Rostock Max V2 to V3 troubles

A place to talk about user-made mods and upgrades to their machines
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Rostock Max V2 to V3 troubles

Post by Stimpy »

I recently (last weekend) upgraded my V2 to a V3 with EZR struder, HE 280, and wiring whip.

I made a couple of adjustments to the V3 firmware settings after some trial and error. Flip Y-axis, adjust Max Z, etc. To make it line up better with the V2 settings.

After the upgrade the calibration from the LCD panel ran fine once or twice. I attempted to use the Delta Calibration to fine tune it but after the second or third probe it simply started dragging across the build plate. I didn't change any settings during the failed Delta Calibration.

Now whenever I run the calibration from the LCD panel it will probe fine at the tower bases however when it probes the center it digs in hard and slides to the right. It also sounds like a belt may be slipping (tensioned them again with similar results and inspected for wear). It then simply repeats this same probe several times with the same results until I shut it off.

It will occasionally (when completely cold) complete the calibration routine properly however that first center probe will still usually dig in hard once or twice before it proceeds. I have yet to have it complete the process when it was warmed up (even with the heat turned off).

I have also frequently seen I2C errors in MatterControl and MatterControl touch (latest releases of both).

I am contemplating trying to put a ferrite bead on the wiring near the Rambo Board and possibly in the top of the printer as well to see if this may prevent the I2C recv errors I am seeing.
latif
Plasticator
Posts: 6
Joined: Wed Apr 27, 2016 11:54 pm

Re: Rostock Max V2 to V3 troubles

Post by latif »

Hi Stimpy,
Anything new on your end, I have the same issue where the hotEnd will dive hard into the plate.
I thought that the idea with the upgrade was to prevent especially this from happening, so not sure...
I also had error messages when sending configuration.h to the board did you get that too?
Thanks
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

I didn't have issues with uploading the firmware once I got the correct port selected. I did accidentally send a reset EEPROM to my Bluetooth headphones though. They still seem to work. Still poking at some things and will likely pick up some ferrite beads tomorrow or tonight to try on the I2C lines. I will let you know how that goes.
geneb
ULTIMATE 3D JEDI
Posts: 5362
Joined: Mon Oct 15, 2012 12:47 pm
Location: Graham, WA
Contact:

Re: Rostock Max V2 to V3 troubles

Post by geneb »

Ferrite beads on the i2c lines won't help, and will likely damp out the legitimate signal on those lines.

g.
Delta Power!
Defeat the Cartesian Agenda!
http://www.f15sim.com - 80-0007, The only one of its kind.
http://geneb.simpits.org - Technical and Simulator Projects
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

The first thing to try is to see if the Z height and travel is correct. Do this by using MatterControl terminal or similar to send G28 then a G0 Z20 command.

That should send the hot end to 20 mm above the bed. If that is close then you know you have the Max Z height good and the step size correct and that the steppers are working well.

Let us know how that goes. Then we can try other steps to diagnose.

Dale Eason
User avatar
thechewiestbacca
Printmaster!
Posts: 50
Joined: Tue Dec 27, 2016 5:53 pm

Re: Rostock Max V2 to V3 troubles

Post by thechewiestbacca »

Dale,
I am having the same issue, and when I ran "G0 Z20" as you suggested, the nozzle continued to drive into the bed until I turned the printer off. My guess was that I needed to change the Z-Height in configuration.h to the original RM2 setting (350). When I ran "G0 Z20" with the new config, the same thing happened. What is the next step if that code sent the hotend crashing into the bed?
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

Right so lets look at the eeprom settings. Either the Z Max Length or the step size is wrong. Use MatterControl to read the EEPROM values. On my V2 to V3 with a 1/8 inch PEI build plate my Z Max Length is 362 mm. Steps per mm is 80. Yours should be somewhere similar for Z Max Length and 80 for steps per mm.

These setting are initially set by the values in configure.h only if the EEPROM has been cleared before running the software for the first time I think.
But you can change them using MatterControl or Octiprint.

