If I think about it again he probably asked SmoothOn themselves. I guess they must have done a lot of testing of their product for different compounds such as coolant.
Nitro RC you have such a variety of coolants for so many different applications however I don't think there would be any inherent problems with most of the one's commonly used by most machines. Unless you get some backyard wizard who mixes up something really acidic that eats it up and swears it's the best coolant ever made. Lol
Love it. I've always been amazed how clean you keep your stuff, but this is a great idea for a guy like me that would need as much as I can get to make cleanup quicker. Nice to know about the Smooth-on too...
I didn't make complete doughnut circles, but I made some posts that are similar to the posts in the centers. This was on a manual milling machine. First I milled off the top of the piece so that I had square (more or less) posts sticking up. Then I positioned my spindle where I wanted a round post center, and in one cut brought the tool down and left the round post. I used a boring bar, and turned the cutting edge inside so it reduced the size to a precise diameter post instead of the opposite operation of enlarging a hole. I was cutting plastic. I believe that with the right cutting tool in a boring bar, you could do the same thing to make your mold like drilling a standard hole, instead of the tiny end mill.
hey did the persons vac work to degas? I would love to use shop air and get rid of my vacuum pump that is just taking up space for the few times we do mold poring (or wood stabilizing)
John my first thought would the hot chips burn through? I know you all do mostly aluminum but running steel with indexable tooling and no coolant is throwing off blue chips. Might melt. There has to be a heat resistant rubber right? Awesome stuff brother! Appreciate your videos.
Great video. Seems like everytime i am working on a project you or grimsmo releases a video dealing with the same issue i may be having. Just made my first mold for vacuum forming some thermoplastic.
A better idea than the magnetic vehicle signs I used. They worked great but cleaning the chips from them was a pain. How does this material hold up to hot chips? Likely not an issue with flood coolant but could be if you run dry.
I wish o had cnc and a fixture plate. I will be adapting fast fail, cheap fail to my life, and thank you my friend foe Mr he awesome video. I don't comment much, but I had to share the love I have here
Smooth-On is great if you only want to make one or two parts but if you are doing it in bulk they are not the best most cost effective company to use. BJB would be a step up in terms for mass producing them and there are better companies up from there. For optical clarity you need a rigid resin, degas the resin, place the mold in a pressure chamber for cureing. Optical clarity really isn't a thing in flexible materials.
Nice, Not sure there is any benefit in going to metal moulds for a poured/cast end product. I typically would cut this out of machine wax for the master positive and pull out a silicone mold. Wax is Fast and reusable if you change your mind. You can then pour 10 silicone moulds off the master wax for ramping up to production, eliminating extra machining and any part release issues. Thanks for the video. I always enjoy watching
John, Something I have been wanting to learn more about, I wish you would do a video on thread wires/measuring over wires. I wish i knew more about it, and how it all works. I have always had the luxury of ring and plug gauges. I also think some of the guys starting out could learn how to check threads without needing an expensive gauge.
I think Edge Precision discussed it some in one of his vids. It's has a great CNC channel if you aren't already aware of it. I'm sure Mr. Pete has covered it also at some point also. There's really not much to it. The hardest part is holding the damn wires with one hand!
light media blast mold to get an even finish, go fine as you can and still get an etch. that way later on you can easily touch up imperfections in the mold. or high polish if you think sticking may be an issue.
Realy nice, what shore hardness did you use? I'm lacking a little bit of feel for that, I have some understanding for the soft stuff from working with silicone, but PU rubber is often more on the harder side. And how did you manage air bubbles that result from the casting process itself? I imagine that it could trap some air when flowing into those small pockets and I saw you poking around there with a stick. I heard you can compress those in a pressure chamber to a size that's barely noticable, maybe that's a good idea for you.
Great idea for a product. I bought the Tosa plate for my 770 before I knew you had an offering. They have a bazillion plugs and I ditched that idea right off in favor of just placing aluminum plate on the uncovered spots. I've not compared the threaded and milled hole specs between theirs and yours. I'm hoping they're the same and I can buy some of these from you. Being able to cut sacrificial or special shapes is great. Thanks for the video and new product.
I'm in the same boat. Both plates have the same hole spacing which is good. But the tosa plates have a .562 reamed hole, while Johns has a .500. So I'm guessing it will work, but the nubs that go into the holes wont be snug. I doubt flood coolant will be much of an issue, but if you have air blast like me, then they will probably move.
Thanks, Shawn. I had just gone back to check spacing and it indeed looks the same with the exception of how many holes are threaded, like you say. If SMW does come out with the product I'll give it a try and report back on its fit with Tosa.
