Chemistry Electronics Services
ARCHIVE OF REPAIR TRICKS, USEFUL INFO, ETC.

HP/AGILENT HP59864B Micro-ion Ionization Gauge
also Granville-Phillips 342, 343 series vacuum gauge

TYPICAL FAILURE:

On the main 12VDC switching power supply, the cluster of three large 470uF filter caps next to the high-freq transformer will eventually go bad. They take a couple amps of 100KHz and run too warm, also being positioned right against the large hot TO-220 high freq output rectifier. This 12VDC switcher supply runs both the filament and 180VDC switchers as well as the front end op amp supply rail, so a few volts of 100KHz ripple isn't so good. Replace with better ones: digikey P12389.

The failed filter caps may kill the two protection zeners located between the caps and the high-freq transformer secondary. If one zener shorts, both will fry, so always replace both 1N4746 zener 18V.

SCHEMATICS:

120V LINE INTERFACE: The schematic for the main switching supply is pretty close to fig. 30 on the UC3845 datasheet "suggested application," off line regulator. (Also similar are the 170VDC and 2.5VDC ion-tube supplies found on the same board.)

UC3845B pwm current-mode controller
http://onsemi.com/pub_link/Collateral/UC3844B-D.PDF

On the main 120VAC supply the input network on pin-2 and pin-1 is chosen so the feedback winding on the large 100KHz transformer will put 16.5V on pin-7 of the UC3845 (and also puts 12V on the main secondary winding.) Rather than using the 56K resistor from 160VDC for the bootstrap supply shown in the specsheet, a common-source FET regulator with a zener is used to give 15V. Once the system wakes up, this FET bootstrap supply gets shut down by the higher supply coming from the 16.5V winding. And during normal operation, the main secondary winding supplies 13VDC to the three large 470uF electrolytics (13VDC initially, 12VDC when active and driving the ion-tube filament.)

A second UC3845 circuit converts the main +12VDC to +170VDC for the ion-tube grid. A third UC3845 circuit produces the ~2VDC for the filament (via a toroid transformer and dual diode, for floating filament pins.)

The filament control section takes the 10uA emission current as input from one of the filament pins, through 1N4007 diode, 499 ohm resistor, 1N4750 27V zener, then through a 250K resistor to ground. Voltage across the 250K is limited by a 9.1V 1N5239 zener. 10uA through the 250K gives 2.5V feedback signal. This is applied to a non-inverting buffer op amp and then to pin-2 of the UC3845 through 10K resistor. (The pin-2 node is designed to run at 2.5V: half the 5.0V internal reference in the UC3845.) This PWM switcher chip produces the filament volts (max 2.5V out) by driving the primary of a small toroid transformer, with the floating C.T. secondary leading to the filament pins via a dual diode for fullwave DC output.

The +30V floating filament potential isn't produced by another supply, but instead via the 10uA filament common-mode emission current to ground through that 27V 1N4750 zener (added to the 2.5V across the 250K.)

The front-end schematic is nearly identical to "6-decade LOG amp" circuit found on op-amp datasheets using the MAT-01 matched NPN transistor pair. See one example on p13 of LTC1050...

LTC1050 precision zero-drift op amp
http://www.linear.com/docs/1770

On the collector front-end, the HP gauge uses OPA129P op amp as input, an OP07, a 2.5VDC reference and OFFSET trimmer to calibrate the gauge to a pressure ref at 1.0 x 10^-4 Torr (zero volts sent to the AD section for 10^-4 pressure reading.) The AD input voltage comes from pin-6 of the OPA129P. Max 10^-7 vacuum gives +2.78V, while highest pressure 9.9^10-2 is approx -4V.

FAKE ION-TUBE TEST JIG

A zero input on the collector coax pin gives negative infinite log, (so, out of range, all zeros displayed.) To see a real reading we must apply some nanoamps to this input. A 10M or 22M ohm resistor between the 12VDC rail and the collector input will apply a microamp and give a reading between 10^-2 and 10^-1 Torr.

This gauge unit detects whenever no gauge-tube is plugged in and shuts itself down after 30 sec. To keep it turned on, load the filament connector pins with two 0.33 ohm 5W resistors in series. Create 10uA emission current using 10M + 3.9M of resistors in series, connected between the grid pin and either of the filament pins. That's 135V from filament to grid, over 14M ohms for approx 10uA.

MISC DATA:

Gauge tube filament:           2amps, 1.2V ( about 0.6 ohms)

Gauge tube emission current:    10uA across 135V

Gauge tube collector electrode: +7.0pA to +7.0uA

AD input volts, pressure, collector current:
  -?4V?  = 1 x 10-1 Torr = 7.0uA
  -2.79V  = 1 x 10-2 Torr = 0.70uA
  -1.39V = 1 x 10-3 Torr = 70nA
  +0.0V  = 1 x 10-4 Torr = 7.0nA
  +1.4V  = 1 x 10-5 Torr = 0.70nA
  +2.4V  = 1 x 10-6 Torr = 70pA
  +2.78V  = 1 x 10-7 Torr = 7.0pA











Created and maintained by Bill Beaty.