Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
BattleMaster
Aug 14, 2000

I went and looked up if someone else had identified the lead screw and it does have 2 starts for an effective pitch of 4mm. I guess I wasn't examining it as closely as I thought :negative:

Adbot
ADBOT LOVES YOU

Ambrose Burnside
Aug 30, 2007

pensive

Some Goon posted:

Is there a goon-recommended starter (~$300) CNC machine? Google is recommending a Chinese 3018, but the review sites themselves don't inspire much confidence. I'm mostly looking to fabricate enclosures for electronics projects and trying to decide between a CNC machine or 3d printer, but leaning towards CNC.

Assuming you want a fully-fledged cnc mill that can handle engineering materials/metals- you need a higher budget, significantly higher when you remember that tooling and fixtures will tack a couple hundred at minimum to the mill price. I wouldn't say you need to spend Tormach-tier money, there's always Taig and Sherline machines, but we're still talking "one to several thousand dollars" all in. I have a CNCed Taig micro mill and I think it's run me about $3k canadian since I first picked it up.

That said, I wouldn't buy a mill -or- a 3d printer for making electronics enclosures, I would get a laser cutter. They're the best pick for doing "2D" work like this- instead of producing the enclosure as a single monolithic part, you break it up into panels, cut those out, and assemble. Reasons you should do this:
- you can produce parts very quickly
- most machines have generous X/Y working areas compared to similarly-priced 3d printers/cnc mills
- They do accurate and precise work
- it's a non-contact process so you barely need to think about workholding, gravity is usually sufficient. and tooling? what's tooling?
- their feedstock- wood/plastic sheet or board stock- is absurdly cheap compared to 3d printer filament and you can get it from any hardware store
- things you make have excellent mechanical properties compared to FDM printed parts
- a laser cutter will also double as an engraver, letting you burn raster images or vectory tracery into parts in the same program set up for cutting

downsides are that it's pricier than an entry-level 3d printer- although still much cheaper than a cnc mill, i would budget $500-1000 for an entry-level laser machine- and that the nature of CO2 lasers means you can't work with metal or specific engineering materials that offgas toxic poo poo when burnt. also it's a noxious/hazardous process compared to 3d printing, albeit still perfectly safe if used responsibly- you can't run them unattended because of the fire risk and you'll need to put serious thought into either filtering the laser fumes or blowing them outside.

Fantastic Foreskin
Jan 6, 2013

A golden helix streaked skyward from the Helvault. A thunderous explosion shattered the silver monolith and Avacyn emerged, free from her prison at last.

Really all I was thinking was wood. I'm not really looking to do anything you couldn't bang out in a garage without much hassle, but a. I don't have a garage, b. I don't have tools, and c. letting a computer handle the measuring and cutting is very appealing, especially considering the dimensions some panel-mount components want to work with. A 3D printer would be ideal for these kinds of things but I'd rather avoid the waste if possible.

Its a bit moot though, doing some reorganizing and I don't have the space for either right now. I'll take a look into laser cutters when I get the space though.

JEEVES420
Feb 16, 2005

The world is a mess... and I just need to rule it

Ambrose Burnside posted:

Assuming you want a fully-fledged cnc mill that can handle engineering materials/metals- you need a higher budget, significantly higher when you remember that tooling and fixtures will tack a couple hundred at minimum to the mill price. I wouldn't say you need to spend Tormach-tier money, there's always Taig and Sherline machines, but we're still talking "one to several thousand dollars" all in. I have a CNCed Taig micro mill and I think it's run me about $3k canadian since I first picked it up.

