To order, e-mail clarke@windyhillfoundry.com with the subject line "Clough" to make sure your message doesn't get lost. Machining drawings are here: bit.ly/46s3xOC
My summer job in a pattern shop (late 70's) suggests that you 3D print a pattern to cast an aluminum pattern, which will last a good long time in production. The 3D print will have to be enlarged slightly in size to account for the shrinkage of the aluminum, if you desire that level of size accuracy. Love your channel!
The problem isn't Fusion, the problem is the font itself. Fonts generally are horrible when worked with like this, they aren't designed with precision. My typical path for any logo (not just for bringing into CAD) is to convert it to outlines. Then manually go and edit all the nodes, adding and deleting as necessary, removing any overlapping areas, and duplicate nodes, adding in constraints (smooth) for all nodes on the splines when transitioning from a line to a spline to ensure tangency. I do this until I get a minimal set of nodes that cleanly represents the original, which should be a single outline for any character (or joined set of characters), not including cut-outs. (basically what you ultimately did in Fusion, but doing it all in the illustration package of your choice). You can then export that as a SVG, and bring it into just about anything without problem, it should correctly scale then too, without eating itself as you saw.
Fonts are artistic madness confined to vector rigidity. I spent so much time trying to convert a font to small caps in FontForge and it was a horrible ordeal to make the kerning match correctly with scaled letters. Recreating it in cad is such a good solution (for typefaces like this).
One fairly quick strategy I use with fonts is to fill them with black, then take a screenshot of your text (or export as jpeg, bmp, etc..). Then open inkscape and use the bitmap trace feature. This will give you a connected single vector around each letter (or the entire word if you are using connected fonts.)
The "normal" production sand casting process is to print or machine the pattern from something soft. We used a material called Renboard. That was used to cast an aluminum pattern. If the aluminum pattern wore out, we would cast a new one from the original master. That said, the renboard patterns were often used to cast hundreds of prototype parts without issue. It's pretty fun to machine compared to any kind of metal.
Just wanted to say Thank You James! I talked with Clarke earlier this week and got one ordered. I can’t wait to get it installed. I don’t think I have told you in a few years, but your ELS conversation was a game changer and makes the lathe a joy to operate!
I'm so glad you posted this. i thought i just sucked at Fusion because I'm always struggling with text. Now I know its not just me. I wish there was an easier way to trace curves than just drawing over though.
It's truly inspiring to see the dedication and hard work you've invested in perfecting the toolpost riser casting over two years！😯 Thank you for choosing Siraya Tech's Fast Mecha and Blu Mecha resins for your project. We're thrilled to see how they've contributed to your incredible work. Looking forward to seeing what’s your next!
You must have heard me through time and the internet on that font. I used to be a full time desktop publisher. Illustrator, Photoshop and Quark XPress were my life. 😂Good fonts are expensive and are expensive for a reason. The foundries would spend hours perfecting issues like the ones you demonstrated with the "Mrs" problem by creating custom pairings and other tricks. Those issues didn't exist in them. But they were hundreds of dollars each, sometimes each style - Bold, Italic, Bold-Italic etc. We had one customer that insisted on using a provided font file - something they bought cheap - and I spent more time, and their money, fixing the problems we had using it than it would have been if they bought a professionally designed font.
I have a simpler solution. Print the pattern without the text. Print the letters as a seperate prints and glue in place. I have done this a number of times. it also makes it easier to fill smooth the main pattern without the text in place.
I'm an engineer that works in product development. Frequently we have to put logos/text on sand casting bucks and try and get them to draft. As you found: its not easy! We developed a great way to approximate a draft angle on text (often "close enough" is "good enough" for sand casting). Rather than doing an offset of the surfaces, copy the surfaces you want and translate them in the same direction of draft (this eliminates any opportunity for undercuts to form) Then you create a new sketch normal to the text and offset each letter inwards a few thou. Use that sketch to split/trim away the offset text so that it is a little bit smaller than the parent surfaces. Once you are happy with the offset, you simply create a loft between the original surface and the offset surface. If you are trying to hit a specific angle you can start playing with the distance you offset/trim off of each letter until you get close. The result is a letter that will have a consistent appearance across all of the letters and be in draft!
Brilliant! I just scored the King version of this lathe at an auction, looking forward to attempting all your upgrades and more. Thanks for undertaking these mods for our benefit!
Thanks for pointing out the font problem AND a solution to it. I tried something similar a couple of weeks ago and it wouldn't work and figured I had done something wrong. Now know to go back to the drawing board and try again. Will try different fonts to see if I can find something that will work without having to recreate the entire alphabet and numbers! You have given me some hope. Thanks.
Good to know the fusion this deeply. When I made my own pilow block pedestal and I needed my brand name on curve what I did was I made the pattern without the logo and made pattern from wood and pasted my logo on curved surfaces and the letters were made of plastic and problem solved. And thank you for the alternative approach for fusion. You have always thought me as new ideas.
