Absorption and Attenuation Meter
Contents
1. n Csig Cret x 6 The ac 9 pathlength is a fixed constant typically at 25 cm and can be found in the device file Now we require the determination of In N x to accurately measure e This value is derived for each channel and is supplied with each instrument s calibration data sheet as the Water Offset value and is referred to as Cog in equation 8 The temperature correction is applied using the temperature from the reference line and the channels correction table from the either the Calibration Sheet or the Device File The approximate correction value is linearly interpolated from the table First the correct temperature bin is determined by finding the two bin temperatures To and Ti that bracket the current temperature Then using the values A and A from the table we obtain 7 T 7 B T Ama An 7 A Ar T 20 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs where Ai compensation constant T current temperature To first bin temperature T second bin temperature Ar first value Ar second value This temperature correction is automatically applied by our WET View software If you are manually processing the raw data stream this correction must be applied to arrive at the temperature corrected absorption and attenuation coefficients WET View reads in the raw binary data and applies an algorithm that 1 Computes uncorrected engineering units2. 0 001m 1 Hz Dynamic range 0 001 10 m 5 1 Power Requirements Voltage Input 10 18 VDC Power Requirements may differ by 10 percent Lamp 1 2 5 Watts Lamp 2 2 5 Watts CPU board 2 0 Watts Other Electronics 0 5 Watts DC DC Converter 1 5 Watts TOTAL 9 0 Watts 30 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs 6 References The following references have been cited in this manual Moore C In situ biochemical oceanic optical meters Sea Technology Vol 35 No 2 10 16 1994 Zaneveld J R V J C Kitchen A Bricaud C C Moore Analysis of in situ spectral absorption meter data Ocean Optics XI Proc Soc Photo Optical Instrum Eng SPIE Vol 1750 1992 Moore C J R V Zaneveld J C Kitchen Preliminary results from an in situ spectral absorption meter Ocean Optics XI Proc Soc Photo Optical Instrum Eng SPIE Vol 1750 330 337 1992 Pegau W S J R V Zaneveld Temperature dependence of the absorption coefficient of pure water in the visible portion of the spectrum Ocean Optics XII Proc Soc Photo Optical Instrum Eng SPIE Vol 2258 597 604 1994 Roesler C S J R V Zaneveld High resolution vertical profiles of spectral absorption attenuation and scattering coefficients in highly stratified waters Ocean Optics XII Proc Soc Photo Optical Instrum Eng SPIE Vol 2258 309 319 1994 Zaneveld J R V J C Kitchen and C C Moore Scattering error correction
3. The packet length includes all bytes sent in the packet except for the four byte checksum and the padding bytes The packet length should be 634 bytes The serial number is a standard four byte long integer The first two bytes of the serial number denote the instrument type The next two bytes are status bytes used for ancillary options The floating point scan rate usually 5 8 6 1 scans per sec can be calculated from the 2 integer bytes receivedInt that are sent with this formula SampleRate 1 0 0 00003160 receivedInt The 2 byte status of the meter is currently unused 18 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs To calculate the depth from the two integer depth bytes this formula is used D mDy b where D depth in meters Draw raw number of counts sent by the device m multiplier determined in the laboratory for each depth sensor unit and is unique to that unit stored in the configuration file b offset determined by calibrating the device at sea level this can easily be done in the field stored in the configuration file As stated above the signal data is sent as eighteen groups of three characters Each group of three characters represents a 24 bit number This number is the actual signal measurement for a particular wavelength The order of this data is the nine a channels and then the nine c channels The order of the wavelengths within the a and channels is co
4. ac 9 User s Guide ac 9 Revision T 31 July 2008 21 WET Labs Experimental results show that ars the water specific absorption correction due to temperature at 715 nm is approximately 0 0035 per m deg Various researchers Pegau Moore have derived results ranging from 0 0024 to 0 0035 The corrected 715 nm absorption is then determined using 4715T a715m at Tm Tap 8 where azisr is the water temperature corrected absorption at 715 nm a7ism is the measured absorption at 715 nm Tm is the water temperature at the time of measurement and Tea is the water temperature at time of calibration Tea is recorded at the time of calibration and is provided with the instrument calibration sheet Note WETView does not perform this water based absorption temperature correction 3 3 5 Scattering Correction As described in the Instrument Description section of this manual the absorption meter uses a reflective tube to collect scattered light Because of the reflective tube and detector design some of the light scattered at wider angles than 41 7 degrees with respect to the optical axis is lost This results in a scattering error for the absorption meter This error was empirically determined by measuring apparent a and values in the presence of increasing concentrations of a scattering agent Maalox The error in all channels was between 18 and 19 percent throughout most of the a meter s dynamic range The relationship
5. Cleaning Remove flow tubes and all O rings Remove the collars from the flow tubes Wash with a mild detergent diluted with distilled reverse osmosis filtered RO or de ionized DI water to gently wash all of the windows and rinse the flow tubes Use Kimwipes or other lint free tissues to wash the windows Rinse off the meter completely with water to ensure no soap residue is left inside the flow tubes or on the windows Dry the meter Place the instrument in a protected area where it can dry completely Using a small heater to blow warm air over the meter may help speed the process Using dry nitrogen to blow dry the meter and remove water from the small grooves around the windows will also help speed the process It is suggested that the instrument be left overnight to dry out completely Reassemble the meter Carefully replace O rings and slide collars back on to the flow tubes Replace O rings around the windows Clean windows using a Kimwipe or lens paper Place a couple of drops of methanol or ethanol on the Kimwipe With firm pressure gently wipe the windows with methanol This should remove any visible streaks on the windows If necessary follow with a dry wipe in one direction across the window face Blow off any lint or dust with a dry air source Clean the flow tubes by putting a few drops of methanol on a Kimwipe and using a wooden dowel rod carefully slide the Kimwipe through the flow tube Repeat this procedure with both flow
6. We can use equation 14 to determine the error in an absorption meter provided hypotheses 1 2 and 3 are true kg is the scattered light included in the attenuation measurement This depends in a collimated system on the acceptance angle of the detector We should know this parameter to within 10 percent of the scattering coefficient The remaining error is then determined by am ArVCml r In a recent ac 9 test a 715 ci 715 was 0 33 k was guessed to be 0 12 0 05 We would then set k 0 03 0 88 0 05 and ka 0 29 0 017 This implies that the absorption meter did not receive 29 percent of the scattered light and that we can correct for this effect to within 1 7 percent of the scattering coefficient ac 9 User s Guide ac 9 Revision T 31 July 2008 25 WET Labs 4 Calibration and Characterization One can think of the ac 9 as a single beam spectrophotometer whose output values are related to a specific reference medium that is clean water The instrument is calibrated to provide a reading of 0 00 for each channel in clean fresh water The offset value determined during our calibration process is the number which when added to the raw instrument output in clean water provides zeroes for all wavelengths with the meter at a specific temperature This offset value is referred to as N in the preceding sections This is an important consideration for users who want to compare data from other transmissometer
7. in inverse meters from the signal and reference values 2 Applies a linear temperature correction for the meter s internal temperature using constants supplied in the instrument s device file 3 Applies clean water offsets supplied from the instrument s device file that provide a value referenced against clean water Combining these steps into one formula c A Coer I x In Cyig Cre i Ar 8 and a A aorr 1 x In Asig Aref Ar where c A a A attenuation coefficient and absorption coefficient respectively in m Coffs Aoff water offset value provided on the Calibration Sheet in m Csig Asig measured amount of light power that reaches the receiver detector from the data stream in raw digital counts Cret Cret amount of light power measured by the reference detector from the data stream in raw digital counts X sample volume pathlength in meters Ay internal temperature compensation correction value in m derived from Eq 8 3 3 4 Temperature dependent Absorption Correction Absorption of water shows a temperature dependency that is wavelength dependent While throughout the visible portion of the spectrum this effect is negligible See Pegau and Zaneveld 1992 1994 at infrared wavelengths the effect must be taken into account With regard to the ac 9 the water absorption temperature effect must be considered to correct the absorption and attenuation at 715 nm
