Drinking Water Treatment Plant 2

[alienitmeca]

Hello everyone

When desinging a water treatment plant, what are the main process methods used.
I should mention that the plant is set to produce a regular flow of 1 cubic meter per second (Q = 1 m3/s) and it could go up to 1.25 m3/s
Coagulation and flocculation are going to be used in the process

Any information will be very helpfule at this stage.

Thanks

 

[ashtree]

I am sorry to say it but you will have provide more information.

While it is good that you have provided the flow rate required the other two important requirements are:

1) What is the quality or the range of quality parameters of your water source: pH , alkalinity , turbidity , suspended solids, conductivity, TOC, DOC are just some of the parameters that should be determined. Historical information that records the variability of those parameters would be most helpful.
2) What are the quality requirements for the treated water: Do you require a drinking standard?

Once that information is known you can then start to shape what sort of process options are available. You might also want to look at risk management within the process if the quality and quantity requirements are strict and require a high degree of reliability. This might also guide your process selection. Other factors that might help guide you are:
1) Availability of power
2) What chemicals are available
3) Are skilled staff available to operate and maintain the plant
4) What options are available for the treatment and disposal of residuals.
5) Budget

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[QualityTime]

I agree with Ashtree. What is in the water? You need to understand your source water and what can affect the source water before a treatment method can be selected. There are a whole bunch of well known water sample testing parameters that must be done

 

[bimr]

Source and quality of the raw water is needed.

 

[alienitmeca]

Thank you all for your replies,

I believe my questions should have a troubleshooting aspect, therefore, I am going to ask about Coagulation:

The main parameters in the design of a coagulation unit are as follows:

- G (s[sup]-1[/sup]): Velocity gradient

- T (s): Retention Time of the mixture; water-coagulant in coagulation tank

As I understand, and coorect me if I am wrong, untreated water enters into the coagulation tank, a coagulant (Aluminum sulfate) is simultaniously added to the unit, whereas an agitator must make sure that the mixture is homogeneously agitated as water flows through to the next stage of treatment (Flocculation).

Is there a a method to calculate the retention time of the mixture in the coagulant unit, or it must be determined by jar test during start up ?

I found several documents about CSTR (continious stirring tank-reactor) but I still can't figue it out.

 

[ashtree]

Mixing energy is probably more important than retention time in the rapid mix stage of the process. The actual time it takes for coagulation to occur depends on pH and the type of coagulant being used. For example the formation of Al (OH)2+ takes only a fraction of a second. For practical purposes generally rapid mix process would generally have at least a few seconds to perhaps 30 secs of retention time. Mixing energy is normally at least about 1000Gs-1 but is often more. Once again this depends on the coagulant and process aims. If you are trying to form sweep floc short times are less critical.

Jar testing will be of some assistance but its generally difficult to test really short coagulation times on a jar tester. Typically you would use between 5 and 30 secs of intense mixing on a jar tester, and would try to simulate the actual plant conditions(which can be quite difficult as well). Ideally whatever you design should be designed to be variable. This can be done by altering the dose point , or using variable speed mixers, or having flow paths that can be arranged in series or parallel as required to alter detention times.

Although getting the design numbers right is important it is ultimately about what works and this can involve a bit of trial and error sometimes.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[bimr]

RE: Is there a a method to calculate the retention time of the mixture in the coagulant unit, or it must be determined by jar test during start up ?

The retention time of the flash mix basin should be from 1 to 5 minutes.

The flocculation basin should be sized to provide a retention period of from 20 to 40 minutes.

The size of the basins is determined during preliminary engineering design of the treatment plant based on the flow and raw water quality.

Once the water treatment plant is operational, the dosage of coagulant should be determined each day through jar testing.

https://www.usfa.fema.gov/downloads/pdf/publications/water_supply_systems_volume_i.pdf

 

[ashtree]

bimr is correct .Just to qualify my comments i was referring only to the initial mixing of the coagulant chemical with the incoming water. This is something that many designers overlook. However this should be done correctly if chemical usage is to be optimized. There are many plants particularly, older ones and many package type plants do not impart enough mixing energy quickly enough at the beginning of the process and the downstream process suffers.

Typically you should have:

1) A few seconds of high intensity mixing at >1000Gs
2) 1-5 minutes depending on water temperature of moderate intensity mixing about 300Gs
3) 20-40 minutes of low intensity mixing 50Gs.

If your water temperature is around 25 degrees C or higher than aim for the shorter end of the time range. But at 10 degrees C or lower than you need to be at the longer end of the time. Ideally the setup will allow a gradual reduction in energy between stages , and a wide degree of variation in detention times and mixing energy.