Hey James, A process my son has been using is coating the PLA print in resin. I print the object in PLA on my Bambu then he dunks it in resin and cures it. Fills in almost all of the layer lines. You might try that, might save a bunch of time.
I'm glad i'm not the only one struggling with fonts in CAD... I was often wondering if it wouldnt be easier to do the part modeling in some design software like Fusion/Alibre/Onshape.... and then just taking it to blender or plasticity for adding the fonts.
My strategy for complex geometry chamfers is to draw a triangle on a perpendicular plane and sweep it around the part. This still runs into problems, mostly self intersecting geometry, but it has gotten me through many otherwise "impossible" situations. This doesn't work well for adding material to create the chamfers, it's best suited for situations where you can subtract the swept geometry from the target body. That said, you can sometimes oversize the triangle (and use something other than a triangle) to try to avoid gaps between the sweep and target body when you can't subtract.
The font problem is a real eye opener. I assume similar problems would and did occur when you tried filets instead of the chamfer. I am glad you persevered and found a solution.
I have a slightly smaller 9x20 lathe which could use your tool post holder with a bit of modification. Rather than milling off some of the tool post base height, Have you given some thought to maybe making a couple of different height bases for different size lathes. This could accommodate well known tool post sizes such as AA, A, B, C & D. This covers the whole range of the most popular lathes from minis 8's up to, maybe 14". Just a thought. I don't know how invested you are in this product as it seems you are doing this more as a service to your fans than as a side product. I appreciate your efforts and enjoy the detail that you put into all of your productions.
Regarding Fusion problem - fonts are always pain in the ass. I tend to resketch everything using native geometry, proper splines/tangency etc. Other tip is projecting a sketch to a surface, then using surfaces to really build whatever you want to have there. This way you have 100% of control and you are not at mercy of emboss command.
You should probably stick an uv-led into those holes and cure the inside before plugging them. Hollow resin prints with uncured resin inside have been known to split after a few months
You probably already know, but us carbohydrate foam jockeys use various shaped blocks of wood to make sure we're only sanding the high spots. It definitely helps make things go faster than using the soft pads of your fingers which somewhat follow the contours of the ridges. Great video, i'll keep these resins in mind for my upcoming castings.
Not just fonts, svg's in general. I had so much pain with importing files, until I just started importing and either repairing or redrawing. The add in, sketch checker, does a very good job of identifying those small errors.
Twenty plus years ago I worked for a while as a graphics designer for a sign company. I ran into your Font troubles almost daily. Sad that nothing has changed. Happy that you figured out a solution. In the software that I used back then we could link all the parts of a broken font into a group. The other nasty thing I ran into was multiple layers of lines. 4,5 even 7 layers that you needed to delete down to only one layer. DXF files were the absolute worst when you wanted to turn them into a raster file SVG was your version. I felt for you man. Nice presentation Hoping you and Clarke the best in this adventure. Mark
A lot time when you export / import from Illustrator, you end up with some points on the outlines are doubled or overlapping. Like the start / end is usually not closed, but consists of two points. The same in sharp corners, which you need t clean manually. I am doing 3D design for over 25 years and it can be really vexing has you have found out :). The best way is exactly what you did - simplify - redraw.
James, that font tutorial was worth every penny I paid for this video. lol Seriously I was attempting something very similar and gave up. Thanks for going through all the effort to solve this!
As a user of F360 for semi artistic purposes, LS Crossfire Pro plasma cutter end use, I totally get the issue with fonts and the way they sometimes conflict with toolpaths. I cut my hair short so I can't pull out what's left. I can only imagine what it's like with 3D printing or mold making... Oh, I see that your hair is short too! Super cool video, great example of why I've been watching for all these years, thank you.
Very interesting, and thanks for reminding me why I’ll stick with FDM, for now. With a kid in the house I wouldn’t want all that hassle, chemicals etc. Results are amazing though!
I don't know if you have ever seen 'UV Cured Varnish' for wood finishing. It is simply wiped onto the wood surface and then cured with a handheld UV light. It seems to me than it is essentially a 3D printer resin (or variation). I had the thought that it could be applied to FDM prints to give them a smooth tough finish. Cures in seconds and can be layered. Never tried it, probably never will. Just a thought. Thanks again for another great informative video.😀
Nice work. Have been working with fonts forever. I always recommend cleaning or tracing. For the mold, consider making a 3d resin master, make a silicone mastering then casting resin daughter molds.
I'm surprised you didn't try ABS with acetone smoothing as one of the experiments. Whenever I think of "FDM" and "smooth", that's what I think of immediately.
I like what you came up with here! I’m really curious how strong or flexible different resins are…if you ever need some ideas for videos, I’d love to see some strength tests on resin prints too
Back in the mid 1980s I was taking a CNC Technicians Course and one of the CAD/CAM programs we used on a mainframe was designed for generating curved surfaces as com[lex as turbine blades and one of the problems we had was exactly like you had extruding fonts. We finally generated as list of fonts that you could extrude directly into the program. Others we would have to hand generate like you did by copying and deleting. The main problem we ha was the CAD/CAM program was power hungrily and we could only run it at night when most of the other courses were not using it, and it would still crash about once a week. .
