AU665907B2

AU665907B2 - Device for measuring dissolved oxygen demand

Info

Publication number: AU665907B2
Application number: AU63007/94A
Authority: AU
Inventors: Ron Lee; Mervyn Ogston; Denys A. Wickens
Original assignee: Commonwealth Industrial Gases Ltd
Current assignee: Commonwealth Industrial Gases Ltd
Priority date: 1993-05-28
Filing date: 1994-05-10
Publication date: 1996-01-18
Anticipated expiration: 2014-05-10
Also published as: AU6300794A

Description

~i~ 66590.7 1

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIF I CATION FOR A STANDARD PATENT

ORIGINAL

Name of Applicant: I I 'It' r I trfi 'S S ft.., p...r Pt #0 Actual Inventors: THE COMMONWEALTH INDUSTRIAL GASES LIMITED, A.C.N. 000 029 729 Ron LEE, Denys A. WICKENS and 'ervyn OGSTON SHELSTON WATERS Clarence Street SYDNEY NSW 2000 "DEVICE FOR MEASURING DISSOLVED OXYGEN DEMAND" Address for Service: Invention Title: Details of Associated Provisional Application No: PL9072 dated 28th May, 1993 The following statement is a full description of this invention, including the best method of performing it known to us:nommooft I i -2 .i

III.

I qI4* *1s

III

I. I The present invention relates to a method and device for continuously measuring oxygen consumption rate in oxygenated liquids or slurries.

The invention was developed primarily for use in measuring oxygen consumption rates in biological processes such as sewage treatment and will be described hereinafter with reference to this application. However, it will be appreciated that the invention is not limited to this particular field of use.

There are a large number of commercial chemical and biological processes that perform oxidations on natural, uncontrolled, or otherwise variable feed stocks using molecular oxygen as the terminal electron acceptor and source of driving force. Major examples are the 15 oxidation of mineral slurries, as in gold extraction, and solidification of dissolved and suspended pollutants, as in sewage and other waste water treatment.

The major controllable operating expense in these processes is the supply of dissolved oxygen. Typically, system performance improves until some threshold value of dissolved oxygen (DO) is reached, after which an excess does not provide any further gains. Accordingly, designers and operators of such systems attempt to balance the cost of supplying an excess of dissolved oxygen against the penalties of a short fall. This is as important at the design stage as it is during actual process operation.

The technical challenge arises when there is no i 1 i;: 3 rapid, reliable method of judging process performance.

This is almost always the case where the raw material, such as sewage, has uncontrolled variability. For example, the principal test for the effectiveness of biological treatments of waste water, known as the Biological Oxygen Demand (BOD), takes 5 days to complete, yet the oxidation process itself rarely takes more than 24 hours and is more usually 4-8 hours to completion.

In such cases, the only option to judge the correct level of oxygen supply, is to measure the consumption rate of dissolved oxygen while the process is actually operating. At the design stage, this is generally achieved by operating pilot scale equipment.

Various attempts have been made to measure the 15 oxygen consumption rate where oxygen is not limiting. In biological systems this is often termed the aerobic respiration rate, or simply the respiration rate. No previously available device has yet met with complete industry acceptance. In this regard, there have been 20 severe problems with reliability and accuracy, both long For example, most batch sample measuring techniques utilise sample sizes which are too siall for credible final results as these are not representative of the non-homogeneous bulk. Furthermore, there have been excessive delays between sampling and testing.

Accordingly, these results have a limitation in usefulness as a control parameter due to infrequent i i ttt i a,.

''tag 7: 4 measurements, Even with substantially continuous side-stream testing processes and devices, there is a problem with inaccuracy, because high rate data measurement is very much a function of detector response time. Other inaccuracies arise due to undetected instrument drift in the dissolved oxygen sensors that are used and in the mechanical unreliability arising from low rate pumping of slurries containing fibres and large particles.

Another problem with nearly all the prior art equipment are the inaccuracies arising due to fouling of measuring equipment with oxygen-consuming growths.

It is an object of the present invention to substantially overcome or at least ameliorate one or more of the above discussed disadvantages of the prior art.

