Duet WiFi controlled Flashforge Creator Pro



  • Here is my recently modded FFCP with lots of new additions, the best being of course the Duet WiFi.


    Perhaps some glass cleaner in the near future.


    It's now sporting an E3D Legends pack which at the moment just has one hot end attached with an E3D hardened nozzle. The BLTouch sitting next to it works like a champ. I can swap that out in seconds for an inductive probe as the BLTouch just sits in an adapter in the hole for the inductive probe. I'm also using an LED and a couple resistors on the Z-Probe input to act like a Zener diode so I can do 3.3v to 24v inputs and still be around 2.7v on the probe input.


    Back side of the carriage assembly. I think I need to clean that up a bit. The 30w heater plugs in via a Molex Microfit 3 connector as do most of the other items. The cooling fan is running from a JST connector. The spare connector is for a thermistor, I run PT100 sensors for the normal hot end.


    Back of the printer with spool holder bolts , the two bowden extruders and a 24 pin Molex Microfit 3.0 connector for the entire carriage assembly. The connector breaks out everything I could want to go to the carriage which at the moment I'm only using a limited amount of those items. I need to make a cover for that assembly which could attach to the two screws just above it.


    The extuder steppers are 0.9 deg step high torque motors rated at I believe 1.5A. I had them lying around so just opted to go with them. Getting that extra resolution on the extruder really made for nicer prints. If your doing this..go with the 0.9 deg motors.


    I replaced the 850ma rated high inductance motor with a little better 1.2A rated stepper on the Y axis. I also have a 2A rated motor that I tried out, it worked fine, I'm just in the process of trying this other one out as it has a little less inertia mass on the rotor.


    Speaking of mass I also replaced the X stepper with one of these small pancake style steppers rated at 1A. It runs a bit warm but never misses a beat. This lightened the entire Y carriage assembly a bit and also has less inertia mass on the rotor for the X axis. Overall a win win.


    Another mod I've done is to add spring loaded bed clips to hold glass which were printed in polycarbonate. The glass is just normal window glass I picked up from the local home improvement store. Hundreds of prints on this without issue. Just let it cool a little before pulling it off. I replaced the temp sensor on the bed with a semitec glass thermistor which is mounted about half way inside the aluminum plate with some really careful drilling through the PCB and into the bed. This has really helped.


    Another picture of that spring loaded bed clip. The springs I got from Harbor Freight. The heater PCB is held in place as normal and then the clip runs in a grove on top of the nut and then another nut and thin washer keeps the clip on. All held with blue thread lock so everything stays put. They will eventually melt if printed with ABS and using ABS bed temps, so I went with PC.


    The back clips are fixed in place and keep the glass from sliding around. All the screws are held with blue thread lock to keep them from coming loose.

    I"m currently working on better active cooling ducting. Why is everything Red ABS?, because I had rolls of it sitting around that I got from a screwed up order. It's not even good quality ABS but it did a great job for all this experimenting. I may go with blue, not sure yet, or just not worry about it and keep the red.

    I have a small web cam on a custom ball mount in the top left corner for watching prints remotely. As I print lots of ABS this thing spends much of it's time in the garage.

    I do normal prints now at 80-90mm/s and have had good luck up to 150mm/s with 2500 acceleration. No records here but it's sooo much faster and accurate then the stock machine.

    Comments always welcome,

    Jeff


  • administrators

    Hi Jeff

    Thanks for the write up! Do you have any pictures of how you mounted the DuetWifi?

    Cheers

    Tony



  • @T3P3Tony:

    Hi Jeff

    Thanks for the write up! Do you have any pictures of how you mounted the DuetWifi?

    Cheers

    Tony

    Haaa, that's a funny subject. I actually printed the bottom of a case off TV for the Duet WiFi in ABS which has 4 screw mount points. I carefully pulled the stubs from the bottom of the FFCP (brute force) that were in my way and used 3M super strong double sided tape (grey in color, red liner) to hold the mount in 5 spots. I then hot glued the entire thing where ever I could for added comfort. It's not going anywhere!!. Black tape to cover the holes from the studs. Looks OK.

    I just used the base of this http://www.thingiverse.com/thing:1802496

    Wiring wise I used all the correct crimping tools and connectors, my long time hobby is being an EE so I have the gadgets needed for this stuff. I was a bit too cheap to buy wire however when I have tubs of it floating around. It just needs to be safe, not pretty.

    Got my fan duct done after much issue and it appears to work ok. Now for ATX like power control of the 350W Meanwell PSU in the FFCP. Ohh the LED lighting I just connected the 3 cathods of the LED strip to negative and the one anode to +24V. This gives me a mostly white light. I miss the RGB lighting, however that could be added.

    Thanks,

    Jeff



  • Ok so I took a pic of the belly of the beast. Not all that pretty I know. The red box I just added to power the board and switch on and off the 350w supply. It's a 3A 5 volt switcher. I'm using a normal 5v relay with a flyback (snubber) diode to keep the spikes from killing the driver FET.

