Categories: Jump down the page to:
How does SC relate to coagulant dose and clarified water turbidity?
How does SC relate to zeta-potential?
Linear relationship, but the slope and offset from zero depends on the condition
of the particular piston and sensor. (It also varies between units)
What do the displays show?
The upper display is the most important. This shows how far away the measured SC
is from the target SC. This is what other single display SCMs show.
So "it's all relative"?
Yes
How do you find the target SC?
Use jar tests to determine the best dosage at a certain time, then set the SC
target on water with this dosage. The "set zero" sets the target to the
measured SC so that the upper display reads zero. As the raw water changes the
SCM can be used to automatically adjust the coagulant dosage to maintain the SC
at this level.
Why can't you set the target to filtered or de-ionized water?
SC from different types of water can't be usefully compared since SC is caused
by colloid particles in the water. The measured SC of a solution with no
particles is the SC of the piston itself with has no relationship with the
process.
Can it be used on raw water and the dose determined from that?
No. It saturates with high negative or positive readings and is extremely
sensitive to sensor wear and raw water changes. This type of feed-forward
control was used in the early days and didn't work that well.
How much change in turbidity can it handle without needing to be re-zeroed?
In theory it can handle almost unlimited changes in turbidity without having to
be re-zeroed, provided the nature of the turbidity particles is consistent.
However in practice, raw water sources vary not only in the quantity of
turbidity particles but also in both the nature of the suspended solids (eg
different clays have different colloidal characteristics) as well as in pH and
alkalinity. In addition, the sweep-floc effect (which is not related to charge)
is more pronounced at higher dosages.
We try to avoid giving a direct answer to the question of how much turbidity
variation requires a re-zero because it is so plant specific that any answer
could be dangerously misleading. The best thing a plant operator can do to
answer this is to ask themselves 'How much turbidity change signals that we are
getting different water?'. Such as water which takes a different path into the
river, or will have picked up different contaminants, rather than just more of
the same dirt.
Having said this, even if things have changed significantly, the coagulant
dosage determined by an SCM is still a great improvement over constant ppm
dosage, or dosage based on raw water parameters. The re-zeroing is only
required because it is no longer optimal.
Our raw water doesn't experience much variation, will a SCM be useful to us?
If you never, ever, need to adjust your coagulant dose then it will not be
helpful. However consider the following:
What about direct filtration or air flotation plants?
Unlike conventional clarification plant, these plants don't always want the
biggest and densest flocs possible. They will tend to work best with a specific
type and size of floc, this type of floc will normally correspond to a SC that
is a bit lower (more negative) than normal. The fact that they have a
relatively narrow band to aim for makes online control particularly useful, but
it is more difficult to establish the target SC.
Can you use it to control polymer or flocculent aid dose?
Not really. Unlike coagulant, flocculent aid has a mostly physical effect on
clarification. They entangle and mesh the colloids together to form flocs. Even
anionic or cationic polymers have only a small fraction of the effect on charge
as the coagulant. The effect of the flocculent on zeta-potential can normally
be ignored. In conventional plants the best coagulant dosage is normally
selected independently of flocculent dosage.
Can the SCM be used to control coagulant dose to achieve removal of organics or
colour?
The amount of dissolved material does not directly relate to streaming current.
Therefore if the main reason for coagulant dosing is to remove organics,
colour, taste and odour then a SCM will not be as helpful as it is for clay
suspensions. However if there is both turbidity and dissolved material in the
raw water then it can be beneficial to set the target SC to a slightly higher
value than required for optimal turbidity removal.
Why do we suggest people still do jar tests when some other scm sellers implyit
is unnecessary?
These people are oversimplifying. You should still do them from time to time,
just not nearly as often - maybe monthly, instead of daily. The jar tests are
now being used to determine the SC target rather than control the plant
directly.
Do we really need a controller?
