Re-ARM from Panucatt

So I promised a review after a few weeks.

There isn't too much to write as once you've got it setup its fairly smooth sailing. I won't say more about setup, a lot of it is covered above.

Print quality is definitely better, it just seems more accurate. Prints themselves are improved.
I have had no issues with endstops, sensors etc.. these all seem fine and its a 5v tolerant board for powered endstop devices which cannot run on the 3.3v supplied on the endstop VCC pins. There is a

The setup process is quite easy provided it is happy with your memory card, if not try another (or two). Getting the pins right takes a bit of thinking through and referencing them with the diagrams above of a ramps board and the re-arm.
Compiling the latest edge firmware is again straightforward if you follow the guide on the smoothiware website. Getting it on the re-arm is just putting firmware.bin on the memory card and inserting it, it changes to firmware.cur when its installed. It does make me wonder why a pre-compiled version can't be hosted by smoothie on a weekly/daily basis it would save users some hassle, but then as re-arm customer I am not paying smoothie for their firmware, so enough said.

It does seem a bit reluctant to connect to octoprint on Raspberry Pi, but it will connect and print just fine.
The default web interface is reasonable, but download the enhanced version from smoothieware its massive around 280mb but this includes all the documentation and whole lot of stuff, memory card storage space is cheap, right? This version works quite nicely, has a console and enables you to upload files, albeit quite slowly even on wired ethernet.
I haven't tried the SD card slot in my reprap full graphics display to see if that works too. I do have a grip about the LCD in that theres a lot of space on screen but I cannot view bed, extruder, and chamber temps (2nd extruder thermistor) temps at the same time it cycles between them.

It takes a little getting your head around the jerk (junction deviation) used in smoothie, but its a minor thing.

The grid levelling seems fairly good, its not perfect but it works, if you print a reasonably thick first layer its not a problem.

All in all well worth the money if you are upgrading an existing machine already using an ardunio/ramps board. That being said I am not sure the cost of a re-arm+decent ramps board and 4/5 stepper drivers would be lower than a 32 bit integrated board such as the other offerings from Panucatt (Azteeg), genuine smoothieboard, or the chinese version the MKS-Sbase.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Considering the chance to have a bad RAMPS boards with underrated power connectors, wrong MOSFets, bad polyfuses and/or 16V capacitors and the fact: firmware is not plug_and_play, I would lean to a 32bit board where these issues are already ironed out. Maybe Panucatt will sell a re_arm bundle with a known good RAMPS and SD-Card with preinstalled sample files for most common printer types.

Would it be possible to write a "auto-config" routine?
You'd connect the old Mega board and Re-Arm and the routine would read out the Mega-Eeprom and generates a new smoothie-config file.

Then there would be only a few things you'd have to setup manually.

I'm sure it would be possible in a way, but not in the way you suggest, as you can't extract a config from the rom of an arduino. You can get the data stored in eeprom but this is far from the whole config. Plus, you'd need to have a version which worked for each version of Marlin (they change it all the time) and Repetier. You could do it from the config files on your pc.

It might be better to go down the route that Marlin, RRF and Repetier all have and that is a configuration wizard for re-arm that just asked you to enter all the stuff you have and then generates you a config file, with all the right pins etc...

Edited 1 time(s). Last edit at 03/10/2017 03:35AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

DjDemonD, you are using the Ramps fan extender successfully? It doesn't have an issue running with the 3.3v instead of 5v?

Doesn't seem to 3.3v is obviously enough to switch the mosfet/transistor on the fan extender. Im not 100% but I read somewhere some outputs are at 5V, might have to check that with the DVM.

Edited 1 time(s). Last edit at 03/10/2017 02:59PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Regarding integrated versus rearm plus ramps if panucatt sold ramps plus rearm it would be the same price as a 32bit board. IMO this product is for those who have an existing machine and want to go 32bit for 49.99.

If I was going for a new build I'd buy an integrated board.

Edited 2 time(s). Last edit at 03/11/2017 07:21AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Well I got my Re-Arm working on the bench today.

The cable that Panucatt sell for the graphic LCD has both keys on the wrong side. This wouldn't be an issue if it didn't have the extra 5V line hanging off... but it's there. I had to cut off both tabs and flip both connectors. Then the display worked fine.

I have to say that the menus for Smoothie is not a patch on Repetier. Quite a step backwards in terms of functionality.

Hmm I didn't have to do that but if it works, it works. Yeah the menu system is pretty basic there's a custom section which I haven't played with maybe that might be more useful. It'd quite a bit more basic than marlin too.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Just wanted to post this link with the table of pin numbers if anyone wants to add to it. It's on the other thread on General forum so best to have it here too.

[1drv.ms]

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Has anyone gotten the SD card on a graphic display to work? I've been playing around with it, it just doesn't seem to get recognised.

I'm also surprised that the smoothie devs seem to think that printing via SD card on the panel is unreliable. I've been printing for years on my RAMPS setup, using the SD card on the panel, with no issues.

Got everything working now. Trap for young players: if you notice that one of the web interfaces wants to use a file called "config" without the .txt suffix, and you put one onto the SD card, and you then try modifying config.txt, then the .txt file is not read on startup. Blech.