Excuse me if I missed this in another video but can you share your thoughts on the horizontal offset you designed into this solid mount? Is just as simple as keeping the cutting tool as close to the work piece as possible? Thanks, Erik
Thanks James, I love your videos - good solid information, well articulated, with flawless production. The font solution in Fusion is a keeper. As I remember being told as an apprentice “It’s great to learn from your mistakes - even better to learn from someone else’s” All the very best from the UK.
Have you thought about just matching the pattern once and for all out of metal, steel or even Ali would be good enough probably? That would be my go to or even make a bigger printed pattern to cast an aluminium one?
Haven’t gotten all the way through, but I’m curious if you’re going to talk about ABS or PLA vapor smoothing instead of sanding or bondo. Edit: Ok, no mention but your result is great obviously. Still curious about vapor smoothing. If it works it should be cheaper and faster.
Interesting. I've done a lot of work with importing True Type fonts and other artwork into CAD/CAM tools, and haven't seen these sorts of artifacts in the fonts. The stuff i wrote was a v-engrave plugin for CamBam that walked the outline and computed a 3D path for a v cutter to dive into the outline such that the cutter rides tangent to both sides of the outline. One option that might be easier for cleanup is sucking the outline into Inkscape and cleaning it up as an SVG, which can then be converted back to dxf and reimported into CAD software. I did a lot of that for a recent project.
Would you happen to know if this fits PM1340 lathe? I contacted Clarke and he said they are available but I know you designed it for Grizzly 1022 and I am hoping it fits on my PM1340.
When I saw the deletes in the timeline I knew your pain, though I don't think I've gone quite that far... I'm also trying to cast a solid tool post, but using class-C grout, until my printer stopped a few days ago.
woah 20min cure seems like a lot. I generally cure my prints for a 30-60s per side, but then I don't use opaque filled resins like that. I've only used Elegoos standard resin for a few prints, but then only Ameralabs XVN-50, and it's great to work with. Good resin really makes it so much nicer to work with
Nice work on the pattern. Clark will thank you for the care you take on draft angles. Have you used the oil based paint that Tractor supply has with their catalyst added sets up very hard and with a nice finish.
Can you convert the font geometry to mesh and clean it up there, there are some automatic mesh tools that might make this easier like smooth, simplify, patch etc
At around 9:50 when you were editing the text outlines, did you know you can double click the text lines and it will show you how continuous the line is? Generally I only use the Fusion360 built in text function to label mold parts that are not seen on the final part. If you can't afford Adobe even a free drawing program like Inkscape will produce high quality type for 3D uses. Nice detailed video, I watched the whole 33 mins.
The problem with the fonts isn't just a problem with fusion360. I often use sketchup too and I ran into the same issues. I do it like you and recreate the writing or use a geometrical "simpler" Font like Arial...
Thanks James ... very informative. True Type fonts are an issue with CNC engraving from my seat ... and I've given up on finding good single-stroke engraving fonts (Hershey Fonts are already in residence, thanks). I'm going to have to look further into resin printing ... I still haven't found a good use-case for FDM printing ... but enjoy your experiences, especially with the carbon-xxx filaments. Maybe I missed something ... but can't you do an extrusion with a constant reduction/enlargement ratio? Regards, R
I remember that collaboration! I would love one for my 7x14. Also still looking for someone here in Boise with a surface grinder who is willing to dust off my little grinding vises so they are matched hight.
I would make this pattern out of a hardwood such as Hard Maple using a CNC router, sand the final product smooth as glass then coat it with epoxy some sort of easy flow epoxy.
An easy way to create the embossing is to create an offset face of the surface you want to emboss, then use the split body command and use that offset face as the splitting tool. Then simply merge the objects and you're set. The other simple way is to extrude from within the body you want to emboss on, use the "replace face" tool on the text face, then move the text out the same distance needed. If a visual guide helps i'm happy to make a quick tutorial on embossing on complex geometry in fusion, just let me know :)
@@GodardScientific I'd be interested in seeing that. I've tried most of the techniques shown in videos all over the Internet, and most struggle with this particular geometry, though it would be worth trying them again with good sketch geometry.
In Autodesk Inventor there's a "ruled surface" command that I've had a lot of luck with achieving a specific draft angle around the perimeter of complex embossed logos on complex surfaces in injection mold designs. I wonder if Fusion has something similar. It's still a tedious process, and I've never seen a tutorial detailing this approach. Unlike the software world where everyone shares ideas on Stack Exchange, the CAD world seems a bit more guarded, especially when you go beyond the basics. I wish I could share a picture of the part I had designed to give you an idea what I'm talking about.
Those Siriya Tech resins looks great. I just got into resin 3D printing and have been looking for some good mechanical resins. Have been looking at some of the 3DMaterials resins (specifically the SuperRigid, SuperPCS, and SuperFast) and the various Loctite resins (although pretty expensive).