That said, I wouldn't buy a mill -or- a 3d printer for making electronics enclosures, I would get a laser cutter. They're the best pick for doing "2D" work like this- instead of producing the enclosure as a single monolithic part, you break it up into panels, cut those out, and assemble. Reasons you should do this:
- you can produce parts very quickly
- most machines have generous X/Y working areas compared to similarly-priced 3d printers/cnc mills
- They do accurate and precise work
- it's a non-contact process so you barely need to think about workholding, gravity is usually sufficient. and tooling? what's tooling?
- their feedstock- wood/plastic sheet or board stock- is absurdly cheap compared to 3d printer filament and you can get it from any hardware store
- things you make have excellent mechanical properties compared to FDM printed parts
- a laser cutter will also double as an engraver, letting you burn raster images or vectory tracery into parts in the same program set up for cutting

downsides are that it's pricier than an entry-level 3d printer- although still much cheaper than a cnc mill, i would budget $500-1000 for an entry-level laser machine- and that the nature of CO2 lasers means you can't work with metal or specific engineering materials that offgas toxic poo poo when burnt. also it's a noxious/hazardous process compared to 3d printing, albeit still perfectly safe if used responsibly- you can't run them unattended because of the fire risk and you'll need to put serious thought into either filtering the laser fumes or blowing them outside.

$500-$1000 gets you roughly the same working area as a 3018 and a 40w laser tube. Instead of tooling you need lenses and CO2. You're limited on what materials (no metals at all at that price) and thickness you can work with and have to assemble flat cut parts. You will have to think about glue/screw/tabs, how the parts can fit together, and durability.

Lasers have their place, and is next on my purchase list, but I really don't see it compared to a CNC/3d printer in the type of work it can do.

NewFatMike
Jun 11, 2015

Ambrose Burnside posted:

Assuming you want a fully-fledged cnc mill that can handle engineering materials/metals- you need a higher budget, significantly higher when you remember that tooling and fixtures will tack a couple hundred at minimum to the mill price. I wouldn't say you need to spend Tormach-tier money, there's always Taig and Sherline machines, but we're still talking "one to several thousand dollars" all in. I have a CNCed Taig micro mill and I think it's run me about $3k canadian since I first picked it up.

That said, I wouldn't buy a mill -or- a 3d printer for making electronics enclosures, I would get a laser cutter. They're the best pick for doing "2D" work like this- instead of producing the enclosure as a single monolithic part, you break it up into panels, cut those out, and assemble. Reasons you should do this:
- you can produce parts very quickly
- most machines have generous X/Y working areas compared to similarly-priced 3d printers/cnc mills
- They do accurate and precise work
- it's a non-contact process so you barely need to think about workholding, gravity is usually sufficient. and tooling? what's tooling?
- their feedstock- wood/plastic sheet or board stock- is absurdly cheap compared to 3d printer filament and you can get it from any hardware store
- things you make have excellent mechanical properties compared to FDM printed parts
- a laser cutter will also double as an engraver, letting you burn raster images or vectory tracery into parts in the same program set up for cutting

downsides are that it's pricier than an entry-level 3d printer- although still much cheaper than a cnc mill, i would budget $500-1000 for an entry-level laser machine- and that the nature of CO2 lasers means you can't work with metal or specific engineering materials that offgas toxic poo poo when burnt. also it's a noxious/hazardous process compared to 3d printing, albeit still perfectly safe if used responsibly- you can't run them unattended because of the fire risk and you'll need to put serious thought into either filtering the laser fumes or blowing them outside.

What lasers are you thinking around that price range? Are K40s no longer expensive firestarters?

Ambihelical Hexnut
Aug 5, 2008
K40s are great if you know what you're getting into.

There's a freeware program called K40 whisperer that eliminates the need to use the wacky usb dongle and cracked chinese coreldraw with most of the modern K40 controllers. Have to watch out as there are narrower K40s which look very similar but have shorter tubes; a standard "40 watt" K40 is actually like 32-35w, and these new ones are slightly less (26-28 I think). I bought one of these after destroying my DSP K40 and it works great despite being a bit less powerful and arrived with way better build quality than my original 2014ish one.

NewFatMike
Jun 11, 2015

If y'all have some models to link, I'd love to take a look. I realized I could plot directly from SOLIDWORKS Drawings to Universal lasers and I'd love to catch that high on a small home machine.

Ambihelical Hexnut
Aug 5, 2008
If you want to do it on the cheap look up the list of supported controllers for K40 Whisperer and ensure the item description says that. There are not a lot of standards for these. Every K40 is a project.