8. Sensor optional Use and 07 27 00 Care DCR 49 H Van Zee J 10 09 00 Add reference to deep unit DCR 62 H Van Zee K 8 13 01 Renumber graphics reorganize sections 3 6 DCR H Van Zee 135 L 1 10 03 Review and update Section 1 correct depth rating D Stahlke H Van Zee DCR 265 M 6 29 04 Update specifications DCR 401 D Stahlke N 1 13 06 Clarify warranty statement DCR 481 A Gellatly S Proctor O 2 23 06 Correct initial warranty period S Proctor H Van Zee P 2 19 07 Change acceptance angle to 0 93 degrees DCR 510 H Van Zee R Zaneveld Q 2 5 08 Clarify internal temperature equations DCR 562 A Barnard J Bell R 5 20 08 Correct internal temperature correction equation on p A Barnard J Bell 22 DCR 595 S 5 22 08 Update user s guide DCR 597 H Van Zee A Derr H Van Zee T 7 31 08 Correct instrument weights DCR 607 Walsh ac 9 User s Guide ac 9 Revision T 31 July 2008 33
9. The flow tube assemblies are integral to the optical behavior of the ac 9 They are optical components that help produce a very precise measurement and thus they need to be dealt with accordingly Before using inspect both tubes and make sure they are free of stains and dust The reflective flow tube for the absorption measurement operates using the principle of internal reflection To maintain its reflective properties it requires a thin air gap between the outer wall of the quartz tube and the inner wall of the surrounding sleeve The reflective tube should be periodically checked for leaks To determine if the tube is maintaining its reflective properties immerse it in water and point towards a fairly light background The inside of the tube should appear uniformly bright If the tube has leaked call the factory for repair instructions or tube replacement To clean the absorption path s reflective tube carefully plunge an alcohol soaked tissue through the tube and rinse thoroughly with distilled water Whenever plunging a tissue through the tube use a wooden or plastic dowel to prevent scratching the sides of the tube After rinsing dry the tube either by blowing dry nitrogen through it or by plunging a soft tissue The attenuation path flow tube is virtually maintenance free except for occasional cleaning Follow the same basic procedures supplied for cleaning the absorption path tube Remove the flow tube sleeves when drying the flow tub
10. This means that if you clean the windows and flow tubes and make an air measurement and if that air measurement matches the factory ac 9 User s Guide ac 9 Revision T 31 July 2008 27 WET Labs supplied air values the resulting clean water measurements should provide zero values Equally important if the air values taken in the field differ from factory air values then that difference may be applied as a correction factor to the original water calibration values contained in the DEV file a clean Aclean T air E aair 1 where Aclean 1S the clean water offset value supplied with the ac meter clean IS the corrected clean water offset a air IS the factory supplied air value a ir is the field obtained air values Alternatively the offset air value difference can be applied directly to the data as a post processing step Obtaining air values in the field is a much more straightforward process than obtaining clean water values This is a powerful technique that can track and compensate for instrument drift and filter aging Care must be taken when implementing this technique since a drop of water or smudge on the windows will result in a reading quite different from the factory air values This difference might be interpreted as instrument drift By iterating this process several times re cleaning the instrument and obtaining a new set of air values you will gain experience in differentiating true instrument drift from
11. case We recommend a neoprene spacer between the unit and its frame or clamp At the very least any contact area should be taped carefully to assure mechanical and or electrical isolation Do not apply torsional stress to the instrument housing The optical path is encased in a rigid housing but is still subject to distortion if the unit is subjected to undue stress The instrument has a delicate optical path that is subject to misalignment if stress is applied unevenly to the upper and lower cans Make sure the unit is mounted on at least two points and that neither point is a stress point Make sure you have provided for an unobstructed upward flow through the flow chambers and the pump Figure 3 Mount pump above flow tubes so water is drawn up through system Avoid potential air locks by my lt assuring the tubing moves water upward Outlets from flow assemblies may be tied to a Y or T fitting to provide parallel flow Use stainless steel screen filters to prevent intake of particles larger than 1 mm Figure 3 Proper connection of ac 9 flow tube 1 6 3 Deployment Tips The following suggestions will help you obtain the highest quality data from your ac 9 The instrument is extremely sensitive and should be handled carefully ac 9 User s Guide ac 9 Revision T 31 July 2008 7 WET Labs e A sturdy shipping transport container should be used to transport your instrument to the field The instrumen
12. checksum is the sum of all characters sent in the packet including the identifier FF 00 FF 00 Several padding characters are sent at the end of the packet before the next packet identifier is sent ac 9 User s Guide ac 9 Revision T 31 July 2008 19 WET Labs 3 3 3 Post Processing Once accumulated by a host computer the data must be post processed to yield meaningful scientific data The primary transfer equation for yielding the attenuation coefficient or the absorption coefficient a for a given wavelength is Tr eX 1 where Tr is the transmittance c is the attenuation coefficient and x is the pathlength of the water volume being measured The following treatment applies to the absorption case if is replaced by a The pathlength of the ac 9 is fixed at either 25 or 10 cm In the case of the ac 9 the transmittance is computed by taking the ratio of the signal value to the reference value Tr Csig Cref N 2 Substituting equation 2 into equation 1 we get Csig Cref IN oo 3 where Csig and Cref are the signal and reference count values from the instrument and N is an instrument specific calibration constant obtained in the laboratory using clean water For more information about how N is derived see Section 4 Calibration and Characterization We therefore can solve for c or a by c 1 x In Coig Creg In N 4 ore 1 x In Cgig Crep 1 x in N 5 or In N x
13. following components ac 9 with one absorption flow tube and one attenuation flow tube four flow sleeves with four protective black plastic caps dummy plugs with lock collars this manual CD with o This User s Guide o ac 9 Protocol Document o WETView User s Guide o WETView software and instrument specific calibration information e Three meter test cable e double Y de bubbler tubing with stainless steel intake screens Familiarize yourself with the ac 9 remove the black plastic flow tubes by grasping the flow tube sleeves and sliding them away from the ends of the flow tube toward the middle of the flow tube You only have to slide the collars about 4 in to unlock the flow tube from its fixed position The flow tube will lift out exposing the transmitter and detector windows on the lower and upper flanges respectively Observe the flow tubes The attenuation tube is different from the absorption tube Its flow chamber is plastic and the two sleeves on the tube are identical This tube installs on the c side of the instrument the side with the identical looking windows The c tube has no up or down orientation The absorption tube is lined with a quartz tube and one of the two sleeves is flat on top the lip present on all the other sleeves is missing This tube installs on the a side of the instrument which can be identified by the a detector on the upper flange and is the only window which is
14. from the flow tube entrances and obtain values using WET View or your own custom program Record and save about 5 minutes worth of data Examine your data in a spreadsheet At one second binning 6 samples the data should maintain a standard deviation of 0 001 m or less If you have a substantial slope to the data over the acquisition period your meter may have some residual moisture in the flow tubes or on the windows If the optical path isn t getting dry use a small flow of dry nitrogen through the tubes while sampling Remove flow tube and re clean the windows The single wipe technique with ethanol will work fine at this point Repeat steps D F until mean values stabilize over three cycles Use these values as air calibration values ac 9 User s Guide ac 9 Revision T 31 July 2008 29 WET Labs 5 Specifications Mechanical Size Weight Environmental Temperature range Rated depth Pressure sensor Electrical Input Current draw Serial output Sample rate 69 x 10 4 cm diameter in air 5 45 kg acetal copolymer 8 kg aluminum in water 0 85 kg 0 30 deg C 500 or 5 000 meters optional 10 18 VDC 0 75 Amps 12 V nominal RS 232 or RS 485 6 scans sec nominal Optical Spectral Range 412 nm 715 nm Bandpass 10 nm channel Pathlength 25 cm 10 cm optional Beam cross section 8 mm diameter nominal Linearity gt 99 R Output wavelengths 9 Al Accuracy 0 01 m Precision H O0 0O03 m 6 Hz 5i
