Update 1.05 and firmware 0,92 (?) issue

Discussions about firmware/slicing software, tweaks and settings
Post Reply
elshout
Posts: 2
Joined: Fri Oct 10, 2014 7:23 pm

Update 1.05 and firmware 0,92 (?) issue

Post by elshout » Mon Oct 20, 2014 9:39 am

Just had a HARD WEEKEND behind me... : (windows 8.1 / FW .92 / Rhost 1.04 and 1.05)

Updated to 1.05 during the time was also experimenting with setting in the manual-settings-menu
to fine-tune a print that had issues.

The new 1.05 installed the latest COM-USB drivers, but since they have a slightly different name also the older version - installed by 1.04- remains...(has to be manually removed).

Since i also - just before installing 1.05 - had adjusted the flowrate slider i got into SEVERE problems.

After a restart i could not get the printer to work anymore...only manual positioning commands
sometimes worked .. starting print DID warm up bed and extruder but once oparational temperature was reached nothing happend (although at very close range with your ear some sound was heard !)

I suspected a communication error between PC en Felix since the two drivers (as mentioned above)
might had caused the conflict....and yes the filament got stuck several times so i had to disassemble
the (dual)head twice,so had to got back to virgin install on different PC.

It took till last night when i finally discovered the issue seems to be firmware (0.92) in the Felix...
Someone please try setting flowrate to 95% or so and try to start a print....
If you have the same issue at least my machine is OKE...

Thanx, Gerard

frankjoke
Posts: 150
Joined: Sat Dec 28, 2013 6:07 pm

Post by frankjoke » Mon Oct 20, 2014 10:33 pm

Downgrade to 0.91, 0.92 did not work with me as well.

Dreide
Posts: 176
Joined: Sun Sep 29, 2013 12:32 am
Location: Lausanne, Switzerland

Post by Dreide » Tue Oct 21, 2014 9:39 pm

Felix 2.0->3.0dual * Repetier (host+firmware) * KISSlicer Pro, Simplify3D * Cubify Design

elshout
Posts: 2
Joined: Fri Oct 10, 2014 7:23 pm

Post by elshout » Sat Oct 25, 2014 4:34 pm

BUG confirmed by Felix. Now printing with 0.91 fine !

frankjoke
Posts: 150
Joined: Sat Dec 28, 2013 6:07 pm

Post by frankjoke » Thu Oct 30, 2014 12:12 am

Ok, there is new 0.92 firmeware out since October 28th. What did change? Did it work for you?

User avatar
DDME-Marc
Posts: 71
Joined: Sun Nov 24, 2013 11:04 pm
Location: New Zealand

Post by DDME-Marc » Thu Oct 30, 2014 2:00 am

Hi Frankjoke,

I am running the October 28th release of 0.92 (actually I have used all 0.92 releases).

There were a couple of major issues which came to light with the first 0.92 release. First the decoupling and secondly a firmware coding mistake which would cause a complete printer lock up if you adjusted feed rate percentage through the Repetier-Host user interface.

Solutions for both these major issues have been found and the subsequent 28th firmware release has been recoded to implement the fix. I see that the decouple test period is now down to 15 seconds (from 30) with the latest release.

Scouring through the various firmware sections I see several minor changes with coding (and or coding removed).

And somewhat amusingly the word 'killed' has been replaced with 'power off'. Perhaps to avoid new users from thinking they have destroyed their printer?

So apart from using different extruder PID values (to suit my materials) and various printer settings, I have made no major changes to the latest firmware release, and I have no problems to report to date.

Cheers,

Marc
Felix 3.0 Dual Head * E3D Titan V6 * Repetier-Host V2.1.3 * Repetier-Firmware 0.92.9 (01/08/17 - Modified) * KISSlicer Pro - 1.6.3 * Arduino 1.8.5 *

frankjoke
Posts: 150
Joined: Sat Dec 28, 2013 6:07 pm

Post by frankjoke » Fri Oct 31, 2014 9:20 am

Ok, I tried this version (0.92 from 28th) now and it is the first 0.92 which works with me :)

You speak about different PID values. How do you find and set them? Do you mean you have different ones for PLA and ABS because you need different temperatures?

frankjoke
Posts: 150
Joined: Sat Dec 28, 2013 6:07 pm

Post by frankjoke » Fri Oct 31, 2014 7:20 pm

So I know now that 0,92 works but I also know that PID values are different and BAD!

The extruder takes now endless to get to it's temperature. If I want to heat up to 195 it heats @ 100% until 180 and then steps back to 20-30% and it takes endless (now already 15 minutes) to reach 195°!

How can I change this or can I just use old PID values or did they change timing of loops and we need to calculate new ones?

User avatar
DDME-Marc
Posts: 71
Joined: Sun Nov 24, 2013 11:04 pm
Location: New Zealand

Post by DDME-Marc » Fri Oct 31, 2014 9:08 pm

Hi Frankjoke,

Like with most problems your specific symptoms could be a result from of a number of issues.

Firstly I would note Guillaume is using higher gain/higher damping PID values in the standard 0.92 firmware than I have personally have set. Additional to this is that he is allowing the PID to seek mathematical convergence within the full power band (e.g. 0 to 100 percent of the integral drive).

The benefit of standard 0.92 firmware settings is that that you can achieve a more stable temperature quicker, across a much broader range temperature settings, especially so if you heavily utilise fans.

If I understand correctly a possible down side of standard 0.92 settings is the increased noise as a result of the higher convergence initiated by the PID settings. And reading through earlier postings noise has been documented as an issue in the past...

In answering your question, yes you can use the old PID values, and these can be found/set in the configuration.h file (in the firmware). And yes I used the old 0.91 values initially with the first release of 0.92 (with reasonable success).

Before you embark on changing settings make sure you make a copy of your firmware, and always document changes as you go, since your personal printer may well vary to some extent.

You can type an M303 P0 S250 (where P0 is extruder 1, and S250 is the target temp.) command using your Repetier-Host. This will initiate an automatic goal seek for optimal extruder PID values, which it reports back at it conclusion a few minutes later.

Although I used the M303 command, I personally found the automatic goal seek values to be somewhat conservative because the test is done without fluctuating load (as it would encounter in normal operation).

I have also restricted the integral drive PID minimum and maximum values. This has allowed me to achieve a very relaxed (less convergence). Typically I have ~3 degrees overshoot at the start, and then progressively temperatures settle down over the next half a minute. Extruder's reach 250 degrees in just over two minutes, and once at temp it stays rock solid, fluctuations less than 1 degree regardless of retracts and travels.

For starters it is possible that noise may be part of your problem, so I perhaps running a few tests surrounding the PID values.

Cheers,

Marc
Felix 3.0 Dual Head * E3D Titan V6 * Repetier-Host V2.1.3 * Repetier-Firmware 0.92.9 (01/08/17 - Modified) * KISSlicer Pro - 1.6.3 * Arduino 1.8.5 *

frankjoke
Posts: 150
Joined: Sat Dec 28, 2013 6:07 pm

Post by frankjoke » Fri Oct 31, 2014 10:05 pm

I switched back to 0.91. I got heater decoupled messages and strange temperature behaviour. The PID is totally different between the two and the old one did not work as well. So some more things changed.

What are the changes and pro's what 0.92 should actually bring? I noticed some different text on the screen but that's it.

Post Reply