Here’s the new (narrower) one I bought:

https://www.ebay.com/itm/Preenex-12...4.m46890.l49292


Here’s an example of one that APPEARS TO have the classic more powerful tube (wider form factor, look at the electronics panel door on the right).

https://www.ebay.com/itm/40W-Co2-Digital-Control-12-x-8-K40-Wood-Laser-Engraver-Cutter/402270344893?hash=item5da92e56bd:g:aTsAAOSwzH9f1r1v

NewFatMike
Jun 11, 2015

There was a brief, beautiful moment where I thought the price was $420.69 instead of $420.62 :negative:

Thanks very much! I imagine things get less hinky around the $1k mark?

Ambrose Burnside
Aug 30, 2007

pensive

JEEVES420 posted:

$500-$1000 gets you roughly the same working area as a 3018 and a 40w laser tube. Instead of tooling you need lenses and CO2.
Granted, but those consumables aren't nearly as expensive, proportionally-speaking, as mill tooling will be. You're right that a K40 isn't particularly big, I was more thinking of scratch-builds, because the belt-drive aluminum extrusion design + the lack of any cutting forces can be scaled up quite a bit without incurring much additional cost.

quote:

You're limited on what materials (no metals at all at that price) and thickness you can work with and have to assemble flat cut parts. You will have to think about glue/screw/tabs, how the parts can fit together, and durability.

Lasers have their place, and is next on my purchase list, but I really don't see it compared to a CNC/3d printer in the type of work it can do.

The work laser machines are suited for is different, sure, but I'd say they're better-suited to fabricating most enclosures than 3d printing is- boxes are an ideal use-case for "assemble it from slices" unless you have very particular geometry needs, and the bigger the enclosure gets the less sense it makes to print it. I don't see the joinery as a big downside unless you're absolutely disinterested in assembling anything- you can automate it and use sth like an illustrator/inkscape parametric finger-joint plugin to automatically generate the edge joinery, and you can build very rugged small enclosures from glued-together slices with no joinery at all; more sophisticated approaches like snap-together glueless designs, 'live' fasteners cut into the panel stock, knock-down designs using nuts and bolts, etc offer their own advantages if you wanna do the design work.


All that being said, I'm approaching this from the perspective of someone who designs and builds prototypes professionally, I have lovely perspective into how a hobbyist would approach this. A properly-made laser-cut enclosure will generally be stronger, more durable, have a better surface finish w no porosity, be cheaper, and be dramatically faster to produce, so that's what I reach for for this sort of thing wherever possible, 3D printing is a last resort for if you really can't swing a laser-cut enclosure, and even then I'd usually try to laser-cut as much of the enclosure as possible and only send the components with complex geometries to get printed. Hobbyists probably won't weight Effective Enclosure Fabrication nearly as heavily as, say, ease of use or a gentle learning curve, or just getting something that *works* directly from the printer bed.


So I guess it depends on how you wanna make use of the machine, and what your resources are. if you want a simple one-stop shop solution to enclosures, don't wanna get bogged down in enclosure design, are making small enclosures with complex geometries, or wanna also make a wide range of things that aren't suited to 2D slice assemblies- you're right, laser cutting doesn't make much sense. If you're integrating it into a broader workshop setup, are fine with designing to suit the Laser Cutting Design Paradigm, want to produce large and/or batches of enclosures, or want a more finished end-result for stuff you're selling, I would go with the laser cutter.
From what Some Goon has said, a 3d printer does sound like the right pick for his needs.

Ambrose Burnside fucked around with this message at 19:18 on Jan 16, 2021

Aurium
Oct 10, 2010

JEEVES420 posted:

$500-$1000 gets you roughly the same working area as a 3018 and a 40w laser tube. Instead of tooling you need lenses and CO2. You're limited on what materials (no metals at all at that price) and thickness you can work with and have to assemble flat cut parts. You will have to think about glue/screw/tabs, how the parts can fit together, and durability.

Lasers have their place, and is next on my purchase list, but I really don't see it compared to a CNC/3d printer in the type of work it can do.