15. instrument should be lowered to just below the water s surface Turn on the instrument and pump and check to ensure that the pump has primed and is operating properly Lower the package to a depth of 10 20 meters Run the instrument for 3 5 minutes to allow the motor controller to stabilize the flow tubes to clear and the instrument to begin to equilibrate with the water temperature e After the warm up period raise the package to just below the surface and begin data collection The initial depth will be dependent on the natural surface conditions and the amount of bubbles that the ship itself is generating Steadily lower the cage through the water column e The upcast can proceed immediately after the downcast It is the user s choice whether to open a new data file for the upcast or include the down and upcast data in a single file e Once the cage is back to the just below the surface stop the data collection and turn off the pump Carefully bring the cage on deck and lash it down Give the cage and instrument a fresh water wash down after every cast If this is not practical wash the instrument at the end of each data collection day Holding a hose low pressure over the pump discharge port will flush the tubing and the flow tubes Rinse the flanges and connectors If leaving the instrument on deck for more than a few minutes cover the cage with a tarp to avoid over heating the instrument due to solar insolation At the end of eac
16. running the meter for an interval of time and determining the standard deviation of the signal We perform this measurement in air With one second binning of the signals nominal performance is approximately 0 001 m for a 25 cm path instrument Since the absorption or attenuation measurement is pathlength dependent precision values increase approximately by a factor of 2 5 for the 10 cm pathlength version of the ac 9 It should be noted that both air value and water value measurements can be difficult to perform If running the tests in air you must insure that the flow tubes and windows are completely clean and dry Do not allow any ambient light penetration into the cells Use a cap or black tape to seal the flow tube nozzle openings When operating the instrument in water use clean water free of bubbles and do not exceed flow rates of 1 2 liters per minute 26 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs 4 3 Absolute Calibration Absolute calibration of any underwater optical absorption or attenuation sensor is difficult as there are no absolute standards that can readily be used The most obvious choice for a calibration medium pure water is difficult to obtain is unstable once it is made and its absolute optical properties are not well known Nevertheless pure water is the baseline for all oceanic property measurements At WET Labs we produce our calibration water with a custom de ionization and filtrati
17. s clear aperture or completely blocked The CPU sampling period on the A D is about 3 msecs It only samples in the middle of the filter s aperture This means that at a 6 Hz rotation speed the CPU samples and averages about 100 samples per filter or blank per pass Both the signal and reference detectors are sampled simultaneously during this period Once averaged the dark values are then subtracted from the raw light values in order to create a given datum point Csig Csighight Csigdark Cref Crefiight Crefaark ac 9 User s Guide ac 9 Revision T 31 July 2008 15 WET Labs The light sample period and the dark sample period are separated by 0 013 to 0 027 second interval depending upon the rate of filter wheel rotation This interval defines the limiting frequency response of the electro optical signal processing within the unit Trace three shows the analog signal from the attenuation detector The absorption and attenuation beams are located 45 degrees out of phase with respect to each other in order to provide optimum switching and sampling by the CPU Trace four shows sampling of the attenuation signal Input from the primary gain stages is switched into the subsequent processing stages by the CPU Thus processing for the a and c channels remains virtually identical The CPU collects and buffers one revolution of raw signal data and then begins to output the data as it is sampling the next revolution Refe
18. samples of both signal and reference channels accumulates them through the sampling period and then averages the values at the end of the sampling period Once averaged light and dark values are collected for each channel the CPU takes the difference of these values to derive its output value Reference values are subsequently averaged for ten filter wheel scans Once signal data is accumulated over a given filter wheel rotation it is output is transmitted through the RS 232 RS 485 port Every ten rotations the CPU sends averaged reference values as well as temperature and status information The data output is sent as raw 24 bit hexadecimal averaged values representing A D counts Hex 0 FFFFFF 33 1 Data Format The table below shows a serial output record from the ac 9 If you have a terminal emulator program or are using your own software package this is how the data will appear The comments appearing after the semicolons are not part of the data stream Note The instrument sends data in binary format You must be sure your data collection program is set to read the binary bit stream The characters in the table are shown in hexadecimal for clarity A semicolon denotes a comment and is not seen in the data stream All two byte integer words are sent in low byte high byte order The three byte data words are unsigned and sent in low byte high byte decimal fraction order ac 9 User s Guide ac 9 Revision T 31 July 2008 17 WET L
19. smudged optics Caution Back up the factory DEV file before making any modifications to it Effective air calibration requires 1 The instrument s optical path is clean and dry 2 The instrument s optical path is completely shielded from ambient light To perform air tracking we recommend the following basic procedures a Remove flow tubes and clean and dry completely Remove the sleeves from the flow tubes to assure there is no trapped moisture Allow the flow tubes to sit in a dry environment with good air flow b Clean and completely dry the optical windows of the instrument If the optics have not been cleaned for some time clean with a mild solution of detergent and distilled water rinse and wipe dry Follow with a single wipe across each window with an ethanol soaked lens tissue or other low water content alcohol type solvent and then a single wipe with a dry tissue If the windows are already fairly clean simply use the ethanol wipe procedure You may wish to apply a stream of pressurized air or nitrogen around the windows to force moisture from the window mount WARNING Beware of using standard air compressor air It can blow oil all over the optics c Allow the unit to sit open for an hour or two to assure that it dries d Turn on meter and allow it to warm up about 15 minutes 28 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Reassemble and insert flow tubes Block ambient light
20. tubes Examine each flow tube when you are through to ensure there are no streaks or small pieces of lint left on the inside of the flow tube Dry the windows Since small amounts of moisture can affect the air readings it is important to ensure the meter is completely dry Using nitrogen to blow dry the windows immediately before replacing the flow tubes works very effectively This will remove any water or methanol trapped in the small grooves around the window Replace the flow tubes Carefully slide the flow tubes into place without sliding dirt across the windows Slide the sleeves up around the windows and over the O rings making certain they are firmly in place and aligned correctly Use small black caps or black electrical tape over each of the nozzles on the flow tube to provide a dark environment and to keep the meter clean and free of moisture while obtaining data Turn the meter on and allow it to warm up for at least 15 minutes When the meter is stable you should be able to collect 10 minutes worth of data and the values should not vary more than 0 005 m over the 10 minute time period ac 9 User s Guide ac 9 Revision T 31 July 2008 3 WET Labs 8 Collect data Record a one to two minute file and save data Repeat steps 4 6 until you can collect three data files cleaning after each file such that the average values for each channel vary by no more than 0 005 m The meter is calibrated over a specific range