Dale
User avatar
thechewiestbacca
Printmaster!
Posts: 50
Joined: Tue Dec 27, 2016 5:53 pm

Re: Rostock Max V2 to V3 troubles

Post by thechewiestbacca »

Well, I cleared my EEPROM, uploaded the firmware, and ran the calibration process. It worked fine. The only reason I could think of as to why it didn't work before, is because I didn't clear the EEPROM. I didn't realize that it was necessary to clear each time the configuration was uploaded.
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

I added ferrite beads to the I2C lines near the RAMBO and at the top of the wiring whip and my initial attempt at running calibration did not have the same problem as before. I haven't fully tested it yet but this was a promising result. I will update again after I attempt a print or two and after I hook a Matter Control system back up to it with information on if I see the I2C recv error again.
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Ok when running the calibration from the LCD and the printer mostly cool it works fine. If I run it while the printer is heated up it probes fine at the tower bases then bottoms out on the first probe to the bed center multiple times until I shut it off. After running the calibration while semi-cool from the LCD panel I then heated things up and attempted a simple print and the nozzle was completely buried into the BuildTak and digging a light furrow into it. I have attached my EEPROM configuration if anyone wants to look it over.

Update: Also still occasionally seeing the I2C recv error occasionally in MatterControl when connected.
Attachments
RosieEEPROM.txt
(1.99 KiB) Downloaded 218 times
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Also if I run the calibration from the LCD with the Bed on and heated to 100C it will crash even harder than it did at the center when warm but at the tower bases which are fine when warm.
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

I see your EEPROM setting for Z MAX LENGTH at 369.xx is 7 mm larger than mine at 362.xx I replaced my glass plate with a PEI plate about the same thickness as the glass so I would expect your values to be closer to mine assuming you just have a glass plate. As an experiment I would change it to something smaller and then try the G29. The G29 of course will change it to what It thinks it finds. But at least then you can see if that value is the issue.

I suppose there could be other reasons for your Z height being larger than mine. Anyway reducing the value may give us some info if you want to try it.

Dale
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Hi Dale,

I did a manual check of the Z Max by driving the hotend down to the platform via the LCD and the 369 height is actually accurate to within a couple of tenths.

So I started playing around with the M261 command to see if I could replicate the intermittent accelerometer I2C recv error in various states. I didn't do a good experiment yet because while doing that I found something else. If the following turns out to be a rose colored fish I will do a proper experiment testing for I2C errors at various temperatures and with the hotend and bed in different states (on/off)

What I did find when I was about to start testing with the M261 command was an interesting bit of information in the connection log. Please see the attached. The "Error:expected line 1 got 64" line appears to very the last number somewhat. I saw it also at one point with a number in the 1000s range don't remember the exact value though.
ConnectionLogFull.txt
(4.81 KiB) Downloaded 218 times
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

Could you set the EEPROM value to 360 and then try the G29 to see if you still get a crash right after doing that. You can always set it back to 369 value afterwords. If you still get a crash then we know that is not the problem and it is more related to the probe not working than something else.

Dale
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Just tried with a fresh install of the firmware Printer set to 2 Confirmed Z Max was 350. Nothing heated and when I ran the Full Calibration from the LCD Menu (G29 from console did nothing) it crashed at the base of the X Tower on the first probe attempt. I really think there is something wrong with my accelerometer board such that it stops communicating intermittently.
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

You did not say if you cleared EEPROM before installing new firmware. That is an important step. I too found that it would crash if I did not clear EEPROM using the clear EEPROM program.

It also puzzles me why G29 does not work from the console. What console (mattercontrol?) maybe? Does the G28 work from the console?

Dale
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

The important part of communicating with the accelerometer is the "int" line. That line indicates that the accelerometer value has exceeded the preset value meaning the acceleration has crossed the preset threshold. That threshold is set at power on of the printer. The i2c interface is then used to clear that value. So you can test that as you started earlier by sending the M261 commands and then you can thump the hot end with your fingers to see if the int line changed states.

Dale
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

I looked up the "expected line " line error you got and it is just part of communicating with the USB. It just did not get the response it expected from the host computer. That could just be trash left in the serial buffer of either machine that caused it to be out of sync communicating. It probably is not an issue unless it happens a lot and commands are lost.

Dale
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

I did clear the EEPROM before loading the firmware. The main reason I was concerned about the message during boot is because I also get Accelerometer recv errors occasionally in Matter Control.

