See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/282334141

TESTING & OPTIMIZATION OF THE MATCHING RESPONSE TIME FOR THE

REAL TIME FEEDBACK CONTROLLED ICRH AUTOMATIC MATCHING
NETWORK (AMN) SYSTEM OF SST-1 WITH HYBRID COUPLER & TOTAL
TRANSMI...

Technical Report · December 2008

CITATIONS READS

6 253

10 authors, including:

Ramesh Joshi Manoj Singh

Institute for Plasma Research Institute for Plasma Research
55 PUBLICATIONS   62 CITATIONS    30 PUBLICATIONS   87 CITATIONS   

SEE PROFILE SEE PROFILE

Kumar Rajnish Siju George

Birla Institute of Technology, Mesra Institute for Plasma Research
66 PUBLICATIONS   262 CITATIONS    36 PUBLICATIONS   148 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Software estimation: An Empirical and Analytical Approach View project

Development of Microwave Coupling System for SYMPLE View project

All content following this page was uploaded by Ramesh Joshi on 01 October 2015.

The user has requested enhancement of the downloaded file.

 Technical Report

TESTING & OPTIMIZATION OF THE MATCHING

RESPONSE TIME FOR THE REAL TIME FEEDBACK
CONTROLLED ICRH AUTOMATIC MATCHING
NETWORK (AMN) SYSTEM OF SST-1 WITH HYBRID
COUPLER & TOTAL TRANSMISSION LINE

RAMESH JOSHI1, MANOJ PARIHAR, KUMAR RAJNISH, H M JADAV,

SIJU GEORGE, D RATHI, RAJ SINGH, D PUROHIT, S V KULKARNI &
ICRH GROUP
1
Email: rjoshi@ipr.res.in

INSTITUTE FOR PLASMA RESEARCH,

NEAR BHAT VILLAGE, GANDHINAGAR- 382428. INDIA

1
1. Introduction:
Ion Cyclotron Resonance Heating (ICRH) is a promising heating method for a
fusion device due to its localized power deposition profile, a direct ion heating at
high density, and established technology for high power handling at low cost. For
the same reason 1.5-Megawatt ICRH system is developed for steady state
superconducting tokamak (SST-1). (ref: 10.1)

SST-1 is a large aspect ratio tokamak with wide range of elongation (1.7-1.9) &
triangularity (0.4- 0.7). The maximum plasma current is expected to be 220kA.The
maximum toroidal magnetic field at the center is 3.0T. The major & minor radii are
1.1 meter & 0.2 meter. 1.5 MW of RF power is to be delivered to the plasma for
pulse lengths of upto 1000 second [1].

Since plasma-loading impedance is generally small as compared to the

characteristics impedance of the transmission system, a significant amount of
power will be reflected from the antenna back towards the generator giving very
high reflections to SWR sensitive RF tube. Matching network is used to match the
total Tx-line system to the antenna so that the RF generator sees a matched load
and can operate at high efficiency. In the ICRH system, coarse matching network &
the automatic matching network designed for the impedance matching of the
system to on-line transfer maximum power during the plasma formation. (all
figures has shown in ref. Appendix XI at last page.)

Slow on-line impedance matching (say in order of second) would be achieved with
the coarse-matching network. It consists of following components.

Coarse Stub: Mechanical shunt tuners are added at appropriate positions on the
main transmission line. The reactance of such shorted stub is varied by varying
length of the stub.

Coarse Phase-shifter: Mechanical phase shifters are added to the transmission line.
The reactance of such shorted stub is varied by varying length of the phase shifter.
As it involves mechanical movement, matching is expected to achieve in few
seconds.

Automatic Matching Network (AMN):

Automatic matching network would match on-line impedance within milliseconds

order of time according to the variation of various plasma parameters like plasma
density, plasma position etc. (ref. 10.3)

AMN consists of two motorized 0.5-meter stubs & two motorized vacuum variable
capacitors (VVC) connected in parallel to each other. This network is located
between the generator & the coarse matching network. In the beginning of any
plasma shot operation AMN systems are kept in NULL position. At Null position

2
 of the system, inductive reactance (XL) provided by the stub movement is equal in
magnitude but opposite in direction of capacitive reactance (XC) provided by the
VVC, thus net reactance in the system cancels out each other’s effect.

During tokamak operation plasma load impedance may be inductive and capacitive
depending on the plasma parameters & its position of plasma in vacuum vassal.
Hence, AMN system has to operate in such a way so it can match both inductive &
capacitive loads. Keeping the 0.5meter stub to lower limit position & VVC at its
6th turn of rotation ensures null position in the system. Movement of VVC motor
towards lower turn from its null position provides inductive reactance to the system
& towards higher turn provides capacitive reactance to the system. So in AMN
operation only VVC motor moves & stub remains static [1][2][3].