21. 8 6 mm quartz pressure window 4 38 mm singlet lens 9 Reflective flow tube 5 Interference filter 10 Diffuser Signal detector Figure 6 Schematic Representation of a beam optics The a beam and c beam optics are similar The a beam light is 45 degrees out of phase from that of the c beam Beam splitter optics and aperturing of the beam are identical with the c beam source optics The sample water volume is enclosed by a reflective flow tube Light passing through the tube is both absorbed by the water itself and by various pigments contained in particulate matter within the sample volume Forward scattered light is reflected back into the water volume by the reflective tube The light is then collected by a diffused large area detector at the far end of the flow tube The flow tube uses the internal reflection principle in reflecting light back into the water volume A clear quartz tube is employed The outer perimeter of the tube is enclosed by a thin annular volume of air Using the Fresnel Equation one can see that with an index of refraction of 1 33 in water and index of refraction of 1 in air the total internal reflection is achieved to 41 7 degrees with respect to the optical axis With our deep units or upon request we employ an aluminized flow tube with an oil filled gap that provides pressure equalization over the rated depth of the instrument 2 3 Electronics The primary electronic components of the spectral absorption meter incl
22. Overview SE EER ER EN Re ee SG ERG eters 11 2 2 OPCS se EE EE E Ge oat ee 12 2 3 ElectroniCS cesoie OR OE E E E E RE EAEE 13 2 4 Pressure Housing Material soek EEN SR MEER GN Eb OS DRS ED Ee 14 2 5 Bo COS RE EE A E O 14 2 6 Signal CSS IR RE OR OE OE N 15 3 Data Pro essind seed ekke EE Re Ge Be DE N oe ES ee AIS 17 3 1 Analog es Lei Ie EA N Re RE OE 17 3 2 Analog to Digital Conversion oe EE SE RE GER se ee De ES 17 3 3 Digital Processing and Data Output esse esse ee RA Re ee Re ee ee RA ee Re ee 17 4 Calibration and Characterization sees es ee ee ee Re 26 4 1 Temperature OLE SCLION vies EE be ek bien see eek GE ERA SN Ges ee Ge 26 4 2 ee GE EE EE EE eee ees 26 4 3 Absolute Calibration ES GE DR EG Ee EE RE oe id 2T 4 4 Linearity and Dynamic RAGE seks di REEN ee KI ee si Ee ese ee Gee 27 4 5 Air Calibration and Tracking sc sesse ee BR SESSE RGN Ge Ee ENG ee SEE 27 5 Selde ie AE AE N OR EE EE EE OE N 30 5 1 Power Requirements ee ee Ge ee ee RD de ee ER ie 30 6 Referents se EK EE ee wen OE ee 31 Appendix A Optional Pressure SeNSOF iese ee ee se ee RR EE 31 Appendix B Legacy ConnecCtOF iese sesse ek ee ee ee ee EE ee 32 ac 9 User s Guide ac 9 Revision T 31 July 2008 i WET Labs 1 Operation This section provides an introduction to the ac 9 and describes how to perform air tracking and bench testing prior to deploying the ac 9 The ac 9 is delivered in a sturdy wooden shipping container with the
23. WET Labs Absorption and Attenuation Meter ac 9 Users Guide WET Labs Inc P O Box 518 Philomath OR 97370 541 929 5650 www wetlabs com ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Attention Return Policy for Instruments with Anti fouling Treatment WET Labs cannot accept instruments for servicing or repair that are treated with anti fouling compound s This includes but is not limited to tri butyl tin TBT marine anti fouling paint ablative coatings etc Please ensure any anti fouling treatment has been removed prior to returning instruments to WET Labs for service or repair ac 9 Warranty Standard Warranty This unit is guaranteed against defects in materials and workmanship for one year from the original date of purchase Warranty is void if the factory determines the unit was subjected to abuse or neglect beyond the normal wear and tear of field deployment or in the event the pressure housing has been opened by the customer To return the instrument contact WET Labs for a Return Merchandise Authorization RMA and ship in the original container WET Labs is not responsible for damage to instruments during the return shipment to the factory WET Labs will supply all replacement parts and labor and pay for return via 3 day air shipping in honoring this warranty Annual Servicing Extended Warranty WET Labs will extend the warranty on this unit to five years if it is returned annual
24. abs This is the header section of the packet 00 FF 00 FF Packet registration EE 00 Record length of full packet not including chksum 00 00 01 05 serial number 00 00 status reserved 00 00 filter wheel rotation period 00 00 depth opt 00 00 reserved This is the data section of the packet Time Ch 1 Ch2 Ch3 Ch4 Ch5 Ch6 Chl8 0001 B2A1 C4 8123 59 7777 2A 7A3D 20 3B20 10 2345 DO 1000 DF Time 2 bytes eighteen groups of three bytes each Each group represents a 24 bit number Channels are transmitted in the order in which they appear in the DEV file 0001 B2A1 C4 8123 59 7777 2A 7A3D 20 3B20 10 2345 DO 1000 DF The above time and data lines are sent a total of 10 times per packet This is the reference section RCh 1 RCh2 RCh3 RCh4 RCh5 RCh6 RCh18 9919 99 3333 55 2136 40 5211 32 2000 Al 2999 D2 3333 55 These are the reference channels Same format as the data fields which is eighteen groups of three characters forming 24 bit numbers Reference Channels are transmitted in the order in which they appear in the DEV file 0001 temperature 2 bytes B715 checksum 00 00 00 00 00 00 padding 3 32 Primary Processing When receiving binary packets from an ac 9 the first thing to look for is the packet identifier which is four characters The characters in hex are 00 FF 00 FF After successfully receiving these characters the packet header is the next item to be received
25. ad Connector Pin and Socket Arrangements Waterproof patch cables to adapt one connector style to the other are available from WET Labs Figure A 2 shows the patch cable to connect an instrument with the newer connector to an older sea cable Figure A 3 show the patch cable to connect an instrument with an older connector to a newer sea cable SOCKET 1 SOCKET 1 mq MCIL 6 FS WITH VMK 6 MP WITH LOCK SLEEVE MCDLS F LOCK SLEEVE K MLS P WIRED 1 1 PIN 1 ED VMK 6 FS WITH MCIL 6 MP WITH LOCK SLEEVE K FLS P LOCK SLEEVE MCDLS M WIRED 1 1 Figure A 3 Older Instrument Connector to Newer Sea Cable WET Labs dwg 210052 32 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Revision History Revision Date Revision Description Originator A 10 07 99 Begin revision tracking H Van Zee B 01 11 00 Update document DCR 4 D Hankins C 01 17 00 Clarify packet information DCR 13 D Romanko Change bulkhead connector and pin information DCR D 02 17 00 14 A Derr Add drawing showing top flange bulkhead connector E 04 18 00 for pump DCR 25 ECN 113 A Derr Change pump connector functions in table DCR 36 F 06 12 00 A Derr Correct pump connector wiring DCR 42 ECN113R2 G 06 26 00 A Derr H 07 11 00 Correct temperature correction equation DCR 46 C de Lespinasse Add Appendix B Pressure
26. ansmitted via both RS 232 and RS 485 which requires a serial interface on the host PC or data logger Host Data Logger The ac 9 can be interfaced to any PC or data logger capable of supporting a 19 200 baud rate serial interface Software The WETView host software package interfaces directly to the ac 9 via your computer serial port WET View allows real time graphical data output as a function of time depth or wavelength WETView automatically applies calibration constants temperature corrections and generates a tab delimited ASCII text file that can be imported into programs such as Excel or MatLab for post processing or manipulation The data output format of the ac 9 is defined in the Data Processing Section of this manual 1 6 Operating the Meter 1 6 1 Basic Power On 1 Connect the test cable to the proper COMM port on the host computer 2 Attach the power leads to a stable power source that supplies 10 18 VDC to the ac 9 Make sure the polarity is correct before switching on the power supply Connect the RS 232 connector to the desired serial port of the data collection computer If your instrument is sending data in the RS 485 format an RS 485 to RS 232 converter is required to allow proper operation Plug the submerged wet end of the cable into the ac 9 Applying a small amount of silicone grease or equivalent to the base of the instrument bulkhead makes the plug insertion easier and provides greater assurance of a good seal Us
27. ant to save the data To quit the program choose QUIT from the File menu At this point you have successfully completed a bench test of the instrument 1 3 Air Tracking We provide an air calibration CAL file similar to the device DEV file that can be applied in WETView in the same manner The DEV file provides the clean water offsets so that when measuring clean fresh water the instrument s output should be very nearly 0 0 for all channels The CAL file provides the offsets that provide 0 0 values when the instrument is clean and dry and measuring air values This is a useful tracking tool for catching instrument drift filter aging and improper cleaning 2 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs When a DEV file is opened in WETView it will display DEV in the dialog box which will list all the DEV files on the default drive If you change the DEV to CAL the available CAL files will be displayed Select the latest CAL file and start collecting data Make sure the black plastic caps are installed on the flow sleeve nozzles so no ambient light can enter the flow tubes If the instrument is clean and dry the values displayed in WETView should be very close to 0 0 within 0 005 0 01 If the values are within this range the instrument is clean and ready to deploy If the values are outside this range the first step is to re clean the instrument and then reapply the CAL file offsets 1 4