Attached is my config for a delta. Not tested in a printer, but everything seems to work on the bench, including the SD card.

I have not tried using the SD card on the LCD, I've just been mounting the on board SD card in Octoprint and copying files that way.

I need some info on BLTouch with Re-ARM. I have seen the servo code on Smoothiware. I'd like to know if we can use the same servo pin (1.23) for re-ARM. My BL-Touch is connected to the first (leftmost) servo connector.

I assume we also have to enable Z probing, using pin 1.29 for Z min.

Yeah 1.29 is z min endstop.

Servo D4 (1.18) D5 (1.19) are in the word doc above I'm using them for my fan extender. But they can be configured to output 0-255 pwm.

Edited 1 time(s). Last edit at 03/25/2017 09:14AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Does anyone have this working with the E3D PT100 sensor and amplifier? I have the wiring correct, and I am on the Edge firmware, but the temperature reads 950C. I wired it up on 1.30 and put the entries in the config.txt per the wiki.

Edited 1 time(s). Last edit at 03/26/2017 01:15PM by Traveler0612.

Well this is funny. I pulled the nightly build ( Build version: edge-7796e28, Build date: Mar 25) and it is working now. This forum is magical.

Quote
drmaestro
I need some info on BLTouch with Re-ARM. I have seen the servo code on Smoothiware. I'd like to know if we can use the same servo pin (1.23) for re-ARM. My BL-Touch is connected to the first (leftmost) servo connector.

I assume we also have to enable Z probing, using pin 1.29 for Z min.

To get the servo to work, you need to provide the centre rail on the servo ports with 5V. The normal jumper on RAMPS only provides 3.3V, so I added an extra 7805 regulator to give it some 5V dedicated oomph.

The leftmost servo connector is 1.20.

I use an actual servo, but this code should work for the BLTouch as well. You might need to tweak the S numbers to get the BLTouch to do what it needs to (I have one, but gave up on it a while ago due to it not being accurate enough for me).

Here's my servo setup code:

switch.servo1.enable				true		# Enable servo1
switch.servo1.output_pin			1.20		# SER1 pin
switch.servo1.output_type			hwpwm		# PWM output
switch.servo1.startup_value			0		# Start with servo disabled
switch.servo1.max_pwm				100		# Max PWM
switch.servo1.pwm_period_ms			20		# PWM period
switch.servo1.input_on_command			M280		# G code to set PWM period.  M280 S4 is full one direction, S7 would be midway, S10 is the other extreme.
switch.servo1.input_off_command			M280		# G code to set PWM period
switch.servo1.fail_safe_set_to			0

And the Z probe code is as follows:

## Z-probe
# See [smoothieware.org]
zprobe.enable                                true           # Set to true to enable a zprobe
zprobe.probe_pin                             1.29^          # Pin probe is attached to, ^ means pullup, invert with ! if the sense goes the wrong way
zprobe.slow_feedrate                         5               # Mm/sec probe feed rate
#zprobe.debounce_count                       100             # Set if noisy
zprobe.fast_feedrate                         100             # Move feedrate mm/sec
zprobe.probe_height                          5               # How much above bed to start probe
#gamma_min_endstop                           nc              # Normally 1.28. Change to nc to prevent conflict,

You may need to invert the probe pin with a ! character.


Good luck

Edited 1 time(s). Last edit at 03/29/2017 10:19AM by nebbian.

Are you guys having any issues with PID tuning and hotend temperature control? I have a genuine e3d V6, and on Marlin my hotend temperature control was spot on.

On Smoothie, after running Autotune the same way I did on Marlin, I'm having pretty bad temp control. On initial heatup, the temperature always overshoots by 10-12 degrees, then takes a few minutes to come back down. Then after the first layer and the fan kicks on and print speed goes up, the temp drops 10 degrees below target, and takes a few minutes to come back up. On hotter materials like PETG, the temp constantly bounces between target and 5 degrees below target throughout the print.

I Autotune to 210 degrees with the fan on full blast with filament in the hotend for both Marlin and Smoothie.

I haven't had a problem with auto tune on rearm, it's a bit slow on the hotend getting to temp taking maybe 30 seconds to climb up from 10 degrees below set temp. Fans don't seem to bother it. The bed on pwm is very stable.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
Pheneeny
Are you guys having any issues with PID tuning and hotend temperature control? I have a genuine e3d V6, and on Marlin my hotend temperature control was spot on.

On Smoothie, after running Autotune the same way I did on Marlin, I'm having pretty bad temp control. On initial heatup, the temperature always overshoots by 10-12 degrees, then takes a few minutes to come back down. Then after the first layer and the fan kicks on and print speed goes up, the temp drops 10 degrees below target, and takes a few minutes to come back up. On hotter materials like PETG, the temp constantly bounces between target and 5 degrees below target throughout the print.

I Autotune to 210 degrees with the fan on full blast with filament in the hotend for both Marlin and Smoothie.