24 probe signals are taken out from the probe section of the ICRH-transmission
line. These signal are acquired, digitized in high speed analog board at the rate of
1ms. Software programming written in VME system does the calculation from
these signals as per the RF error signal equations & generates the directional signal
to motor-controller for the movement of the VVC motor in case of mismatched
load.

Figure 1: Real time feedback controlled ICRH – AMN System for SST-1

Magnitude of the probe signals change because of the new VVC movement &
again new error signal calculations are performed which causes new directional

3
 signal to be generated to the motor-controller for the VVC movement. This keeps
continuing till the error signal becomes zero (it ensures that load is matched).

So AMN is feedback control system as shown in figure 1(ref: 10.4), which is

controlled by the real time VME system and its response time is expected to be in
milliseconds. Hybrid coupler equally divides the input power to two ports if the
load connected at both port is of same value, otherwise all power will dump into
the fourth port (dummy load). Matching network is used to match the total Tx-line
system to the antenna so that the RF generator sees a matched load and can operate
at high efficiency which includes DAC system with motor controller software
coordinately working with driver software program using computer system.

2. ICRH System overview: The block diagram of the ICRH system as shown
in the figure2 having following sub-systems.

Probe section: From this section 22 voltages probe signals are taken out for
VSWR plot. Two signals are used for error signal calculation of AMN & hence for
matching. The distance between two probe positions is 210 mm.

Direction Coupler: From direction coupler forward & reflected power signals are
taken out for error signal calculations of the AMN.

RF Automatic Coarse
source Tx. Probe Direction
Matching Matching Interface
Antenna
1.5 MW Line Hybrid
section coupler
Network Network
Port6
Coupler
Automatic Coarse
Dummy Load Probe Direction Antenna
Matching Matching Interface
section coupler Port14
Network Network

Figure 2: ICRH System

Automatic Matching Network:
Two motorized .5-meter stubs (movement controlled by stepper motor).
Two motorized Capacitors (movement controlled by Servo motor).

Coarse Matching Network:

Single 1.5 meter stub (movement controlled by stepper motor).
Single Phase shifter (movement controlled by stepper motor).

Interface:
Interface section provides the separation between the vacuum of the tokamak &
atmosphere pressure of the transmission line by two RF vacuum feed-throughs. It
also provides mechanical support to the antenna box.
3. Aim of Testing:

4
The main objective of this experiment is to optimize the response time of both
Automatic Matching Network (AMN) lying with each transmission line with the
following related sub-objectives.

• Testing of the coarse-stub operational logic & position limit Interlock.

• Testing of the phase shifter operational logic & position limit Interlock.
• Testing of the static-stub operational logic & position limit Interlock.
• Probe signals acquisition.
• Error signals & reflection coefficient calculation.
• Generation of CW/CCW signal for the AMN operation.
• Evaluation of associated hardware & software performance and optimization.
• In first phase tested tx-line1 with single AMN system ref:10.5
• Second phase for testing tx-line2 with AMN system singly which has same
identical result as above point.
• Test Both AMN System on both tx-line for achieve optimum response time for
matching with plasma impedance to antenna impedance.

4. System requirement for Interfacing with the DAC:

(4.a) Motor & its Driver:

Motor & its driver are selected as per required performance from the system
considering its physical, mechanical, electrical requirement & interfacing with
DAC.

(4.b) DAC Hardware:

Isolation card: As the isolation between system side and DAC side has to be there,
so all field signals are firstly conditioned in the isolation cards placed in the signal
conditioning rack.

Fiber Optic Tx/Rx cards: The entire motor driver control signal is fiber optic
based. It serves dual purpose of providing noise immunity & isolation to signals.

RF-Detector: It converts the RF signal (45.6 MHz) picked from the transmission
line through disc probe to DC signals.

VGD4 card (Analog Input): VME based card digitizes all probe signals obtained
from the signal conditioning cards.

VMIC –2528 DIO Card: VME based card used for Digital Input/Output signals.

VMIC-9660 Card: Carrier board for Timer card (IP-480) & Analog Output card
(IP-220).

(4.c) DAC Software:

5
VME Control Programming: The programming on the VME system includes
card driver programming, system integration programming for the ICRH-SST-1
operation & control. In this AMN control loop operates within 5ms.

VVC Motor Driver Programming: Motor control programming is needed to

control the motion profile of the motor e.g. velocity, acceleration, moved step size,
S/W limit of VVC motor in the motor driver/controller.