The controller can be either an external unit, in the PLC or (optionally) built
into the Accufloc SCM. It is the unit responsible for taking the SC reading and
converting it into the dosing pump speed necessary to obtain the target SC.
If you want to use the SCM as more than just an operator advisory tool, then you
need a controller. The greatest benefits of having an SCM is the automation of
coagulant dosing, although, of course, it is still useful for detecting raw
water changes even by itself.
Why doesn't your unit compensate for pH changes.
pH effects the SC reading because pH effects the surface charge on the
particles. So, of course, pH changes can also alter the effectiveness of the
coagulation and clarification processes. The actual effect of changing pH will
depend on the water, the plant and the coagulant used. It is not possible to
compensate for pH within the Accufloc because none of this information is
available to the unit. All SC/pH compensation schemes are only applicable to
the specific conditions for which they were designed.
Big changes in pH may require the SCM be re-zeroed, because SC is not a perfect
indicator of the optimal point.
Still, it is much better to have a scm than not to have it. Dosing at a constant
ppm or basing dosage on raw water turbidity are still generally worse.
What range of pH / alkalinity / conductivity, etc will it work over?
A SCM can be used over the same ranges over which coagulation can occur. The
underlying physical processes are related. For example, coagulation fails at
less extreme pH levels than this measurement method does.
Is it necessary to control pH before coagulation to use a SCM?
No, it is not necessary, but it does produce better more consistent results.
If you experience significant pH variation then it is normally a good idea to
control pH anyway, whether a SCM is used or not.
Can your unit be used for titrations?
Probably. A pump in the sample bucket would be required to continuously
circulate the sample through the SCM. Just watch out for the effect where
dissolving gases gradually change the pH and (therefore SC) of the sample.
Sample Point Selection
Modern best practice generally calls for inline blenders or rapid mixers with a
'high mixing intensity'. This is good for us because it allows a well mixed
dosed-water sample to be taken as soon as possible after the mixer.
What is wrong with using a big open basin in the sample line as a grit trap ?
Long delay makes controller tuning harder and control response slower.
Also exposure of a freshly dosed alum solution to air changes both the SC and
the pH readings. Inline grit traps and filters are a much better option.
Does the flow rate through the instrument affect the reading?
Not significantly. The instrument measures the average charge on particles in
the water, how fast they arrive is not important. Higher flow rates result in a
faster response and less chance of clogging. This is why 2L/min is the
recommended minimum flow rate, little improvement is achieved at flow rates
above 4 L/min.
Can / should the SCM be switched off when the plant is not running?
Yes. It will last longer if turned off when not in use, the electronics can warm
up in a few seconds, while the time for a reading to full accuracy is 5-15
minutes after any significant process change.
It can also run continuously without problem, however if the flow rate through
the sample chamber falls to zero there is a elevated risk of damage to the
piston and sensor from particles and heating.
The sensor and piston should be cleaned prior to being allowed to dry.
What range of temperature does it work over?
The water temperature can be 5-35 degrees C.
The ambient air temperature can be in the range of -10 to 50 degrees C. Care
must be taken when installing the unit in very small, air-tight boxes - the
industrial motor generates quite a lot of heat.
How much shelter from the weather does it need?
The motor part of the instrument has a protection rating of IP56, therefore it
should not be exposed to direct weather or sprayed with water. In most
installations a rain-shield is sufficient, a fully water-tight enclosure is
only required if high pressure spray is expected.
How flexible are the orientation requirements?
The water can run in either direction through it, as the sensor is sideways
symmetrical.
However, the unit must be installed in the vertical orientation shown. This is
to avoid air becoming trapped in the sample chamber and being sucked into the
annulus during operation, this can result in an unstable reading.
Should a low flow alarm be installed on the sample line?
Sample lines in coagulated water can become clogged with floc buildup relatively
easily. If there is no manual monitoring of the sample flow rate then a low
flow alarm is advisable.
An insufficient flow through the sensor causes a very slow response. A zero flow
will result in a slow random drift of the reading, which will be meaningless.