There is a difference between degassing and outgassing, if you pull too much vacuum you actually evaporate the chems inducing air bubbles or cloudiness. Low/mild vacuum for epoxy type chemicals works best. I do like the black.
He's using a silicone product. I can't read the part number on the bottles, but I'd guess it is platinum based. Smooth-On likely has detailed instructions on proper degassing for the product. www.smooth-on.com
Still wondering why you’re so against helical ramping in small pockets... Probably could have done those cavities in 1/2 the time and 1/10 the code size. I know time wasn’t an issue. But a helical ramp would have also evacuated the chips well too. And scotch brite that main flat surface too to get the tooling marks out. 😁
Nevin Zimmerly Possibly. But they’re suppose to be priced to be sacrificial anyway. And it looks like a silicone rubber which holds up to 500F without melting.
Occams, are you talking about a helical boring op? Looked like he did helical ramp, he didn't plunge. If you are I was thinking that too. MtD, what you said makes no sense to me Nevin, dont have to worry bout melting when you use coolant.
I think a better explanation of smoothing is to eliminate discrete line segments by fitting arcs to the geometry. You can often eliminate hundreds of discrete [approximated] line segments and replace them with a single arc, or few arcs. The tolerance determines how approximate the arcs follow the geometry. The upside is allowing the CNC control to do the math at runtime instead of at post time. Many newer controls can then apply internal accuracy models which can trade time for accuracy, or surface finish for time. This goes back to last week when you explained the tesselated tool path the software generates; the native toolpath is line segments instead of arcs.
Can't it also replace straight lines, for instance like a jagged within smoothing tolerances? Don't you find on older machines like your Fadal to keep as much math out of the control and in the post? I know I do.
I take exactly the opposite approach. I want the code size to be as small as possible, so I force the post to generate as much of the geometry as native G-code. The control is plenty fast enough, they aren't doing floating point, they are doing fixed point math with lookup tables for sin/cos. With a more modern control there is less of a penalty for code size, but if you are lazy then you can get in a situation where the control has to process a bunch of lines per second, and when the vectors are short, the path planning can get confused due to not having long enough vectors to properly calculate smooth accel/decel points. This is where _lookahead_ become crucially important. The lazier the post, the larger your lookahead needs to be in order to avoid jerk and bad surface finish. If you have a small fixed lookahead, then having tight code is important to getting good performance from the motion controller.
Interesting, I agree with a lot of that but my Fadal has barely enough memory to hold a small probing cycle so file size doesn't really matter to me since I'm DNCing w/ my Calmotion no matter what. I have an 88HS controller without AFF or anything and I don't use a ton of HSM toolpaths but my machine seems to run smoother with line approximated arcs for the most part. You say "lazier post" and I'm certainly not claiming to know best on this subject, far from it but I figured the machine has to do the math on the radius then do the line approximation on its own because it can only move in straight lines in practice and all that takes processing time. I'm prob wrong, but all I know is it seem to be how my machine runs best for some reason. Thanks for the info.
I have the CNC88 control, it processes 1 block at a time, with small pauses between them. My machine has 156KB, which is just enough to do something useful. I'm planning to replace the control with LinuxCNC and Mesa hardware, retaining the stock drives and resolvers. Tormach based their control on LinuxCNC, with a different UI -- they also contributed a better trajectory planner, which really shows when watching these NYC CNC videos.
I think I can see how you got that initial file over 7MB. But the machine I work on. That file will be reduced to 256Mb. Because I don't have a choice. :(
Hi john. Why not get a mold made by protolabs. Probably only $3 -4k for the mold and piece part price would be super low. Plus you can get 5000 odd pieces from a mold and if parts go out of tolerance they re cut the mold for free. Huge variety of materials
I am not a fan of this solution as it is going to take too much messing around cutting and fitting this stuff for each configuration. I have the aluminum plate, it is going to take me some time but I am going to make .5 x .3 steel slugs to fit each hole. They will sit flush in the holes, I will simply use a magnet to snag it out and put in a stud when I need to use a hole. That means even under a vise every single hole will be covered and below the surface of the table so it will never bother fixturing.
Have you tested the casting material for breakdown resistance against coolant ?
I'd like to know that also.
If I think about it again he probably asked SmoothOn themselves. I guess they must have done a lot of testing of their product for different compounds such as coolant.
Nitro RC you have such a variety of coolants for so many different applications however I don't think there would be any inherent problems with most of the one's commonly used by most machines. Unless you get some backyard wizard who mixes up something really acidic that eats it up and swears it's the best coolant ever made. Lol
I thought the goal was just to prototype with the Smooth-On, and then you could injection mold them after you're happy with the design.
I dont think the tormach flood coolant is strong enough but I wonder if a VMC flood would knock or lift those covers out?
I learn something new every time I watch one of your videos. Keep 'em coming!