Data Acquisition: MDS+ based data acquisition module has been implemented. It
can acquire data at the rate of 1ms for 1000-seconds operation.

Data Analysis: Jscope based Data analysis software module has been implemented
for off-line (post shot) analysis.

(4.d) DAC Signals:

Detail list of the motor driver signals of one arm of ICRH system, interfacing with
the DAC is shown in appendix I.

5. Test-setup:

The block diagram of the ICRH system for the testing & optimization of the AMN
system’s response time is shown in figure-3.

Figure 3: ICRH Test Setup for AMN Testing

(5.a) Hybrid coupler:

Hybrid coupler equally divides the input power to two ports if the load connected at
both port is of same value, otherwise all power will dump into the fourth port
(dummy load).

(5.b). Dynamic Load (R || C) preparation:

As the plasma environment is very dynamic in nature and keeps changing during
plasma formation, so to simulate the same condition as of plasma load, a very fast-
motorized variable load is made from the variable capacitor (35pf-1000pf with 48

6
rotation) & resistor (10 Ohm/ 1000W) as shown in figure4. To minimize the
radiation the load assembly is made to co-axial type. This variable load is
characterized at ICRH operating frequency of 45.6 MHz with the help of VNA.

GND GND R o ta tion Z lo a d

35-1000pf
10 ohm

0 (C m in ) 018 + j106
4 137 + j490
8 040 - j64.8
12 031 - j34.4
16 030 - j25.6
20 029 - j21
24 28.7 - j18.9
VNA : :
44 27.8 - j13.8
4 8 (C m a x ) 27.7 - j13.2

Figure 4: Variable load & its characterized value

(5.c) Integrated System with the DAC:

Figure 5: VME Based DAC controlled matching network of ICRH System

7
Figure5 shows the block diagram of the DAC system integrated with both ICRH-
Transmission line section.

6. Acquisition & Control:

(6.a) Equations used to control the AMN:

The AMN consists of .5meter stub (provides inductive reactance to system) &
VVC (provides capacitive reactance to system). In the beginning, the system is kept
at Null position (inductive reactance & capacitive reactance is equal i.e. XL = XC.
This ensured by keeping the static stub to lower limit position & capacitor at 6th
turn of rotation).
Any load mismatching at system will generate an error signal for both the antenna
end VVC and generator end VVC.

Error A (Ea) = (V1- Vf )* Sin2bL2 - (V2 - Vf )* Sin2bL1

Error G (Eg) = (V1- Vf )* Cos2bL2 - (V2- Vf )* Cos2bL1
where
b = 2* pi* freq / C
Frequency (operating frequency of ICRH)=45.60E6 Hz
C (velocity of light) = 3.0E8 meter/second
(Ea & Eg) >0 then move the VVC motor M1 & M2 in CW direction
(Ea & Eg) <0 then move the VVC motor M1 & M2 in CCW direction
Ea= Eg = 0 then No movement

Local Voltage1: V1 (voltage at distance L1 from AMN ref. point)

Local Voltage2: V2 (voltage at distance L2 from AMN ref. point)
Forward power: Vf (from directional coupler)
Reflected Power: Vr (from directional coupler)

(6.b) Vacuum Variable Capacitor (VVC) operation:

H/W Limit H/W Limit

1 2 3 4 5 6 7 8 9 10 11 12

20pf 35pf 50pf 65pf 80pf 95pf 125pf 140pf 155pf 170pf 185pf 200pf
110pf

Figure 6: Capacitance of VVC per Rotation

VVC provides capacitance of 20pf-200pf in full 12 rotations i.e. 15pf per rotation
as shown in the figure6.

8
 As the plasma load may be inductive and capacitive, so the VVC has to operate
such that it can provide both inductive and capacitive reactance to the system when
load changes during plasma formation. For this purpose the AMN operation always
starts from the Null position of the system (Null position is ensured by keeping the
0.5meter stub to lower limit position & capacitor at its 6th turn of rotation). Thus in
the AMN operation only VVC motor moves and 0.5meter stub remains set at fixed
position.
6- 0 turn: movement of VVC motor will provide inductive reactance to the system.
6-12 turn: movement of VVC motor will provide capacitive reactance to the
system.

(6.c) Signals for the VVC motor driver:

Programmable motor driver controllers are responsible for the VVC’s motor
movement (motions parameter like velocity, acceleration, minimum step count,
direction, limit etc can be uploaded & downloaded from the serial communication
in the driver from PC) as shown in figure 7. Driver requires following digital input
signals from the VME system (generated from the AMN control equations) to run
the motor [4].