According to a first aspect of the invention there is provided a device for continuously measuring oxygen i consumption rate in oxygenated liquids or slurries by extracting and measuring a representative test stream, said device comprising:

I

a plug flow reactor in the form of a predetermined length of pipe connected between liquid inlet means and liquid outlet means; means to pump a test stream of liquid through said inlet, reactor and outlet means; an oxygenator to admit dissolved oxygen to said liquid test stream in or adjacent the inlet means; at least two dissolved oxygen sensors spaced apart to measure dissolved oxygen levels at predetermined points along the reactor; means to measure the liquid flow rate past said sensors; flow reversal means to selectively reverse the flow of licjuid through said reactor and past said sensors; and control and monitoring means to compare the output from said spaced apart sensors and selectively actuate flow reversal.

According to a second aspect of the invention there is provided a method for continuously measuring oxygen r consumption rate in oxygenated liquids or slurries, said 4rfI method comprising the steps of: *4*4 extracting a representative test stream of said 15 liquid; passing said test stream through a plug flow .sIjl reactor in the form of a predetermined length of pipe .connected between liquid inlet means and liquid outlet means; 20 injecting dissolved oxygen into said liquid test stream in the inlet means; measuring the dissolved oxygen levels with oxygen sensors disposed at at least two spaced apart points along the reactor; measuring the liquid flow rate past said sensors; comparing the output from said spaced apart sensors with control means and generating a signal indicative of the difference; and 6 selectively reversing the liquid flow through said reactor.

Desirably, the flow is reversed on a periodic basis and/or when the signal indicates a sudden change in respiration rate, to verify the readings from the spaced apart sensors and/or identify sensor drift.

Desirably, the control and monitoring means calculate the dissolved oxygen drop across the reactor and determines the reactor residence time from the measured fluid flow rate and relevant reactor volume. It is preferred that the liquid flow rate measuring means is located prior to the oxygenator.

Preferably, the minimum pipe reactor residence time is two minutes in order to establish a reliably S 15 measurable dissolved oxygen drop 0.5 mg/l) across the reactor at the minimum expected consumption rate, but not so much that unreasonable starting levels of dissolved oxygen might be needed 15-20 ppm).

In a preferred embodiment suitable for use with 4 rr O..o 20 sewage, a submersible centrifugal chopper pump is selected, preferably with a recessed impeller.

ao o oFor use in biological systems, the system is desirably designed to provide a scouring velocity in the pipework of at least 0.3-0.45 m/sec, and preferably 0.8-1.0 m/sec to maintain biological solids in suspension and to prevent slime growth on the reactor walls.

Preferably, the oxygenator takes the form of a U-tube pipe dissolver with bubble coalescence and 7 7 recycle. This is chosen to maintain plug flow and to minimise the residence time between sampling and testing, although any dissolver which injects an oxygenated essentially bubble-free liquid or slurry into the reactor could be used.

Desirably, the oxygenator is located immediately before the reactor to ensure that short term effects immediately after oxygenation are included in the measurements. It is preferred that pure or substantially pure oxygen is used as a source of dissolved oxygen to produce dissolved oxygen levels in the range of 10-40 I- :mg/l, and to minimize or eliminate carry-over of inert rrrr ,gases, such as nitrogen, into the reactor.

(lot at,' In a preferred embodiment, the flow reversal means a 15 comprises two paired three-way valves operated in tandem to direct flow in either direction through the reactor, i A preferred embodiment in the invention will now be described, by way of example only, with reference to the accompanying schematic drawing.

i 4 20 The device shown generally at 1 comprises a plug I flow reactor 2, in the form of a predetermined length of too# o pipe 3, connected betweer il.quid inlet means 4 and liquid outlet means 5. The outlet 5 (not shown in full) is connected back with the liquid aeration basin 8.

A submersible centrifugal chopper pump 7 is provided in the liquid aeration basin 8 and has connecting pipework 9 joining to inlet 4 via an oxygenator shown generally at

I

The oxygenator 10 is in the form of a U-tube dissolver 11 having an oxygen feed conduit 12 located at the uppermost horizontal section 13. Adjacent the exit of the dissolver, a recycle pipe 14 is provided for returning undissolved bubbles of oxygen to the top of the dissolver through feed conduit 12.