Love it. I've always been amazed how clean you keep your stuff, but this is a great idea for a guy like me that would need as much as I can get to make cleanup quicker. Nice to know about the Smooth-on too...
I didn't make complete doughnut circles, but I made some posts that are similar to the posts in the centers. This was on a manual milling machine. First I milled off the top of the piece so that I had square (more or less) posts sticking up. Then I positioned my spindle where I wanted a round post center, and in one cut brought the tool down and left the round post. I used a boring bar, and turned the cutting edge inside so it reduced the size to a precise diameter post instead of the opposite operation of enlarging a hole. I was cutting plastic. I believe that with the right cutting tool in a boring bar, you could do the same thing to make your mold like drilling a standard hole, instead of the tiny end mill.
I miss Smooth On! It was so handy when I lived in the US...
Great idea! You're really making a nice set of products around those fixture plates!
The reason we call it smoothing is it doesn't just create arc segments. It can also create long line segments through a series of shorter lines.
hey did the persons vac work to degas? I would love to use shop air and get rid of my vacuum pump that is just taking up space for the few times we do mold poring (or wood stabilizing)
I just bought 5! I’ve been wanting something like this ever since my first big project with the fixture plate.
Great idea John!! You will sell tons of those! Thanks for all you do!
John my first thought would the hot chips burn through? I know you all do mostly aluminum but running steel with indexable tooling and no coolant is throwing off blue chips. Might melt. There has to be a heat resistant rubber right? Awesome stuff brother! Appreciate your videos.
Great idea. And another awesome video. Thanks for all you do John.
Great video. Seems like everytime i am working on a project you or grimsmo releases a video dealing with the same issue i may be having. Just made my first mold for vacuum forming some thermoplastic.
A better idea than the magnetic vehicle signs I used. They worked great but cleaning the chips from them was a pain. How does this material hold up to hot chips? Likely not an issue with flood coolant but could be if you run dry.
I wish o had cnc and a fixture plate. I will be adapting fast fail, cheap fail to my life, and thank you my friend foe Mr he awesome video.
I don't comment much, but I had to share the love I have here
Smooth-On is great if you only want to make one or two parts but if you are doing it in bulk they are not the best most cost effective company to use. BJB would be a step up in terms for mass producing them and there are better companies up from there. For optical clarity you need a rigid resin, degas the resin, place the mold in a pressure chamber for cureing. Optical clarity really isn't a thing in flexible materials.
Nice, Not sure there is any benefit in going to metal moulds for a poured/cast end product. I typically would cut this out of machine wax for the master positive and pull out a silicone mold. Wax is Fast and reusable if you change your mind. You can then pour 10 silicone moulds off the master wax for ramping up to production, eliminating extra machining and any part release issues. Thanks for the video. I always enjoy watching
I loooove sikablock.
Thank you for the extra metric data.
John, Something I have been wanting to learn more about, I wish you would do a video on thread wires/measuring over wires. I wish i knew more about it, and how it all works. I have always had the luxury of ring and plug gauges. I also think some of the guys starting out could learn how to check threads without needing an expensive gauge.
I think Edge Precision discussed it some in one of his vids. It's has a great CNC channel if you aren't already aware of it. I'm sure Mr. Pete has covered it also at some point also. There's really not much to it. The hardest part is holding the damn wires with one hand!
Steve just Steve a little grease holds them well. ;)
Nice making mold's is really good stuff to see . mold cap for fixture's awesome :) happy new year :)
you should make some that can go on a bridgeport mill cut to fit the slots and pockets. and so you can cut them away for a vise
Great editing in this video
Fantastic video and product John! Happy New Year!
my jaw dropped down to the core of the earth when I saw 122 m/min then I realized it was not your feed but your tangential tool speed haha
light media blast mold to get an even finish, go fine as you can and still get an etch. that way later on you can easily touch up imperfections in the mold. or high polish if you think sticking may be an issue.
Realy nice, what shore hardness did you use? I'm lacking a little bit of feel for that, I have some understanding for the soft stuff from working with silicone, but PU rubber is often more on the harder side.
And how did you manage air bubbles that result from the casting process itself? I imagine that it could trap some air when flowing into those small pockets and I saw you poking around there with a stick. I heard you can compress those in a pressure chamber to a size that's barely noticable, maybe that's a good idea for you.
Great product, and great price point!
Great idea John!
Congrats on realising RPM is already plural, or has someone been getting on your case?
How much air are you pushing through the fogbuster here?
Nice, I love using Smooth On products.
