I was trying to use the slope corrections from the 'Advanced' menu in the application window. However, no matter which value that I inserted into the Slope X and Slope Y fields of Scan characteristics section, the system won't provide an affect on the scan.
Is there a solution method that you can recommend to overcome this issue ?
Thanks in advance.
All the best,
Bugrahan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
for that to answer I do need to know your exact configuration:
MK2 or MK3 and how is your Z-scan and Z-offset configuration? What are you Piezo amplifier's capabilities? Analog Z adding input at gain 1 (not full scale, but all depends on your system and requirements to determine what is best? Or do you only use one fixed gain, etc etc...
The best scenario is if you Piezo Amplifier has a 1x gain analog adding input for Z and you configure a dedicated analog output for the Z-slope component.
Anyhow it can also be added internally... as there various options you need to do some configurations.
For the MK3 you may have some thing like this, take a look at the startup info dialog (or also the monitor or message log file for the current MK3 signal output configuration:
** CURRENT MK3 OUTMIX CONFIGURATION **
OUT[0] := Null-Signal + Null-Signal - Null-Signal
OUT[1] := Null-Signal + Null-Signal - Null-Signal
OUT[2] := Z Offset + Z Offset from XY Slope - Null-Signal
OUT[3] := X Scan Rot + X Offset - Null-Signal
OUT[4] := Y Scan Rot + Y Offset - Null-Signal
OUT[5] := Z Servo Neg + LockIn Ref * AC Ampl Aux - VP Z pos
OUT[6] := VP Bias + Null-Signal + LockIn Ref * AC Ampl
OUT[7] := Motor + Null-Signal - Null-Signal
Here
OUT 3,4 are XY Scan with Offset
OUT 5 Z Scan ONLY from the feedback Z Servo
and on
OUT2 you have the Z Offset + Z Offset computed from the XY Slope component mapped.
Thus if you do not connect OUT2 to you amplifier with analog adding to Z-Scan for a final Z-Piezo signal the slow in Gxsm will not have any effect.
Anyhow, you can reconfigure the OUT5 mapping and replace the Z modulation option here
"LockIn Ref * AC Ampl Aux" with the "Z-Offset from XY Scan" signal.
For this you need to go into the mk3_spm_configurator.py and open the "Patch Rack" App...
-P
Last edit: Percy Zahl 2022-01-12
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for the suggestions. I am adding the settings that we got as screenshots to not miss a point while configuring the new settings and receive your comments based upon them.
Thanks again for your time and help.
Regards,
Bugrahan Guner
OK You have a MK3 as I see you have signals to select as feedback input.
I need to know your hardware, specifically the high voltage piezo amplifier's capabilities and if it can add external analog signals to Z in addition to the Z offset and if it can do that at all.
Simple question:
Do you have more than Z-Scan attached to you AFM's controller?
Do you have more input option for Z? Gain(s)?
if none of those options are available we have to do the simplest way. But this may limit your effective Z range depending on the slope and scan ranges -- thus will potentially clip/saturate if you are close to the +/- Z full scale (or +/-10V) limits. This is why an external analog adding is preferred if possible.
This reconfiguration is more low level and currently is not supported via the GXSM preferences, thus not show in there, but via a DSP configuration tool:
1) You need to find the mk3_spm_configurator.py script:
in "Gxsm-3.0 SVN" it is here, the Ubuntu system may provide a link to it:
Gxsm-3.0/plug-ins/hard/MK3-A810_spmcontrol/python3_scripts
Start that up!
Then see my two screenshots with annotations here.
Find the "3 SPM Signal Patch Rack" App Button and (green marked in the screenshot) click/start it, Then the window as shown next to the App list on left opens -- may take a moment to load the current configuration and build the control window....
Here I highlighted the of interest signal in the current or default configuration (left) window with a long list of all DSP signals. Scroll down carefully (do not scroll on the menus by accident!! as that may change settings!!) to have the signal destination (right column) OUTMIX CH2 ... CH5 is view.
This is what I described briefly in my previous message! BTW there is a whole section on this in the manual explaining the mechanism...
