History of Modifications to Model 1130

1
History of Modifications to Model 1130 Model
1135

• Frank Schaedlich

330 Nantucket Blvd. Toronto, Canada M1P 2P4
Rev 1.12 Aug 2004
2
Brief History

• Advent of Model 2537A in 1992 allowed automated
  TGM measurements
• In 1995 it was discovered that RGM, not Hg0 was
  the source of mercury problems in the Florida
  Everglades
• Researchers immediately wanted to apply the
  capabilities of the 2537A to the measurement of
  RGM
• Tekran started research on RGM characteristics
• Feb. 1995 High levels
• April 1996 Low level ambient RGM

3
Just how much mercury are we looking for?
Pipe Filled with Air
240,000 miles
Assume that one could build a pipeline from the
surface of the earth to the moon
Analogy courtesy Tom Brown, US DOE

• Ambient air global background level Total Hg
• Pipeline would be 240,000 miles long
• Background is 1.7 ng/m3 0.2 ppt (200 ppq)
  (v/v)
• Approximately 0.003 inches of the pipeline would
  be filled with mercury vapor

4
About this 0.003 length of mercury vapor
Pipe Filled with Air
240,000 miles
Assume that one could build a pipeline from the
surface of the earth to the moon

• Model 2537A performs this measurement
• Continuously, every 5 minutes
• With a precision of better than 1
• Mercury researchers now often refer to this
  background atmospheric level (1.7 ng/m3) as bulk
  mercury

5
RGM and HgP Levels
Pipe Filled with Air
240,000 miles
Assume that one could build a pipeline from the
surface of the earth to the moon

• HgP and RGM levels are hundreds of times lower
• Pipeline is still 240,000 miles long
• Typical RGM value is 6 pg/m3 (1 ppq (v/v))
  (300x lower)
• Approximately 0.00001 inches of this pipeline
  would be filled with RGM
• Model 1130 and 1135 take 2 to 3 hours for this
  measurement

6
Original Automated Approach

• Originally envisaged as a dual port sampling
  system
• Pyrolyzer port would measure total Hg
• Scrubbed port would measure elemental Hg
• RGM determined
• by difference
• Only available auto-
• mated solution
• based on current
• knowledge (1997)

7
Known Problems

• Looking for a very small RGM component by
  subtracting two large values
• RGM HgT Hg0
• Data would not be time simultaneous
• What about particulates?
• It wasnt known at the time just how large a
  problem these were

8
Newly Discovered Problems

• Method required that the RGM scrubber
• Absorb all RGM
• Pass all elemental Hg
• Work for a period of at least one week
• KCl only lasted a few hours before ceasing to
  work
• Saw just how difficult it was to transport HgCl2
• The Low point a presentation at a Tekran User
  Group Workshop in May 1997 entitled
  Speciation Simple Doesnt Work

9
New Method

• In May 1997 Tekran discovered thermal
  regeneration of KCl
• Characteristics
• Quantitatively trapped RGM while passing
  elemental Hg
• Quantitatively desorbed previously captured RGM
  when heated to 500 C
• RGM was desorbed in elemental form
• Regenerated KCl surface so that it would work for
  another cycle

10
Original RGM Apparatus
11
Model 1130-P Prototype

• Three units delivered May 1998
• Two have been upgraded
• (Requires replacement of DM and PM)
• Prototype limitations
• Could not work with 1135
• No viewing window
• Smaller sample pump

12
Model 1130Denuder Pump Module
13
Model 1130 Model 1135 Sample Flow Path
14
Model 1130 Model 1135 Desorption Flow Path
15
Quartz Denuder Assembly
Zero Air Inlet Tee
Sample Outlet
Annular Denuder
Elutriator
Impactor Housing
12 Scale
16
Changes to Model 1130 System

• RGM Unit

17
List of Abbreviations

• DM Denuder Module
• PM Pump Module
• CM Control Module (Controller)
• T/C Thermocouple (sensor)
• PID Temperature controller
• RPF Regenerable Particulate Filter

18
Generations of Heaters 1 2

• 1st Generation Stainless steel core
• Denuder fit inside stainless steel tube
• Fast to heat cool
• Shorted to ground
• Relatively short useful life
• 2nd Generation ceramic core
• Somewhat slower to heat cool
• Lasted longer
• Windings still sometimes shorted to T/C sensor

