key design parameters shows that there are some inconsistencies in the information that is available on the Janucki omniprocessor. It would be good if someone from Janucki, Gates or ONAS could provide some further information to clarify the situation, clearly set out the design parameters and say something about how experience in the field has either confirmed or led to modification in these design parameters.

One last point. I said in an earlier post that I would provide some information on the land requirements for a more conventional approach to septage treatment. I have made some rough calculations, based on an assumed 100,000 population, 80% with on-site sanitation, of which 50% are availing of regular (once every 4 years) pit/tank emptying services. These assumptions suggest that the amount of septage to be dealt with would be of the order of 40 cubic metres per day. This would include a large volume of supernatant water and so would be fairly weak and watery. I have assumed septage reception and screening, followed by hopper bottomed sedimentation tanks, an anaerobic baffled reactor and facultative and maturation ponds, with sludge drying beds for solids removed from the sedimentation tanks and baffled reactor. My calculations suggest that the land requirement would be of the order of 3000 square metres (0.3hectare), something over 10 times the footprint of the omniprocessor but still not very large. This arrangement is not electricity dependent. Most of the land is required for the ponds and drying beds and it would be possible to look at alternatives to both – perhaps high-rate filters to further reduce the liquid strength and planted drying beds, which require less area than unplanted drying beds.

Finally, it would be interesting to know what arrangements are in place for an independent assessment of the Dakar initiative. This is presumably fairly high on the list of priorities for the BMFG. If so, would it be possible to have an idea of when a detailed assessment will be generally available. This should, of course, include assessment of running costs, income from the sale of electricity and water and a discussion of problems encountered and the extent to which it was possible to overcome those problems.

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
You need to login to reply
  • pkjha
  • pkjha's Avatar
  • Long-term forum user
  • Working for over 30 years in the fields of sanitation, biogas from human wastes, septage management, waste water treatment in rural as well as urban areas in India and other developing countries.
  • Posts: 117
  • Karma: 10
  • Likes received: 37

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Dear Kevin

Your detail description on septage is very informative and useful. You are requested to kindly elaborate the required land. You mentioned that for 40 cum sepatage daily, total land requirement would be of the order of 3000 m2. It appears too high. What would be number and size of filtration beds (planted and unplanted)? For filtrate of less than 40 cum daily, land requirement for anaerobic sedimentation tanks, facultative and maturation ponds should not be so high. Such filtrate has very high BOD. If there is any published results, you may like to share.
Regards

Pawan

Pawan Jha
Chairman
Foundation for Environment and Sanitation
Mahavir Enclave
New Delhi 110045, India
Web: www.foundation4es.org
Linked: linkedin.com/in/drpkjha
You need to login to reply
  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Dear Pawan and others

Here are my assumptions and a summary of my calculations for my assessment of the area needed for a conventional 'low tech' septage treatment plant. The calculations are approximate and Pawan is right to say that the calculated area is probably on the high side - I was being quite generous with my calculations as I did not want to be seen to be slanting the calculations to favour this approach.

I have already briefly described how I arrived at the figure of 40m3/d of septage to be treated so will start with that figure. My calculations also showed that a high proportion of this septage would be supernatant water so I have assumed a 2% solids content by weight.

Other assumptions are as follows:

Sedimentation tank area calculated on a flow figure intermediate between instantaneous discharge from a single tanker and likely maximum hourly rate of discharge. (This could be refined but does not greatly affect the calculated area)
Anaerobic baffled reactor area assumed - this could also be refined but again does not make a big difference
BOD removal 30% in hopper bottomed sedimentation tanks and 70% in ABR
70% of influent suspended solids reaches sludge drying beds
Area of facultative pond calculated using Arthur's equation = Load (kg/ha) - 20T - 60 where T is the centigrade temperature - which I would normally assume is the average temperature over the coldest three months of the year. For this calculation, I assumed a design temperature of 20 degrees centigrade - it would be higher in Indonesia, the Philippines and West Africa and somewhat lower in northern India and other places where there is a relatively low winter temperature.
Loading on sludge drying beds - 200kg dry solids/m2.yr
I did not calculate the area of maturation ponds required but assumed that this would be the same as that of the facultative ponds - this is almost certainly an overestimate.

Based on these assumptions, I arrived at the following areas:

Inlet works, including screening and space for trucks to turn and back up 150m2
Hopper bottomed sedimentation tanks (2no 4m x 4m + civil works 50m2
Anaerobic baffled reactor 150m2
Facultative ponds 500m2
Maturation ponds 500m2
Sludge drying beds (560m2 + allowance for dividing walls 600m2
TOTAL AREA FOR UNITS 1950m2
In my original rough calculation, I assumed a further 50% for circulation, office building etc, bringing the total area up to just under 3000m2. This could probably be reduced to say 25%, reducing the total area required to just under 2500m2.

