Add service for AimTTiPL-P #149

ivalaginja · 2023-12-10T00:35:10Z

Fixes #148.

Manufacturer info:
https://www.aimtti.com/product-category/dc-power-supplies/aim-plseries

Note how "The New PL-P Series is the programmable (remote control) version of the New PL Series [...]", since the above website lists both variants (PL and PL-P).

Usage example for the dcps library taken from here:
https://github.com/sgoadhouse/dcps/blob/master/dcps/AimTTiPLP.py

Edit: Could also be interesting to look into https://msl-equipment.readthedocs.io/en/latest/_api/msl.equipment.resources.aim_tti.html#module-msl.equipment.resources.aim_tti

Todo:

Generalize to be able to control all channels of one device
Add methods to measure current and voltage
Check current command units
Test on hardware

Notes:

The device automatically applies a remote interface lock when it is commanded for the first time (see manual, page 23). This is also noted in the dcps library. Even after setting the interface lock to LOCAL, this will instantly be overwritten back to REMOTE when a new command is issued.
All remote commands are listed on page 34 of the device manual.
I only implemented a minimum of the commands listed in the manual. More commands can easily be added in the future as needed.

ivalaginja · 2023-12-10T00:45:14Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+from dcps import AimTTiPPL
+
+
+class AimTtiPlp(Service):


Not sure I like this class name...

@ehpor should I use this abomination of a class name or do you have other suggestions? :P An alternative would be AimTTiPLP like in the dcps library, although I think that might override it in the name space (same name)?

Name doesn't matter to me. I would not use the same name as the library one.

ehpor · 2023-12-10T04:12:42Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+            try:
+                # Get an update for this channel
+                voltage = self.voltage.get_next_frame(10).data[0]
+                current = self.current.get_next_frame(10).data[0]


This won't work as expected. If you don't get an update frame on the voltage and the current in the same iteration of this loop, then the setVoltage() and setCurrent() will not be called. You should use threads to monitor both. Then, you'd also wanna use a lock to avoid simultaneous access to the physical device from those two threads.

Yup, like any other service that needs to monitor several data streams. I also need to extend this to several output channels with two data streams each, but I stopped so I can think about how to do that.
Sounds like it's as simple as having twice as many (monitored) data streams/threads as output channels (each requiring a voltage and current data stream). Is that right? Is there any better way to do that than to call them 'voltage_{channelnumber}' and 'current_{channelnumber}' and spin off a monitoring thread for each?

ivalaginja · 2023-12-11T15:12:44Z

@ehpor this should now be functional (albeit untested) code, do you agree? I decided to split monitoring methods in two separate ones to save myself the if...elif... to differentiate between the voltage and current data streams, but I don't know if there is any reason to not do it like this. I'd love to hear your opinion.

catkit2/services/aim_tti_plp/aim_tti_plp.py

ivalaginja · 2023-12-12T15:53:25Z

I might have gone a little too fast with this implementation - it is currently very limited. The power supply in question has various modes it can be run in and I don't think I will be able to implement all possibilities (they're not endless, but about 6 too many compared to what we need).

Being able to switch between constant voltage and constant current mode seems to be the absolute basis, but even so I feel like I would have to add respective data streams and mode switches. Not sure if I should move this over to thd-controls instead and just implement the one mode we're using right now.

I will put this on hold to think about it.

ivalaginja · 2024-03-14T15:11:03Z

Hey @ehpor could you help me out with the couple of weird problems I have? I am testing this on hardware right now and I feel like I am very close to make things work.

I manage to establish a connection to the device with this service and get returns from the commands (that do not use data streams). I am equally able to access the data streams of the service, submit to and read from them, however, my device shows no reaction. Since I am perfectly able to control it just by using the code from within the service in a separate terminal through a connection without using catkit2, it looks to me like I messed up the threads and how they're set up with the main function. Do you maybe see what is wrong?

Also, when I Ctlr+C on the server, I can see that this service gets closed down, but it immediately tries to restart which yields a bunch of:

RuntimeError: The testbed is shutting down. Starting a new services is not allowed anymore.

Eventually, the restarting request seems to time out and everything closes all right, but I do not know what is causing this. This also causes the GUI to get stuck, which I have to force shut down.

ehpor · 2024-03-14T18:38:05Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+            try:
+                # Get an update for this channel
+                frame = self.current_commands[channel_name].get_next_frame(10)
+                value = frame.data


Note: value is a Numpy array here. You probably want frame.data[0].

ehpor · 2024-03-14T18:39:38Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+        self.device._instWrite(str)
+
+        # Give some time for power suppluy to respond
+        time.sleep(self.device._wait)


Prefer self.sleep() since this also quits sleeping if the service is being asked to shut down (ie. it checks the shutdown flag internally).

Same for other places in this file.

ehpor · 2024-03-14T18:46:27Z

Hey @ehpor could you help me out with the couple of weird problems I have? I am testing this on hardware right now and I feel like I am very close to make things work.

I manage to establish a connection to the device with this service and get returns from the commands (that do not use data streams). I am equally able to access the data streams of the service, submit to and read from them, however, my device shows no reaction.