28. between a and remains linear throughout most of a given wavelength channel s dynamic range gt 998 For very high c values gt 12 m we noticed a slight roll off in a The imperfect reflectivity of the reflecting tube and other imperfections cause scattered light errors in the absorption meter The magnitude of these errors must be corrected to obtain accurate absorption values There are several alternatives in performing the scattered light correction The simplest involves subtracting the temperature corrected a715 measurement Assuming that water is the only species present that absorbs light at 715 nm and assuming that the scattering error is spectrally independent we can use this wavelength to provide a base for the scattering error correction This technique is commonly used in bench top spectrophotometric measurements We present one recommended method developed by Dr Ron Zaneveld at Oregon State University SPIE 1992 that has thus far proven to be a robust correction mechanism For what follows we assume that pure water absorption and attenuation have been subtracted For a given wavelength the absorption tube overestimates the absorption coefficient as it does not measure all of the scattered light In what follows we will designate the proportion of the total scattering coefficient that the device does not receive by ka A If the shape of the scattering function does not change much as a function of wavelength we
29. clearly different from the other three The flat flow tube sleeve goes on this detector You may want to mark the tubes and their orientation with tape or marking pen before using the instrument at sea so that there is no confusion on reinstalling the tubes after cleaning the optics Reinstall the flow tubes before bench testing your instrument If you have removed the plastic caps from the stainless nozzles replace them at this time 1 1 Installing WETView WETView displays data produced by WET Labs instruments It runs on PC compatible computers with at least 16 Mb of memory and 3 Mb free hard disk space 1 Insert the CD with WETView in the host computer 2 Double click on the SETUP EXE icon Setup will guide you through the rest of the installation process Caution If you have old device files from previous calibrations you should rename them or archive them in a different directory so that they will not be overwritten ac 9 User s Guide ac 9 Revision T 31 July 2008 1 WET Labs 3 Copy the airxxyyy cal and ac9xxx dev files from the CD to the host PC These are instrument specific calibration files xx is the calibration number yyy is the instrument number 1 2 Bench Testing The following items are needed to bench test the ac 9 e A clean solid lab table or work bench e The ac 9 with test cable or sea cable e A 12 15 volt power supply the ac 9 requires 10 18 VDC e A PC with WETView installed 1 Connec
30. e a connector lock ring if one is available ac 9 User s Guide ac 9 Revision T 31 July 2008 5 WET Labs 3 Turn on power supply To verify basic operation when not hooked directly to a computer remove one of the flow tubes and in a darkened environment place a white piece of paper into the beam path You should be able to see the beam image on the piece of paper You should be able to hear a faint whirring of the filter wheel motor if you place your ear directly against the lower can If you neither hear the motor nor see the beam the unit is not working In this case check your connections and your power supply If the instrument still does not run you may want to seek technical assistance from the factory 4 For optimum stability allow the instrument to warm up for 3 5 minutes before acquisition While this is not an absolute requirement the instrument s electro optics are subject to an initial stabilization period 1 6 2 Mounting Meter ac 9 operation is optimized for a vertical to forty five degrees off vertical orientation with the bulkhead connector facing upward Figure 2 If your application requires a horizontal mounting take care to provide proper pump priming and to avoid trapping air bubbles in the flow tubes This can be accomplished by taking the system to a depth of 20 meters and allowing the required in water warm up period to occur at depth This helps the pump to prime properly and compresses small air b
31. ed by an A aj A aj Ap bjA j Ay 12b This is the desired result that allows us to correct the scattering error of the absorption measurement Note that we can thus determine the true non water absorption coefficient at any wavelength using the three hypotheses We do not need to know the fraction of scattered light not received by the attenuation meter i e we need not know Ke For spectrophotometric work one normally cannot take into account the spectral variations in b A One simply subtracts the absorption at the reference wavelength This 24 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs then would imply that bm A b A from equation 12 an unnecessary assumption in our case The above correction method is thus more accurate than the usual spectrophotometric approach We only assume that the shape of the volume scattering function be independent of wavelength whereas in the spectrophotometric approach it is assumed that the absolute value of the scattering function be independent of wavelength The accuracy of the absorption meter may be expressed in terms of kg From the above we can derive an expression for kg Substitution of 6 into 9 gives aj 715 kg 1 ke ka bm Ay 13 solving for ka then gives kalaj tbm Ay HC ke aj Ay The scattering coefficient of pure water at the reference wavelength is very small so that ay Ap tbm Ar Gr and ka 1 ke aj r ci Ay 14
32. es 4 Storage The ac 9 should be stored and transported in a shock protected environment Typically units are shipped in a sturdy wooden crate Using the crate will assure that you can safely transport the instrument providing it is handled in a reasonably careful fashion 10 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs 2 Instrument Description This section provides a general description of how the ac 9 operates It provides a general discussion of the primary instrument configuration as well as a description of the optical and electronics system 2 1 Instrument Overview Figure 4 shows a diagram of the ac 9 The unit consists of two pressure housings separated by three stand offs The shorter of the two pressure cylinders houses the light sources filter wheel and transmitter optics The longer of the two cans houses the receiver optics and the control and acquisition electronics for the unit The absorption and attenuation beam paths and flow tube assemblies are between the receiver and transmitter housings Power to the unit and signal out of the unit are provided via the bulkhead connector at the end of the long receiver housing lt Bulkhead Connector Power RS 232 or 485 output _ Acquisition and Control Electronics c beam Collimated Receiver a beam Diffuser Receiver C beam Non reflective Flow Tube a beam Reflective Flow Tube W Transmitter Optics Rotating Fi
33. h data collection day remove the tubes and carefully clean and dry both 8 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs the flow tubes and windows Failure to flush the instrument with fresh water may cause corrosion damage over time e For further information on deployment techniques see the ac 9 Protocol document 1 6 4 Data Acquisition This section describes how to collect data from the ac 9 WET Labs offers several output configuration options to provide flexible interfacing to different systems These various output protocols are discussed at length in the Data Processing Section of the manual Unless supplied with a custom output protocol the instrument powers up in a free run mode This means that when turned on the unit automatically begins acquiring data and outputting that data in its appropriate format Typically the instrument comes supplied with RS 485 and RS 232 output operating at 19 200 baud Consult the Specifications Section for the output format RS 485 For longer cable lengths and maximum data integrity RS 485 protocol is the preferred method of data transfer Data is transmitted from the instrument in a binary format To view this data you must have a program capable of reading binary data If you are using our WET View software package the binary read is done automatically If you do not plan to use WETView or a WET Labs supplied data logger consult the Data Processing Section of this manual fo