D-16000 D16000

CCW Direction CW Direction

-5 -3 -2 -1 +1 +2 +3 +5

-4 +4
m
om

Zoo

S/W Limit S/W Limit

zo

-6 0 +6
H/W Limit

D-3800 D-1000 D-800 D-600 D-400 D-200

0 D200 D400 D600 D800 D3600 D3800 1
-1
D-4000 D4000

CCW Direction CW Direction

1 turn : d4000 (distance travel in 4000 counts) 1 turn : d4000 (distance travel in 4000 counts)
1/20 turn : 1/20 turn: d-200 1/20 turn: d200
4 turn: d-16000 (S/W Limit in CW) 4 turn: d16000 (S/W Limit in CCW)
6 turn: d-24000 (H/W Limit in CW) 6 turn: d24000 (H/W Limit in CCW)

Figure 7: VVC motor’s motion controller

• H/W Enable (HE): Enables the controller to drive the motor.

9
• S/W Enable (SE): Enables the controller to drive the motor in program mode.

• Homing Enable (HO): Motor will travel towards the home (initial position /
null position / distance D0 counts) direction from any current position.

• CCW Direction (CCW): Motor will travel in CCW direction & in single sense
of same direction signal; the motor will travel distance to the set count value in
controller programmer (e.g. d-400, d-1600, d-4000).
• CW Direction (CW): Motor will travel in CW direction & in single sense of
same direction signal; the motor will travel distance to the set count value in
controller programmer (e.g. d400, d1600, d4000).

For the safety of the VVC, motor driver is programmed such that motor can move
maximum four rotations in one direction from its home position (system’s null
position). So software puts limit to the VVC of maximum eight rotations.

6-2 turn movement (110pf– 50pf): provides inductive reactance to the system.
6-10 turn movement (110pf – 170pf): provides capacitive reactance to the system.
The VVC motor makes one rotation (15pf capacitance
change) in d4000 counts of the driver. So to move 1/20 rotation (change of 0.75pf
capacitance) d200 count is needed.

(6.d). Flow chart of VVC motor operation:

VVC motion control programming written in the motor driver is given in
appendix II.

(6.e). Flow chart of AMN operation:

Automatic Matching Network (AMN) programming written in VME system is
given in appendix III.

7. Testing Procedure:

(7.a). Calibration of the system:

Movement of static stub, coarse stub, phase-shifter, and the VVC motor has been
calibrated from the DAC-VME system. Each system has been interlocked with the
Hardwire (from the limit switch signals of the system) & Software (max. no of
pulse signals to the system). Calibration results are shown in appendix IV.

(7.b). Calibration of the Direction Coupler:

Direction coupler is calibrated at different load conditions (i.e. reflection coefficient
value from 0 to 1) and error in reading is observed at the every 10 degree of
direction loop as shown in appendix V.

(7.c). Calibration of the probe signals:

10
The entire probe signals are calibrated from the source end (RF detector) up to the
destination (VME Analog input terminal) end as shown in figure 8.

Tx Line
Probe Probe Probe Probe Probe Probe
disc disc disc disc disc disc

VGD4
RF Inverter Isolation
I/V Card AI
detector card card Terminal
Figure 8: Probe signal calibration in electronic cards

This is ensured by
1. By keeping the penetration depth of disc probe same for each signal.
2. Electronics card is calibrated.

(7.d). Matching from the Coarse N/W:

The motorized variable load is varied from its 0 turn to 48 turn and by varying the
movements of the Coarse matching network (1.5 meter coarse stub & phase shifter)
remotely from the VME system, impedance matching of the system is achieved.
Matching is seen from the VNA before the AMN system as shown in figure 9 and
matching results are shown in appendix VI.
AMN SYSTEM

VVC2 VVC1
COARSE
PHASE Dynamic
SHIFTER Load

RF
Source
T

COARSE
VNA STUB

Figure 9: Load impedance matching from the Coarse matching networks

(7.e). Error signal calculation in matched load:

Dynamic Load is set at 16th turn of capacitor (ZL=30 - j25.6) and the starting of the
coarse matching network matches it up to the most possible value (It ensured by
monitoring forward & reflected power at the generator end in the power meter as
well as from the probe signals obtained from direction coupler at different input
power level).
In matched load cases all probe signals should have identical values (i.e. V1= V2 =
Vf) for VSWR=1,so error signal should be zero as per error signal equations.

11
 Ea = (V1- Vf)* sin2bl2- (V2 -Vf) sin2bl1
Eg = (V1- Vf )* cos2bl2 - (V2 -Vf) cos2b l1

But it is observed that even in matched case the Error A (Ea) & Error G
(Eg) has non-zero values. So AMN motor will continuously rotate alternatively in
both directions. Effect of this on the AMN system is shown in appendix VII.