19
Generations of Heaters - 3

• 3rd Generation User Replaceable
• Ceramic core with tube for T/C probe
• No chance of shorts from coils to T/C
• Replacement of heater is now possible by end user
  in the field
• T/C probe may also be replaced separately
• Hi-temp terminal blocks within oven to terminate
  heater leads
• Heaters are available only for 115 VAC

20
User Replaceable Heater

• Denuder Module must be returned to Tekran in
  order to be retrofit to accept user replaceable
  heaters
• Once upgraded, users can change heater TN-305

21
Zero Air Canisters

• Original units were plastic
• Two layers of different carbon adsorbent
• Could be refilled by users
• New units are sealed metal canisters
• Disposable
• Give better zeros (Esp. in hot weather)
• Last for years

22
Zero Air Canisters

• Factory upgrade requires moving mounting
  brackets TN-302
• Same canister is used for the Model 2537A
• All canisters will cause problems if they become
  wet inside! TN-311

23
Keeping Canisters Dry

• Simple user upgrade alters plumbing inside PM
  TN-313
• Same upgrade kit also includes smaller restrictor
  for DM to lower zero air purge flow rate during
  sampling

24
Pump 1st Generation

• Used in Model 1130-P
• Closed loop MFM, feedback circuit and variable
  speed pump controller
• GAST DC pump with brushes
• Could barely maintain 8.5 l/m through denuder
  under optimal circumstances
• No chance it would work with an RPF

25
Pump 2nd Generation

• Used in Production Model 1130
• Larger version of same GAST pump
• Large enough to maintain flow
• Brush life was a problem
• GAST was useless in helping to resolve the
  problem
• Dug up a specification stating that expected
  brush life was 2000 hours (3 months)
• Couldnt even meet that
• Brushes very expensive (sold individually!)

26
Pump 3rd Generation

• New Thomas pump used in all Model 1130s
• Available as a field installable retrofit kit for
  older PMs TN-303
• Brushes last far longer
• Smaller physical size
• Larger pumping capacity
• Brushes are less expensive

27
Pump Upgrade

• Old GAST Pump

• New Thomas Pump

28
External (Impactor) Heater

• Original systems used heating tape and a separate
  T/C probe
• Tape rated for 70 VAC
• Tape could easily overheat if sensor not
  installed properly
• Difficult to change and rewrap

29
New External Heated Boot

• Custom made flexible silicone heater
• Integral T/C sensor
• Simply snaps into place
• Extra insulation may still be required
• Accepts higher PID power limit
• Field installable upgrade TN-301

30
Back Pressure Regulator - 1

• BPRs are not common and are hard to source
• 1st Original was plastic Airtrol unit
• Seemed to work OK in the prototypes
• Sometimes NPT ports cracked in new PMs
• Problem traced to vibration of pump
• Unit suffered from other reliability problems
• 2nd Refit Airtrol unit with soft tubing
• Still suffered from some reliability problems
• 3rd Airtrol with 10-32 ports
• Sometimes oscillated (whistled)

31
Back Pressure Regulator - 2

• New unit had to fit in available space
• German BPR manufacturer finally located
• Custom made, all metal unit
• Retrofit requires enlarging rear panel mounting
  hole (Factory refit)

32
Sealant Cord

• 1st Generation Buna (black)
• Became hard brittle at cold temperatures
• 2nd Generation Cold temp buna (black)
• Wider temperature range
• Stiff at all temperatures
• 3rd Generation Silicone (red)
• Better, but still did not seal very well
• 4th Generation CFlex (white)
• Wide temp range and soft and pliable
• User Retrofit TN-308

33
Mass Flow Meters

• No operational differences or improvements
• Common electrical interface
• 1st Tylan
• 2nd Sierra
• 3rd Hastings

34
DM Zero Switching Valve - V1

• 1st Generation
• Valve was mounted on back panel
• Occasionally froze if wet zero air was passed
  through it
• 2nd Generation
• Original valve mounted on standoffs to prevent
  freezing
• 3rd Generation
• New, larger orifice valve found
• Factory refit

