Posts made by Metallikahn

So, I'm working on getting my smart effector calibrated and I've been running into a few issues.

I'm currently running the 360mm Haydn magball arms and the standard v6 hotend (tried running ball cup arms before on a custom deisgned mount, but couldn't get it to work at all.) I've got a rostock max V2 maching and I've got a 309mm bed with plans to print out to about 305mm diameter.

Running the stock Bed.g file from the reprap configurator at 6 points and 3 inner points at 6 factors at 135mm radius and instantly got me down around 0.05-0.08 on the deviation. Good! I played around and increased it up to 10 outer, 6 inner, and 6 factors, and then filled in individual height corrections. Since doing so it seems like every probing gets progressively worse. Best I managed was a deviation around .22.

After eliminating a couple potential mechanical issues from the mix, things didn't improve. I went back to the stock bed file (10 spots at 135mm radius) and also applied 9 factor (just for the heck of it) and my deviation was .039! Ok... so maybe the outer bed is really screwing it up. I moved the radius out to 150mm and it calibrated at .023 on the deviation. Huh? I switched back to 6 factor, rebooted, (to clear out the 9 factor data)and it came out at .11. Moving back in to 135mm brought this back down to .088.

In the guide I know it said 6 factor was the best for a typical delta and that 7-9 factor was only useful if one could get outside the triangle of the towers. How far outside? At 150mm and 9 factor I seem to be getting good results. My diagonal rods are hand measured at 360.24mm. The 9 factor calibration reports back a diagonal of 372.020mm

Here are my current m665 after running my last g32.

1/13/2020, 3:27:18 PM	M666
Endstop adjustments X-1.88 Y-1.07 Z2.95, tilt X0.48% Y0.85%
1/13/2020, 3:26:26 PM	M665
Diagonals 372.020:372.020:372.020, delta radius 143.252, homed height 258.302, bed radius 137.0, X 0.659°, Y 0.060°, Z 0.000°
1/13/2020, 3:26:16 PM	G32
Calibrated 9 factors using 10 points, deviation before 0.122 after 0.023

Does this look right, given what I mentioned for spec's on my printer above? Or should I just call it good enough and print away?

I'm more than happy with .023 on the deviation (If that's accurate) as long as it gets me a nice first layer out to the edges of the bed for some of the larger prints I intend to do and stays dimensionally accurate doing so.

Thoughts?

Right on. Figured it'd be handy for a dual extrution, Just wanted to make sure I didn't have something set wrong there as well for regular single extrution set up.

Thanks a ton for the Quick responses, btw. Really helped me out!

Bingo! That was it. Not sure how I got those switched around on the board, but I did. I swapped the pins and then swapped the P0/P1 designations in the config.g and now it's working like it's suppose to.

Only Oddity I see now is that in web control it gives a separate fan control (tool fan) that appears whenever the extruder heater is engaged. As far as I can tell it's a copy of the print fan control, seeing as how they will mirror each other whenever either one is changed. Anyway to set it in the tool settings so it just identifies it as the "print fan" instead of creating a whole new useless fan control?

Hi all.

So I've been slowly getting my rostock back together and working with a duet2 wifi board and smart effector. I think I'm finally just about there, but I'm having issues getting the hotend fan to work correctly. As of right now, I have it wired correctly at the effector and running to the fan0 connection on the Duet2 WIFI. I also have it set to turn on at 100% when the hotend reaches 45C. This works exactly like it should. The problem starts when it goes to actually print. As soon as it finishes probing and begins to extrude filament the hotend fan will shutoff. Obviously this has let to more than one jam on me as I've been trying to figure this out.

I also don't believe it to be related to the slicer. I've used both simplify3d and cura. The only start Gcode I use in the slicers is literally:

G32
G29

At the moment I'm not running a print fan (long story, but basically I'm waiting on a berd airpump because I haven't been happy with anyof the other print fan solutions I've seen).

Here's my current config.g file. Am I just missing something super simple here or is it other issues? Any help would be great! Thanks.