Reading and writing from/to datastreams is completely independent from the service being responsive. No service code is ran when interacting with its datastreams.

Since I am perfectly able to control it just by using the code from within the service in a separate terminal through a connection without using catkit2, it looks to me like I messed up the threads and how they're set up with the main function. Do you maybe see what is wrong?

I looked briefly at the code, but the threads look fine by eye. I found one error though with how you're reading from the datastreams. And one related to sleeping, which is very likely unrelated.

Also, when I Ctlr+C on the server, I can see that this service gets closed down, but it immediately tries to restart which yields a bunch of:
RuntimeError: The testbed is shutting down. Starting a new services is not allowed anymore.

This is likely the GUI trying to restart the service.

Eventually, the restarting request seems to time out and everything closes all right, but I do not know what is causing this. This also causes the GUI to get stuck, which I have to force shut down.

ehpor · 2024-03-14T18:48:36Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+        for channel_name in self.channels.keys():
+            self.set_voltage(channel_name, value=0)
+            self.set_current(channel_name, value=0)
+            self.device.outputOff(self.channels[channel_name]['channel'])


I don't think you're joining the threads.

ehpor · 2024-03-14T18:49:07Z

catkit2/services/aim_tti_plp/aim_tti_plp.py

+                value = frame.data
+
+                # Update the device
+                with self.lock_for_current:


I'd move this lock to inside self.set_current() and the same for self.set_voltage(). That offers more protection.

ivalaginja · 2024-03-15T16:48:55Z

Thanks @ehpor! I will be back in the lab on Monday to test on hardware.

ivalaginja · 2024-03-18T14:25:23Z

After some final small fixes, this is now working as expected on hardware.

ivalaginja · 2024-03-18T15:49:59Z

@ehpor thanks for helping me figure this one out! I had quite some troubles considering what a simple device it is...

raphaelpclt · 2024-04-02T18:15:11Z

catkit2/services/aimtti_plp_device/aimtti_plp_device.py

+        str = 'TRIPRST'
+        self.device._instWrite(str)
+
+        # Give some time for power suppluy to respond


Tiny typo in "suppluy"

raphaelpclt

One general remark, I got a bit confused by all the "channels" names. There's a self.channels dict, that has channel_names dict as keys, that have channel, that is a number. Maybe call everywhere this last-stage channel channel_number everywhere, as it is done in some places? (eg L164)

Apart from that, everything looks good, very tiny typos in commented code.

raphaelpclt · 2024-04-02T18:18:23Z

catkit2/services/aimtti_plp_device/aimtti_plp_device.py

+        """Attempt to clear all trip conditions.
+
+        This only works if the device has not been set up so that it can only reset trip conditions manually on the
+        front panel, or by cycling the AC power.
+        """
+        str = 'TRIPRST'
+        self.device._instWrite(str)
+
+        # Give some time for power suppluy to respond


Minor comment: would there be a way to log if it worked or not? Like a feedback from the device saying "Reset trip conditions not possible"

From scanning the manual real quick, it doesn't seem there is. There are ways to get potential errors back however I never tested that.

raphaelpclt · 2024-04-02T18:18:44Z

catkit2/services/aimtti_plp_device/aimtti_plp_device.py

+        str = 'OCP{} {}'.format(channel_number, value)
+        self.device._instWrite(str)
+
+        # Give some time for power suppluy to respond


Also tiny typo

raphaelpclt · 2024-04-02T18:19:06Z

catkit2/services/aimtti_plp_device/aimtti_plp_device.py

+        str = 'OVP{} {}'.format(channel_number, value)
+        self.device._instWrite(str)
+
+        # Give some time for power suppluy to respond


Copy-pasted tiny typo? :P

ivalaginja · 2024-04-02T19:05:58Z

One general remark, I got a bit confused by all the "channels" names. There's a self.channels dict, that has channel_names dict as keys, that have channel, that is a number. Maybe call everywhere this last-stage channel channel_number everywhere, as it is done in some places? (eg L164)

Yeah I get that, however the structure with the dict is necessary because I want to have named channels like in the case of the DMs. There, the self.channels dict uses the channel_names as keys that maps to the respective channel number, which I implemented here as well - except I called the numbers just "channels" in the config. Ok yup I see what you mean, I like your suggestion, give me a sec to change things appropriately.

raphaelpclt

Looks good

ivalaginja · 2024-04-02T19:22:03Z

Thanks @raphaelpclt!