I would stress again that these are approximate calculations, intended only to get an idea of the order of magnitude of the area required I am intending to refine them, and will eventually hope to share the refined calculations on the forum.

Any feedback on this would be welcome.

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
You need to login to reply
  • pkjha
  • pkjha's Avatar
  • Long-term forum user
  • Working for over 30 years in the fields of sanitation, biogas from human wastes, septage management, waste water treatment in rural as well as urban areas in India and other developing countries.
  • Posts: 117
  • Karma: 10
  • Likes received: 37

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Dear Kevin and others

Calculation is perfect. In fact I did not calculate office space. But it is required.
In India, as on date, there is not a single project completed at town level for septage management. First project has just been sanctioned for some town in the Rajasthan State, funded by ADB and State Government. I’ll be involved in that project. One RFP for another project on septage management for a few towns in the same state, funded by the ADB, BMGF and State Government, has been issued recently. Last year a similar project for Warangal town in Telangana State was funded by BMGF to ASCI, Hyderabad. I prepared drawings etc. The project has not been implemented so far. In the absence of technical data based on local conditions, it is a bit difficult to prepare actual drawing. Load of sludge drying bed @200 kg/m2.year is taken in designing, as you also mentioned. BOD of filtrate- an important parameter, is not known. TS is also not known. However, we take higher values in designing.
In designing filtration beds and treatment chambers, one major challenge is that actual quantity of septage to reach at the site is expected to be much lower than the calculated quantity. The reason is simple- calculation is based on 2-3 years cleaning interval of septic tanks. However, in practice people get septic tanks cleaned only after it is completely filled and there is blockage in flow of water from toilet pan. Normally it takes 5 years or more, depending on the size of the tank, to get it cleaned. In such case there would be unnecessary requirement of land. However, quality of treated water would be much better due to higher HRT in different chambers.

Regards
pawan

Pawan Jha
Chairman
Foundation for Environment and Sanitation
Mahavir Enclave
New Delhi 110045, India
Web: www.foundation4es.org
Linked: linkedin.com/in/drpkjha
You need to login to reply
  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Dear Pawan and all

Here is a slightly belated response to Pawan's last contribution. It is interesting that there is an initiative in Warangal. Several years ago, I was part of the consultants team for the Andhra Pradesh Urban Services for the Poor (APUSP) project and Warangal was one of the project towns. At that time, there was a pilot initiative in Warangal to test the constructed wetland option for wastewater treatment. From what I remember, the wetland had a free water surface rather than the more normal arrangement of having the water level below the level of the gravel bed. When I looked at it, my impression was that it needed some work to clarify parameters if it was to be of any real use for testing the constructed wetland approach. Not directly relevant to FSM but it shows that there were efforts to explore new ideas in Warangal.

Now on to my main point. Pawan's observation about actual septage quantities being less than design quantities is consistent with findings from elsewhere. We reached the same conclusions in Indonesia, where people often go for years without emptying pits. The figures quoted in WSUP publications regarding the number of pits emptied by the 'Dream Tean' in Kanyama, Lusaka lead to the same conclusion, the rate at which pits emptied shows that the average emptying interval is much greater than the 2 - 3 days commonly assumed. Even in the Philippines, where there have been serious attempts to introduce scheduled emptying, loads delivered to septage treatment plants have been less than assumed in design. Even in the United States, I came across a reference to the number of septic tanks emptied each year in Florida, which was much less than might be expected on the number of on-site sanitation facilities and a 3 - 5 year emptying cycle. I am putting together some notes on to provide input to a possible book on aspects of septage treatment. The other issue for design is the strength of the septage - the available information, which unfortunately is quite limited, suggests that the strength is often lower than the typical design assumption of 5000mg/l BOD.

These are important issues, which arguably should be receiving as much research attention as efforts to develop completely new technologies. It would be good to hear from others on efforts to obtain basic information on septage quantity and strength for design purposes.

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
You need to login to reply
  • Tore
  • Tore's Avatar
  • Regular forum user
  • Posts: 28
  • Karma: 2
  • Likes received: 8

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

I am amazed at all of the first world solutions to developing country problems. Who maintains the equipment? Who pays for the equipment? Where will the people come from that will maintain the equipment. Where and who pays for the repair parts. We have all seen unusable equipment that was originally installed to accomplish a purpose but was abandoned when it broke. KISS, Keep It Simple Stupid. There are many low tech solutions to accomplish the same thing. A UDDT will purify the excrement and provide fertilizer and soil conditioner.
You need to login to reply
  • F H Mughal
  • F H Mughal's Avatar
    Topic Author
  • Long-term forum user
  • Senior Water and Sanitation Engineer
  • Posts: 750
  • Karma: 19
  • Likes received: 190

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

I endorse the viewpoint. I have myself seen new equipment lying rusting, as people here simply don't know how to operate it. I uphold that motto: Keep it simple.

