Attempting to replicate the SmartEffector for DeltaMaker.



  • Creating a derivative of the Smart Effector to fit a DeltaMaker end effector. I've managed to get the board manufactured by JLCPCB. I am an employee at DeltaMaker with no formal EE education, just a self-taught guy learning what I can.

    I've also done my best to select components on the wheatstone bridge (is this what I'm looking at with regards to B and T tags in the schematic?) according to the documents.

    Attached is the list of components that I've utilized. Some of it is still missing as we didn't feel the lights were needed while we tested the circuitry to make sure we could replicate the work.
    0_1552409605130_Cart_Mar12_1145AM.pdf

    I've flashed both v1 and v2 firmware to the ATTiny45 with an AVR programmer. I took the .elf files from both repositories, and ran the command "avr-objcopy -j .text -j .data -O ihex [elf file] [hex file]" in order to create the HEX file which I then flashed with avrdude. The activity LED lights up appropriately during flashing and the programmer is able to verify the flash file.

    The first version of the firmware, when connected to power - blinks the activity LED faintly.

    The second version of the firmware, when connected to power - keeps the activity LED on constantly, but again - faintly. To be clear, the "activity LED" I'm referring to here is D3 on the board.

    When assembling this board, what are some "pre-flight checks" that I can check with either an oscope or multimeter to ensure I'm within the proper ranges?

    Would Duet3D be willing to work with DeltaMaker in creating a batch of these? I've been fighting to get rid of our previous auto bed leveling system for ages, and this would be a huge benefit for our customers.

    https://drive.google.com/open?id=1Nsqo0SxTigf3paUD8WRTVJ1FaCXu5XvD - Are the current zipped modifications to the Smart Effector. (as a note, I know the 2x8 header is "on the wrong side"; it does actually match our pinout on the top connector though - KiCAD is new to me.)

    If willing, we'd love to work with you in creating this variant so that our customers could benefit from an improved leveling system and hot end ecosystem.


  • administrators

    What is the resistance of the top and bottom strain gauge traces of your effector? On our boards they are around 30 to 40 ohms each. We have the board made using 0.5oz copper instead of the usual 1oz to achieve this. If the resistance is too low then the current consumption will be excessive.

    The voltage at the junction between the top and bottom strain gauge traces should be about 0.5V, i.e. half the output of the 1V regulator.



  • Well that's certainly a detail that I'll need to figure out - I'm not back at work yet but if I remember correctly the strain gauge traces were measuring about 24ohm each. So there's a huge thing I missed. 😛

    Onto attempt #2! 😄

    On the schematic, it lists a 0.8v LDO; am I using the wrong LDO in this case?

    I highly appreciate your help on the matter. All of the DeltaMaker's components are bought from open source companies where applicable and we do our best to support the open source community by buying from them directly. Unfortunately this is one instance where we need a custom solution, so we have to produce a variant. 😕


  • administrators

    We used a 0.8V LDO on prototypes but we switched to a 1.0V LDO when we realised that the trace resistance was high enough.

    The version 2 firmware (using the attiny44a) includes a self test that includes a measurement of the LDO output voltage. This will likely fail using a 0.8V LDO, in which case you would see 9 flashes of the green LED.



  • I appreciate all your help on this. It looks like I've got some revisions to do and some additional testing.

    I saw some other things that I'm going to attempt with your setup - hopefully I'll be able to toss the revisions out there for you guys if they pan out.

    I had run into another guy using thin film resistors, like a 2512 (imperial) surface mount package in order to do his own strain gauge PCB. I'm hoping that once I can get this up and running and understand the circuit a little better - to replace the meandering traces on the PCB with a set of these. This (theoretically) should open up some board space for mounting some other hot ends (Slice Mosquito, E3D Chimera, etc).

    We will certainly pass the changes back upstream if we're successful!

    By chance do you have an updated BOM for the SmartEffector? I'm about to order some parts on mouser, and would like to try to iron out any discrepancies in my version.


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.