(7.f.) Keep error signal window margin in software:

To prevent the motor from running in the oscillating mode an error margin is added
in the software & this error limit doesn’t adversely affect the load impedance
matching as shown in appendix VIII.

8. Optimization of the AMN Response time:

Test setup for the optimization of response time of AMN is shown in figure 10.
RF source set at –5.7dBm(700w Power at 50-ohm dummy Load)
Operating frequency: 45.6 MHz.
Coarse stub & phase shifter kept at fixed position

A M N S Y S TE M

VVC2 VVC1

RF
S ource P ow er
M eter PHASE
S H IFT ER
C O A R SE
STUB
D ynam ic
Load
Figure 10: Load impedance matching from the AMN

(8.a). Changing the speed parameter of VVC motor

As the VVC motor’s speed parameter e.g. velocity, acceleration, move counts are
changed in motor controller the response time of matching is reduced as shown
below in two cases -
CASE 1:
• Motor Parameter set in VVC (i.e. AMN) Driver
Velocity: 20rps
Acceleration: 150 rev/sec/sec
One step count: d400
• Dynamic Load changes (by varying the capacitance) the impedance to
new value.

12
 • Power meter value shows following reading.
Before AMN operation start: Pf = 350W, Pr = 240W
After AMN operation start: Pf = 650W, Pr = 10W
• Matching time: ~1.3second

CASE 2:
• Motor Parameter set in VVC (i.e. AMN) Driver
Velocity: 50rps
Acceleration: 2000 rev/sec/sec
One step count: d400-d4000
• Dynamic Load changes (by varying the capacitance) the impedance to
new impedance value
• Power meter value shows following reading.
Before AMN operation Start: Pf = 350W, Pr = 240W
After AMN operation Start: Pf = 650W, Pr = 10W
• Matching time: ~500msecond
So As the acceleration, velocity & move count increases, matching time decreases.

(8.b.) VME Loop refresh time:

As shown in figure 11, loop refresh time is the time when VME takes new values
from the analog input card’s buffer memory for error calculation & generates the
direction signal for the AMN motor.
In the VME software programming this time is changed from 250ms to 5ms, so
motor controller’s digital input signals (CW / CCW) get refreshed at every 5ms.

M O TO R
VM E DO
V M E S /W F O C T X /R X C A R D C O N T R O L LE R
C A R D T E R M IN A L
I/P S E N S O R

2 5 0 m s (in itia l R e fre sh tim e )

5 m s (N e w R e fre sh tim e )

Figure 11: Adjustment of loop refresh time

After setting loop refresh time to 5ms, with same speed parameter in the motor
driver as in above-mentioned case-2 (8.a), matching time reduced to 200ms.
Matching time: ~300msecond

(8.c) Adjustment of debounce time in motor controller:

As per the theoretical calculation matching time for the above setting of variable
load should be in around ~150msecond.Careful observation showed (as shown in
appendix IX) that VVC motor starts moving every time after 50ms of given AMN
Start/Stop signal (or CW/CCW signal at controller). So there is a time lag between

13
the motor controller input sensor section & its driver section of appx. 50ms .It is
observed that default setting of input debounce time in motor controller is 50ms as
shown in figure 11.

MOTOR
VME DO MOTOR
VME S/W FOC TX/RX CARD CONTROLLER
CARD TERMINAL DRIVER
I/P SENSOR

50mS debounce
5ms (VME Refresh time)
5mS debounce

Figure 12: Adjustment of debounce time in controller

Debounce time: period of time that the Digital input must be held in fixed state before the
driver recognizes it.
Default setting of 50ms time was changed to 5ms in the controller programming as
shown in the figure 12. Observation showed (appendix X) that now VVC motor
starts moving immediately after getting CW/CCW control signal & matching time
reduced to 140msec.

(8.d) Adjustment of step size of count in motor controller:

As the step size in the motor controller changes from d400 count (1.5pf / step) to
d1200 count (4.5pf / step) keeping the other motions parameters same in the motor
controller, matching time reduced to 90ms for the same set value of variable load as
shown in appendix XI.
Matching time: ~90msecond

As the step size in the motor controller changes from d1200 count (4.5pf / step)
count to d4800 count (18pf / step) keeping the other motions parameters same in
the motor controller, matching is not achieved for the same set value of variable
load, because AMN makes the system oscillate between matching & mismatching
region as shown appendix XII.

Matching time: oscillating

As the step size in the motor controller changes from d4800 count (18pf / step) to
d3600 count (13.5pf / step) keeping the other motion parameters same in the motor
controller. Matching time reduced to 45ms for the same set value of variable load
as shown in appendix XIII.

