**Hi all...
I am currently developing a Stm system which is working on the GXSM 2.We are testing our setup with 1-50namp set current.spec**:Ubuntu 16.04.2 LTS
Codename: xenial.
kernel version : 4.4.0-63-generic #84-Ubuntu SMP Wed Feb 1 17:23:56 UTC 2017 i686 i686 i686 GNU/Linux
Hardware: SR-MK2 connect at /dev/sranger_mk2_0
DSP-SoftId : 1001 [HwI:1001]
DSP-SoftVer: 2032 [HwI:2032]
Gxsm Package Version: 3.11.0 "Signal Master SLS HiDPI"
picomotor (coarse motion) Resolution =22nm
piezotube: Vertical Resolution=479A/v
Axial Resolution =103A/v
How can I make autotune in high noise atmosphere?
Thanx in advance.....
happY weeeKend
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi percy...
Now I am doing STM experiments.I can reach into the tunneling region.But the i tunnel is not getting stable with respect to my settle current.I checked manual tuning of the PI control system(feedback gain).But i failed with manual tuning to get stable tunnelling current.So now I am looking forward for autotunning.I suspect is there any provision is there to autotune my feedback control system in gxsm.
thanx in advance....
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
There are many reason for that or any instabilities of the tunnel junction to be checked off and isolated - beyond feedback settings which are the easiest to get right -- just start with a slow feedback.
Question:
do you really use still Gxsm-2.0? MK2 or MK3? Please note Gxsm-2 is phased out at this time and there is not point inm using it any more unless you happen to still use the very 1st gen of "SR-MK1" aka SignalRanger-STD/-SP2 (not supported any more by Gxsm3 or pening a port).
Regarding the Feedback itself:
For a typical STM system using a Z tube piezo here are some numbers to start with:
MK2 and Gxsm2.0 -- start with CP=CI=0.01 .. 0.1 , MK3 CP=CI=10..100
MK2/MK3 and Gxsm-3.0 -- both same now, typically start with CP=CI=10..100
Once that works you may want to set CP about 20..30% higher than CI for enhanced stab and speed.
Consider a current low pass filter setting around 500 Hz (low, low temp) ... 5000Hz faster system response.
A sample bias suitabel for you sample (Si, flashed in UHV, use at least 2V), metals, anything 100mV .. 2V to get started. HOPG 0.5 .. 1V for example.
And make sure all you gain stettings are correct withing GXSM:
XYZ HV gains, bias and current gain and XYZ piezo constants (Ang/V) in preferences.
However, most instabilities are actually NOT caused by the feedback itself, but in response to a instable tunnel junction what can be caused by many issues:
a) bad, dull, dirty, multipe, loose tip -- I'd say this is 90% the cause of troubles! Including picked up "flaky" material by the tip, typically on HOPG you easy pick up rip off tiny flakes then flopping round at the tip end hard to get rid of -- eventually with lot's of patience you can get rid of them...
b) mechanical instability or noise. Loose tip, loose sample, non rigit STM head, etc.
c) bad/dirty sample, (part) oxidized surface, hardly conductive, etc.
d) bad bias contact
e) bad wireing
f) not suitable current amplifiert or gain settings
-P
Last edit: Percy Zahl 2017-11-20
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanx for quick reply percy..
I am giving the bias voltage and set current in accordance with sample.(in my case gold sample Vbiase=.05V;set current=6nA).
Now I could reach into the tunneleing current and see stabilized Zvoltage in panWindow.
But the problem is the I average is varying inbetween 2nA to 80nA. can you figure it out the reason behind it.? second thing is If i am changing into GXSM 3. What are the changes I have to do for initializing the preference window comapring to the GXSM 2.?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
See my previouis post. 6nA seam fairly high so, but depends on my setup if that's work stable. Try with 0.5..1nA first, and may be more 0.1..0.5V to get started -- then work your way "down" closer to the sample -- what is much more demanding on you instrument stability... Learn first.
Gxsm3 is pretty much the same from preferences. But a major modernization of the user interface.
Only note CP/CI for MK2 and MK3 are now similar. Typically a setting of 10..100 fpr CP and CI for MK2 as well.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
hi...
I have some queries which are I listed bellow.
1.how can get iv curve using gxsm?
2.can i figure it out which part of my STM making problems?