For a 40w laser you're not buying multiple lenses. The reason to have more than one is to have a longer focal length to cut though thicker material, and a 40w laser doesn't have the punch to cut though thick stuff in the first place.

You're also not buying CO2 at this price point. The tube is sealed, and is the consumable with a life of around 1500-2500 hours.


NewFatMike posted:

What lasers are you thinking around that price range? Are K40s no longer expensive firestarters?

Good news! all lasers are expensive firestarters. And no, K40s are as bad as they've ever been. They're still fantastic value for money. Get one with an analog meter and a knob, not one with a 7 segment panel and +/- buttons. Yes, this sounds backwards. Yes the digital panel is billed as an upgrade. It is not. The digital panel is also far more common.

A random example. https://www.ebay.com/itm/313082039289

The digital panel ones just set percentage of a pwm output which then sets current to the tube. The problem here is that at 100% pwm is far more current than the tubes can handle, and with no meter you have no real idea what you're actually sending. It's far too easy to burn up a tube with them.

The next level up is the 50w cutters with a "dsp" controller which start at about 1500. They usually have a better build quality, come with an air assist by default, have an adjustable z, and can do combined etch and cut jobs at once as they can automatically control power level. They still don't actually link power level% with laser output percentage, and don't have a current meter, again making it rather easy to burn up a tube.

Ambihelical Hexnut
Aug 5, 2008

NewFatMike posted:

There was a brief, beautiful moment where I thought the price was $420.69 instead of $420.62 :negative:

Thanks very much! I imagine things get less hinky around the $1k mark?

The next range of Chinese low end lasers is usually in the ~1300 to 2500 range with 60-80w and 600x400 work area. Things are pretty static in the low end co2 laser market and those are basically your two options for cheap house fires. If you go that way try to find one that’s $1337.

NewFatMike
Jun 11, 2015

Thanks friends! I've never really looked into home lasers, and the 50W ones look a little more my speed if I feel the need to grab one for home.

My first home shop CNC pickup will hopefully be a Taig lathe - that'll do nicely with the Tormach and ShapeOkos at the makerspace.

App13
Dec 31, 2011

Anyone have opinions on the glowforge? Significantly more expensive I think at around $5k

ante
Apr 9, 2005

SUNSHINE AND RAINBOWS
Slick marketing, not sure they actually live up to the price premium?

I dunno, people got really excited about them for preorders, and then everything was silent once they got them.

sharkytm
Oct 9, 2003

Ba

By

Sharkytm doot doo do doot do doo


Fallen Rib
Our maker space has one. It "just works".

eddiewalker
Apr 28, 2004

Arrrr ye landlubber
I was reading that the glowforge requires a constant internet connection to work. Is that concerning? What happens if the company does under?

sharkytm
Oct 9, 2003

Ba

By

Sharkytm doot doo do doot do doo


Fallen Rib
This is exactly the problem. Thus far, they haven't.

NewFatMike
Jun 11, 2015

I think I've asked before, but does anyone have any preferred shuttles or pendants? I think the Contour Design one is rather popular, but I'd love to have one I didn't have to label

https://www.bhphotovideo.com/c/product/697104-REG/Contour_SXPRB_Shuttle_Xpress_NLE_Multimedia_Controller.html

NomNomNom
Jul 20, 2008
Please Work Out
Is the lowrider cnc going to be a massive disappointment? I'm looking to cut thin mdf and plywood for templates, maybe some engraving. All woodworking applications. I have a worktable I'd like to use, so I'm considering going the dual end stop route so I can have the machine self square when deployed and be easily removed from the table for storage.

mekilljoydammit
Jan 28, 2016

Me have motors that scream to 10,000rpm. Me have more cars than Pick and Pull
I've got one that's still in the tuning stage (my dad and I have a lot of other stuff going on) and tbh within its capabilities it seems pretty slick. Fastest possible? No. But for area per buck it's awesome.

cakesmith handyman
Jul 22, 2007

Pip-Pip old chap! Last one in is a rotten egg what what.