The reactor inlet means 4 and an outlet means 5 are each divided prior to connection to the reactor 2 by means of respective pieces 15 and 16. Each of the resulting divided lines then attach to one of two automatic three-way valves 17 connected to the reactor 2.

J i Dissolved oxygen sensors 18 and 19 are also disposed at either end of the pipe reactor 2 as shown, It$' .and a liquid flow meter 20 is provided in the pipework 9 8 I 4 15 at a location prior to the oxygenator :In use, the submersible chopper pump 7, mounted below the liquid level in the aeration basin 8, delivers a pumped test stream of sample liquid into the connecting pipe work 9 and past the flow meter 20, which preferably I 20 measures the liquid flow before any gases are admitted to the system. Oxygen is then added to the test stream in the U-tube dissolver 10 by injection of gas through inlet 12 which is fitted with a check valve (not shown). As the test stream of liquid and injected oxygen descends 25 the second vertical section of the U-tube, oxygen is dissolved in the liquid and any undissolved bubbles are returned to the top of the dissolver via the recycle pipe 14.

I

The resulting oxygenated test stream then passes through inlet 4 and is divided at piece 15 into two separate streams which connect to the two three-way valves 17. Also connected with the valves 17 are the two split sections of the outlet 5 which join at piece 16. The automatic three-way valves 16, flow meter 20 and dissolved oxygen sensors 18 and 19 are all connected with a suitable controller (not shown).

The controller selectively switches the valves 16 to permit the flow of test stream liquid in a predetermined direction through the pipe reactor 2. This flow passes past oxygen sensors 18 and 19, the output 22. from which is monitored by -:he controller.

I* "Every twenty-four hours, or any other convenient S 15 prolonged period of operation, the automatic valves are switched, via the controller, to reverse the direction of ~flow through the pipe reactor 2.

e. The significant effect of this direction change is to convert oxygen sensor 18 from reading, say, the inlet i 20 concentration to reading the outlet concentration. The other meter 19 is then likewise changed from reading the outlet concentration to reading the inlet concentration.

Control of the system is effected in a fairly standard manner. Limit switches (or other position indicating or output monitoring devices) are provided on the automatic valves 17, the flow meter 20, and both dissolved oxygen sensors 18 and 19 which in turn are connected to electronics, such as a programmable logic i0 controller (PLC). This unit calculates the oxygen consumption rate from the difference in dissolved oxygen readings from the two meters and determines the relevant flow-rate from an average velocity reading and the main reactor volume, which may be calculated from the pipe dimensions (length and internal diameter).

The PLC may also be linked to alarms to attract operator attention if flow fails, or if the change in calculated oxygen consumption is excessive when flow reverses, thereby indicating a need for non-routine sensor re-calibration.

The controller can also be made to reverse the flows Pt any time the respiration rate changes suddenly or radically to ensure that meter drift is not the true j 15 cause.

Optionally, the PLC may also be used to control the oxygen flow to maintain the steady dissolved oxygen level coming into the reactor, or to ramp the feed of dissolved oxygen up and down to monitor the systems response to a i given dissolved oxygen level. The control system can i also be used to switch off the oxygen flow for safety tiff reasons if the liquid flow stops.

When designing a system specifically for use in sewerage treatment, the minimum sample flow rate is likely to be about 3m 3 /hr, which is dictated by size availability for subn.ersible centrifugal chopper pumps.

This is likely to be adequate for most waste water treatment basins.

L 1 i sil i i i; 1 i _I -II 11 Furthermore, the minimum pipe residence time for slurries of this nature is approximately two minutes.

This is necessary in order to establish a reliably measurable dissolved oxygen drop 0.5 mg/l) across the reactor at the minimum expected consumption rate mg/l/hr), but not so much that unreasonable starting levels of dissolved oxygen might be needed 15-20ppm).

The system was also preferably designed to provide a scouring velocity in the pipe of approximately 0.8-1.0/sec, to maintain biological solids in suspension and to prevent slime growth on the reactor walls.

This minimum scouring velocity requirement sets the reactor pipe diameter to less than about 65mm. This in turn sets the minimum reactor pipe length to around (based on 2 mins residence time and 3m /hr pump rate in a 50mm diameter pipe).