3.with my setting some i got good result. But after some time I am struggling more.can you suggest any paths?
thanx in advance
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
**Hi all...
I am currently developing a Stm system which is working on the GXSM 2.We are testing our setup with 1-50namp set current.spec**:Ubuntu 16.04.2 LTS
Codename: xenial.
kernel version : 4.4.0-63-generic #84-Ubuntu SMP Wed Feb 1 17:23:56 UTC 2017 i686 i686 i686 GNU/Linux
Hardware: SR-MK2 connect at /dev/sranger_mk2_0
DSP-SoftId : 1001 [HwI:1001]
DSP-SoftVer: 2032 [HwI:2032]
Gxsm Package Version: 3.11.0 "Signal Master SLS HiDPI"
picomotor (coarse motion) Resolution =22nm
piezotube: Vertical Resolution=479A/v
Axial Resolution =103A/v
How can I make autotune in high noise atmosphere?
Thanx in advance.....
happY weeeKend
Hi!
Sorry, but I am not sure what you mean with this, please explain in more detail what you like to accomplish:
"How can I make autotune in high noise atmosphere?"
You talk about a resonance peak find for (NC) AFM ?
Or feedback settings?
-P
Hi percy...
Now I am doing STM experiments.I can reach into the tunneling region.But the i tunnel is not getting stable with respect to my settle current.I checked manual tuning of the PI control system(feedback gain).But i failed with manual tuning to get stable tunnelling current.So now I am looking forward for autotunning.I suspect is there any provision is there to autotune my feedback control system in gxsm.
thanx in advance....
Well, welcome to the real world of STM ;)
There are many reason for that or any instabilities of the tunnel junction to be checked off and isolated - beyond feedback settings which are the easiest to get right -- just start with a slow feedback.
Question:
do you really use still Gxsm-2.0? MK2 or MK3? Please note Gxsm-2 is phased out at this time and there is not point inm using it any more unless you happen to still use the very 1st gen of "SR-MK1" aka SignalRanger-STD/-SP2 (not supported any more by Gxsm3 or pening a port).
Regarding the Feedback itself:
For a typical STM system using a Z tube piezo here are some numbers to start with:
MK2 and Gxsm2.0 -- start with CP=CI=0.01 .. 0.1 , MK3 CP=CI=10..100
MK2/MK3 and Gxsm-3.0 -- both same now, typically start with CP=CI=10..100
Once that works you may want to set CP about 20..30% higher than CI for enhanced stab and speed.
Consider a current low pass filter setting around 500 Hz (low, low temp) ... 5000Hz faster system response.
A sample bias suitabel for you sample (Si, flashed in UHV, use at least 2V), metals, anything 100mV .. 2V to get started. HOPG 0.5 .. 1V for example.
And make sure all you gain stettings are correct withing GXSM:
XYZ HV gains, bias and current gain and XYZ piezo constants (Ang/V) in preferences.
However, most instabilities are actually NOT caused by the feedback itself, but in response to a instable tunnel junction what can be caused by many issues:
a) bad, dull, dirty, multipe, loose tip -- I'd say this is 90% the cause of troubles! Including picked up "flaky" material by the tip, typically on HOPG you easy pick up rip off tiny flakes then flopping round at the tip end hard to get rid of -- eventually with lot's of patience you can get rid of them...
b) mechanical instability or noise. Loose tip, loose sample, non rigit STM head, etc.
c) bad/dirty sample, (part) oxidized surface, hardly conductive, etc.
d) bad bias contact
e) bad wireing
f) not suitable current amplifiert or gain settings
-P
Last edit: Percy Zahl 2017-11-20
Thanx for quick reply percy..
I am giving the bias voltage and set current in accordance with sample.(in my case gold sample Vbiase=.05V;set current=6nA).
Now I could reach into the tunneleing current and see stabilized Zvoltage in panWindow.
But the problem is the I average is varying inbetween 2nA to 80nA. can you figure it out the reason behind it.?
second thing is If i am changing into GXSM 3. What are the changes I have to do for initializing the preference window comapring to the GXSM 2.?
See my previouis post. 6nA seam fairly high so, but depends on my setup if that's work stable. Try with 0.5..1nA first, and may be more 0.1..0.5V to get started -- then work your way "down" closer to the sample -- what is much more demanding on you instrument stability... Learn first.
Gxsm3 is pretty much the same from preferences. But a major modernization of the user interface.
Only note CP/CI for MK2 and MK3 are now similar. Typically a setting of 10..100 fpr CP and CI for MK2 as well.
hi...
I have some queries which are I listed bellow.
1.how can get iv curve using gxsm?
2.can i figure it out which part of my STM making problems?
3.with my setting some i got good result. But after some time I am struggling more.can you suggest any paths?
thanx in advance
Please have a look in the manual even it's the gxsm2 manual, it still works the same way.
Page 140ff:
http://gxsm.sourceforge.net/Gxsm-main.pdf