Matching time: ~45msecond

(8.e) Observation:

14
 •It observed that load is set such that VVC has to rotate (9/10)th rotation
(d3600 count) to match it.
• At this setting of load, experimental matching time & calculated matching
time is same.

Velocity=50 Acceleration = 2000

step to Match the Load = D3600(9/10 rotation)

Step count move time No of move Total time

1/10 rot.(1.5pf) D400 15ms 9 135ms
3/10 rot.(4.5pf) D1200 25ms 3 75ms
9/10 rot(13.5pf) D3600 43ms 1 43ms
1.2 rotation(18pf) D4800 49ms over step Oscillating

9. Conclusions:

In Previous report (ref 10.5) we had given the results of single transmission line
using automatic matching network. In this report we present the results of the total
transmission system which includes RF Generator, Hybrid coupler & two
transmission lines with two variable loads to be simultaneously using automatic
matching network unit.

The design, development & testing of the AMN is very important aspect for the
ICRH experiments in tokamak, where plasma impedance changes dynamically.
This report explains all the technical details of the matching achieved with dynamic
load in a very short duration. It is also observed that this matching time is shorter
than many matching systems in other tokamaks. Following are the important
conclusions.
• All the H/W and S/W delay from the DAC is optimized up to ~2ms.
• Mechanical inertia time of this VVC motor is negligible (~ .1ms)
• VVC motion is optimized to 1.5pf/15ms.
• Motor will move as fast as set acceleration & velocity in controller.
• Once the motor-controller get CW/CCW signal, motor will move one
set count value (d400/d4800).
• Small count can match the load but matching time will be more (d400
count preferably).
• Large count will make system oscillate.
• If time scale or magnitude of plasma impedance variation is known then
count value can be optimized further to minimize the matching time.
• Set count (distance) into motor controller from the VME system is not
possible, because controller has its own programming (compu-motor

15
 S/W) which can be uploaded/downloaded from serial communications
only.
• Total matching time is 135 ms with d400 count of motor controller
program & vme response time for feedback loop is 2 ms.

10. References:
1. Engineering design report (EDR) of ICRH-SST1
SST-2.03-050199-RF Group

2. Durodie F., Veriver M.

European physics society (EPS), 1992,p 81

3. D. Bora, Sunil Kumar, Raj Singh, K. Sathyanarayana, S.V. Kulkarni

A. Mukherjee. …& RF group.
Cyclotron resonance heating systems for SST-1
Published in Nuclear Fusion:Volume 46 Number 3 March 2006

4. Manoj Singh, Kumar Rajnish, Ritesh Sugandhi………….& RF Group.

VME based On-Line Impedance Matching system for ICRH experiment on SST1
Presented in 21st National Symposium on Plasma & technology

5. Manoj Singh, Ramesh Joshi, Kumar Rajnish………….& RF Group.

Technical Report of Automatic Matching Network for Txline1 for ICRH
experiment on SST1

Appendix-I
Driver/Controller signals for Motor

16
 Digital output signals:
S.N. Signals Name Signal Type Controller Logic No. Of Channel
1 Static Stub1 CW TTL Active High 1
2 Static Stub1 CCW TTL Active High 1
3 Static Stub1 CW TTL Active High 1
4 Static Stub1 CCW TTL Active High 1
5 Coarse Stub PULSE TTL Active High 1
1
6 Coarse Stub SIGN TTL Active High 1
7 Phase shifter PULSE TTL Active High 1
1
8 Phase shifter SIGN TTL Active High 1
2
9 Motor-1 H/W Enable TTL Active Low 1
10 Motor-1 S/W Enable TTL Active High 1
11 Motor-1 CW TTL Active High 1
12 Motor-1 CCW TTL Active High 1
13 Motor-1 Homing TTL Active High 1
2
14 Motor-2 H/W Enable TTL Active Low 1
15 Motor-2 S/W Enable TTL Active High 1
16 Motor-2 CW TTL Active High 1
17 Motor-2 CCW TTL Active High 1
18 Motor-2 Homing TTL Active High 1
1. Sign pin will be +5V for CW signal, 0V for CCW signal
2. To enable the signal, controller needs 0V(Active Low)

Digital input signals:

S.N. Signals Name Signal Type Limit Signal No. of Channel
High/Low
1 Static Stub1 Upper Limit High/Low NO/NC 1
2 Static Stub1 Lower Limit High/Low NO/NC 1
3 Static Stub1 Upper Limit High/Low NO/NC 1
4 Static Stub1 Lower Limit High/Low NO/NC 1
5 Coarse Stub Upper Limit High/Low NO/NC 1
6 Coarse Stub Lower Limit High/Low NO/NC 1
7 Phase shifter Upper Limit High/Low NO/NC 1
8 Phase shifter Lower Limit High/Low NO/NC 1