CH2 is the Z-OFFSET you most likely do NOT sue currently. And why nothing happens.
I marked all the slots of interest for you information only (green).
Yellow: is the Z-Servo signal mapped to CH5.
The purple and cyan marked signal mixing we need to reconfigure:
We need the signal S53 "Z Offset from XY Slope" and remap it to be added to Z-SCAN...
BTW: you can leave the Z-OFFSET mapping unchanged or disable it or reuse the channel for anything else...
All you need to do now:
Refer to the right screen shot list "Reconfigured MK3..."
a) Change the input signal (left pull down) of
OUTMIX_CH5_ADD_A to Sig: S53 "Z Offset from XY Slope"
b) Disable the input for OUTMIX_CH5_SMAC_A:
select "DISABLED" (instead of S70...) (did not marked this in green).
2) As this is NOT stored on the DSP automatically a power cycle or reset will restore the original defaults. Thus you do should save the configuration to a file and then later if you restart the DSP (leave it running unless not used for a long time) reload it form file.
3) The "SPM Signal + DSP Manager" App will help you here. As show in the other screen shot.
Simple press "SAVE DSP CONFIG TO FILE" to save to file.
Later come back here and press "LOAD DSP CONFIG FROM FILE" to restore it after power cycle or reset.
Thank you so much for the help about the slope correction. It works now. If you have the chance I want to ask an additional question towards a similar problem with LDC (Linear Drift Correction). Right now it is not working when we want to correct the Z value by changing dz parameter to a non-zero A/s value when we are in an engaged situation.
I thought it may be solved with a similar approach so I wanted to ask for your opinion.
Thank you so much in advance for your time and effort.
All the best,
Bugrahan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes, we checked that box, which is in the advanced tab of the DSP Control menu. Even though we open the LDC feature, for values that we give for dx, dy and dz, nothing happens. We tried both positive and negative values in a range starting from reasonable values to excessively large ones to at least see a difference. Overall, we still observe a drift in the Z value throughout the scanning procedure after we approach to the surface and the values used have zero affect on that. We also made sure that we didn't saturated to the limit in Z.
The value is changing for the Offset ZO parameter but for the Z value where we approached to the surface stills drifts (with the same rate in both LDC on and LDC off cases) and eventually cause an inconsistency according to the obtained results.
Is it possible that the LDC also requires a reconfiguration ? Is there a possibility that it is closed from another control menu other than the 'SR DSP Control' tab since we already checked the box in the 'Advanced' section of it?
Thanks for your time once again.
Bugrahan
Last edit: bugrahan guner 2022-01-27
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes as you are not suing the "X/Y/Z-Offset" signals or separated analog outputs. This is a similar problem.
Before you are enabling this may I ask you to rethink if that really makes sense for you:
the Z-Servo -- I assume you are running with some kind of Z-feedback -- or open loop!?!?? -- is taking care of the Z-drift or what ever your problem is is a more precise way. And you can always later background correct the image data.
I am say this, as you won't gain really much when you operate the Z-drift (LDC) on the Z-Scan (Servo Signal). It's a different story when you run the LDC on a full scale offset, not accessible to the Z-Servo! You see where I am going here?
In contrast to the slope (fixed sample tilt) correction required for perfect const height scans in open loop mode -- this Z LDC won't do any good but adding a complication with saturation of the Z-Servo signal.
This said, I won't recommend doing this. Better address the problem at the instrument's root cause. Shield, thermalize, make symmetric, use low or zero expansion constructions materials, build small and compact, etc...
And if really necessary and linear (usually exponential on long term) I strongly recommend to add the Z Offset outside the box analog -- may be even wit ha simple passive resistive adding network if your amplifier's input is high impedance that may do!
-- However it is possible to do --
You can enable it as like this:
I have one more OUTMIX_CH5_SUB_B option similar to this one you recently added to Z-Scan:
OUTMIX_CH5_ADD_A to Sig: S53 "Z Offset from XY Slope"
It is currently configured to operate with the Z-Probe - - you can swap that to
"Sig 57: Z Offset".
However, this will disable the option to do Z-manipulations via GVP / probing. But with a larger drift this is very trouble some anyways.