If you are going to use alum you need to give careful consideration to your pH and alkalinity as well.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[alienitmeca]

bmir, you said that "The size of the basins is determined during preliminary engineering design of the treatment plant based on the flow and raw water quality"

I am going to consider my project as an example:

-Turbidity of the raw water = 2000 TNU (high turbidity water)
-flow rate needed is = 1 m3/s

I am wondering about two things:

- If the water turbidity decreases, does the size of the flash mixer basin decrease as well ? or should I think about the mixer velocity and coagulant injections and keep the basin with the same size ?
- How may determine the size of the flash mixer in the first place ?

Thank you in advance for your help

 

[Compositepro]

What is Gs? The units for shear are reciprocal seconds.

 

[bimr]

If the water turbidity decreases, does the size of the flash mixer basin decrease as well? No

or should I think about the mixer velocity and coagulant injections and keep the basin with the same size ? Yes

- How may determine the size of the flash mixer in the first place ?

A very large plant would have the smaller 1 minute retention. The cost of the mixing basin tends to be expensive as the plant size increases.

 

[alienitmeca]

Thank you for your responses,

In our project, we are opting to 3 coagulation chambers mounted in series, we still do not have reference to the size of the chamber,

hears an example for 2 m3/s water treatment plant,

Size of the chamber: H x L x l = 4.7 m x 4.9 m x 4.7

Then volume of the Coagulation chamber is then 110 m3,

This seems to be over-sized, but again, I this is the fist time I design a water treatment plant, is there a method to verify whether it is an optimized sizing

Thank you in advance.

 

[bimr]

For a system of that capacity, it would be worth the expense to run a pilot or bench top study to confirm the process design.

 

[QualityTime]

In municipal water treatment plants a flocculation tank hydraulic retention time of 20 – 40 minutes is typical. Thirty minutes is more typical. That means each stage of a tapered 3 stage flocculation tank has a hydraulic retention time of 10 minutes. You need to be conservative because water temperature has a big influence in the chemistry of flocculation. You can always take tanks off line if you have too much flocculation tankage capacity but if you are short you are scuppered. Flexibility costs money.

If you are using hydraulic spiral mixing, typically, the water velocity discharging into each tank is 6 ft/s in the first stage, then 3 ft/s in the second stage and finally 1.5 ft/s going in the third stage. Basically in the first tank you are mixing the flocculant in the water. By the time you get to the third tank you do not want to shear the flocc that has formed and hence the low discharge velocity. By the time you have exited the third tank you will have formed the pinpoint flocc that you want.

If you are using mechanical mixing, the same principals apply. G Factors are kind of a “black magic” art with respect to the mixer manufacturers. The shape of their mixing blade means everything. Talk to them and they will give you a practical explanation of G factors and how they go about it

 

[QualityTime]

hears an example for 2 m3/s water treatment plant,

Size of the chamber: H x L x l = 4.7 m x 4.9 m x 4.7

Then volume of the Coagulation chamber is then 110 m3,


The numbers say if there are 3 coagulation chambers that each have a volume of 110 m3 you have a 55 second HRT in each tank at the design flow of 2 m3/s. Am I missing something or are there more sets of coagulation tanks. Not only that the coagulation tanks should have square sides in plan

 

[ashtree]

I think the problem is that the language being used is "mixed". That was why i went back and clarified what i said initially.

Coagulation by definition is the addition of a chemical coagulant or coagulants for the purpose of conditioning the suspended, colloidal and dissolved matter for the subsequent processing by flocculation or to create conditions that will allow for the subsequent removal of particulate and dissolved matter.

Flocculation is the aggregation of of destabilized particles and precipitation products formed by the addition of coagulants into larger particles known as flocculant particles (floc).

The OP talks about a "coagulation" unit. I suspect the OP is referring to what would be often referred to as a flash mixer or rapid mix unit.
Coagulation as a chemical reaction is quite quick but is probably only a few seconds at most depending upon temperature and the chemical used. Flash mixing normally involves short periods of high intensity mixing with a gradual reduction in energy as the water moves into the flocculation stage.

So for "coagulation" the 3 x 110m3 chambers is probably about right(depending on a few things). The chemical probably should be added somewhere up the pipe and be followed with an inline mixer. Each chamber than should have a mechanical mixer with variable intensity with each successive chamber being less than that preceding it. Ideally it would be good to be able to take one or more of the chambers off line if its not required under some conditions. These coagulation chambers would then feed the flocculator which has the 20-40 mins of detention time that is being talked about.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[QualityTime]

Personally I have never heard the use of a coagulation tank but I have used inline pipe flash mixers in front of tapered 3 stage flocculation tanks to get the mixing of the chemical with the raw water.

If you think about it, a flash mixer is supposed to mix the coagulant instantanously with the raw water right then and there. The purpose of the tapered 3 stage flocculation tanks is to provide the time to grow the flocc. The flocc tanks also provide further mixing of the coagulant in the raw water

 

[ashtree]

2000NTU is not that unusual for a dirty river particularly during flood times. Putting the water in a storage with a few weeks detention time might get a lot of it to settle out. Some colloidal clay just never settles though.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"