35
PM Zero Switching Valve V2

• 1st Generation
• Square (rectangular) valve
• 2nd Generation
• Cylindrical shape
• Less restrictive than original V2
• Requires mechanical changes and electrical
  changes to PCB (uses more power)
• Factory refit
• Not an essential upgrade

36
Support (Vibration) Bracket

• Used to reduce vibration of lines when pump
  running at certain frequencies
• Requires new V2 as prerequisite TN-312

37
System Controller

• Original controller
• Also used for Model 1110 (aka flux box, TADS)
  and 1120 (aka Standard Addition Unit, SAU)
• Subject to static damage
• Operating at limits
• of design capacity
• Required 3 separate
• connectors to PM

38
System Controller

• New controller
• Completely redesigned inside (same outer shell)
• More robust
• Can store several methods
• Requires only one connector to PM
• Can be downloaded in the field with new operating
  program TN-200
• Also now used for Model 1110 Model 1120
• User Upgrade TN-306

39
DM Internal Filters

• Zero air supply filter is the usual 0.2 um Teflon
  membrane (also used for 2537A)
• Incoming sample filter (after denuder or RPF) is
  1 um glass membrane
• 0.2 um Teflon is too restrictive at 10 l/m
• Quartz would be OK but is needlessly expensive
• Original recommendation was for 0.5 um membrane

40
Sodalime Trap

• Prevents occasional passivation of gold after
  desorption TN-304
• Passes elemental Hg
• Not always required
• Not always 100 effective
• May not be best material for the job but has been
  commonly used in analytical applications (eg
  Method 1631)
• Change weekly (along with denuder)

41
DM Internal Case Heater

• Dual set point PID controller tries to maintain
  case temperature
• Internal heater comes on if case lt38C
• Cooling fan comes on if case gt40C
• Original screw base heater was 100 W
• New case heater is 200 W
• Easy user retrofit (P/N 35-13833-200)

42
DM Case Insulation

• Original material was yellow fiberglass
• New material has aluminum foil protective layer
• Less likely to absorb moisture
• Neater sheds fewer fibers when touched
• Electrically grounded foil
• Factory upgrading generally not required

43
New Exterior Vents

• Original oven and case exhaust vents were thin
  gauge aluminum
• New units have custom made stainless steel vents
  with S/S flapper
• Requires some drilling, so factory upgrade is
  recommended

44
Drip Shields Channel Case Only

• Rain shields located above door on Model 1130 DM
  and Model 1135
• Prevents water from dripping directly onto top of
  door seal
• Factory refit

45
NEMA Case

• New NEMA style case
• Waterproof
• Flat seal gaskets on doors
• Mechanically interchangeable with older style

46
NEMA Case
NEMA Case
Old Style Case
47
Changes to Model 1135 System

• Particulate Unit

48
RPF Tail Bracket

• Bracket supports tail at right angle fitting
• Reduces chance of tail breakage once RPF is in
  place
• Thumbscrews allow removal when changing RPF
• Factory refit

49
RPF Changes

• Less quartz wool at top of pyrolyzer
• Less restrictive
• Better zeros
• Looser packing of quartz wool at RPF
• Less restrictive

50
Method Changes

• Higher pyrolyzer RPF desorption temperatures
  (800 or 850 C)
• RGM heat broken into two steps
• RGM-Ht RGM, RPF and Pyro heaters all on
• RGM-H2 RPF and Pyro heaters turned off
• Why?
• Reduces power consumption
• Heaters stay warm enough to keep 1135 components
  transparent to mercury
• Less thermal shock on heaters when cooling begins
• Reduces cool down time

51
New Style PID Controller

• New PID Type
• Functionally equivalent
• Has data communications capability
• New Model 1135 have data-com cable
• For future use

Data communications cable
52
Other Changes to Model 1135

• User replaceable heater (F)
• Stainless steel exterior vents (F)
• 200 W case heater element (U)
• Exterior rain drip shield (F)
• New NEMA style case (PC)
• New insulation (n/a)
• Key
• F Factory Upgrade
• U User installable upgrade
• PC Product Design Change (not refittable)

53
Conclusion

• Work is ongoing to improve products
• Customer input is an important part of product
  development