ivalaginja self-assigned this Dec 10, 2023

ivalaginja added hardware Integrate new hardware collaborators Worked on by external collaborator. Might need some extra help on code integration. labels Dec 10, 2023

ivalaginja commented Dec 10, 2023

View reviewed changes

ehpor reviewed Dec 10, 2023

View reviewed changes

ivalaginja commented Dec 12, 2023

View reviewed changes

catkit2/services/aim_tti_plp/aim_tti_plp.py Outdated Show resolved Hide resolved

ivalaginja commented Dec 12, 2023

View reviewed changes

catkit2/services/aim_tti_plp/aim_tti_plp.py Outdated Show resolved Hide resolved

ivalaginja force-pushed the feature/aim_tti_plp_service branch 2 times, most recently from f34b639 to 1068198 Compare March 14, 2024 12:47

ivalaginja requested a review from ehpor March 14, 2024 15:07

ehpor reviewed Mar 14, 2024

View reviewed changes

ivalaginja added 12 commits March 15, 2024 17:47

Add service for AimTTiPLP DC power supply

0480915

Add new dependency

784a065

Remove unused import

f246ada

Set current and voltage to zero before closing

26a970f

Extend to all power supply channels, each with both voltage and current

f4bb3f9

Add reminder to check for max voltage and current

41d1af4

Update device

5ddbc88

Extend comment

bb2728b

Add missing imports

6e49e72

Split between current and voltage streams

a27a931

Change variable name

0fdadcd

Fix typo

4844080

ivalaginja added 7 commits March 15, 2024 17:47

Current commands are in Amps

00af75a

Keep measuring voltage and current

555606f

Extract float from numpy array

c38b2ee

Prefer self.sleep over time.sleep

bd681a9

Join threads in close()

d8115b5

Move locks to respective methods

febd7c8

Change service name

bc06d45

ivalaginja force-pushed the feature/aim_tti_plp_service branch from b7c38bb to bc06d45 Compare March 15, 2024 16:47

ivalaginja added 3 commits March 15, 2024 17:49

Remove unused import

b3d56a2

Rename folder and filename also

bab2795

Specify frame dtype

bf96e24

ivalaginja requested a review from ehpor March 18, 2024 14:24

ivalaginja marked this pull request as ready for review March 18, 2024 14:24

Fix underscore in folder and filename...

65b6781

ivalaginja mentioned this pull request Mar 18, 2024

Add simulated service for AimTTiPL-P device #168

Merged

ivalaginja changed the title ~~Add service for AimTTiPLP~~ Add service for AimTTiPL-P Mar 18, 2024

ivalaginja requested a review from raphaelpclt April 1, 2024 20:58

raphaelpclt reviewed Apr 2, 2024

View reviewed changes

raphaelpclt requested changes Apr 2, 2024

View reviewed changes

Fix typo everywhere

6e335b2

Change "channel" values to "channel_number"

3d6b136

ivalaginja requested a review from raphaelpclt April 2, 2024 19:08

raphaelpclt approved these changes Apr 2, 2024

View reviewed changes

ivalaginja merged commit e320094 into develop Apr 2, 2024
6 checks passed

ivalaginja deleted the feature/aim_tti_plp_service branch April 2, 2024 19:21

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add service for AimTTiPL-P #149

Add service for AimTTiPL-P #149

ivalaginja commented Dec 10, 2023 •

edited

Loading

ivalaginja Dec 10, 2023

ivalaginja Mar 10, 2024 •

edited by THD2-team

Loading

ehpor Mar 14, 2024

ehpor Dec 10, 2023

ivalaginja Dec 10, 2023

ivalaginja commented Dec 11, 2023

ivalaginja commented Dec 12, 2023

ivalaginja commented Mar 14, 2024 •

edited

Loading

ehpor Mar 14, 2024

ehpor Mar 14, 2024

ehpor Mar 14, 2024

ehpor commented Mar 14, 2024

ehpor Mar 14, 2024

ehpor Mar 14, 2024

ivalaginja commented Mar 15, 2024

ivalaginja commented Mar 18, 2024

ivalaginja commented Mar 18, 2024

raphaelpclt Apr 2, 2024 •

edited

Loading

raphaelpclt left a comment

raphaelpclt Apr 2, 2024

ivalaginja Apr 2, 2024

raphaelpclt Apr 2, 2024

raphaelpclt Apr 2, 2024

ivalaginja Apr 2, 2024

ivalaginja commented Apr 2, 2024

raphaelpclt left a comment

ivalaginja commented Apr 2, 2024

Add service for AimTTiPL-P #149

Add service for AimTTiPL-P #149

Conversation

ivalaginja commented Dec 10, 2023 • edited Loading

Todo:

Notes:

Choose a reason for hiding this comment

ivalaginja Mar 10, 2024 • edited by THD2-team Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ivalaginja commented Dec 11, 2023

ivalaginja commented Dec 12, 2023

ivalaginja commented Mar 14, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ehpor commented Mar 14, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ivalaginja commented Mar 15, 2024

ivalaginja commented Mar 18, 2024

ivalaginja commented Mar 18, 2024

raphaelpclt Apr 2, 2024 • edited Loading

Choose a reason for hiding this comment

raphaelpclt left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ivalaginja commented Apr 2, 2024

raphaelpclt left a comment

Choose a reason for hiding this comment

ivalaginja commented Apr 2, 2024

ivalaginja commented Dec 10, 2023 •

edited

Loading

ivalaginja Mar 10, 2024 •

edited by THD2-team

Loading

ivalaginja commented Mar 14, 2024 •

edited

Loading

raphaelpclt Apr 2, 2024 •

edited

Loading