The preferred oxygenator takes the form of a U-Tube pipe dissolver with bubble coalescence and recycle as described. This is chosen to maintain a plug flow and to minimise the residence time between sampling and testing. It is also preferred that pure or substantially pure oxygen is used as a source of dissolved oxygen to produce levels in the range of 10-40mg/l.

It will be appreciated that the choice of pump (and hence, the mininum practicable sample flow rate) is determined by the characteristics of the fluid. The ideal sample flow rate is determined by the ratio to the bulk flow and bulk mixing effectiveness to get 1 r 12 representative sampling. The final consumption rate value is critically dependent on the stability of the dissolved oxygen sensors, particularly since a differential value is required from two independent units. The characteristics of dissolved oxygen analysers are that the zero does not drift, but that the span may.

This is the reason for adding flow reversal to normal routine calibration checks for the device. If, say, the span has drifted by plus 10% on the inlet meter only, and the true dissolved oxygen change is from 10 mg/l to 6 mg/l, the apparent net reading will change on flow reversal from (10 1.1 6 5 mg/l to (10 6 1.1 3.4 mg/l. This will be easily detected by the control system.

A 15 The method and device for continuously measuring oxygen consumption according to the invention, offers a substantial number of advantages over the prior art.

9 For example, by providing a reliable continuous ,cc measuring process, the errors associated with non-representative sampling are substantially reduced.

Furthermore, oxygen-consuming slime and growths are eliminated in the reaction chamber by ensuring that the flow rate is sufficient to maintain the predetermined ,,,fla scouring velocity. As the system is also very i mechanically simple, mechanical reliability problems resulting from pumping low speed flows of liquid containing large solids are substantially reduced.

One of the most important advantages is that the l I I I I I. 2 13 method and device provides for automatic testing for meter drift in the dissolved oxygen sensors, which is achieved by the provision for selective flow reversal and comparison of the sensor outputs.

Also of significance is the reduction or elimination of the importance of meter. response time. In the system according to the invention, the meters are required to change only as fast as the change in consumption rate and not as fast as the consumption rate itself.

For example, a consumption rate change of mg/l/hr/hr from, say 30 mg/l/hr to 60mg/l/hr in minutes (very high value for such a change) only requires the exit meter to change at 2 mg/l/hr in a two minute residence time unit built according to the invention, which is far less than typical meter response capabilities.

In contrast, the prior art batch measurement techniques may require meter response times to be up to about a hundred times faster than required for the present invention.

The present system also reduces the time between sampling and testing to very low levels, particularly when contrasted with laboratory-based prior art batch measuring techniques.

The device of the present invention is also self-cleaning by the scouring action of the pump liquid.

This contrasts with the majority of prior art devices Oh pr >r 14 that either use unreliable mechanical cleaning means or consume appreciable maintenance resources to ensure the requisite degree of cleanliness.

The reliable measurement capacity of the device according to +he invention is also substantially improved.

In this regard it is estimated that in a two minute residence time the unit according to the invention, the upper practical limit for measurement is about 500 mg/l/hr, or even higher if a multi-stage dissolver is fitted.

Comparing this with the prior art batch-type units utilising currently available sensors, distortions of the results due to sluggish meter response time could easily occur at rates greater than about only 60 mg/l/hr.

tit In addition, the device of the present invention can be used to determine relative toxicity and the relative i strength of the load of an influent to an acclimatized biomass. This is achieved by forwarding a mixture of I C: .acclimatized biomass and influent in various ratios through l the device and analysing the variation in the oxygen 20 consumption rate for each ratio such that for increases in I mg-0 2 /l1-influent, the strength of the load is increasing and toxicity is not an issue; whereas for decreases in mg-0 2 /l1-influent, toxicity is an issue.

The invention has been described with reference to a preferred example, it will be appreciated by those skilled in the art that the invention may be embodied in many other forms.