I would do a clean vector file projected onto two construction surfaces. One raised and one slightly below the casting surface. Offset the one but the distance needed for draft. Then use loft tool to create letters. Then use combine feature later with casting. I have found that creating a multi body part and combining at very end is the only way to get there effiently sometimes. Sorry if I was hard to follow.
One other thing you can do. Go to manufacturing space. Cam the original part with no chamfer. Use original model as stock model. Cam the chamfer. Save Cam machining simulation as stl body. Cheater way of doing it.
Most work I did on my CNC was engraving. Making the chamfers for the V-bit had always been hell due to the font issues. So I didn't want to make a business out of it. Engraving can be profitable, but it's not worth the frustration.
Did you ever try increasing the height of the font by about 20 thou to see if it could compute that way? When you said that the chamfer would compute after you set it to a certain height I was almost certain that you were then going to say that you simply just made the letters a bit taller to compensate.
I would have used model and your CNC mill to make pattern out of Aluminum or you 3D printed to make one casted part polished part and have a pattern. Some times the old ways still work.
This issue is not unique to Fusion in the slightest though, it's the same in every CAD package I've used, in every vector drawing program I've used. You need to make sure the source material you're using is good, that's always the first step, otherwise you get issues like these. As the saying goes, garbage in, garbage out. I routinely have to teach this to the engineers I work with, even those who have worked with CAD for decades. I admire your effort in recreating the logo in Fusion, though honestly I think it would have been far less of a headache to just redo it properly in Illustrator, cleaning up errant points, fixing inconsistencies, etc.
I think for part of this, you perhaps lost sight of your end-goal, in the pursuit of making your master meet the requirement of being "quality and durable and printed".. You didn't necessarily need all 3 from one piece. Print one that takes into account 2x shrinkage, then cast sub-masters so it only needs to survive a few of those castings, store it, clean sub-masters to an acceptable level. Then you'd have actual metal castings to use as masters for the rough life of being sand mold-formers, so sand can be pounded against them without excessive degradation.
Taking in to the account that its just font on casting metal(so it always have some small imperfections and dont need to be 100% accurate) I would just load main body in Blender or ZBrush merge font with body, smooth edges to get proper slope.... and done. Is it dirty workflow? Yep. Is it supper accurate? For casting and just for font/logo its perfectly fine. :D
There’s a video where a user uses a microscope to see how CF and GF filaments embed shards on the skin. Could you make a video about how safe those filaments are? In-depth safety when handling fiber filaments and prints. Thanks!🙏🏻
What's the dimensional stability of the resin print? Because it is a thick opaque part my assumption is that resin inside the part is not fully cured, which may result in deformation of the part in future.
I would for sure make a female pattern and then cast it with Xencast P6 polyurethan plastic.. It cures in around 1h. In that way you only have to do the work once for the female mold and you can reuse the mold for many many times. Especially if you make the mold in silicone.
You're going to want to wear a dust mask (or even a proper respirator) when sanding resin. Pneumoconiosis from processing resin has been documented since at least the eighties, and is no joke.
To order, e-mail clarke@windyhillfoundry.com with the subject line "Clough" to make sure your message doesn't get lost. Machining drawings are here: bit.ly/46s3xOC
Pro Tip: If you want to cleanup your wash IPA just expose it to UV. The uncured resin will cure and settle out leaving clean IPA
Cant you make a video of these being cast? This was such a amazing watch!
@@fleischer444 I'm not casting them. That would be up to Clarke at Windy Hill Foundry.
My summer job in a pattern shop (late 70's) suggests that you 3D print a pattern to cast an aluminum pattern, which will last a good long time in production. The 3D print will have to be enlarged slightly in size to account for the shrinkage of the aluminum, if you desire that level of size accuracy. Love your channel!
The problem isn't Fusion, the problem is the font itself. Fonts generally are horrible when worked with like this, they aren't designed with precision. My typical path for any logo (not just for bringing into CAD) is to convert it to outlines. Then manually go and edit all the nodes, adding and deleting as necessary, removing any overlapping areas, and duplicate nodes, adding in constraints (smooth) for all nodes on the splines when transitioning from a line to a spline to ensure tangency. I do this until I get a minimal set of nodes that cleanly represents the original, which should be a single outline for any character (or joined set of characters), not including cut-outs. (basically what you ultimately did in Fusion, but doing it all in the illustration package of your choice). You can then export that as a SVG, and bring it into just about anything without problem, it should correctly scale then too, without eating itself as you saw.
Fonts are artistic madness confined to vector rigidity. I spent so much time trying to convert a font to small caps in FontForge and it was a horrible ordeal to make the kerning match correctly with scaled letters. Recreating it in cad is such a good solution (for typefaces like this).
One fairly quick strategy I use with fonts is to fill them with black, then take a screenshot of your text (or export as jpeg, bmp, etc..). Then open inkscape and use the bitmap trace feature. This will give you a connected single vector around each letter (or the entire word if you are using connected fonts.)