I read a lot of blogs and reviews of the mpcnc and lowrider this weekend actually, I specifically looked for bad reviews and people who couldn't get it working and scrapped it, there's loads and most seem to be problems with the electronics or not understanding basic fabricating so I'm confident if I had the space I could get one working. Plenty of reviews from people running small businesses who can't get rid of this homemade junkpile because it makes them money nonstop too.

mekilljoydammit
Jan 28, 2016

Me have motors that scream to 10,000rpm. Me have more cars than Pick and Pull
Yeah, they intend on using Marlin boards and that annoys me when GRBL is sorta better suited, but that's kind of a "whatever" thing. Oh, I remember one thing that I came across that I don't like - the Z droops when power isn't on or when power is on but the steppers aren't enabled yet. I swapped the Z steppers with one with brakes, and will eventually put them to a relay controlled from the controller board - stepper drivers enabled? Then turn off the brakes. But I was chasing a weird Z issue before that.

I'd recommend getting the side plates CNC routed or something, IIRC they just sell them and doing them by hand is possible but will be another issue to chase. Also the Z axis movement needs a little fiddling to move freely. None of these are big deals considering the absolutely unrivaled bang per buck IMO.

NomNomNom
Jul 20, 2008
Please Work Out
Yeah the z droop seems like an annoyance for sure, especially since I envision this thing living at the end of my assembly/outfeed table. Could I swap a stepper with brake for the included steppers from the kit? I haven't committed to anything yet since the kit is currently out of stock.

Good to know about the side pieces, I was going to print the templates and cut those myself.

Karia
Mar 27, 2013

Self-portrait, Snake on a Plane
Oil painting, c. 1482-1484
Leonardo DaVinci (1452-1591)

Figured this might be a good place to ask. Does anyone have any recommendations for a good dial-based tool offset gage (like this one)? I see a bunch of Amazon options, but it's hard to know how much to trust any of them. The goal is to train students with them (don't want to start out with the probes so they can get an idea what the offset numbers mean), so low-cost would be good since I guarantee they'll smash them eventually.

mekilljoydammit
Jan 28, 2016

Me have motors that scream to 10,000rpm. Me have more cars than Pick and Pull

NomNomNom posted:

Yeah the z droop seems like an annoyance for sure, especially since I envision this thing living at the end of my assembly/outfeed table. Could I swap a stepper with brake for the included steppers from the kit? I haven't committed to anything yet since the kit is currently out of stock.

Good to know about the side pieces, I was going to print the templates and cut those myself.

Yup, 100% easy bolt-on for the Z steppers. It's sort of... hm. I was seeing "oh just don't turn it off" from a quick google of the solutions and I just... thought that was a ridiculous solution and figured out what seemed to me a better one almost instantly. Like why not just fix the problem?

My lowrider has a few little issues I haven't troubleshot because it's really my dad's lowrider and both of us have had other stuff going on... like I think the printed parts that hold the Z steppers should really be flipped 180 degrees and there's some stuff binding because they're not. Or maybe my dad assembled it backwards.

It's perfectly possible to get a working router by printing the templates and cutting them yourself... what happened when my dad did that is that the nut for the Z axis leadscrew was slightly misaligned so that close to the table, it would bind up. Temporary solution was just space everything up about an inch to where stuff didn't bind and then CNC cut accurate side pieces using the lowrider itself.

All told if I put in probably a solid workday I bet I could have it working right; I full well admit that I just haven't done that.

Ambrose Burnside
Aug 30, 2007

pensive

mekilljoydammit posted:

All told if I put in probably a solid workday I bet I could have it working right; I full well admit that I just haven't done that.

i think im going to write this on a plaque and put it by my workbench

mekilljoydammit
Jan 28, 2016

Me have motors that scream to 10,000rpm. Me have more cars than Pick and Pull

Ambrose Burnside posted:

i think im going to write this on a plaque and put it by my workbench

Yeah - I'm getting to a point where I'm learning how to prioritize projects by them being gating items on other stuff, rather than in order of coolness.

