Hot end autotune question

[v8tony]

I re-ran the autotune on my hotend and used the last set of values that popped up instead of averaging all three sets and it seems to be working better... I used to get +/- 3 degree swings while in use and a +6 degree overshoot on the initial heatup... These issues seem to be gone now... Can anyone else confirm that they used the final set of numbers of did everyone use the average... I found other forums that implied just to use the last set of numbers (which is why I tried it)...

Tony

[Polygonhell]

I just use the last set of numbers, but it shouldn't matter much.
I wouldn't use the average, averaging the values, doesn't make very much sense.
It is an iterative process, in that running the process multiple times should get you closer.
Practically you should see minimal overshoot <5degrees and it should maintain temperature within 1 degree when tuned appropriately.

[fredini]

Interesting- I was averaging the three values. I'll try just using the last set.

Also- is anyone else getting these error messages in the console after the values:

< 5:33:25 AM: PID Autotune start
< 5:36:41 AM: bias: 68 d: 68 min: 198.36 max: 205.89
< 5:37:36 AM: bias: 68 d: 68 min: 197.86 max: 202.81
< 5:38:24 AM: bias: 64 d: 64 min: 198.49 max: 202.23
< 5:38:24 AM: Ku: 43.51 Tu: 48.24
< 5:38:24 AM: Clasic PID
< 5:38:24 AM: Kp: 26.11
< 5:38:24 AM: Ki: 1.08
< 5:38:24 AM: Kd: 157.42
< 5:39:13 AM: bias: 64 d: 64 min: 198.39 max: 202.10
< 5:39:13 AM: Ku: 43.93 Tu: 48.63
< 5:39:13 AM: Clasic PID
< 5:39:13 AM: Kp: 26.36
< 5:39:13 AM: Ki: 1.08
< 5:39:13 AM: Kd: 160.21
< 5:39:59 AM: bias: 64 d: 64 min: 198.82 max: 202.10
< 5:39:59 AM: Ku: 49.65 Tu: 46.53
< 5:39:59 AM: Clasic PID
< 5:39:59 AM: Kp: 29.79
< 5:39:59 AM: Ki: 1.28
< 5:39:59 AM: Kd: 173.27
< 5:39:59 AM: PID Autotune finished ! Place the Kp, Ki and Kd constants in the configuration.h
< 5:39:59 AM: Error:Line Number is not Last Line Number+1, Last Line:2154
< 5:39:59 AM: Resend:2155
< 5:40:00 AM: Error:Line Number is not Last Line Number+1, Last Line:2154
< 5:40:00 AM: Resend:2155
< 5:40:01 AM: Error:Line Number is not Last Line Number+1, Last Line:2154
< 5:40:01 AM: Resend:2155
< 5:40:02 AM: Error:Line Number is not Last Line Number+1, Last Line:2154
< 5:40:02 AM: Resend:2155
< 5:40:03 AM: Error:Line Number is not Last Line Number+1, Last Line:2154
< 5:40:03 AM: Resend:2155

[dgirard]

Yes...I get those same messages...

I'm assuming it's a firmware bug where it's not counting the lines it outputs correctly...

When I got them, I just used the values, then hit the reset button on the controller to clear everything up.

Is anyone *not* getting those messages?...

[Polygonhell]

Yes...I get those same messages...

I'm assuming it's a firmware bug where it's not counting the lines it outputs correctly...

When I got them, I just used the values, then hit the reset button on the controller to clear everything up.

Is anyone *not* getting those messages?...

I don't see them, it's usually an indicator of USB errors.
The firmware assumes that it gets gcode lines with sequential line numbers, if it doesn't see them it throws the error, and the host can resend the data.
Make sure you're USB cable is away from the power lines, you can also try reducing the baud rate in the firmware and the host.

[dgirard]

Strange...

I've done several test prints, and several calibration runs...

I *always* get those line number messages at the end of the calibration run (after the third set of outputs), and I've *never* seen one of these messages during a print....I'll check my usb cable, but the experience is much to predictable to be related to interferance or a bad cable IMO...

It's interesting that some are getting the messages, and some are not...could we be using different releases of the firmware or a different version of repetier?...

The firmware I have is "RELEASE CANDIDATE 2 FOR MARLIN 1.0.0 by seemecnc" (from readme.md) dated November 23,2012.

I'll have to check my RepetierHost when I get home, but it's running on windows 7 (the installer was setupRepetierHost_0_82_50b39f394178e.exe)

[fredini]

I am on a Mac, so the diference in Repetier versions across platforms could also be the culprit....

[dgirard]

Hmmm.

I'm running it on a PC (win7) and I see the errors too.

I have a mac--so I can test there also...I also see that there's a newer firmware (release candidate 2) updated just a few hours ago...time to do some further testing on this.

[Polygonhell]

I did my calibration in Prontrface, so it's possible I didn't see it for that reason.
I use Repetier to print, but it for whatever reason it seems to suppress some TTY output, even when I think I have all the channels enabled.

[msdevstep]

Chances are that if your not seeing the errors it because you have the suppress errors button unchecked right above the bottom status area.

[dgirard]

All:

related to the line number error messages...I updated my firmware to Marlin 1.0.0 RC2 from github that John uploaded recently. It seems to have corrected the line number errors after autotune...