     ; set Z probe trigger value, offset and trigger height
M557 R135 S40                                ; define mesh grid

; Heaters

M307 H0 B0                            	     ; enable bang-bang mode for the bed heater and set PWM limit
M305 P0 X201                                 ; configure PT100 for heater 1
M143 H0 S120                                 ; set temperature limit for heater 1 to 120C
M305 P1 X200                                 ; configure PT100 for heater 2
M143 H1 S310                                 ; set temperature limit for heater 2 to 310C

; Fans
M106 P0 C"Hotend Fan" S1 I0 F500 H2:2:1 T45   ; set fan 0 name, value, PWM signal inversion and frequency. Thermostatic control is turned on
M106 P1 C"Print Fan" S0.6 I0 F500 H-1        ; set fan 1 name, value, PWM signal inversion and frequency. Thermostatic control is turned off

; Tools
M563 P0 S"e3D V6" D0 H1 F0                   ; define tool 0
G10 P0 X0 Y0 Z0                              ; set tool 0 axis offsets
G10 P0 R0 S0                                 ; set initial tool 0 active and standby temperatures to 0C

; Custom settings are not defined

; Miscellaneous
M501                                         ; load saved parameters from non-volatile memory

Yep. It's more or less the thinned down machined version of the effector mount I printed. The Problem is the barbells are the same distance from the center of the hotend as what I've got. In order to get my full build plate without hitting the towers I need the barbells to be mounted the same distance from the hotend center as the mag balls would be mounted. With the way the smart effector is designed for the electrical connections, I would need to design a mount that puts the barbells far enough above to not interfer with the wiring, thus lowering the z height even further. If I mount the barbell underneath the effector then the arms hit the smart effector before they hit the tower.

The Magballs seem to be the best compromise between build area and height. I like the fact that the ballcups have the springs for backlash compensation and won't come unhooked during a print versus the mag ball arms. Would it be possible to get mag balls the same diameter as the balls on the Barbells and use the ballcup arms with the stock mounting location on the smart effector?

So, in my plan to upgrade, I decided to stay with the Rostock Max’s ballcup carriage/effector setup. One of my main goals in upgrading was to maximize my build area. I have a 310mm bed and with my old tricklaser/accelerometer setup I estimated I could build out to 304-305mm before the arms started clipping the towers/belts. (Though, my 290mm arms prevented me from getting there reliably)

I added longer, 340.5mm, carbonfiber arms made by SeeMeCNC and printed a smart effector ballcup adapter someone pointed out to me on thingiverse. The mount didn’t fit the newest version of the smart effector very well, so I modified it by adding some pillars. Not only did this allow the smart effector to fit, but I was able to raise it up enough to recover more Z height and contain the layer fans/wires nicely.

I mounted it up and discovered that due to the smart effector size and how the barbells have to matchup I actually lost build area! I’m now at around 290mm before the arms hit the towers.

So now I’m left scratching my head wondering what to do. If I go to mag arms I should recover enough to get back to my full build area, but at the risk of platform disconnects and a bigger loss of Z height (due to the longer arms and standard effector mount). I could design a new barbell mount, but it would need to mount above the effector to keep the maximum build area and it would need to be high enough to not interfere with the effector wiring, meaning even greater reduced z height, which is already bordering on the minimum height I’m willing to go to.

Are mag arms my only realistic way to increase build area with the smart effector, short of rebuilding the whole frame or sacrificing way to much z height?

Thanks for the response! Been hot out here the last few days so I haven't been getting much done on my upgrade. (Will probably be a few weeks before my custom bed heater arrives, so no big hurry, I guess).

The more I think about it, the more I'm leaning towards just running with the wires I have and running a couple new wires for the hot end heater. The wires I got are already the right size and run in separate bundles up the center of individual towers, making it easy to keep them isolated.

Should I run an additional 2 wires to turn my e3D PT-100 into a 4 wire connection? I saw how someone else accomplished this and it was basically putting a Y split on the 2 PT-100 wires. At that point it didn't look like it made a difference whether that split was done at the Smart effector or the daughterboard. Or is the Smart Effector essentialy handling the split in it it's circuit trace?

Should I leave it as a 2 wire connection all the way to the daughter board, would I need to solder/bridge the pads next to the PT-100 Connector on the smart effector that say "jumpered for 2 wire"?

At this point, It's not a lot of effort to run an extra 2 wires for the PT-100 with the 2 I already had to do that for. Is there really any appreciable benefit to doing so? (fwiw, the AC heating pad I have coming will have an integrated 4-wire pt-100 sensor all wired up and ready to go.)

