BElt tension graphs are not plotting

[RocksaltX]

When running the belt test I am getting this: Traceback (most recent call last): // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 259, in // main() // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in main // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 22, in parse_log // if not header.startswith('freq,psd_x,psd_y,psd_z,psd_xyz'): // UnboundLocalError: local variable 'header' referenced before assignment 19:19:23 // Command {plot_graph} finished

The 2 files are being generated in home/pi/klipper_config/input_shaper raw_data_belt_a_20220315_191922.csv raw_data_belt_b_20220315_191922.csv

Klipper version: v0.10.0-312-g81283423

[zellneralex]

What happens if you ssh in your PI and then run

cd klipper_config/input_shaper/
/home/pi/klipper/scripts/graph_accelerometer.py -c raw_data_belt*.csv -o  resonances_belts_test.png

This is the native klipper script call to generate the picture. If that fails than there is a problem in your installation. Have you checked your raw files do they look as expected?

[zellneralex]

Mine start with:

#time,accel_x,accel_y,accel_z
4155.430577,0.000000,76.491870,8796.565050
4155.430898,611.934960,76.491870,7802.170740

I did run the test last time 26.2.2022 but my klipper is always up to date so that I can not tell the klipper version. But I run the picture creation today at the same klipper version as yours while preparing the instructions above.

[zellneralex]

The klipper script was last updated 3 month ago and changed from python2 to python3. Have you followed the steps shown at Measuring_Resonances.html#software-installation

That part has changed and need to be executed if your installation is older than 3 month

sudo apt update
sudo apt install python3-numpy python3-matplotlib

[RocksaltX]

when I run: TEST_RESONANCES AXIS=1,1 OUTPUT=raw_data TEST_RESONANCES AXIS=1,-1 OUTPUT=raw_data

Then run: ~/klipper/scripts/graph_accelerometer.py -c /tmp/raw_data_axis*.csv -o /tmp/resonances.png

It works properly.

Only when I run this auto generated graphs it does not produce the graphs and gives: // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 259, in // main() // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in main // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 22, in parse_log // if not header.startswith('freq,psd_x,psd_y,psd_z,psd_xyz'): // UnboundLocalError: local variable 'header' referenced before assignment 19:19:23 // Command {plot_graph} finished

[zellneralex]

Please answer my question and do what I ask for.

[zellneralex]

Or send the raw files so that I can analyze them.

[RocksaltX]

I thought I did answer the best I could. I have taken the generated .csv files from the belt_test files and used: ~/klipper/scripts/graph_accelerometer.py -c /tmp/raw*.csv -o /tmp/resonances.png and this generates the graphs.

raw_data_belt_b_20220315_191922.csv

raw_data_belt_a_20220315_191922.csv

[RocksaltX]

/home/pi/klipper/scripts/graph_accelerometer.py -c raw_data_belt*.csv -o resonances_belts_test.png

This works if I SSH to PI and generates the graph in /home/pi/klipper/. however, if I am not in the folder /home/pi/klipper_config/input_shaper/ It will give the same errors. So it seems it can't get to the folder from the current folder in the script or I have something wrong on my end?

[zellneralex]

That is what I wanted to know. The klipper scrips are working with the generated files with a and b in the name. Than it must have to do with the collecting the 2 files in the bash script. I will take a look into that.

[laurienzu]

I confirm the same behaviour happen to me

[s0ybean]

was this issue ever looked into? i get the same error. manual input through ssh works fine, and so does the resonance test macro

[zellneralex]

Yes it was I tried a lot to reproduce it but for me it works fine

could you add a Pause before https://github.com/zellneralex/klipper_config/blob/master/input_shaper.cfg#L123 (G4 P2000) to insure that the files are really ready

[s0ybean]

thanks for following up. unfortunately, the issue remains :(

17:09:41 // Command {plot_graph} finished 17:09:40 // if not header.startswith('freq,psd_x,psd_y,psd_z,psd_xyz'): // UnboundLocalError: local variable 'header' referenced before assignment 17:09:40 // main() // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in main // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 218, in // datas = [parse_log(fn, opts) for fn in args] // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 22, in parse_log 17:09:40 // Traceback (most recent call last): // File "/home/pi/klipper/scripts/graph_accelerometer.py", line 259, in 17:09:40 // Generate Graph for belts ... 17:09:39 // Running Command {plot_graph}...: 17:09:39 // BELT TEST: Generate graph in backround 17:09:39 // BELT TEST: Resonance Tests done

[zellneralex]

I will look in that again but since it runs on my machine it is really hard to debug.

[zellneralex]

But I see you have run the belt test and not the input sharper test. If that’s true than you need the G4 P2000 before https://github.com/zellneralex/klipper_config/blob/master/input_shaper.cfg#L143

[s0ybean]

I added it to both at the time of testing for consistency :)

although the resonance test always works fine and provides .png images.

[zellneralex]

Ok the belt files are a lot bigger. I will check if a sleep will work before https://github.com/zellneralex/klipper_config/blob/master/script/plot_graph.sh#L68. I rename the files to get the date in the name. So maybe my SD Card is faster.

[zellneralex]

Ok I scrolled thru my old notes. A member of the Mainsail crew get it to run with G4 P5000 before https://github.com/zellneralex/klipper_config/blob/master/input_shaper.cfg#L143. So it seams really to be depending on the write speed and compute power of the PI.

[zellneralex]

Oh I guess I understand now why it works for me. My /tmp is Setup es as a RAM disk. So the time to write the files is nearly 0

[s0ybean]

Unfortunately that did not resolve it. I don't think it's a delay issue because a second run of the belt test doesn't work even if the are already csv files present from the previous run. It seems to be a file path issue?

[zellneralex]

You Devine your own path in that line https://github.com/zellneralex/klipper_config/blob/master/script/plot_graph.sh#L11

are your files still at /tmp? And if yes does a manual call of the script generate the picture.

[s0ybean]

Sorry, I do know to define the directory. That is why resonance test works :)

Manual call to the script with belt variable results in the same error

Files are not in temp. They are all moved to the defined directory

[zellneralex]

Never seen that. Are you sure the 2 files are in /tmp when calling the script. Until now it works always as long the files where there.

[s0ybean]

I swapped sd cards for an unrelated reason. The issue is resolved. Possibly a bad SD causing write errors? Either way, thank you for the troubleshooting help!

[zellneralex]

I will close it.

Bottom line: I can not reproduce the issue. Some helped to add a G4 P5000 before line https://github.com/zellneralex/klipper_config/blob/master/input_shaper.cfg#L143