Test and Measurement Forum

HIGH PRECISION PROGRAMMABLE VOLTAGE REFERENCE - PART 2

Forum for discussing PicoLog

HIGH PRECISION PROGRAMMABLE VOLTAGE REFERENCE - PART 2

Postby Glovisol » Mon Dec 14, 2015 10:01 am

SYSTEM DESCRIPTION & PERFORMANCE DATA


The Math Servo optionally works in two different operating modes, as will be clarified by the examination of the schematic diagram.

- In OPERATING MODE #1 resistor R5 is wired in and switch Q2 only provides quick charge to C3 at start-up
- In OPERATING MODE #2 resistor R5 is omitted and switch Q2 provides on-off charge current to C3 on LOW END commands.


This description based on Picolog 1012 DA, but valid for most other machines. For example the 1016 will afford even closer control & precision, as well as two more outputs.

Suppose an output voltage of 1 V is required. Voltage level across storage capacitor C3 is buffered by IC1a. Output of IC1a reaches DA input channels 7 and 9, which are set as follows.

OPERATING MODE #1
CHANNEL 9 – LOW END
Metering: Volt c.c.
Scan time: 10.000 uS
OPTIONS
Format: V
Width:3
Positions: 5
Min. value: 0.6
Max. value: 3.0
Scale Equation: X*0.9982 (this is the calibration factor for this particular 1012 DA)
Alarm
Lower threshold:1,99
Upper threshold: 2.20
Digital output 1 (NOTE: this corresponds to output D0)

CHANNEL 10 – HIGH END
Metering: Volt c.c.
Scan time: 10.000 uS
OPTIONS
Format: V
Width:3
Positions: 5
Min. value: 1.99
Max. value: 2.01
Scale Equation: X*0.9982 (this is the calibration factor for this particular 1012 DA)
Alarm
Lower threshold: 0.00
Upper threshold: 2.00020
Digital output 2 (NOTE: this corresponds to output D1)

CHANNEL 7 – TEST MONITOR
This channel used for system monitoring only and needs not be occupied in normal use.

At turn-on C3 will be fully discharged and needs to be quickly brought up to 2 V. This is done by output D0, which is in the ALARM state, e.g. at level “1”, because channel 10 reads below 1.99 V. In this condition Q1 saturates, saturating Q2 which quickly charges C3 thru R4. As soon as Channel 10 reads above 1.99V, output D0 drops to level “0”, but charge to C3 carries on at a reduced rate thru R5 (220KOhm).

As soon as C3 voltage reaches 2.0002V, Channel 10 enters the ALARM state, D1 goes to level “1” (HIGH END) and Q4 saturates, bleeding charge from C3 thru R11. From now on the charge on C3 and hence its voltage, is “trapped” below 2.0002 V and stays there with a negligible see-saw effect.

Useful output at pin 1 of IC1b can be quickly turned off by removing the Vcc supply voltage. Output will drop to zero, even if capacitor C3 is still fully charged. If one needs to change output voltage quickly, an external positive voltage (e.g. pushbutton or external logic) to the base of Q3 will fully discharge C3 in no time.

In OPERATING MODE #2 R5 is omitted and the LOW END Channel works all the time, first to quickly bring charge to C3 to the required level and secondly to replace charge bled off by R7 and internal parasitic loss of the storage capacitor. In this example channel 9 would be modified as follows.

OPERATING MODE #2
CHANNEL 9 – LOW END
Metering: Volt c.c.
Scan time: 10.000 uS
OPTIONS
Format: V
Width:3
Positions: 5
Min. value: 0.6
Max. value: 3.0
Scale Equation: X*0.9982 (this is the calibration factor for this particular 1012 DA)
Alarm
Lower threshold: 2.0000
Upper threshold: 2.20
Digital output 1 (NOTE: this corresponds to output D0)

Hyb.PSU 21 - 1.0 V REF..plw
1 V Output PLW File
(22.04 KiB) Downloaded 43 times


Hyb. PSU 20 - 1.0 V ref. (Large).jpg
1 Hour test run @ 1 V output
Glovisol
Advanced User
Advanced User
 
Posts: 114
Joined: Sat May 11, 2013 10:03 am
Location: San Marzano Oliveto, Asti, Italy

Re: HIGH PRECISION PROGRAMMABLE VOLTAGE REFERENCE - PART 2

Postby Glovisol » Tue Dec 15, 2015 11:57 am

Hyb. PSU 24 - 15V ref. (Large).jpg
Math Servo Operation @ 15 V output
FURTHER RESULTS & DATA

The Math Servo has been tested in OPERATING MODE #2 at 15 V output. Conditions were:

- Resistor R5: omitted.
- Vcc Supply: 24 V.
- 10:1 resistor netwok to bring 15 V down to 1.5 V for the inputs of Picolog 1012. R101 = 4.700 Ohm; R102 = 42.300 Ohm; R101 + R102 = 47.000 Ohm. R102 made with a selected 3.300 Ohm in series with a selected 39.000 Ohm.

Accuracy and stability are excellent, considering that any error (the DA works @ 1.5V) is multiplied by a factor of 10.

Test results and PLW file enclosed, as well as a graph of voltage fluctuations @ 1 V output.

Hyb. PSU 24 - 15 V REF..plw
15 V out. pLW File with all settings
(23.12 KiB) Downloaded 38 times


Fluctuations @ 1V output.jpg
Excel graph showing output fluctuations @ 1 V output


The attachment Fluctuations @ 1V output.jpg is no longer available
Glovisol
Advanced User
Advanced User
 
Posts: 114
Joined: Sat May 11, 2013 10:03 am
Location: San Marzano Oliveto, Asti, Italy


Return to PicoLog

Who is online

Users browsing this forum: No registered users and 0 guests