Analog signals for monitoring and control:

S.N. Signals Name Frequency Action No. Of Channel

1 Probe signal DC M/A/C 22

2 Forward Power DC M/A/C 1
3 Reflected Power DC M/A/C 1
Monitoring-M, Acquiring-A, Controlling-C

17
 Appendix-II

Flow chart of VVC Operation (As written in Motor Controller)

Enable the motor drive

Motor is at home Position(D0)

Set the CW rotation limit – 4 turn(D16000)

Set the CCW rotation limit – 4 turn(D-16000)

Direction control signal( generated by VME) in motor controller

Is CW Is CCW Is home
signal ? signal ? signal ?
yes

yes

yes
No CW VVC is in No CCW VVC is in
yes yes
movement CW Limit movement CCW Limit

Move towards home

NO
NO

position (D0) without

checking any condition
Move 1/10th rotation CW(D400) Move 1/10th rotation CCW(D400)

18
 Appendix-III

Flow chart of AMN Operation (As written in VME System)

AMN START

Probe signal Acquisition & Digitization

Reflection coefficient calculation

Refl. coeff Generate Home

is within operational NO signal for motor1
Range & motor2
YES

CALCULATE ERROR SIGNAL

(.2v)<Ea > (.2v)<Eg > Eg <-.2v

Ea >.2v Ea <-.2v Eg >.2v
-(.2v) -(.2v)

Generate Generate Generate Generate

No signal for CW direction CCW direction No signal for CW direction CCW direction
motor M1 for motor M1 for motor M1 motor M2 for motor M2 for motor M2

VVC in VVC in VVC in VVC in

5 ms CW Limit CCW Limit CW Limit CCW Limit
Loop
time No CCW
No CW No CW No CCW
movement movement movement movement
NO

NO

NO

NO

1/10 th CW 1/10 th CCW 1/10 th CW 1/10 th CCW

Rotation Rotation Rotation Rotation

19
 Appendix-IV

Calibration of the system

System Motor Moveable Effective Pulse Total Max. Limit

length length /mm pulse Frequency
(mm) (mm) (speed)

Static Stepper 500 330 125 40K 500Hz Lower (CCW)

Stub1 Upper (CW)

Static Stepper 500 330 125 40K 500Hz Lower (CCW)

Stub2 Upper (CW)

Coarse Servo 1500 1320 500 660K 2000Hz Lower (CCW)

stub Upper (CW)

Phase Stepper 1500 1470 650 955K 1500Hz Upper (CCW)

Shifter Lower (CW)

VVC Servo 12 8 rotation 4000 16000 Programmed

rotation 4 rot. Count Count velocity,
(CW) Per Per acceleration
4rot. rotation Direction
(CCW)

20
 Appendix-V

Calibration of the Direction coupler:

•I/P Power is fixed at 600 watt.

• Load is set at different value of Г.
• Vf & Vr is measured at every 10 degree.
• In match load ~8% error comes out.
• Good results of Direction coupler come at angle of 40 degree.

Angle Г=0 Г=. 2 Г=. 4 Г=. 54 Г=. 69 Г=. 89

0 12% 36% 47% 51% 76% 100%
10 12% 28% 47% 54% 77% 100%
20 11% 36% 42% 53% 70% 100%
30 10% 35% 42% 56% 78% 103%
40 8% 22% 41% 56% 73% 94%
50 16% 21% 47% 60% 82% 94%
60 32% 22% 52% 68% 82% 95%
70 62% 36% 77% 76% 83% 97%
80 107% 100% 88% 89% 96% 92%
90 142% 136% 104% 100% 96% 89%
100 262% 205% 133% 104% 102% 97%
110 336% 420% 178% 126% 116% 106%
120 461% 569% 188% 144% 125% 97%
130 522% 531% 208% 152% 122% 101%
140 777% 569% 239% 161% 131% 101%
150 1167% 400% 221% 164% 121% 98%
160 973% 435% 218% 182% 124% 104%
170 1090% 381% 220% 180% 127% 100%
180 1350%
416% 227% 178% 127% 104%