Hi all! I'm currently in the process of Upgrading my Rostock Max V2 to a full duet, high precision Monster. I'm doing the Duet Wifi, Smart Effector, pt100 daughter board, 0.9 degree steppers in the towers, a 500 Watt AC Bed Heater (SSR controlled) and a full 24V upgrade for everything else.

I Just finished swapping out the stepper motors and pulling out the old PSU. Now I'm looking at cleaning up my wire routing and seeing what, if any, wires I have to add/replace/etc... A quick look at the wiki shows that I should need right at 12 separate wires to connect to and control all the features of the Smart Effector. I currently have 10 available. My initial thought is to just run a new long cable for the Heat Cartridge (I already have 2 separate cables run for the e3D PT-100 sensor). Since I'm upgrading to a 24v system, I purchased e3d's new 24v/40W Premium cable (Green! ). It comes with a nice long cable that is still about 2 meter's short. Is it Ok to just extend it with 18-20 gauge wires from the top of the printer? (I'm already basically doing that for a 12v/40w standard cartridge, so I'm not overly concerned about it)

Anyway, the fact that I have to run 2 new lines to meet the wiring requirements got me thinking about finally adding on an external wire routing channel up the back side of the Z tower. It Gives me the option for heavier gauge wire and easier upgrade-ability down the road (Should I decide to add a filament sensor or, god help me, dual extrution! )

My Questions are :

1.) Given that I'm running from the bottom of the printer, to the top, and need enough to reach almost back to bottom, how heavy should I go? Personally, I think It's better to go heavier whenever given the option and would be fine just running everything as 18-20 gauge stranded copper. My only reason for going lighter gauge would be due to the heavier gauge having a negative effect on something (like the probing or pt100 sensor, which I know can be a bit finicky to those kinds of things.)

2.) Should I be running all wires to the effector in a single bundle, or even in close proximity, all the way from board to board? Is there any chance of interference I should have to worry about?

or

3.) Should I just run the 2 extra for the heater and use what I've got? It's been running my current setup just fine, but it's not the prettiest thing ever when I pull a panel off and that severely bugs my OCD! lol

I'm not in any super big hurry to get it back together. I Ordered a Keenovo AC heater pad and probably have 2-3 weeks at the very minimum before that arrives. In the mean time I just want to get everything spliced together, tested and good to go so that the bed heat is no more than a 5 minute install, when it gets here.

@phaedrux said in Help with custom firmware settings for Rostock Max V2 upgrade:

@metallikahn said in Help with custom firmware settings for Rostock Max V2 upgrade:

throwing in a PT100 sensor for me

Don't forget that to use a PT100 you will need the daughter board.

Yep! Included in my first post. Just some of the many $ I've spent with Filastruder in the last couple weeks!

@dc42 said in Help with custom firmware settings for Rostock Max V2 upgrade:

You need different types of SSR depending on whether the bed heater is low voltage DC or mains voltage AC.

got one specc'ed exactly the same as SeeMeCNC sells for running their AC heaters.

@dc42 said in Help with custom firmware settings for Rostock Max V2 upgrade:

My delta isn't a Rostock, but it has a 300mm diameter printable area, 360mm rods, 24V power, and 0.9deg 1.68A motors. It can do 250mm/sec travel speeds. It can't quite manage 300mm/sec at the edges of the bed using my standard acceleration settings.

I'm shooting for between 300-304 (Over 304mm and the ballcups Juuuust touch the belts!) printable with SeeMeCNC's 340.5mm Carbon fiber arms. Can just manage 300 with the 290mm arms I currently have, but it's really hit or miss and the steppers are definitely slipping by that point. Definitely looks like 24V is the way to go. That'll give me a near identical hardware setup as yours in that particular area. As I mentioned, I rarely use a print speed over 50-60mm/s (even with a .6-.8 nozzle), but if it can handle that kind of speed it should really do a great job on the lower speeds for me, which is exactly what I'm after!

It's better than that. 0.9deg motors give you neatly twice the torque per unit angular error than 1.8deg motors, so you get better resolution. Whereas increasing microstepping from x16 to x32 typically doesn't improve resolution, because of the drop in incremental torque per mitostep.

I've been looking at doing the .9 upgrade as long as I've been eyeballing the Duet wifi upgrade! Always assumed the two should go hand in hand. If increased microstepping ends up being a wash on results, then x16 interpolating is definitely the way to go!