The G28 command that I ran before attempting G29 did in fact home the printer but I waited about a minute after sending G29 and nothing happened. I have had G29 work via the Matter Control console after hooking up the HE280 but for some reason several attempts today did nothing with the G29 command. Even after a restart of the printer. G28 worked fine every time G29 never did anything. Running the calibration from the LCD bottomed out hard at the X tower even with Z Max at 350.

I will try the thump test tomorrow to see what happens.
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Life got a little busy for a few days but I just got done testing the accelerometer. After homing the printer and running M261 I got INT PIN: 1, this INT value of 1 seemed to hold for multiple runs of M261 without any other action taking place. I then thumped the cold nozzle lightly and ran M261 again, this time INT PIN: 0 was the result. Another M261 and INT PIN returned to 1. This was repeated about 10 times. The M261 after each thump always returned INT PIN: 0 and subsequent M261 without thump returned INT PIN: 1.

I then started heating the bed to 100 C. And continued with occasional thump tests. Every test at this point has resulted in INT PIN: 0 after thump and INT PIN: 1 prior to thump or after 2 M261's post thump (1 for the INT PIN: 0 output and another for the reset output).

At around the 80 C temperature Matter Control began reporting I2C recv errors again in the UI and occasionally in the terminal.

I then decided to start heating up the hotend. This very quickly began reporting more I2C recv errors in the terminal and a significant number of the M261 tests had errors mixed in or replacing the register output lines. The thump tests seemed to continue working properly though with the INT PIN going to 0 after thumps however there were some lighter thumps (maybe under the threshold) that did not cause INT PIN to go to 0.

I have attached the terminal output below and cleaned it up (since MatterControl seems to include a lot of garbage output these days even with filtering enabled).
Terminal.txt
(35.3 KiB) Downloaded 109 times
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

Well it sounds like you accelerometer is working and being read fine. The I2C circuit shares a line with the thermistor so it is not supposed to be used with the hot end heat turned on. The G29 command turns the heat off before it tries calibration so that is not an issue for it. The M261 command just plows along without doing anything about the hot end heat.

I'm not sure I have ever seen the I2C error you are seeing. I will pay more attention and do some testing with my machine. But it looks too me like somehow it is heat related. Maybe as the Rambo or its terminals heat up since this happens even when only the bed is being heated. I'm sorry I can't help more on that.

Maybe others can comment on ways to diagnose further.
Dale
Stimpy
Plasticator
Posts: 14
Joined: Sun Apr 17, 2016 10:58 pm

Re: Rostock Max V2 to V3 troubles

Post by Stimpy »

Main main concern at this point is that I can calibrate the machine (usually) when it is cold but when everything heats up the clearance between the bed and nozzle is much lower and when I attempt to print the bed is blocking the extrusion.
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

That kinda happens on mine as well. What I do is just readjust the EEPROM setting for MAX Z Height. Others have suggested changing the EEPROM value for prob height. I think that is more appropriate and could be what it's purpose is for in the first place. The bed is going to be at a different height when expanded by heating. My problem is I never took the time to understand if the probe height should be lengthened or shortened to account for a shorter Z distance.

Dale
User avatar
thechewiestbacca
Printmaster!
Posts: 50
Joined: Tue Dec 27, 2016 5:53 pm

Re: Rostock Max V2 to V3 troubles

Post by thechewiestbacca »

I've had the calibration process work with the MAX Z Height at the V3's value of 395 (on accident) and the V2's value at 350.
Why would the Z MAX HEIGHT matter? Doesn't the probe detect where the bed is, no matter what the EEPROM value is stored as?
Dale Eason
Printmaster!
Posts: 191
Joined: Fri Mar 06, 2015 8:56 pm

Re: Rostock Max V2 to V3 troubles

Post by Dale Eason »

The probe is not on except during a special section of the calibration. The first thing the code does is to move the head down to 20 mm from the max z height without the probe being turned on. If the Max Z height is set 20 mm larger than the bed distance it will crash into the bed.

It depends on when the wrong max z height was set. If it was set in the configuration.h but you did not clear the eeprom before loading the firmware back in then it will use the old value already in the EEPROM. So there may be times when you thought it was using the 395 value but it was not.

Dale
Post Reply

Return to “Mods and Upgrades”