Interesting, I found the autotune to give really good results, way better than either Marlin or Repetier. It only needed a very slight tweak for me to get great results. Sounds like the numbers really need tuning on your end. Out of interest, what are your numbers for the PID settings? You might have your I clamp (not sure what it's called in Smoothie) set way too high (or not enabled at all), meaning that the I term suffers from "integral windup". That's what it sounds like to me.

Edit: found it, it's called "i_max".
[smoothieware.org]

I would lower i_max, and raise your P value.

Edited 1 time(s). Last edit at 05/03/2017 08:42PM by nebbian.

Quote
nebbian

Quote
Pheneeny
Are you guys having any issues with PID tuning and hotend temperature control? I have a genuine e3d V6, and on Marlin my hotend temperature control was spot on.

On Smoothie, after running Autotune the same way I did on Marlin, I'm having pretty bad temp control. On initial heatup, the temperature always overshoots by 10-12 degrees, then takes a few minutes to come back down. Then after the first layer and the fan kicks on and print speed goes up, the temp drops 10 degrees below target, and takes a few minutes to come back up. On hotter materials like PETG, the temp constantly bounces between target and 5 degrees below target throughout the print.

I Autotune to 210 degrees with the fan on full blast with filament in the hotend for both Marlin and Smoothie.

Interesting, I found the autotune to give really good results, way better than either Marlin or Repetier. It only needed a very slight tweak for me to get great results. Sounds like the numbers really need tuning on your end. Out of interest, what are your numbers for the PID settings? You might have your I clamp (not sure what it's called in Smoothie) set way too high (or not enabled at all), meaning that the I term suffers from "integral windup". That's what it sounds like to me.

Edit: found it, it's called "i_max".
[smoothieware.org]

I would lower i_max, and raise your P value.

Turns out I'm just an idiot. I ran the autotune and set the values multiple times without noticing the three factor settings in config.txt were still commented out. I haven't ran a print with them enabled, but just setting a target temperature it behaved much better, although not perfect.

Must the power jumper be set to Internal for normal usage? I would like to leave it set to USB power, so that the board stays on when I turn off 12v power. This would allow the board to control an ATX power supply or relay to turn the printer on/off.

I have had the same issue for quite some time as my other printer I can control it using Octoprint to turn it on and off and would like to do the same with the Re-Arm board. I finally got around to looking at the schematics using diptrace as I was not sure how JP1 int / usb power jumper worked and whether it could be powered by USB alone. I also had issues where the Re-Arm board would disconnect from the USB which meant I could not print via the USB serial and also could not have the USB connected whilst printing.

So after looking over the schematics I decided to use an Arduino 12V 2A PSU connected into the Re-Arm board and that seems to have cured my USB connection issues. Now I am going to use a Arduino single relay board wired in as per this instructables article convert led PSU with remote controlled power

I have now put a remote relay inside my led psu but now I find that if I issue an M80 power on I get a kill button pressed message. Sometimes an M81 gets me unhalted message but this is really odd as I have the kill button set to disabled. Anyone else seen this?

I have now fixed the issue. Just in case anyone else has this issue I found that with the kill button disabled it is still set to pin 2.12 by default, which is the same as the PS_ON pin for the Re-Arm. Pin 2.11 is the encoder push switch and I found that setting the kill button to pin 1.22 which is the stop button on the RR GLCD works fine now.

Just joined the forum and nearly finished my first build. Nearly everything working

Has anyone had a problem with the temperatureswitch on the Re-Arm?

I have the hotend cooling fan on the fan extender. It works with the M245 M246 commands.
The other output on the fan extender also works with M42, and the partcooling fan works on D9 with M106.

I have configured a temperatureswitch to turn the fan on off at 50 degrees. (Attached).

No response. Seems an important function to get working rather than rely on start codes in the start gcode file or each print file. I like the fans off when not needed.

It seem strange to me to have

switch.fan.input_on_command M245
switch.fan.input_off_command M246

and

switch.fan.output_pin 1.20 # FANEXTENDER

If you have a on command you dont need a pin, and if you have a pin surly you don't need a on_command

The one eg I can find [www.instructables.com] show the commands are commented out

this is just an educated guess... (cant test it at the moment)

Thanks Dust

Isn't the output pin no required to tell the firmware where the fan wires are connected ?

There seems no reason to comment out the on / off commands unless some feature of smoothie doesn't like M codes and switches doing the same thing.

Anyway I tried commenting out the lines, also changing the M codes to M42 and M43 in case there is some reserved use for some codes. Also tried M106.1 someone suggest elsewhere but that failed (some syntax I don't understand perhaps.)

Todays workaround is to rely on M245 Snnn codes in the print file and also I have added M245 and M246 as custom controls in the panel.

The other issue, is it possible to copy print files to the external SD card? The external card is not visible to fdisk but the internal one is.

I can copy files to the internal SD card using either Octoprint or simply by mounting the internal card and direct access with the file manger and printing from the LCD.

As you can tell I'm a novice floundering around at the moment although everything else, nearly, is working.

Edited 1 time(s). Last edit at 12/01/2017 08:35PM by garis.

I've gone back to Marlin on my Re-Arm. It was working really well, up until I updated the firmware to enable advanced pausing, and bricked my board