-P
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Percy,
I was trying to use the slope corrections from the 'Advanced' menu in the application window. However, no matter which value that I inserted into the Slope X and Slope Y fields of Scan characteristics section, the system won't provide an affect on the scan.
Is there a solution method that you can recommend to overcome this issue ?
Thanks in advance.
All the best,
Bugrahan
Hi,
for that to answer I do need to know your exact configuration:
MK2 or MK3 and how is your Z-scan and Z-offset configuration? What are you Piezo amplifier's capabilities? Analog Z adding input at gain 1 (not full scale, but all depends on your system and requirements to determine what is best? Or do you only use one fixed gain, etc etc...
The best scenario is if you Piezo Amplifier has a 1x gain analog adding input for Z and you configure a dedicated analog output for the Z-slope component.
Anyhow it can also be added internally... as there various options you need to do some configurations.
For the MK3 you may have some thing like this, take a look at the startup info dialog (or also the monitor or message log file for the current MK3 signal output configuration:
** CURRENT MK3 OUTMIX CONFIGURATION **
OUT[0] := Null-Signal + Null-Signal - Null-Signal
OUT[1] := Null-Signal + Null-Signal - Null-Signal
OUT[2] := Z Offset + Z Offset from XY Slope - Null-Signal
OUT[3] := X Scan Rot + X Offset - Null-Signal
OUT[4] := Y Scan Rot + Y Offset - Null-Signal
OUT[5] := Z Servo Neg + LockIn Ref * AC Ampl Aux - VP Z pos
OUT[6] := VP Bias + Null-Signal + LockIn Ref * AC Ampl
OUT[7] := Motor + Null-Signal - Null-Signal
Here
OUT 3,4 are XY Scan with Offset
OUT 5 Z Scan ONLY from the feedback Z Servo
and on
OUT2 you have the Z Offset + Z Offset computed from the XY Slope component mapped.
Thus if you do not connect OUT2 to you amplifier with analog adding to Z-Scan for a final Z-Piezo signal the slow in Gxsm will not have any effect.
Anyhow, you can reconfigure the OUT5 mapping and replace the Z modulation option here
"LockIn Ref * AC Ampl Aux" with the "Z-Offset from XY Scan" signal.
For this you need to go into the mk3_spm_configurator.py and open the "Patch Rack" App...
-P
Last edit: Percy Zahl 2022-01-12
Hi Percy,
Thanks for the suggestions. I am adding the settings that we got as screenshots to not miss a point while configuring the new settings and receive your comments based upon them.
Thanks again for your time and help.
Regards,
Bugrahan Guner
OK You have a MK3 as I see you have signals to select as feedback input.
I need to know your hardware, specifically the high voltage piezo amplifier's capabilities and if it can add external analog signals to Z in addition to the Z offset and if it can do that at all.
Simple question:
Do you have more than Z-Scan attached to you AFM's controller?
Do you have more input option for Z? Gain(s)?
if none of those options are available we have to do the simplest way. But this may limit your effective Z range depending on the slope and scan ranges -- thus will potentially clip/saturate if you are close to the +/- Z full scale (or +/-10V) limits. This is why an external analog adding is preferred if possible.
This reconfiguration is more low level and currently is not supported via the GXSM preferences, thus not show in there, but via a DSP configuration tool:
1) You need to find the mk3_spm_configurator.py script:
in "Gxsm-3.0 SVN" it is here, the Ubuntu system may provide a link to it:
Gxsm-3.0/plug-ins/hard/MK3-A810_spmcontrol/python3_scripts
Start that up!
Then see my two screenshots with annotations here.
Find the "3 SPM Signal Patch Rack" App Button and (green marked in the screenshot) click/start it, Then the window as shown next to the App list on left opens -- may take a moment to load the current configuration and build the control window....
Here I highlighted the of interest signal in the current or default configuration (left) window with a long list of all DSP signals. Scroll down carefully (do not scroll on the menus by accident!! as that may change settings!!) to have the signal destination (right column) OUTMIX CH2 ... CH5 is view.
This is what I described briefly in my previous message! BTW there is a whole section on this in the manual explaining the mechanism...