The "normal" production sand casting process is to print or machine the pattern from something soft. We used a material called Renboard. That was used to cast an aluminum pattern. If the aluminum pattern wore out, we would cast a new one from the original master. That said, the renboard patterns were often used to cast hundreds of prototype parts without issue. It's pretty fun to machine compared to any kind of metal.
I've often walked by pallets of renshape big enough to mock up automotive body panels at work. It's amazing what you can do with it.
WWW in the house, you have that CNC machine up and running yet? Nice work James thanks for the detailed explanation on Fusion 360.
I use those siraya tech resins and have had good luck with all of them .
Just wanted to say Thank You James!
I talked with Clarke earlier this week and got one ordered. I can’t wait to get it installed. I don’t think I have told you in a few years, but your ELS conversation was a game changer and makes the lathe a joy to operate!
I'm so glad you posted this. i thought i just sucked at Fusion because I'm always struggling with text. Now I know its not just me. I wish there was an easier way to trace curves than just drawing over though.
@@crashkg felt!
It's truly inspiring to see the dedication and hard work you've invested in perfecting the toolpost riser casting over two years！😯
Thank you for choosing Siraya Tech's Fast Mecha and Blu Mecha resins for your project. We're thrilled to see how they've contributed to your incredible work.
Looking forward to seeing what’s your next!
You must have heard me through time and the internet on that font. I used to be a full time desktop publisher. Illustrator, Photoshop and Quark XPress were my life. 😂Good fonts are expensive and are expensive for a reason. The foundries would spend hours perfecting issues like the ones you demonstrated with the "Mrs" problem by creating custom pairings and other tricks. Those issues didn't exist in them. But they were hundreds of dollars each, sometimes each style - Bold, Italic, Bold-Italic etc. We had one customer that insisted on using a provided font file - something they bought cheap - and I spent more time, and their money, fixing the problems we had using it than it would have been if they bought a professionally designed font.
I have a simpler solution. Print the pattern without the text. Print the letters as a seperate prints and glue in place. I have done this a number of times. it also makes it easier to fill smooth the main pattern without the text in place.
That font solution is an absolute game changer James, I'm definitely going to remember that so thank you! 👍👍
I'm an engineer that works in product development. Frequently we have to put logos/text on sand casting bucks and try and get them to draft. As you found: its not easy!
We developed a great way to approximate a draft angle on text (often "close enough" is "good enough" for sand casting). Rather than doing an offset of the surfaces, copy the surfaces you want and translate them in the same direction of draft (this eliminates any opportunity for undercuts to form)
Then you create a new sketch normal to the text and offset each letter inwards a few thou. Use that sketch to split/trim away the offset text so that it is a little bit smaller than the parent surfaces.
Once you are happy with the offset, you simply create a loft between the original surface and the offset surface. If you are trying to hit a specific angle you can start playing with the distance you offset/trim off of each letter until you get close.
The result is a letter that will have a consistent appearance across all of the letters and be in draft!
Brilliant! I just scored the King version of this lathe at an auction, looking forward to attempting all your upgrades and more. Thanks for undertaking these mods for our benefit!
Thanks for pointing out the font problem AND a solution to it. I tried something similar a couple of weeks ago and it wouldn't work and figured I had done something wrong. Now know to go back to the drawing board and try again. Will try different fonts to see if I can find something that will work without having to recreate the entire alphabet and numbers! You have given me some hope. Thanks.
Good to know the fusion this deeply. When I made my own pilow block pedestal and I needed my brand name on curve what I did was I made the pattern without the logo and made pattern from wood and pasted my logo on curved surfaces and the letters were made of plastic and problem solved. And thank you for the alternative approach for fusion. You have always thought me as new ideas.
Hey James, A process my son has been using is coating the PLA print in resin. I print the object in PLA on my Bambu then he dunks it in resin and cures it. Fills in almost all of the layer lines. You might try that, might save a bunch of time.
I'm glad i'm not the only one struggling with fonts in CAD...
I was often wondering if it wouldnt be easier to do the part modeling in some design software like Fusion/Alibre/Onshape.... and then just taking it to blender or plasticity for adding the fonts.
My strategy for complex geometry chamfers is to draw a triangle on a perpendicular plane and sweep it around the part. This still runs into problems, mostly self intersecting geometry, but it has gotten me through many otherwise "impossible" situations.
This doesn't work well for adding material to create the chamfers, it's best suited for situations where you can subtract the swept geometry from the target body. That said, you can sometimes oversize the triangle (and use something other than a triangle) to try to avoid gaps between the sweep and target body when you can't subtract.
The font problem is a real eye opener. I assume similar problems would and did occur when you tried filets instead of the chamfer. I am glad you persevered and found a solution.