BattleMaster
Aug 14, 2000

I'm building a new Z-axis for my 3018 router to turn it into a 4-axis drag knife cutter. I am using a NEMA 17 stepper as the tool rotational axis and I bought an ER11 collet chuck for a 5mm shaft. This matches up with both the specs and my measurement of the NEMA 17's shaft. However, the chuck just doesn't fit onto the shaft. It is barely too small and won't go on even with force. What should I do? Was this the wrong part for the job or is it just bad and has poor tolerance?

BattleMaster fucked around with this message at 00:06 on Jan 31, 2021

Sagebrush
Feb 26, 2012

Do you have a micrometer? Measure the shafts to find out what you're working with. You say the measurements check out, but like how close are they to a hundredth of a millimeter? It's likely that the collet chuck is bored to the exact figure so that you have less runout to worry about, but that means you don't have any slop to work with either.

If the values you measure are within say .003 (.08mm) of each other, that's probably enough for a shrink fit. Put the motor in your freezer and the chuck in your oven at 400 degrees and assemble them immediately after taking them out.

Also, it is possible that there is a burr on either the shaft or the bore. Take some fine crocus cloth and polish the edges of both, see if that helps.

BattleMaster
Aug 14, 2000

I don't have anything more accurate to measure it with than a pair of cheap digital calipers and a ruler with metric holes in it, unfortunately, so if either is off by a tiny amount I would have the ability to gauge it. I was hoping that because the chuck has set screws in it that it wouldn't be the kind of thing intended for a permanent fit - I was hoping it would be the sort of thing I could remove if I wanted to use the motor or chuck for something else. Drag knifing isn't something I'd expect to need incredibly high performance from the chuck for so I don't mind if it's just held in by set screws.

edit: Maybe I should get a small piece of 5mm steel rod and do the shrink fit with that, and use a 5mm to 5mm coupler to attach it to the motor?

BattleMaster fucked around with this message at 00:34 on Jan 31, 2021

Sagebrush
Feb 26, 2012

If that's the case, I would try polishing the edge of the bore as I suggested, and if it still doesn't work, buy a 5mm drill bit and go to town on it. You're correct that a drag knife won't care (much) if the tool is a fraction of a millimeter off center, so if your drill work is less than perfect it won't matter.

You could also get a 5mm reamer, or like a 5.05mm one, but that's more expensive than a drill bit.

e: sure, you could also make a coupler

LightRailTycoon
Mar 24, 2017
Those er-11 collets are supposed to be a shrink fit on the shaft. Freeze the motor, and heat the collet to 350-400f, and it should slide right on.

BattleMaster
Aug 14, 2000

There's always something I miss, I wasn't envisioning a permanent fit - at least, I don't expect there's an easy way to get it off once the shrink fit is done. I really was hoping that the set screws were an indicator of easy swapping :negative:

But maybe there's hope...

Sagebrush posted:

If that's the case, I would try polishing the edge of the bore as I suggested, and if it still doesn't work, buy a 5mm drill bit and go to town on it. You're correct that a drag knife won't care (much) if the tool is a fraction of a millimeter off center, so if your drill work is less than perfect it won't matter.

You could also get a 5mm reamer, or like a 5.05mm one, but that's more expensive than a drill bit.

e: sure, you could also make a coupler

I'm going to try a bit of polishing once I get my hands on some crocus cloth - best I can do right now (and I don't think I'll try with it) is 2000 grit sandpaper.

I am able to fit M5 screws into the chuck's shaft hole with even a little bit of give so I'm thinking maybe the shaft is the part that's out a bit. There's such a tiny difference and I have no way of quantifying it though.

BattleMaster fucked around with this message at 00:52 on Jan 31, 2021

Karia
Mar 27, 2013

Self-portrait, Snake on a Plane
Oil painting, c. 1482-1484
Leonardo DaVinci (1452-1591)

BattleMaster posted:

There's always something I miss, I wasn't envisioning a permanent fit - at least, I don't expect there's an easy way to get it off once the shrink fit is done. I really was hoping that the set screws were an indicator of easy swapping :negative:

But maybe there's hope...