34. h half maximum FWHM filters that are spaced around the perimeter at approximately a 3 1 ratio with associated blank spaces This configuration provides a chopped output for the detectors which compensates for temperature coefficients in the detector and amplifier circuitry as well as providing low level ambient light rejection Once the light has passed through the filter wheel the beam passes through a beam splitter creating a primary beam and a reflected beam The reflected beam intensity is measured by a reference detector Using a ratiometric scheme with the reference and signal detectors we compensate for long term lamp drift The primary beam then passes through a pressure window into the sample water volume A flow tube encloses the water path Scattered light that hits the blackened surface of the flow tube is absorbed and therefore does not contribute to the measurement of transmitted intensity Light radiated through the flow path is therefore subject to both scattering and absorptive losses by the water 12 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Once through the water path the light passes through another pressure window and then is re focused through a 30 mm lens upon a receiver detector A 1 mm aperture is placed directly in front of the detector creating a 0 93 degree acceptance angle in water 2 2 2 a Beam Optics 1 Lamp 6 Beam splitter 2 1 mm aperture 7 Reference detector 3 6 mm aperture
35. it To understand the exact nature of signal processing it is first necessary to better understand the primary data sampling Figure 7 is a timing representation of the ac 9 signal sampling through a single filter wheel rotation so that each of the nine interference filters are brought in line with the optical path once per revolution Acquisition Timing Diagram Figure 7 Acquisition Timing Trace one represents the optical signal from the absorption detector as the filter wheel spins through its cycle During the filter wheel rotation signal output from the absorption detector is continuously monitored and amplified through an analog current to voltage amplifier circuit The current to voltage amplifier serves as the primary gain stage for the signal Typical gains for the channels are set at 5 Mb After the primary gain stage the signal is passed through two more analog stages for level shifting and voltage inversion At this point the signal is ready for digitization The Analog to Digital Converter A D continuously samples the incoming voltage level at a rate equal to approximately 80 kilosamples per second The A D runs off its own clock and is in effect autonomous from the rest of the system The CPU determines when to sample the A D Trace two shows the sampling periods in which the CPU obtains signal from the A D Input from the motor encoder tells the unit when the source beam is either within a given filter
36. logging unit 2 4 Pressure Housing Material The ac 9 is housed in a robust pressure can made from either acetal copolymer 500 meter depth rating or type 7075 aluminum 5 000 meter depth rating 2 5 Connectors The ac 9 uses the connectors shown below CBH 6 M SEA TEST CABL Sea Test Cable Host Port Connector Pin Functions GND RS 232 RX RS 485 V 10 18 VDC RS 232 TX to host RS 485 SIES SIE Pump Port Connector Socket Functions GUIDE SOCKET GND V 3 3 N C D 2 Voltage supplied to the instrument is internally jumpered to provide power output to the pump port connector Power is applied to the pump connector whenever the meter is powered 14 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs WARNING If the meter is deployed without a plug in the pump connector socket the socket contacts will suffer rapid corrosion Eventually the corrosion could travel through the connector causing the meter to flood Always put a pump plug or dummy plug in this socket 2 6 Signal Processing The purpose of the ac 9 signal processing circuitry is to take a raw optical signal and make it into a physically meaningful measurement ready for output Signals from the absorption path and attenuation path detectors go through several levels of analog and digital processing before they are registered as output from the un
37. lter Wheel Dual Light Sources Lamp Housings Motor Encoder lt gt 10 25 cm Figure 4 ac 9 diagram ac 9 User s Guide ac 9 Revision T 31 July 2008 11 WET Labs 2 2 Optics The ac 9 performs concurrent measurements of the water s attenuation and absorption characteristics by incorporating a dual path optical configuration in a single instrument Each path contains its own source optics and detectors appropriate to the given measurement The two paths share a common filter wheel control and acquisition electronics For purposes of description we refer to the beam performing the attenuation measurement as the c beam Figure 5 and the beam used to make the absorption measurement as the a beam Figure 6 2 2 1 c Beam Optics 2 3 3 2 a gc 1 i 11 38 mm 100 or 250 mm 30 mm 1 Lamp 6 Beam splitter 2 1mm aperture 7 Reference detector 3 6 mm aperture 8 6 mm quartz pressure window 4 38 mm singlet lens 9 Flow tube 5 Interference filter 10 30 mm singlet lens 11 Signal detector Figure 5 Optical Path Configuration for c beam Light from a DC incandescent source passes through a 1 mm aperture The light is then collimated with a 38 mm lens followed by a 6 mm aperture The collimated light passes through bandpass filters mounted upon a continuously rotating filter wheel creating a narrow band spectral output The filter wheel holds nine 12 5 mm diameter 10 nm full widt
38. ly for servicing This includes calibration standard maintenance and cleaning Charges associated with this annual service work as well as shipping costs are the responsibility of the customer Shipping Requirements for Warranty and Out of warranty Instruments 1 Please retain the original shipping material We design the shipping container to meet stringent shipping and insurance requirements and to keep your meter functional 2 To avoid additional repackaging charges use the original box or WET Labs approved container with its custom cut packing foam and anti static bag to return the instrument e f using alternative container use at least 2 in of foam NOT bubble wrap or Styrofoam peanuts to fully surround the instrument e Minimum repacking charge for ac meters 240 00 3 Clearly mark the RMA number on the outside of your shipping container and on all packing lists 4 Return instruments using 3 day air shipping or better do not ship via ground ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Table of Contents Es EE ONE RE OE 1 1 1 Installing WET Ee RE ub N EG DR ee ee RD GR Ee EE 1 1 2 Bench PSU EE OE EO EE EA 2 1 3 Ait Trackin AE OE Ee N NE EE 2 1 4 Clean S OE EE EE LE EE EE OE ER OR Ed 3 1 5 System RedUiF MERS ESE en EG E E R E Ge 4 1 6 Operating the MERET iese N ed nai NAN E Ne AEE es eN RE dee 5 2 Instrument Description sesse EERS SESSE REGEER SE ER WERD EED vee 11 2 1 nstrument
39. may assume that kg A is not a function of 22 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs 3 3 5 1 Hypothesis 1 The fraction of scattered light not received by the absorption meter is independent of wavelength We designate the true absorption coefficient at a wavelength by a A The true absorption coefficient at a given wavelength is the sum of the absorption due to pure water ay A and the absorption due to dissolved and suspended matter ap 2 n for non water so that ad ay A an A la The measured absorption coefficient is designated by am A The measured absorption coefficient consists of the sum of the absorption due to pure water ay A and the instrument output a A so that am aw A ai A 1b Similarly c4 A and cm A are the true and measured attenuation coefficients respectively We define the scattering coefficients similarly be A ed al 2a bm A Em A am 2b bi ci aj A 2c Dm by A bi 2d Using Hypothesis 1 we then set ad A ayn A kaby A 3 and c4 2 em A keb A 4 We have assumed here that the scattered light received by the attenuation meter is also independent of wavelength so that it is a constant proportion of the total scattering coefficient 3 3 5 2 Hypothesis 2 The fraction of scattered light received by the attenuation meter is independent of wavelength We now subtract equation 3 f
40. ntained in the device file included with a particular meter To convert the three characters that are received for each channel to a floating point number use this formula csig double charl char2 256 char3 256 0 Each group of signal data is preceded by a two byte time This is an integer in milliseconds Time starts at zero when the meter is powered on The time and data groups with 56 characters per group are repeated 10 times each packet After the time and data groups are sent a reference packet is sent This reference packet contains reference signal values for each wavelength in the a and c channels The reference packet uses the same format as the data packet in that nine groups of three characters are sent for a and c This means a total of 54 characters are sent for reference information The 3 byte 24 bit reference word is converted to a floating point value by using this formula cref double charl char2 256 char3 256 0 After the reference is sent a two byte temperature word is sent The temperature is given as a reading from a thermistor The manufacturer of the thermistor provides a table correlating the reading counts to temperature That table fits the polynomial equation given above Using 271 counts we get 10 61831 0 045113 271 4891 32 1 271 208130 2 1 271 1171473 1 271 7 69 degrees C At the end of the entire packet a four byte checksum is sent This
