Maurici´s Switch(ish)ware vs.573

[Maurici]

So... I´ve been super entertained during the 3 past days converting one of my smaller Cartesians (formerly an Ender 3 S1) to switchwire...

After finishing my build on the V0.1 decided that one voron wasn´t enough but I didn´t really want to replace my finely tuned cartesians with a V2... next best fiddling option was install the switchwire motion system to one of the Cartesians.

This has been rather budgetly (knock off omron sensor, a few screws and bolts of missing sizes from my kits, 2 meters of belt and a few idlers... all under 70 quid) redesigned the X carrier to reuse my extruder, and the machine was already fully linear railed, so it was all really cheap, quick and easy to do.

Kept the original unorganised wiring as I didn´t wanted to redo harnesses and printed with leftovers of the purple used for the V0...

Sorry guys no LED lights in this one.

Oh. and in record time. 4 hours form the first bolt stripped to the first successfully leveled bed and first print!

anyway. being that simplistic approach... I´m guessing isn´t realistic to request a serial, isn´t it?

Shame she will never be for real, but she will be for me.

Find a few pictures!

 

Edited March 7, 2023 by Maurici

[mvdveer]

3 hours ago, Maurici said:

she will never be for real,

She is for real - printing is it not? 

[smirk]

9 hours ago, Maurici said:

Oh. and in record time. 4 hours form the first bolt stripped to the first successfully leveled bed and first print!

Wow! That's impressive. You'll perhaps start a new craze. Forget printing a benchy in sub-20 seconds - it's the the speed-wire challenge. I'm definitely not entering that

 

I have to admit I do like the Core-XZ mechanics and consider it for my cartesian printers.

[Maurici]

12 minutes ago, smirk said:

it's the the speed-wire challenge. I'm definitely not entering that

well. again, the whole printer was already running an SKR e3 V3, Klipper, and linear rails... and no harness has been adapted/cut to size , but just "tucked away". Sounds pretty impressive, but taking in account that I only assembled 2 motor mounts 4 idlers assys and an X carriage I now think its actually been WAAAAAAY too long! Also helps that the massive Voron comunity had already a printer.cfg file that needed minimal debugging, and that I´m already pretty familiar with klipper.

 

I´m now kind of thinking that I should tidy up the wiring and request the SN. Bollocks.

Edited January 16, 2023 by Maurici

[smirk]

7 hours ago, Maurici said:

I´m now kind of thinking that I should tidy up the wiring and request the SN. Bollocks.

Sounds like a plan - no reason why it shouldn't have a serial #. Then you'll be on the slippery slope to another Voron conversion

[mvdveer]

9 hours ago, Maurici said:

I´m now kind of thinking that I should tidy up the wiring and request the SN. Bollocks.

Do it..

[claudermilk]

18 hours ago, mvdveer said:

Do it..

FTFY. 

[Maurici]

a bit more of fiddling today (nope, sorry, not the wiring nor serial yet).

Following my budgetish style conversion I did cut yet another corner.

I used to previously have a posh  bed mains powered with a glass surface that was perfect for engineering materials, but... if it was going to be a voron, had to be capable of going much faster therefore glass and mains had to go to reduce weight (double intention as I needed steel for the inductive probe) and recycled one of my old Ender 3 PCB beds.

I was shocked this morning when I´ve set up the 4x4 auto mesh, and I´ve seen the deviations... OMG I can´t really believe that this previous 3 days this thing has been printing like this and actually managed to get anything stuck in the PEI... this textured sheets are friking impressive. trully. The range was like 0.25. Tried another bed I had, and it was 0.3. No joy either

I have a massive OCD with bed health and leveling... In my biggest slinger I have a 400x400 glass bed that will consistently show +/-0.08 in a 9x9 grid.

 

So... pulled the old trick book and started shimming. I´ve not seen doing this to many people... (I´ve not found much bed leveling information around either, that probably means I give it WAAAAAY too much relevance).

First step. Crack open a Porter (sweet if possible) and if you happen to smoke, be sure you have plenty of supplies on hand. It will be boring and long.

Step two. Fire up spotify. Put at least 3 hours of your favourite relaxing music.

Step 3. take a look at your bed. It looks flat, but you know it isnt.

 

Step 4 Take an strip of aluminium/foil tape and partially stick it in the frame, and start placing VERY SMALL portions of this strip on the low points between the magnetic base and the spring plate. ONE BY ONE. That is, put a shim, test, redo the test and yada yada. every shim will affect the springiness of the whole bed, this is why it needs to be tested one by one. In my case, I can count 11 strips, plus at least 7 I´ve removed/replaced. That means like 20 ABL processes... 

 

Step 5... warm it up to printing temperature, let it soak for 10 minutes, and recheck.

Step 6... Enjoy the results (note that I´m not specially happy with the range for such an small bed 220x220, but I´m guessing this is what I´m stuck with. I´ve to say that works phenomenal).