I

Claims

2. A device according to claim 1 wherein the control and monitoring means calculates the dissolved oxygen drop across the reactor and determines the reactor residence time for the measured fluid flow rate and relevant reactor volume so as to determine the oxygen consumption rate per unit volume. I 16
3. A device according to claim 1 or claim 2 wherein the control and monitoring means actuates flow reversal on a predetermined periodic basis and/or when a sudden change in the measured dissolved oxygen levels is indicated.
4. A device according to any one of the preceding claims wherein the liquid flow rate measuring means are disposed at a location downstream of the oxygenator. A device according to claim 1 wherein the system is configured to provide a minimum pipe reactor residence time of two minutes.
6. A device according to any one of the preceding claims wherein the system is configured to provide a scouring velocity in the pipe reactor of at least 0.3m/sec.
7. A device according to claim 5 wherein the system is configured to provide a scouring velocity of at least 0.8m/sec.
8. A device according to any one of the preceding claims wherein the flow rate of the test stream through the reactor is at least 3m3/hr.
9. A device according to any one of the preceding claims wherein the control and monitoring means are also used to control the oxygen flow rate into the test stream. A device according to any one of the preceding claims wherein the pump means comprises a submersible centrifugal chopper pump.
11. A method for continuously measuring oxygen consumption rate in oxygenated liquids or slurries, said method comprising the steps of: c :II 17 extracting a representative test stream of said liquid, passing said test stream through a plug flow reactor in the form of a predetermined length of pipe connected between liquid inlet means and liquid outlet means, injecting dissolved oxygen into said liquid test stream in the inlet means, measuring the dissolved oxygen levels with oxygen sensors disposed at at least two spaced apart points along the reactor, measuring the liquid flow rate past said sensors, comparing the output from said spaced apart sensors with control means and generating a signal indicative of the difference, and selectively reversing the liquid flow through said reactor.
12. A method according to claim 11 including the step of using the control and monitoring means to calculate the dissolved oxygen drop across the reactor and to determine the reactor residence time for the measured fluid flow rate and relevant reactor volume so as to determine the oxygen consumption rate per unit volume.
13. A method according to claim 11 or claim 12 including the step of programming the control and monitoring means to actuate flow reversal on a predetermined periodic basis oxygen levels is indicated.
14. According to any one of claims 11 to 13 including Fr 18 the step of ensuring a minimum pipe reactor residence time of about two minutes. A method according to any one of claims 11 to 14 including the step of ensuring a scouring velocity within the pipe reactor of at least 0.3 m/sec.
16. A method according to any one of claims 11 to including the step of ensuring a scouring velocity in the pipe reactor of at least 0.8 m/sec.
17. A method according to any one of claims 11 to 16 including the step of ensuring that the flow rate of the test stream through the reactor is at least 3 m 3 /hr.
18. A method according to any one of claims 11 to 17 Ii including the step of using the control and monitoring means to control the oxygen flow rate into the test stream.
19. A method according to any one of claims 11 to 18 including the step of pumping the test stream of liquid via a chopper pump or using other means to ensure a substantially homogeneous slurry. A device for continuously measuring oxygen consumption rate in oxygenated liquids or slurries substantially as herein described with reference to tha it¢ accompanying dr wing. o 21. A method for continuously measuring oxygen consumption rate in oxygenated liquids or slurries by extracting and measuring a representative test stream substantially as herein described with reference to the accompanying drawings. DATED This 6th Day of May, 1994 THE COMMONWEALTH INDUSTRIAL GASES LIMITED Attorney: IAN T. ERNST Fellow Institute of Patent Attorneys of Australia of SHELSTON WATERS -I i 19 ABSTRACT A method and device for continuously measuring oxygen consumption rate in oxygenated liquids or slurries by extracting and measuring a representative test stream. The device includes an oxygenator (20) to admit dissolved oxygen to the test stream which is then passed through a plug flow reactor The flow rate of liquid is determined with an appropriate measuring means and the dissolved oxygen level measured with at least two oxygen sensors (18) and (19) each respectively disposed preferably at the inlet and outlet to the reactor. Flow reversal means (16) and (17) are also provided to selectively reverse the flow direction of liquid through the reactor and past the oxygen sensors. The flow reversal means, oxygen sensors and liquid flow rate measuring means are connected to control and monitoring means. These compare the output from the spaced apart oxygen sensors and selectively actuate the flow reversal to verify the readings from the sensors r r 20 and/or identify sensor drift. St., It.