I'm going to try a bit of polishing once I get my hands on some crocus cloth - best I can do right now (and I don't think I'll try with it) is 2000 grit sandpaper.

I am able to fit M5 screws into the collet with even a little bit of give so I'm thinking maybe the shaft is the part that's out a bit. There's such a tiny difference and I have no way of quantifying it though.

Got a decent number drill bit set? If so, you can use the shanks as a cheap alternative to plug gages (nowhere near as accurate, of course, but better than nothing.) They're probably pretty accurate if it's an OK set. The gaps between the sizes are too big to judge accurately, but you can get much better resolution by using three of them like this:


Just play around until you find a set that just barely fits and you can trig out the size of the hole based on the nominal diameters. Better than using the ID section of calipers, at least.

LightRailTycoon
Mar 24, 2017
To take it off, kiss it with a torch real quick!

BattleMaster
Aug 14, 2000

I'll keep the 3-drill bit thing in mind of next time, but I managed to get it on and it won't come off - and somehow any of my attempts to get it off just got it on further. So I got it into position and just decided that's where the chuck's home will be - neither the motor nor the chuck were expensive enough for me to fret over so whatever, I probably should have just done it properly with the shrink fit to begin with.

No torch on hand so I'll just order a replacement stepper for the extruder drive when I get around to building a 3D printer. I guess that way I won't have to salvage the knife cutter for parts later then!

Thanks for the advice everyone!

BattleMaster fucked around with this message at 02:22 on Jan 31, 2021

Dominoes
Sep 20, 2007

Hey dudes. My CNC code is getting unweildy, even though it's only doing a few things! Working on a Python script that's tailored to my current use cases, and expand as necessary. Will probably switch to Rust and add sanity checks... seems like a more approp language for something that has spinning sharp objects.

Example function that can be executed (mainly) in an independent way:
Python code:
    def ramp_line(self, start: Coord, end: Coord, ratio: float, depth: float) -> list[str]:
        """
        Execute a linear ramp, with a fixed ratio of length to height.
        Includes start, entry, and stop. It overshoots the desired depth to always
        end on either the start or end point.
        
        `ratio` is in length-to-height. Ie higher means shallower.
        """
        # todo: Do we want an angle in degrees/radians, or this ramp ratio?
        result = [
            f"G01 X{start.x} Y{start.y} Z{self.start_stop_height}",
            "M03",
            f"G01 Z{self.buffer_height}"  # This is where cutting (roughly) begins
        ]

        # `depth_step` is how far to descend each length-wise run. Note that
        # it doesn't take the buffer height into account.
        depth_step  = dist(start, end) / ratio

        active_coord = Coord(start.x, start.y, self.buffer_height)
        while active_coord.z > depth:
            active_coord.z -= depth_step
            if active_coord.x == start.x and active_coord.y == start.y:
                active_coord.x = end.x
                active_coord.y = end.y
            else:
                active_coord.x = start.x
                active_coord.y = start.y
                
            result.append(active_coord.linear_move())

        result.extend([
            f"G01 Z{self.start_stop_height}",
            "M05",
        ])

        return result
Thoughts on this approach? The end result is a separate script that strings together the results of these, and outputs to a file. Going to do a helical-entry circle-cutter next.

Dominoes fucked around with this message at 16:41 on Feb 9, 2021

babyeatingpsychopath
Oct 28, 2000
Forum Veteran


Dominoes posted:


Thoughts on this approach? The end result is a separate script that strings together the results of these, and outputs to a file. Going to do a helical-entry circle-cutter next.

Are you making a gcode postprocessor from first principles for a reason? Why aren't you making shapes in CAD and then running your software's CAM option and then just hitting "go" on the machine?

edit: edit edit: moved.

babyeatingpsychopath fucked around with this message at 17:00 on Feb 9, 2021

Adbot
ADBOT LOVES YOU

Dominoes
Sep 20, 2007

More like a g-code generator. I don't have a CAD model, and am bad at CAD software. (Was messing with FreeCad earlier this week for a diff proj, but have a long way to go) Maybe that's the answer. Hand-written G-Code seems to be a unsustainable.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply