FSR conversion?

Discussion in 'OpenBeam Kossel Reprap and OpenBeam Kossel Pro' started by roadblock, Sep 16, 2015.

  1. roadblock

    roadblock Member

    Hi all,

    Is there a complete thread on how to convert to FSR? I am not sure what it's other than replacing the probing actions...I can't even search for it...

    I want to convert to the ed3 print head but the designed some of the members came up with doesn't have the probe and was made to work with the fsr...
    Last edited: Sep 16, 2015
  2. grainiac

    grainiac Active Member

    1. Order the FSR kit from UltiBots.
    2. Print the bed mounts V2 from @Taylor (the one with off center FSR holes)
    3. Install everything into the printer
    3.1 Replace the original bed mounts with Taylor's
    3.2 Glue sensors to the bed mounts and install the silicon spacers on top
    3.3 Find a good place for the sensor board, set it to lowest sensitivity and mount it somewhere in the printer
    3.4 Connect sensors with board
    3.5 Connect sensor board with probe connector on Brainwave controller​
    4. Remove old probe from end effector
    5. Adapt Marlin for the bed probe, see 1st post in E3Dv6 end effector
    6. Test
    7. Done
    Corey Morreale likes this.
  3. Rick Gordon

    Rick Gordon Member

    yeah, that's pretty much it I think. I'm going through this now (see my thread about replacing the hot end if you want to follow along - i guess i should have started a new one.) Watch out for any contacts with the spring clamps now that you're closer to the bed when probing. Also - if the controller board led is not a bright steady green you may have the connections backwards.
  4. roadblock

    roadblock Member

    Ok thanks guys, very helpful. I will probably have some more questions but I have to order the FSR Kit and ABS to try and print the bed mounts with the current KP set up. Hope it prints the mount ok...
  5. William Siers

    William Siers Member

    I would just order mounts from Matter Hackers. I had EMI problems throwing random readings in so consider some type of shielding on wires that run to FSR board.
  6. roadblock

    roadblock Member

    @William Siers I tried to search for it on their site, I don't see any mounts for sale. Can you provide a link? Thanks
  7. William Siers

    William Siers Member

    Call Taylor at Matter Hackers (949) 613-5838
  8. roadblock

    roadblock Member

    Hi All,
    @grainiac or @William Siers or @anyonethatknows
    What is the post step after installing the FSR and changing the firmware? How do I even know the FSR is working? I tried to print or sent a G28 then a G29, and the head keep skidding into the glass.

    Thanks much...
  9. Rick Gordon

    Rick Gordon Member

    There should be a solid green LED lit on the FSR board, no red LED's.
    When you press on the bed with your hand, one or more red LED's should light up - corresponding to the side(s) that trigger.
    Make sure that the endstop switch on the head is open - that your old probe is not deployed - otherwise that will confuse it into thinking that its making contact.
    Use G1 commands to move the head close to the bed, then start incrementally pushing down 0.1mm at a time.
    When you do the G29 the probing sequence should start.
    Make sure you're not running into the bed clamps at any point.
    Each time the head makes contact with the bed with enough force, some of the the LED's on the fsr board will light up

    that's all that I can think of right now.
  10. roadblock

    roadblock Member

    What specific firmware are you guys using? is it the FSR enabled Marlin on the ultibot product page? I just used the firmware reference in the updating firmware thread, not even thinking I had to use a special fsr firmware?

    Do I need to enable FSR in the openbeam marlin version? I made the changes outlined by @David Boyd in this thread on the open beam marlin version of the firmware.
    Last edited: Sep 24, 2015
  11. David Boyd

    David Boyd Active Member

    Those changes are all that should be required. A common problem is the 3 wire connection to the board may need flipped. Take the bed off and try pressing each sensor with your finger it should light up that sensor's light with just a little pressure. Then test with the bed on. I used the thicker pads that came with the sensor.
    Oh and verify that if you changed the sense for the switch in the firmware you do not have the jumper in place on the board to change the sense polarity.
  12. William Siers