41. of internal temperatures refer to your calibration sheet If the meter s internal temperature is higher than the maximum calibration range it may be operating out of spec Setting the meter in a shallow pan of water immerse most of the lower can can help keep the instrument cool This process should be repeated until the air values are within specification This may take three or more iterations but is very important to be done carefully before taking your instrument to sea The air calibration procedure should be done again at the end of a cruise or data collection period to track the instrument s performance over time 1 5 System Requirements This section explains the basic system components necessary to operate the ac 9 If you are using the instrument in a standard profiling configuration you will generally need the components described below Other configurations such as mooring and underway flow through systems will require additional components such as battery packs data loggers anti fouling canisters and or de bubblers Test cables software and system configuration engineering may be obtained by calling WET Labs Alternatively you will find additional information about the various system components in the Technical Reference Section of this manual This manual concentrates on the use of the ac 9 as an optical profiling instrument The ac 9 was designed for easy use However certain system requirements for power and communica
42. of reflecting tube absorption meter Ocean Optics XII Proc Soc Photo Optical Instrum Eng SPIE Vol 2258 44 55 1994 Appendix A Optional Pressure Sensor An optional pressure sensor is mounted in the top flange next to the electrical connectors A plastic fitting and capillary tube both filled with silicone oil provide a buffer between the pressure transducer and seawater The transducer is both sensitive and delicate WARNING Never touch or push on the transducer The silicone oil in the pressure sensor needs to be checked approximately annually and thus is checked at the factory when you return your ac 9 for service and calibration ac 9 User s Guide ac 9 Revision T 31 July 2008 31 WET Labs Appendix B Legacy Connector Older ac 9 meters used a VSK 6 BCL bulkhead connector The pin and socket arrangements are shown below in Figure A 1 These differ from the standard configuration see Section 5 1 though the pin and socket functions are the same for either connector Sea or Test Cable Connector Pin or Socket Function 1 GND 2 RS 232 RX a ia 3 RS 485 i 2 4 V 10 18 VDC i piia 5 RS 232 TX to host 6 RS 485 ET abe ine etna LOOKING AT SOCKETS LOOKING AT PINS MFG IMPULSE OR SEA CON MFG IMPULSE OR SEA CON P N VSK 6 BCL Fi VMK 6 FS LOCK SLEEVE K FLS P N T SHOWN file 210034 dwg wmf Figure A 1 VSK 6 BCL Bulkhe
43. on system After de ionization the water is processed by a series of filters and held in a 60 liter reservoir that re circulates through an ultra violet chamber and additional filters Water for calibration is drawn through a final 0 01 micron ultrafilter before being used This system allows the highly reactive de ionized water to equilibrate and the ultra violet chamber prevents any biological contamination from entering the reservoir A qualitative but surprisingly accurate test of water purity can readily be obtained using a beaker and a small laser in a darkened room By shining the laser through the water and viewing slightly off axis to the impinging beam one can readily detect scatterers in the water Try this first with some tap water It will show you what to look for Be very careful to not look directly into the beam and do not try this test with a high power laser The pointing lasers used in lectures work very well for this task If you see more than one or two large scatterers in the beam path your water is unsuitable for a clean water calibration WET Labs and researchers from several institutions are presently developing methods for effective field calibration and in situ calibration techniques We provide tech notes on these methods as they are developed and validated 4 4 Linearity and Dynamic Range To determine linearity of the absorption and attenuation detection systems we first vary scattering agent concentrations and plot ap
44. parent a and c Typically linearity in throughout the dynamic range is extremely high with a regression coefficient of 0 9999 Linearity in a exceeds 0 998 Although this test doesn t truly measure a it does give a good indication of response We have also tested linearity in a using an orange food coloring Linearity in the blue channels was on the order of 0 999 The red channels showed very little change due to the food coloring If you try this test in the lab use a dye not likely to permanently stain the receiver One researcher has reported good results using Japanese green tea The scattering tests and dye tests also enable us to determine the instrument s dynamic range We determine dynamic range by dividing the response range for a given channel that exceeds a regression coefficient of 0 998 This relatively conservative criterion provides a worst case dynamic range of about 16 000 to 20 000 in aap The dynamic range of all other absorption channels exceeds 20 000 The worst case dynamic range is typically e412 which is about 50 000 That of the other c channels exceeds 100 000 4 5 Air Calibration and Tracking During the factory calibration clean water values are obtained and used to correct the meter s output We then thoroughly dry the meter and record air values We supply these values with the calibration page shipped with each meter These air values represent the effective measurement of air for a calibrated instrument
45. r a detailed description of the binary data format RS 232 Operating across an RS 232 cable you can obtain binary data from the instrument If you are using WETView read the operational instructions contained in the software manual 1 6 5 Care and Maintenance Built for field deployment the ac 9 requires minimal maintenance However following these simple recommendations will assure optimum data integrity as well as longer instrument life After a field deployment of the ac 9 you should clean the instrument prior to storage Refer to Section 1 4 for detailed cleaning procedures The following steps will help prolong the life of the instrument 1 Pressure housing Begin with a thorough rinsing of the unit and its flow tubes with fresh water If a dummy plug for the connector is available install it on the main bulkhead connector before flushing the instrument After rinsing towel dry the pressure housing and remove the flow tubes 2 Windows The windows should be cleaned with dilute soapy water followed by ethyl alcohol and should receive a final rinse with distilled or reverse osmosis filtered water This will remove any fingerprint oil grease or other contaminants from the windows Use lint free lens cleaning paper to avoid scratching the windows or detectors WARNING Do not use acetone on the windows It will damage the window holders ac 9 User s Guide ac 9 Revision T 31 July 2008 9 WET Labs 3 Flow Tubes
46. rence data is stored and accumulated for ten revolutions since its primary purpose is to compensate for long term drift After ten revolutions the reference channel data is sent along with a temperature reading depth optional and other status information 16 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs 3 Data Processing Data from the ac 9 is acquired and processed through the following steps 3 1 Analog Acquisition Optical radiation at the reference and signal channels is continuously monitored by operational amplifiers operating in a current to voltage configuration The amplifiers are configured for a gain of 5 10 3 2 Analog to Digital Conversion The A D used in the ac 9 maintains a two channel input Between the attenuation beam reference and signal channels and the absorption beam reference and signal channels there are four total channels to sample To meet this requirement an analog switch is placed after the primary gain stage of the inputs During a single filter wheel rotation the switch alternates input into the A D Signal and reference detectors for a given beam are sampled simultaneously by the A D The switch alternates readings between the absorption and attenuation paths The CPU reads the two A D channels and controls the analog switching based on encoder information from the motor that defines the exact filter wheel location 3 3 Digital Processing and Data Output The CPU takes multiple