 

 

 

 

Edited January 17, 2023 by Maurici

[Maurici]

and a bit more fiddling.

To Start with, seems that the mosphet of my part cooling fan in the SKR is blown. I can´t tell by sure if it was due the switchwire conversion, or previous to this...

Either way, swapped the board from another of my cartesians, and started working again. Time to process an amazon warranty... Nevermind.

also... as I´ve seen many complaining, the keyback carabineer thingy holding the gantry when switched off (the yellow bit), is not strong enough (despite being the 20 OZ one... or so it says) and the gantry drops anyway, Not sure if there is different chinesium suppliers or something, but mines (and many other people around the forums) is too weak.

Lukily, I bought two as it was cheaper a pack of two than only one, so decieded to put one on each side. 

Mirrored the STLs and printed them at coarse quality with the V0 (the blue bits)

 

 

I have not payed too much attention to the print quality, and I´ve "risked" printing it a 0.3

Not being a precision critical assembly and the switchwire having to go back to work ASAP, I really wasn´t in the mood for long prints, so tested the boundaries of my V0... Cranked the speed feed in klipper by 250% (basically printing to my firmware limits). managed to cut the 52 minutes "expected" to just under 21. Lovely. I´m still astonished on how fast this small thing can print ABS. Throw a 55minutes on a reasonably fast setting already, crank it up in klipper, go to have lunch. come back to a finished print 30 minutes later. I´m still wondering why did I wait till Stephan´s video to pull the trigger...

The result... well, it seems to work just fine... as it holds the gantry now in a sort of non-gravity position!

 

[Maurici]

Well.

I´ve just demystified the whole Serial request thing...

Say welcome to the VS.573

(and it was the dodgiest submission EVER)

[smirk]

On 2/7/2023 at 10:09 PM, Maurici said:

Say welcome to the VS.573

Welcome! Just remember they grow up fast!

[Maurici]

Well. I don´t seem to be able to leave my printers alone. I´ve finished the Micron+ and one thing only leds to another.

I now want Klicky in my switchwire.

Reasons: The omron inductive... while is not causing major problems has massive deviations, and the difference on readings depending on the chamber temperature is feeding me up.

Also, I´ve worn out pretty quickly my PEI sheet while printing Policarbonate... seems that filament at 340degrees is too hot... so I would like to be able to go back to glass/mirror with PV glue for polycarbonate and PETG.

The inductive probe won´t work with glass...

Also, I´m impressed with the precision of the klicky and how cool the whole dock/undock gimmik looks like. Even better in the switchwire as it needs to work with a servo triggering the dock into a docking position and so.... how cool is this. And Right now I´m already a master of klicky, managed to get the one in the micron working at 2am with plenty of hops in my system...

I´ve just ordered a bunch of switches and two servos from chinesia express. I have plenty of magnets and inserts to do the rest of the thing on my onw. And triggered the whole print in the Micron. First time testing the real boundaries of the thing. I´ve put an "speed benchy" profie, with the right ammount of perimeters, and if it wasn´t because I forgot to re-set the retraction, seems to be printing at mach1 with a pretty decent quality.

Soo cool the micron´s chamber. Seems to be the right combination of volume and heating elements and the temperature is peaking at 65degrees so I expect maximum strenght despite the speed.

 

 

Yeah, Just showing off. something I´ve built goes super quick... not even the macintosh shaped printer of trends prints that quick! how cool am I?

ok, ok. next time I´ll remember to re-set the retraction...

I feel I´ll spend more time cleaning the blobs than letting her print at a normal speed...

 

To the serious bussines tho. Someone here with experience with Servo triggered probes?

It seems to be plenty of instructions in the mod github but I would appreciate first hand experience.

https://github.com/jlas1/Klicky-Probe/tree/main/Printers/Voron/Switchwire

[smirk]

I built that addition for a switchwire convertion thingy that I'm playing with. Really impressed with it. Not really too many snags, mostly just worked. Biggest issue I had was the stock pile of servos that I had were all jiggered for 360 rotation. I had to source some new ones that were "limited" to the regular 180degrees of movement. There's a wee bit of "tuning" where you figure out which servo angle corresponds to deployed/stowed but that was not hugely difficult.

I have not experiemented (without it) but I did put in an additional buck convertor to provide a 5v feed for the servo rather than trying to take the feed from the MCU. I read (think it was actually in the klicky github) the the MCU might not be able to cope with the load (required by the servo). I simply used a separate buck convertor as I did not have a spare 5v into the PI that I could piggy back on. I used the filament detect port on my SKR pico to control the servo.

[Maurici]

 

4 minutes ago, smirk said:

I built that addition for a switchwire convertion thingy that I'm playing with

 

 

Brill. I now have someone to ask for advice.

 

PS.

Parts cleaned out just fine rubbing with the finger. One wonders if retraction is even worth it then... XD.