    William Siers Member

    @David Boyd a little more on the three wire connection needing to be flipped???
    I am back to trying to get mine correct again, just cannot get there, it does the pattern. But when I print it is not level, still high on one side low on the other. Also in my start up it does a G28 after bed warms. Then a G29, however the hot end is warming up during the G29 which causes a slight oozing of filament. This in turn is not helping my calibration.
    Plus now I have little dots of plastic all over the table.
    I have, for now eliminated any / EMI / false triggers prior to touching the plate by twisting the wires to the sensor and putting some aluminum foil around a wire that was connected to the sensor that would be triggered falsely. Then some shrink tube.
  13. roadblock

    roadblock Member

    @David Boyd and @William Siers Ok thanks guys. I manage to get it to light up correctly. I had reverse the connection going into the brainwave board from the fsr board. When I had it reverse it was a dimly lit green light versus a solid bright green light. I now see the red led when pressure is applied. I printed a test cube and it seems to calibrate it like it was supposed to, however the first layer prints too close to the glass that no filament gets extruded. I have to fiddle with it some more, probably the sensitivity settings.

    Others who are trying to do this, the other issue I ran into was the Arduino software. You need to download the 1.06 version.
  14. William Siers

    William Siers Member

    It is in one of the threads here about using 1.06. Lots of info but buried. You probably will still need a Z offset. The bed leveling gives you zero. After bed leveling and prior to a G28, home, use a G1 Z 1, this should put head centered and 1mm above bed. Then go G1 Z0.9 and lower in 0.1 increments until paper slides under the tip with a slight drag. Try that Z value as Z offset. G29 runs before every print, G28 will zero your values to default. I find if I remove the hot plate I need to do everything from scratch. Also running another print after an aborted print the hotend is still hot and when the bed warms up a G28 starts followed by a G29 while the extruder is warming up, resulting in some oozing. This on turn affects the bed leveling as some material is on the tip of the nozzle. Also you may find you have to allow negative Z values in MC setting. I am still having issues with mine. Other option is to run G29 several times, copy the values down and average each position and put in firmware as default. Mine are not consistent enough to do that yet, waste of time. Also your extruded first lines, skirt, should be slightly squished on top. Keep in mind I am a newbie and am just passing along my experience. Let us know what you find!!
  15. Rick Gordon

    Rick Gordon Member

    an old thread - but were you able to reuse the standard spring-based bed clamps with the "lower" threshold setting (and by lower sensitivity - doesn't that mean the head has to press harder on it?) Or did you switch to a different clamping?
  16. William Siers

    William Siers Member

    I used plate holders from Matter Hackers. Made to hold the fsr sensors. But still was able to use the spring clamps from original. However this left one side of clamp plate holder not touching the plate. FSR under one side nothing under other. So I shimmed the opposite side without the fsr. This seemed to help with bouncing and sporatic readings.
    Also I had one sensor that required a lot more force than the others so I shimmed the FSR as well to raise slightly, still think that is a bad sensor. Lower sensitivity should require more force. I ended up with two jumpers to set to high sensitivity. Really the paper leveling system in Matter Control 1.4 works just as well. You dont need every print. The FSRs would also get bad readings caused by EMI. So shield or reroute wires to stop.
  17. roadblock

    roadblock Member

    I was able to use the original spring clamps and it doesn't seem to have any trouble doing the leveling test. Although, I am still battling changing the effector from @grainiac and get it to work correctly. As far as leveling, I set it to lowest settings and using original clamps. It used to have some interference but hadn't done it recently.
  18. grainiac

    grainiac Active Member

    Are you still using the Brainwave or have you also changed the controller? The auto calibration from Smoothieware really helped me to get it dialed in!
  19. Sean Hill

    Sean Hill Member

    Sorry to resurrect such an old thread -- @grainiac, you gave some great steps in getting FSR-based probing up and running. I've got my bed clamps replaced and the electronics wired up, but I'm curious about the firmware part... I'm assuming the stock KP Marlin fork doesn't include FSR support. Do you happen to know which of the branches out there is most appropriate to use here?

    I know you've moved on to Smoothieware for your firmware, but you seem like the right person to ask and I'm not sure what my next steps should be. I can definitely change the constants as necessary and recompile/upload Marlin to the Brainwave, but I'm not sure what my starting point should be.
  20. grainiac

    grainiac Active Member

    Hi @Sean Hill the FSR controller board that does the sensor fusion has a normal signal output you'd expect from a single sensor probe. So electrically there is nothing special. However, I remember that there was a change for the probe offsets which I documented in the first post of http://forums.openbeamusa.com/threads/e3dv6-end-effector.266/. Hope this helps!

Share This Page