47. rom 4 bt bm A iA ketka 5 so that DA bmEMV keka 6 Since using hypothesis 1 and 2 k and kg are constants we see that the measured scattering coefficient spectrum by A is only a constant 1 1 kg kg different from the true scattering spectrum We may the conclude that by AL b4 A2 bm 1 bm 22 7 ac 9 User s Guide ac 9 Revision T 31 July 2008 23 WET Labs We now assume there are some reference wavelengths Ar at which the absorption is due to pure water only 3 3 5 3 Hypothesis 3 There exists a reference wavelength r at which the absorption due to particulate and dissolved material is negligible For the present we set A 715 nm although there may be better choices further in the infrared Thus ar aw Ar 8 Substitution of eguation 8 into 3 gives awlAr Am kabel so that using eguation 1b 0 aj Ay kabir Rearranging gives air kabdAp 9 Hence ka aj Az be Ay 10 Substitution of equation 10 into 3 gives am A ay Ap b A biAz 11 Substitution of equation 7 into the above gives adA am A aj Ap bm A bm Ay After subtraction of the water absorption from both sides we get the desired result an A aj A aj Ap bmA Om Ay 12a Referring back to equation 2d we note that in the vast majority of natural waters bm A gt gt bw A so that in those cases equation 12a may be approximat
48. s that do not read 0 0 in clean water Therefore the final output of the ac 9 s software is the absorption and attenuation with clean water attenuation subtracted out To get absolute or true absorption the post processing steps discussed in Section 3 3 3 should be applied 4 1 Temperature Correction Beyond confirming basic instrument operation and alignment temperature corrections are the first tests performed on the meter Temperature corrections are requisite to all other tests Corrections are performed using WETView WETView employs a correction algorithm that uses multiple offset values Az values obtained by measuring output differences over small temperature increments Instrument values are collected and averaged every one to two degrees Celsius through the operational temperature range of the instrument From these values we generate a table of temperature compensation offsets 4iz This table is contained in each instrument s device file Using the table WETView 5 0 then applies the algorithm a araw Atm for given temperatures in the table For temperatures that fall between table values the program applies a linear interpolation upon the data for further correction By using this scheme we can thus effectively compensate for non linear changes due to temperature in the instruments output For a description of this algorithm please refer to section 3 3 3 Post Processing 4 2 Precision Precision is determined by
49. t the test cable to the proper COMM port on the computer Make sure the lead of the test cable is connected to the terminal of your power supply Make sure the lead of the test cable is connected to the terminal of your power supply Verify that your power supply is providing between 10 and 18 VDC and is rated for at least 2 amps WARNING Power input on this unit is diode protected from reverse polarity power up but this is not 100 percent insurance against damaging the meter nor will it protect it from over voltage 2 Connect the test cable to the instrument Apply power to the instrument and allow it to begin warming up 3 Run WETView When the interface is displayed you will need to provide a DEV file name Choose Open Device File from the File Menu at the top left of the screen The program will ask you to choose the COMM port Note that WETView supports COMM1 through COMM4 only 4 Click on the center button or lt F1 gt to start displaying data After 5 10 seconds tabular data should be displayed on the right side of the screen A real time graph will begin to develop depending on the graph parameters set at the time Refer to the WETView User s Guide for details of running the WETView software 5 After a short time again click on the center button that will stop the data collection and ask for a file name to apply to the data if you choose to save it Press ESC if you do not w
50. tion with the instrument must be met Figure 1 shows a block diagram of the basic system components required for instrument operation An explanation of these various components follows the diagram More detailed information on most of the system the instrument and data output format can be found in the Technical Reference Section of this manual Details about connecting the unit and acquiring data are included in this section WET IE aa SORTWARE POWER SUPPLY i so EE HOST ir COMPUTER BS 232 or R5485 CABLE PUMP ar 9 INSTRUMENT FLOW TUBE INLET Figure 1 Typical ac 9 configuration 4 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Required system components include Instrument The ac 9 and its flow tube assembly form the basic optical sensor Pump The flow through system will typically require an ancillary pump in order to assure proper operation Pump requirements depend upon desired flow rates required depth of operation power availability and existing hardware Cabling The ac 9 requires a minimum of four conductors for power and RS 485 output Three conductors for RS 232 output Power Supply The ac 9 requires a 10 18 VDC supply capable of providing a minimum of 9 watts continuous output If a longer cable is used power losses must be accounted for in determining the power supply voltage and power requirements Serial Communications Data from the ac 9 is tr
51. ts are sturdy but the optics can be jarred out of alignment when subjected to shock e The data will be adversely influenced by bubbles dirt or grease in the flow path or misalignment of the flow tubes Make sure that the flow tubes pump tubing and screens are free of dirt and grease Clean with ethanol or warm soapy water Rinse with clean tap or distilled water Do not allow water to dry on the windows as this will leave a residue that may be hard to remove e Secure the ac 9 to a sturdy cage or lowering frame that will protect the instrument from striking the deck ship s side or sea bed Use a dielectric isolator rubber sheeting or thick tape to isolate the instrument s aluminum case from the steel cage e When clamping the instrument to the cage make certain no torsional stresses are applied Even slight wracking of the instrument can alter the beam alignment This will severely degrade the data quality especially on the c side which requires an extremely tight alignment tolerance e Once every couple of days take a data file in air to track any instrument drift This procedure is outlined in Section 4 5 The instrument must be very clean and completely dry in order to achieve an accurate air calibration measurement Using compressed dry nitrogen or oil free dry air to blow the instrument s flow tubes and windows dry will speed the drying process see Section 4 5 Air Calibration Tracking Method e Upon deployment the
52. ubbles allowing them to be expelled from the ac 9 In addition operating the instrument in the horizontal position could potentially cause small variations in the meter s response due to the different orientation of the filter wheel s rotational plane Therefore we recommend that both air and water calibrations be done in the orientation in which the meter will be deployed to obtain the best results Mount the ac 9 as close to vertical as possible Orientation of ac 9 and pump if horizontal mounting is required Figure 2 ac 9 mounting positions Specific mounting instructions will depend upon implementation of the sensor To assure long term instrument integrity and optimum operation observe the following basic procedures 6 ac 9 User s Guide ac 9 Revision T 31 July 2008 WET Labs Do not make direct contact between the ac 9 pressure housing and a metal frame or hose clamp The ac 9 is available in one of two pressure housing materials Aluminum is used for full ocean depth rated units and an acetal copolymer plastic is used for the standard units rated to 500 meters The aluminum housing is hard anodized with a special plastic impregnation to assure minimum corrosion damage and to provide electrical isolation from the aluminum housing and its surrounding environment However metal to metal contact with the housing can damage this coating and result in possible corrosion of the pressure
53. ude a DC DC converter power supply motor regulation circuitry an optical encoder mounted on the motor amplification circuitry for the detectors an analog to digital converter and a microprocessor controller The filter wheel spins continuously at a nominal rotation speed determined by a pulse width modulation circuit The encoder breaks down a single rotation of the wheel into ac 9 User s Guide ac 9 Revision T 31 July 2008 13 WET Labs 512 steps The position information from the encoder is then read by the controller Signals from the detectors are amplified by a single stage current to voltage operational amplifier configuration After a post gain stage used for signal level shifting the signal is digitized by an 18 bit digital signal processing analog to digital converter A D The A D continuously samples the detector signals at a rate of 80 kilosamples second Its output is sent serially at 4 Mbaud to the controller The controller watches the encoder output to determine when to begin reading the A D Once sampling begins the controller collects and averages approximately 100 readings as a given filter scans through the light beam The encoder once again indicates when to stop sampling and the controller then begins processing the new data This cycle is repeated for each filter and associated blank space through the rotation of the wheel Processed data from the instrument is then sent serially in a binary format to a host data
Download Pdf Manuals
Related Search