Don'r forget that you also need a DC-AC SSR to control an AC bed heater.

Got exactly that coming. Was going to use it to boost my Stock bed to 24v, but the AC heater makes a lot more sense and frees me up a whole 24v supply to run everything else.

Grabbed an e3D 24v 40W Precision heater (should pair well with my pt-100!) a 24v fan for the hot end, and figured... as long as I'm swapping steppers, I might as well go all out and drop in a set of Astrodyn Dampners while I'm at it. This might not be the greatest Rostock Max V2 ever, but It's still gonna be one helluva machine!

Still researching for a good silicone heater that will sit/mount flat between my aluminum spreader and Onyx build plate. Keenovo shot me a quote for $76 delivered on a custom designed (with precut screw holes) 305mm 400 watt AC heater that came with a preinstalled PT-100 sensor and centrally located wires. ( I'd have to cut and enlarge the Onyx hot bed's center hole, but at that point I'm just planning on using it as an insulator for the wood frame anyway.) Only issue is it might take upwards of 4-8 weeks to arrive. For and extra $30, I might be able to get it in 3-4 weeks. As cool and well priced as the custom design is, a 280mm disc would pobably do the same job just as well. I've got them running me a quote on an off the shelf 280mm plate and throwing in a PT100 sensor for me to self install. If it's not too bad, I'll probably jump on it.

@veti said in Help with custom firmware settings for Rostock Max V2 upgrade:

more steps will require more cpu resources. on a delta this could become more problematic at high speed.
If you want more accuracy, 0.9 degree steppers are the way to go.

Ok, now that you say that, I seem to remember reading that a year or two ago. Basically 16x interpolation with .9 steppers gets you the improved motion you might expect from a higher microstepping with 1.8 degree steppers and also saves on the Processing power.

It depends also on the weight of the effector and the acceleration settings. Easiest thing is to try and if you get layer shifts reduce speed or acceleration.
At 125mm/sec you could be limited by your hotend. Higher speed will certainly require something like a volcano hotend.

Yep. The speed concern was mainly for a safe setting in the firmware. Very rare that I use an actual print speed in excess of 60 mm/s. The hotend and smart effector will be the only weight hanging in the middle to start. Depending on how the Titan performs in the new setup, I might persue a remote drive extruder like the Nimble in the future... (so still not much of a factor.

Fans are also voltage specific. 240 watt without heated bed is plenty.
See
https://duet3d.dozuki.com/Wiki/Choosing_the_power_supply

Excellent! I missed that article while browsing earlier. Not too worried about the fans since e3d has lots of 24v fan options and I'll be building a custom layer fan anyway. Looks like if I just grab an AC Silicone Heater pad, a 24v heater cartidge and a couple of 24v fans I should be good to go!

Thanks for the help! I feel like I got a much better handle on how this is going to come together now.

@veti said in Help with custom firmware settings for Rostock Max V2 upgrade:

1. see https://duet3d.dozuki.com/Wiki/Choosing_and_connecting_stepper_motors

So the usual practice is to set the motor current to no more than about 85% of the rated current

2. The duet wifi only supports interpolation at x16. the motors will be a lot quieter with that. At the moment you have disable interpolation

3. Do a Test print and see how the first layer looks.

4. if you dont need ftp or telnet there is no reason to enable it.

I'd seen that document earlier, but it makes a lot more sense now. I've gone back and set it up for 16x (on) in the configurator. Figure 1.2A is a good starting point for my chosen motors. I downloaded the spreadsheet linked in the wiki page and ran the numbers for my steppers (which were pretty close to the defaults in the spreadsheet). I really don't care so much about the stepper noise since I have my printers in a fairly large shop area where it isn't a factor. I'm assuming 16x interpolation is better than just running it at 256x, or somewhere in between, for other reasons than just noise?

As far as stepper Max Speed goes, the spreadsheet shows me a worst case speed of around 125mm/sec (just about spot on with the V4's stock steppers) @ 12V. The V4 firmware shows a max speed of 250mm/sec. Is the general rule of thumb to just double it and plan to print slow near the edge of the plate?

I see I can gain over double the speed by going to 24v. What would you suggest for a Max speed in the firmware at 24v?