I have a slightly smaller 9x20 lathe which could use your tool post holder with a bit of modification. Rather than milling off some of the tool post base height, Have you given some thought to maybe making a couple of different height bases for different size lathes. This could accommodate well known tool post sizes such as AA, A, B, C & D. This covers the whole range of the most popular lathes from minis 8's up to, maybe 14". Just a thought. I don't know how invested you are in this product as it seems you are doing this more as a service to your fans than as a side product. I appreciate your efforts and enjoy the detail that you put into all of your productions.
Order of Operations + Attention to Detail = Clough42
YES!!!!! I was REALLY hoping you'd have these available for sale! THANK YOU!!!
Regarding Fusion problem - fonts are always pain in the ass. I tend to resketch everything using native geometry, proper splines/tangency etc. Other tip is projecting a sketch to a surface, then using surfaces to really build whatever you want to have there. This way you have 100% of control and you are not at mercy of emboss command.
You should probably stick an uv-led into those holes and cure the inside before plugging them. Hollow resin prints with uncured resin inside have been known to split after a few months
You probably already know, but us carbohydrate foam jockeys use various shaped blocks of wood to make sure we're only sanding the high spots. It definitely helps make things go faster than using the soft pads of your fingers which somewhat follow the contours of the ridges. Great video, i'll keep these resins in mind for my upcoming castings.
Not just fonts, svg's in general. I had so much pain with importing files, until I just started importing and either repairing or redrawing. The add in, sketch checker, does a very good job of identifying those small errors.
Twenty plus years ago I worked for a while as a graphics designer for a sign company. I ran into your Font troubles almost daily. Sad that nothing has changed. Happy that you figured out a solution. In the software that I used back then we could link all the parts of a broken font into a group. The other nasty thing I ran into was multiple layers of lines. 4,5 even 7 layers that you needed to delete down to only one layer. DXF files were the absolute worst when you wanted to turn them into a raster file SVG was your version. I felt for you man. Nice presentation Hoping you and Clarke the best in this adventure.
Mark
Fonts never stops surprising. I have been working in the printing industry for many years, and have experienced lots of strange things.
A lot time when you export / import from Illustrator, you end up with some points on the outlines are doubled or overlapping. Like the start / end is usually not closed, but consists of two points. The same in sharp corners, which you need t clean manually.
I am doing 3D design for over 25 years and it can be really vexing has you have found out :). The best way is exactly what you did - simplify - redraw.
Thank you for sharing your lessons learned on Fusion and fonts. I love learning from other's pain so I don't have to learn it the hard way!
James, that font tutorial was worth every penny I paid for this video. lol
Seriously I was attempting something very similar and gave up. Thanks for going through all the effort to solve this!
Nice work, James! 👍
13:07 This would be a funny shirt design.
As a user of F360 for semi artistic purposes, LS Crossfire Pro plasma cutter end use, I totally get the issue with fonts and the way they sometimes conflict with toolpaths. I cut my hair short so I can't pull out what's left. I can only imagine what it's like with 3D printing or mold making... Oh, I see that your hair is short too!
Super cool video, great example of why I've been watching for all these years, thank you.
Very interesting, and thanks for reminding me why I’ll stick with FDM, for now. With a kid in the house I wouldn’t want all that hassle, chemicals etc. Results are amazing though!
I don't know if you have ever seen 'UV Cured Varnish' for wood finishing. It is simply wiped onto the wood surface and then cured with a handheld UV light. It seems to me than it is essentially a 3D printer resin (or variation). I had the thought that it could be applied to FDM prints to give them a smooth tough finish. Cures in seconds and can be layered. Never tried it, probably never will. Just a thought.
Thanks again for another great informative video.😀
Nice work.
Have been working with fonts forever. I always recommend cleaning or tracing.
For the mold, consider making a 3d resin master, make a silicone mastering then casting resin daughter molds.
This makes me feel better for going with Albre over Fusion.
I'm surprised you didn't try ABS with acetone smoothing as one of the experiments. Whenever I think of "FDM" and "smooth", that's what I think of immediately.
I like what you came up with here! I’m really curious how strong or flexible different resins are…if you ever need some ideas for videos, I’d love to see some strength tests on resin prints too
Back in the mid 1980s I was taking a CNC Technicians Course and one of the CAD/CAM programs we used on a mainframe was designed for generating curved surfaces as com[lex as turbine blades and one of the problems we had was exactly like you had extruding fonts. We finally generated as list of fonts that you could extrude directly into the program. Others we would have to hand generate like you did by copying and deleting. The main problem we ha was the CAD/CAM program was power hungrily and we could only run it at night when most of the other courses were not using it, and it would still crash about once a week. .
Great tutorial.
Excuse me if I missed this in another video but can you share your thoughts on the horizontal offset you designed into this solid mount? Is just as simple as keeping the cutting tool as close to the work piece as possible?
Thanks, Erik
Thanks James, I love your videos - good solid information, well articulated, with flawless production. The font solution in Fusion is a keeper.
As I remember being told as an apprentice “It’s great to learn from your mistakes - even better to learn from someone else’s”
All the very best from the UK.