21
 Appendix-VI

Matching from the coarse network system

Rota- Zload Zin before VSWR Return Stub Phase Zin after
tion matching Loss (From LL) Shifter matching
(From LL)
48 27.7 - j13.2 48.4 - j30.8 1.001 -58dB 465mm 894mm 50.031 –j.0823
(Cmax (Up) (Up)
pos)
44 27.8 - j13.8 49.99 -j92.7 1.01 -40.5 No No 49.99 – j0927
movement movement
40 27.9 - j14 49 – j1.22 1.03 -36.2 No No 49 – j1.22
movement movement
36 28 - j15.5 48.45 - j1.9 1.05 -31.9 No No 48.45 - j1.9
movement movement
32 28.2 - j16.0 48 - j3.09 1.078 -28.5 No No 48 - j3.09
movement movement
28 28.4 - j17.2 47 - j4.25 1.025 -25.5 No No 47 - j4.25
movement movement
24 28.7 - j18.9 45.7 – j5.8 1.162 -22.5 No No 45.7 – j5.8
movement movement
20 29 - j21 44.0 – j8.22 1.24/ -19.37 545mm 1028mm 49.9 – j.556
1.002 /-60dB (Up) (up)
16 30 - j25.6 46.7 - j3.6 1.1073 -25.8 No No 46.7 - j3.6
movement movement
12 31.7 -j34.4 41.3 - j10.8 1.354 -16.4 No No 41.3- j10.8
movement movement
8 40 - j64.8 31.3 - j30.21 2.38/ -7.74/- 872mm 810mm 50.06 – j.2
1.003 50 (Up) (Up)
4 137 + j490 60 - j132 7.69 -2.22 No No 27.8 - j13.8
movement movement
0 18 + j106 211 - j204 8.28/ -2.1/-46 1227mm 177mm 49.96 - j.5
(Cmin 1.00 (Up) (Up)
pos)

Variable load is replaced with .75ohm Resistive load and the coarse network does matching.

.75 .204+j4.4 16+j92.1 13.48 -45dB 90mm 1175mm 49.809 +j.54

ohm /1.00 Down (Up)
From
Upper
Limit

No movement: At this setting of the load, coarse matching network (stub & phase shifter)
is not operated for the matching of the load

Conclusion: Coarse matching network can match any load starting from milliohms range
to some ohm range obtained from the variable load.

22
 Appendix-VII

Effect of Error signal in matched load case

•Error signal Ea = -0.17 & Eg = -0.04 in match load.
• It causes both motor continuous rotate in direction signal generated by VME to motor-controller.
•Though they have to standstill because it is match condition

X-axis shows time in millisecond range

Reflection coefficient curve shows it starts downing from value of 0.9 to 0.15 as AMN
Start-Stop Signal generated by the VME (hence VVC motor-1 controller received CW
pulse & VVC Motor-2 controller received CW/CCW Pulse). Now AMN motor should
ideally stop, as it is matched condition (ensured by the power meter reading at generator
side which showing Forward power is 700W & reflected power is 10W & before matching
it was 350W/240W).

Since even in perfect matched load some error signal will generate continuously, which
causes motor driver get receive direction signal from the VME system, hence motor will
alternatively rotate in both direction continuously.

23
 Appendix-VIII

Effect of Error signal window margin

Following Error margin have been incorporated in VME control programming
M1 & M2 in CW direction (Ea & Eg) > 0 .2
M1 & M2 in CCW direction (Ea & Eg) < - 0.2

No movement -0.2 <(Ea & Eg)< 0. 2

X-axis shows time in millisecond range

Reflection coefficient curve shows it is down from value of 0.9 to 0.15 as AMN Start-Stop
signal generated by the VME (hence VVC Motor-1 controller received CW pulse & VVC
Motor-2 controller received CW/CCW pulse).

As error signal window margin added in the VME programming & error signal not
crossing the putted window limit, so motor driver didn’t get any direction signal from
VME system for the same load (as in appendix-7), hence motor remain standstill. (Power
meter showing the reading 650W/10W).

24
 Appendix-IX

Response time in motor controller

X-axis shows time in millisecond range

Reflection coefficient curve shows that after adjusting the response time in VVC Motor
controller for the same case as shown in (appendix IX), it starting to reduce immediately
after receiving the AMN Start/Stop signal. It means VVC motor –1 start moving in CCW
direction as soon as its controller get CCW direction signal from the VME. That time is to
~ 300 msecond.

25
 Appendix-X

Adjustment of step size of count from d400 to d1200 in motor controller

X-axis shows time in millisecond range

Reflection coefficient curve shows that after adjusting the step size of count from d400 to
d1200 in VVC motor controller, keeping the other motion parameter & load condition
same as of in appendix X, the matching time reduces to ~90msecond.

26
 Appendix-XI

VVC

0.5 Meter
STUB

Tx. Line Probe Section

T
Phase 1.5
Shifter Meter
STUB

ICRH Matching Network Disc probe with detector

VVC Motor Motor Controller Motor Connection with VVC VME Processor / Cards

FOC Tx/Rx Card

Conditioning Rack Controller Rack VME Rack Conditioning Card

27

View publication stats