On the subject of 24v, I was planning on keeping it at just 12v to match my current PSU and hotend, but for a few $ I can easily get a 24v heater cartridge and it's not that spendy to just get a good AC Silicone heating pad for the heatbed. I've got a spare 240 watt 24v/10A psu that came stock with my Prusa Mk3. Assuming I run the heatbed off from the AC mains, would 240 watts be sufficient to run my steppers, extruder, 40 watt hotend, and duet3d/fans? I estimate at max Amps on all the motors and hotend I'd be drawing around 9A, leaving me 1A for the board/fans/etc. If I'm running the steppers at the 1.2A I currently have them set, it comes down to around 6.5 +board/fans/etc... Is that Enough acceptable headroom to be safe and sufficiently powered?

I'm already over budget on what I was going to do for an upgrade, but at this point, a little bit extra to go full 24v with an AC heatbed seems to be a much more practical choice (especially since I'll have the whole printer tore down for the switch over anyway). If I can use the PSU I have on hand, however, that would save me a few $ and make me feel a bit better about it.

Ok, I went through the online configurator again. This time I pulled data from the Rostock Max V4 config.g file (and a few spots in my current repetier firmware pertaining to endstops, extruder calibration, and stepper directions).

I've attached a copy of the config.g (wifi data removed)

So, my next round of questions.

1.) Should I be using the full rated amperage listed on the stepper motors or setting them to something lower? My 0.9 degree motors will be 1680 mA. The V4 (I'm pretty sure) has 2000 mA motors, but is running them at 1400mA. ( I also just went ahead and plugged in the values listed for the V4's speed and acceleration, since that's kind of starting point anyway.)

2.) The V4 is still using 16x microstepping on the motors. Is this just something most users eventually adjust up to 32x microstepping after getting everything properly calibrated or is there a specific reason for leaving it at 16x in the case of a delta?

3.) When it came to Bed probing and the homing boxes, I just left everything at default. Should I change any of these now, or just wait till I have everything working after a couple of basic calibrations?

4.) Network settings: Is there any practical reason why I would want to enable FTP or Telnet? I assume anything you would use those for is also handled through Duet Web control, correct?

Anyway, If someone familiar with the Rostock Max (and or anything similar to what I've got hardware wise) could give it a look over and steer me clear of any potential errors, I'd be might grateful! Thanks

0_1561445313727_config.g

Hey all! So... I finally bit the bullet and decided to upgrade from my Rambo and make my Rostock all it can be.

So, to start, My V2 is already kind of a Frankenstein build. After this upgrade the only think stock on it will be the frame and belts! lol

I should have most of my parts by this weekend and thought I'd get started on setting up the firmware ahead of time (as best as I am able without parts in hand, anyway). At the end of the weekend my printer will have the following parts installed.

Starting from the ground up:

Duet 2 Wifi
0.9 Degree Stepper motors (Used on V3.2)
700 watt PC grade PSU supplying 12 V
SSR supplying 24V from a second Powersupply for Bed Heat
GeckoTek Aluminum Heatspreader and Magnetic Flex plate build surface
SeeMeCNC Injected Carriages
SeeMeCNC Carbon Fiber Ballcup Arms (340.5mm)
Duet Delta Smart Effector (adapters used to fit the SeeMeCNC Barbells)
e3D Titan Extruder (using stock Wantai 42BYGHW811P2 Stepper Motor... Calibrated to 405.99 steps/mm)
Capricorn 1.8mm bowden tubing
e3D v6 Hotend with MicroSwiss Heatbreak, e3D Copper heat block, 40w 12v heater.
e3D PT-100 and Duet PT-100 Daughterboard

That's about it. geneb, over on the SeeMeCNC forums suggested using the Rostock Max V4 firmware as a starting point. I can see why, as much of it will apply to my build, with the main differences being related to the Extruder, Hotend/Pt-100, and Smart Effector.

I guess my main questions are:

1.) Should I still use the Reprap Online Configurator and just transpose values and new settings over via that or just make the few changes I need manually to the V4's firmware? Any advice on specific changes to pay attention to?

2.) Is the Config .g file the only one I need to modify for my build?

I know I'll have further questions, but those will probably be trouble shooting and tweaking related. Just hoping to get a basic starting point down that won't completely break the machine the second I try it out.

Thanks all!