Both of these conditions are often very difficult to detect in steady plant
operation. When automatic coagulant control is used, these conditions cause the
controller to become stuck at maximum or minimum dose, or to cycle between the
two.
Can we install the sensor and the analyser in separate locations?
Adding additional wiring to the sensor cable, beyond that installed at
manufacture, is not recommended. The extra length and joinings greatly increase
the chance of electrical interference disrupting the signal.
It is seldom actually necessary to introduce any separation because:
1 The analyser is not physically large compared to the motor and sensor and it
has greater water ingress resistance.
2 Remote monitoring and control can be easily added using the RS485 modbus card
with our software or a PLC, or using a remote 4-20mA display and the digital
input card.
Is it possible to rapidly switch its sample input between two streams, to allow
control of both independently?
This multiplexing is not recommended. Normally one SCM is required for each
independent coagulated water stream. The instrument's response to significant
changes in SC can take more than 10 minutes to settle. Our Accufloc SCM has
advanced signal processing which makes it one of the fastest responding SCMs
available, however as the sensor becomes more worn the response gets slower.
Use with multiple switched streams will result in significant interference
between the two readings, unless a very long time is allowed in between, which
would make it useless for online control.
What do the terms 'Zero point', 'set point' and 'target' mean?
They all mean the same thing. This is the value of SC that has been selected as
corresponding to the desired coagulation. The term zero point is used because
this is where the upper display reads zero. This is internal to the instrument.
What should the external controller SP be?
Zero, which will normally be 12mA. This is also the set-point of the optional
internal controller. Zero on the top display means that the SC is at the
target.
When would you want to adjust the span?
When it is poorly matched to the expected variation of SC. In almost all cases
the default calibration of the unit does not need to be changed.
Can it be calibrated to Zeta-potential units?
Yes, but as the piston gradually wears it will not stay calibrated like that.
Also this doesn't really help much.
Why doesn't the Accufloc scm have a gain adjust knob (eg x10 x100 x100)?
1. We can do a push-button span cal to any sample (span = 1/gain)
2. The need for frequent use of a gain adjust knob means SCM is drifting. This
is better corrected by the cause of the drift, which is probably worn parts.
3. The controller has its own gain setting (PB) which is for adjusting the
effect SC has on pump speed, nobody would consider adjusting the gain on a pH
meter because the pump response wasn't fast enough, yet people often do it for
SCMs.
Why is there a 4-20 span and a SC span? This confuses me!
The 'calibrate span' option does not affect the 4-20mA output. The 4-20 output
is based on what the display says and its range is determined by the 'rd4' and
'rd20' menu options. Leave it at the factory default if there is any doubt.
How should the piston and sensor it be cleaned?
The most useful cleaning tool is a common cheap toothbrush, medium firmness, as
can be bought at any supermarket.
We don't normally suggest use of a cleaning solution as it will leave trace
amounts on the surfaces and affect the reading for a significant time period.
Fluctuations?
Large seemingly random fluctuations in SC reading when the coagulant dose is
steady can be the result of loss of pH control. The SCM responds to pH changes
more quickly than most pH meters do. For correct operation of the coagulant
control the pH correction chemical dosage must change much more slowly than
coagulant dosage does. This is normally the case anyway.
Periodic cycling of the SC and coagulant dose can be caused by cycling of an
external parameter (such as flow or pH correction) with the same period.
Periodic cycling of the SC and coagulant dose can be caused by PID controller
tuning which is too aggressive for the SCM response speed. This may be because
sample line clogging has reduced the flow through the SCM, it should be at
least 2L/min. This can be corrected by ensuring the sample flow is high enough
and, if necessary, by increasing the controller's PB and Tr settings.
Contact us: sales@accufloc.com
Ph: +64 9 579 1888 Fax: +64 9 579 1888
Phone Free Within New Zealand: 0800 50 55 66
copyright © ACCUFLOC 2004