CH2 is the Z-OFFSET you most likely do NOT sue currently. And why nothing happens.
I marked all the slots of interest for you information only (green).
Yellow: is the Z-Servo signal mapped to CH5.
The purple and cyan marked signal mixing we need to reconfigure:
We need the signal S53 "Z Offset from XY Slope" and remap it to be added to Z-SCAN...
BTW: you can leave the Z-OFFSET mapping unchanged or disable it or reuse the channel for anything else...
All you need to do now:
Refer to the right screen shot list "Reconfigured MK3..."
a) Change the input signal (left pull down) of
OUTMIX_CH5_ADD_A to Sig: S53 "Z Offset from XY Slope"
b) Disable the input for OUTMIX_CH5_SMAC_A:
select "DISABLED" (instead of S70...) (did not marked this in green).
2) As this is NOT stored on the DSP automatically a power cycle or reset will restore the original defaults. Thus you do should save the configuration to a file and then later if you restart the DSP (leave it running unless not used for a long time) reload it form file.
3) The "SPM Signal + DSP Manager" App will help you here. As show in the other screen shot.
Simple press "SAVE DSP CONFIG TO FILE" to save to file.
Later come back here and press "LOAD DSP CONFIG FROM FILE" to restore it after power cycle or reset.
Last edit: Percy Zahl 2022-01-14
Hi Percy,
Thank you so much for the help about the slope correction. It works now. If you have the chance I want to ask an additional question towards a similar problem with LDC (Linear Drift Correction). Right now it is not working when we want to correct the Z value by changing dz parameter to a non-zero A/s value when we are in an engaged situation.
I thought it may be solved with a similar approach so I wanted to ask for your opinion.
Thank you so much in advance for your time and effort.
All the best,
Bugrahan
Hi, about the LDC -- did you checked the Enable LDC box?
If I check this and put a non zero value for any of the dXYZ I see the XYZ offsets moving accordingly -- just watch the PanView indicators for XYZ.
And in a way yes, those are applied to XYZ Offset -- NOT XY Scan.
-P
Hi Percy,
Yes, we checked that box, which is in the advanced tab of the DSP Control menu. Even though we open the LDC feature, for values that we give for dx, dy and dz, nothing happens. We tried both positive and negative values in a range starting from reasonable values to excessively large ones to at least see a difference. Overall, we still observe a drift in the Z value throughout the scanning procedure after we approach to the surface and the values used have zero affect on that. We also made sure that we didn't saturated to the limit in Z.
The value is changing for the Offset ZO parameter but for the Z value where we approached to the surface stills drifts (with the same rate in both LDC on and LDC off cases) and eventually cause an inconsistency according to the obtained results.
Is it possible that the LDC also requires a reconfiguration ? Is there a possibility that it is closed from another control menu other than the 'SR DSP Control' tab since we already checked the box in the 'Advanced' section of it?
Thanks for your time once again.
Bugrahan
Last edit: bugrahan guner 2022-01-27
Hi!
Yes as you are not suing the "X/Y/Z-Offset" signals or separated analog outputs. This is a similar problem.
Before you are enabling this may I ask you to rethink if that really makes sense for you:
This said, I won't recommend doing this. Better address the problem at the instrument's root cause. Shield, thermalize, make symmetric, use low or zero expansion constructions materials, build small and compact, etc...
And if really necessary and linear (usually exponential on long term) I strongly recommend to add the Z Offset outside the box analog -- may be even wit ha simple passive resistive adding network if your amplifier's input is high impedance that may do!
-- However it is possible to do --
You can enable it as like this:
I have one more OUTMIX_CH5_SUB_B option similar to this one you recently added to Z-Scan:
OUTMIX_CH5_ADD_A to Sig: S53 "Z Offset from XY Slope"
It is currently configured to operate with the Z-Probe - - you can swap that to
"Sig 57: Z Offset".
However, this will disable the option to do Z-manipulations via GVP / probing. But with a larger drift this is very trouble some anyways.
-P
Hi Bugrahan,
I hope you found a solution (if needed at all).
Please let us know and if we you like further help or tips.
-P