Have you thought about just matching the pattern once and for all out of metal, steel or even Ali would be good enough probably? That would be my go to or even make a bigger printed pattern to cast an aluminium one?
I am always so happy when you and Functional Print Friday, post new videos.
Haven’t gotten all the way through, but I’m curious if you’re going to talk about ABS or PLA vapor smoothing instead of sanding or bondo.
Edit: Ok, no mention but your result is great obviously. Still curious about vapor smoothing. If it works it should be cheaper and faster.
Interesting. I've done a lot of work with importing True Type fonts and other artwork into CAD/CAM tools, and haven't seen these sorts of artifacts in the fonts. The stuff i wrote was a v-engrave plugin for CamBam that walked the outline and computed a 3D path for a v cutter to dive into the outline such that the cutter rides tangent to both sides of the outline.
One option that might be easier for cleanup is sucking the outline into Inkscape and cleaning it up as an SVG, which can then be converted back to dxf and reimported into CAD software. I did a lot of that for a recent project.
Would you happen to know if this fits PM1340 lathe? I contacted Clarke and he said they are available but I know you designed it for Grizzly 1022 and I am hoping it fits on my PM1340.
When I saw the deletes in the timeline I knew your pain, though I don't think I've gone quite that far...
I'm also trying to cast a solid tool post, but using class-C grout, until my printer stopped a few days ago.
Enjoyed…great discussion
Good stuff
woah 20min cure seems like a lot. I generally cure my prints for a 30-60s per side, but then I don't use opaque filled resins like that. I've only used Elegoos standard resin for a few prints, but then only Ameralabs XVN-50, and it's great to work with. Good resin really makes it so much nicer to work with
Do you think these could be tumble polished with ceramic media?
Nice work on the pattern. Clark will thank you for the care you take on draft angles. Have you used the oil based paint that Tractor supply has with their catalyst added sets up very hard and with a nice finish.
So many fingers are you channeling Tony?
Thanks! I have had that problem!
And here I was thinking I was deaf to HF sounds but at least I'm ultrasonic bath level of undeafness
Can you convert the font geometry to mesh and clean it up there, there are some automatic mesh tools that might make this easier like smooth, simplify, patch etc
At around 9:50 when you were editing the text outlines, did you know you can double click the text lines and it will show you how continuous the line is? Generally I only use the Fusion360 built in text function to label mold parts that are not seen on the final part. If you can't afford Adobe even a free drawing program like Inkscape will produce high quality type for 3D uses. Nice detailed video, I watched the whole 33 mins.
The problem with the fonts isn't just a problem with fusion360.
I often use sketchup too and I ran into the same issues.
I do it like you and recreate the writing or use a geometrical "simpler" Font like Arial...
Thanks James ... very informative.
True Type fonts are an issue with CNC engraving from my seat ... and I've given up on finding good single-stroke engraving fonts (Hershey Fonts are already in residence, thanks).
I'm going to have to look further into resin printing ... I still haven't found a good use-case for FDM printing ... but enjoy your experiences, especially with the carbon-xxx filaments.
Maybe I missed something ... but can't you do an extrusion with a constant reduction/enlargement ratio?
Regards,
R
I remember that collaboration! I would love one for my 7x14. Also still looking for someone here in Boise with a surface grinder who is willing to dust off my little grinding vises so they are matched hight.
I would make this pattern out of a hardwood such as Hard Maple using a CNC router, sand the final product smooth as glass then coat it with epoxy some sort of easy flow epoxy.
An easy way to create the embossing is to create an offset face of the surface you want to emboss, then use the split body command and use that offset face as the splitting tool. Then simply merge the objects and you're set. The other simple way is to extrude from within the body you want to emboss on, use the "replace face" tool on the text face, then move the text out the same distance needed.
If a visual guide helps i'm happy to make a quick tutorial on embossing on complex geometry in fusion, just let me know :)
Please do ❤️
@@GodardScientific I'd be interested in seeing that. I've tried most of the techniques shown in videos all over the Internet, and most struggle with this particular geometry, though it would be worth trying them again with good sketch geometry.
@@Clough42 I'll throw something quick together, for your model did you use a fillet or a revolve to create the curvature on the main body?
Video is live now on my channel if anyone's interested :)
In Autodesk Inventor there's a "ruled surface" command that I've had a lot of luck with achieving a specific draft angle around the perimeter of complex embossed logos on complex surfaces in injection mold designs. I wonder if Fusion has something similar. It's still a tedious process, and I've never seen a tutorial detailing this approach. Unlike the software world where everyone shares ideas on Stack Exchange, the CAD world seems a bit more guarded, especially when you go beyond the basics. I wish I could share a picture of the part I had designed to give you an idea what I'm talking about.
Those Siriya Tech resins looks great. I just got into resin 3D printing and have been looking for some good mechanical resins. Have been looking at some of the 3DMaterials resins (specifically the SuperRigid, SuperPCS, and SuperFast) and the various Loctite resins (although pretty expensive).