    Jeff


  • administrators

    Cheers! good to see the external 5V and control interface being used 😄



  • @T3P3Tony:

    Cheers! good to see the external 5V and control interface being used 😄

    You bet, I think it's an excellent feature. Little concerned the 5v supply drops out before the 24v supply cap does when the mains is turned off. Perhaps not an issue.

    I just put a 3.15A fuse inline where Flashforge had the little red jumper wire. I think they thought the mains wiring was just one big fusible link by the looks of things. I've seen far worse, but certainly they could have spared for a bit more copper or more PVC to make it look like they did. Ohh well she's safe now.

    Jeff



  • Update on the printer, I've changed all the red components to a black ABS. I like the all black look a bit better. Will have to post a photo.

    Now to the big step, how to convert the entire printer to a CoreXY setup. I think this can be done with the motors on the back of the printer.

    Jeff



  • @(In)Sanity:

    …........................Now to the big step, how to convert the entire printer to a CoreXY setup. I think this can be done with the motors on the back of the printer.

    Jeff

    TBH, If I was building my big coreXY again, it probably wouldn't be a coreXY - in fact, I may well change it to a simple Cartesian. In theory coreXY gives you more speed because you aren't lugging around one of the motors. In practice, having done a LOT of research on speeds and accelerations and suitable steppers, the big limiting factor on speed is how fast you can melt the filament. Even with something like a Volcano hot end, I wouldn't be able get anywhere near the speed that the machine is theoretically capable of. For sure, keeping the moving mass down is a good thing but is it worth the added complexity? I'm not convinced that it is.



  • @deckingman:

    @(In)Sanity:

    …........................Now to the big step, how to convert the entire printer to a CoreXY setup. I think this can be done with the motors on the back of the printer.

    Jeff

    TBH, If I was building my big coreXY again, it probably wouldn't be a coreXY - in fact, I may well change it to a simple Cartesian. In theory coreXY gives you more speed because you aren't lugging around one of the motors. In practice, having done a LOT of research on speeds and accelerations and suitable steppers, the big limiting factor on speed is how fast you can melt the filament. Even with something like a Volcano hot end, I wouldn't be able get anywhere near the speed that the machine is theoretically capable of. For sure, keeping the moving mass down is a good thing but is it worth the added complexity? I'm not convinced that it is.

    It's one of the reasons I went with the smallest motor I could get away with for the X-Axis as it took away that extra weight on the Y-Axis and it did produce noticeably better results. I just designed a 3 point LM8UU based carriage instead of the stock 4 point. I've found the 4 point to bind towards the ends due to the slight lack of precision in the gantry design. I'm not sure about stability of the 3 point yet, guess I'll print it out and see. I may just move away from the LM8UU bearings all together. I'm often running 70C inside the chamber of the printer printing ABS or PC, this is easy to do when my garage is 50C all on it's own in the summertime. Not sure what bearings will hold up best, the motors of course hate this.

    Jeff



  • got any tip for retract speed/amount with your bowden setup? I just got my E3DV6 and a titan extruder and I'm really considering just going bowden instead of direct. Printing a bowden mount for the titan extruder now so I guess I'm half way there allready. Got parts lying for both direct/bowden but lowering the weight really tempts me.



  • I get excellent results with 4-5 mm retraction (can use less) at 70mm/s and as fast of acceleration on the extruder motors as you can stand. I was running 9000 acl for a while to get the Snap from the retract…never had a single issue...it's just noisy I reduced it recently to quiet down the retracts. I also did Extra Restart distance of - 0.05mm which improves the seams between layers significantly.

    I'll not go back to direct drive anytime soon, I'm very happy with the bowden setup. Getting all that weight off the carriage helped soo much. Faster and more accurate.

    Jeff


  • administrators

    Its worth experimenting with pressure advance on your bowden setup as well, helps to reduce the effect o the elasticity in the bowden tube that is the main disadvantage of bowden.

    Edit: just saw the other thread where pressure advance is being discussed.



  • Printer is offline at the moment, I had to tear it all down as I was getting some sticking on the y-axis. Turns out believe it or not the factory plastic parts were/are placing enough force on the LM8UU bearings as to deform them, this made a nice grove in one of the Y-Axis shafts and was causing it to stick and sound like gravel. So I ordered some good quality hardened, chromed linear rods from VXB along with new LM8UU bearings. This will fix the bent gantry at the same time. Which as I now realize is bent in all directions just a tiny bit.

    Jeff



  • Crazy, but 3 of the 8 LM8UU bearings in my printer are completely shot. This is after degreasing them in an ultrasonic cleaner with Zep citrus and re-lubing. Even with grease I have 2 that bind up and another that are so rough it belongs in the trash. I would say I have 4 of 8 that are OK if I had to be objective. This is from a printer I thought was going to last a bit longer. Bottom line, if you think you have issues….check.

    Edit: Turns out I have LM8SUU (short) bearings in the printer. Now I'm waiting for those to come in so I can print all new parts to replace these stupid things with something normal.

    Jeff



  • What were you useing for grease ? I like to use molly diluted with 3 in one oil . Molly can make anything slippery . I use the oil mix to lube the smooth rods every 4 or so prints .
    May I ask who made the carriage ?
    I was also wondering why no heatsink on the processer and the mosfets ?


 

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