F H Mughal

F H Mughal (Mr.)
Karachi, Pakistan
You need to login to reply
  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Hello everyone

I am returning to this topic after almost a year as I have been looking at the Omni Processor in the context that I am writing on faecal sludge and septage treatment. There is now more information on the internet about how the system work which is helpful in making an assessment. I have been doing some calculations on the energy input and energy output.

In essence, the machine heats water to the point at which it becomes steam. This explains why the water is drinkable. Lets assume that the water starts at 30 degrees CentigraTde. Energy is required to heat it to 100 degrees centigrade and then vaporize it.

The energy required to heat water is 4.186 Joules per gm per degree C
Based on this figure, 293kJ are required to heat 1 kg of water to 100 degrees centigrade
The energy required for vaporization is 2260 joules.gm or 2260kJ/kg
Thus, the total energy required per kg pf water is 293 + 2260 = 2553kJ
The energy required for one cubic metre will be 1000 times this or 2553mJ

Muspratt and others have assessed the energy that can be extracted from dried faecal sludge at about 17.3mJ per kg of total solids

The Omni Processor is installed in Dakar, where the septage is very watery with an average of about 4kg of total solids per cubic metre (0.4%)

So the total solids extracted from one cubic metre of water will be about 4 x 17.3 or 69.2mJ

This is only 2.7% of the energy required to vaporise the water

I am struggling to understand how the system can possibly work without substantial external energy inputs. Am I missing something? Can anyone enlighten me on this?

Kevin

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
You need to login to reply
  • Tore
  • Tore's Avatar
  • Regular forum user
  • Posts: 28
  • Karma: 2
  • Likes received: 8

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

Why do we have such a hard time remembering the KISS rule, Keep It Simple Stupid? This system is not simple. Hoping and praying that a developing country has the expertise and the parts on hand to fix the break downs is ludicrous. If the money that is used to build and transport the device is instead used to educate and build small easy to fix and maintain systems (example UDDT) more people would be helped. We have all seen the relics of past good intentions that are scattered throughout the globe.
Sanitizing feces is not rocket science. Systems that work can be easily taught to the local populations. Education and buy in is crucial.
The following user(s) like this post: canaday
You need to login to reply
  • kevintayler
  • kevintayler's Avatar
  • Long-term forum user
  • I am a water and sanitation engineer, also interested in general urban housing issues. In recent years, I have worked on FSM for various organizations
  • Posts: 64
  • Karma: 9
  • Likes received: 31

Re: Bill Gates blog posts: This Ingenious Machine Turns Feces Into Drinking Water (Omni-processor by Janicki to process sludge) - now in Dakar, Senegal

For some reason, I did not complete the uploading of my first post today so I am resending.

In essence, I am answering my own question of a few days ago. I have now found a powerpoint presentation by Peter Janicki, the man behind the Janicki Omni Processor, which contains calculations that confirm my conclusion that the system is dependent on an external power source if the solids content of the incoming sludge is less than 10% - probably less than around 15% if efficiency losses through the system are taken into account. For solids contents less than this, the system starts to become very expensive to run because it depends on converting water into steam, which requires upwards of 700kWh per cubic metre. Most of this must come from external sources if the solids content of the septage is low.

So, regardless of the complexity, which is certainly a big issue, the maths shows that the system is not self-sufficient for septage such as that in Dakar, where the average solids content of septage is less than 1%. The options are either to import power or to pretreat the septage to produce a sludge with a solids content of at least 15%. both responses mean that the machine can no longer be viewed as an Omni Processor. So, yes there is an element of KISS about the whole thing but the main point that I would take from all this is the absolute need to do some simple sums before embarking on a project that costs millions and perhaps diverts much needed resources away from other initiatives.

On the role of the private sector, all that I would say is that we must remember the need to think in terms of complete sanitation system and the historic evidence is that the public sector has played a major role in dealing with aspects of the complete sanitation service chain that have public good characteristics - particularly treatment.

Kevin Tayler
Independent water and sanitation consultant
Horsham
UK
The following user(s) like this post: goeco
You need to login to reply
Share this thread:
Time to create page: 0.807 seconds