They discontinued the PLA Tough material. Bummer, you can't get it anymore.
You can also set a negative taper angle in your extrude operation, which may have been able to give you a draft angle.
Not against the curved surface.
I would do a clean vector file projected onto two construction surfaces. One raised and one slightly below the casting surface. Offset the one but the distance needed for draft. Then use loft tool to create letters. Then use combine feature later with casting. I have found that creating a multi body part and combining at very end is the only way to get there effiently sometimes. Sorry if I was hard to follow.
One other thing you can do. Go to manufacturing space. Cam the original part with no chamfer. Use original model as stock model. Cam the chamfer. Save Cam machining simulation as stl body. Cheater way of doing it.
For those of us whom already have done the 6 bolt compound conversion, what are the hole spacing?
Most work I did on my CNC was engraving. Making the chamfers for the V-bit had always been hell due to the font issues. So I didn't want to make a business out of it. Engraving can be profitable, but it's not worth the frustration.
Did you ever try increasing the height of the font by about 20 thou to see if it could compute that way? When you said that the chamfer would compute after you set it to a certain height I was almost certain that you were then going to say that you simply just made the letters a bit taller to compensate.
@17:00 I guess you mean 0.08mm 😉
I would have used model and your CNC mill to make pattern out of Aluminum or you 3D printed to make one casted part polished part and have a pattern. Some times the old ways still work.
Just CNC 1 out of aluminum. Last forever. I cut one out of a 4140 block for my diy lathe..
Did you look at painting with a 2 part paint? Both SprayMax and Eastwood make 2K aerosol spray paints that are very tough, and easy to shoot.
This issue is not unique to Fusion in the slightest though, it's the same in every CAD package I've used, in every vector drawing program I've used. You need to make sure the source material you're using is good, that's always the first step, otherwise you get issues like these. As the saying goes, garbage in, garbage out. I routinely have to teach this to the engineers I work with, even those who have worked with CAD for decades.
I admire your effort in recreating the logo in Fusion, though honestly I think it would have been far less of a headache to just redo it properly in Illustrator, cleaning up errant points, fixing inconsistencies, etc.
I think for part of this, you perhaps lost sight of your end-goal, in the pursuit of making your master meet the requirement of being "quality and durable and printed".. You didn't necessarily need all 3 from one piece. Print one that takes into account 2x shrinkage, then cast sub-masters so it only needs to survive a few of those castings, store it, clean sub-masters to an acceptable level. Then you'd have actual metal castings to use as masters for the rough life of being sand mold-formers, so sand can be pounded against them without excessive degradation.
Taking in to the account that its just font on casting metal(so it always have some small imperfections and dont need to be 100% accurate) I would just load main body in Blender or ZBrush merge font with body, smooth edges to get proper slope.... and done. Is it dirty workflow? Yep. Is it supper accurate? For casting and just for font/logo its perfectly fine. :D
There’s a video where a user uses a microscope to see how CF and GF filaments embed shards on the skin. Could you make a video about how safe those filaments are? In-depth safety when handling fiber filaments and prints.
Thanks!🙏🏻
why wouldn't you print and prep 1 master, then make a silicone mold from the master and resin cast a few more gen1 masters from that?
cool!!!
Did you consider printing in ABS and acetone smoothing it?
What's the dimensional stability of the resin print?
Because it is a thick opaque part my assumption is that resin inside the part is not fully cured, which may result in deformation of the part in future.
Why did you choose to cast it, rather than just machine one out of steel from scratch?
I saw a paint that that sets with UV light
Maybe try a 2 part automotive paint for a faster cure time.
Plus 2-parts are much (much!) tougher than rattle can enamel, so double win.
@@d00dEEE I want to see over the next few weeks how the material he used holds up... if a guy doesn't have to paint, all the better
This would be so much easier in Maya!
29:15 James, you do wear a mask, right?
Just order the molds from JLCPCB, they sandblast them and the surface on them would be perfect for your project
Huh, I thought it was a new logo "Clough +2". It wasn't apparant to me what you were doing until you said it.
You have a CNC, can you make a pattern in aluminum to increase durability?
I feel like I missed it, but what was the mix ratio of the two resins? Or was this just two different resins BUT not mixed?
The later
That part would definitely ship very safely wrapped in bubble wrap in a flat rate padded envelope. $11 bucks.
Apologies, F360 noob here. Would a fillet of the letting not have worked? So they end up being rounded in relief, rather than chamfered.
ABS is perfect för standing. PLA you dont want to sand.
The element deletion game at 8:00 reminds me of old Sierra point and click adventure games - Guess what the developer was thinking!
I would for sure make a female pattern and then cast it with Xencast P6 polyurethan plastic.. It cures in around 1h.
In that way you only have to do the work once for the female mold and you can reuse the mold for many many times. Especially if you make the mold in silicone.
You're going to want to wear a dust mask (or even a proper respirator) when sanding resin. Pneumoconiosis from processing resin has been documented since at least the eighties, and is no joke.