Great idea for a product. I bought the Tosa plate for my 770 before I knew you had an offering. They have a bazillion plugs and I ditched that idea right off in favor of just placing aluminum plate on the uncovered spots. I've not compared the threaded and milled hole specs between theirs and yours. I'm hoping they're the same and I can buy some of these from you. Being able to cut sacrificial or special shapes is great. Thanks for the video and new product.
I'm in the same boat. Both plates have the same hole spacing which is good. But the tosa plates have a .562 reamed hole, while Johns has a .500. So I'm guessing it will work, but the nubs that go into the holes wont be snug. I doubt flood coolant will be much of an issue, but if you have air blast like me, then they will probably move.
Thanks, Shawn. I had just gone back to check spacing and it indeed looks the same with the exception of how many holes are threaded, like you say. If SMW does come out with the product I'll give it a try and report back on its fit with Tosa.
I ordered some this AM.
Here's the link: saundersmachineworks.com/collections/tormach-fixture-plates/products/fixture-plate-cover-protectors
Ah, didn't realize he already had the beta units for sale already. Just ordered some myself.
amazing job!!!!
will it fit a tosa tools plate
That's a great idea John :) and cheap to boot ;)
Thank you for saying RPM instead of RPMs. That always drove me crazy.
IT should be "RsPM". Now speak that out loud a few times...
Don't know whether it's a UK thing but we generally use "Revs"
I always think of this at the ATM machine while I'm entering my PIN number.
R W A.. OK. Revs per what?
That first hollowing out action, why not use a larger diameter tool bit?
He explained that.
Err, and tape wouldn't work instead of a molded guard?
What about making some more of those molds and renting them out?
There is a difference between degassing and outgassing, if you pull too much vacuum you actually evaporate the chems inducing air bubbles or cloudiness. Low/mild vacuum for epoxy type chemicals works best. I do like the black.
He's using a silicone product. I can't read the part number on the bottles, but I'd guess it is platinum based. Smooth-On likely has detailed instructions on proper degassing for the product. www.smooth-on.com
Right, for casting resins, you are supposed to use a pressure pot if clarity and bubble free are your criteria.
Still wondering why you’re so against helical ramping in small pockets...
Probably could have done those cavities in 1/2 the time and 1/10 the code size.
I know time wasn’t an issue. But a helical ramp would have also evacuated the chips well too.
And scotch brite that main flat surface too to get the tooling marks out. 😁
chronok
Low HP spindle. Adaptive paths maximize the machines RPM and feed rate and gives a better MRR over traditional paths.
Will hot chips melt in them?
Nevin Zimmerly
Possibly. But they’re suppose to be priced to be sacrificial anyway.
And it looks like a silicone rubber which holds up to 500F without melting.
Probably due to autodesk sponsoring the channel? Doesn't make rational sense to do adaptive on literally everything.
Occams, are you talking about a helical boring op? Looked like he did helical ramp, he didn't plunge. If you are I was thinking that too.
MtD, what you said makes no sense to me
Nevin, dont have to worry bout melting when you use coolant.
What about molding a mold that can mold molds?
1337fraggzb00N That would definitely speed up the molding production. Mold positive molds from negative machined mold to produce the finished molds! 😏
Mold ception
But! How many chucks can a wood chuck chuck if a wood chuck could chuck chucks.
Ron Hunn You might have to contact Jet Lathe for that answer.
We also need a mole. A mold molding mole.
I think a better explanation of smoothing is to eliminate discrete line segments by fitting arcs to the geometry. You can often eliminate hundreds of discrete [approximated] line segments and replace them with a single arc, or few arcs. The tolerance determines how approximate the arcs follow the geometry. The upside is allowing the CNC control to do the math at runtime instead of at post time. Many newer controls can then apply internal accuracy models which can trade time for accuracy, or surface finish for time. This goes back to last week when you explained the tesselated tool path the software generates; the native toolpath is line segments instead of arcs.
Can't it also replace straight lines, for instance like a jagged within smoothing tolerances? Don't you find on older machines like your Fadal to keep as much math out of the control and in the post? I know I do.
I take exactly the opposite approach. I want the code size to be as small as possible, so I force the post to generate as much of the geometry as native G-code. The control is plenty fast enough, they aren't doing floating point, they are doing fixed point math with lookup tables for sin/cos. With a more modern control there is less of a penalty for code size, but if you are lazy then you can get in a situation where the control has to process a bunch of lines per second, and when the vectors are short, the path planning can get confused due to not having long enough vectors to properly calculate smooth accel/decel points. This is where _lookahead_ become crucially important. The lazier the post, the larger your lookahead needs to be in order to avoid jerk and bad surface finish. If you have a small fixed lookahead, then having tight code is important to getting good performance from the motion controller.
Interesting, I agree with a lot of that but my Fadal has barely enough memory to hold a small probing cycle so file size doesn't really matter to me since I'm DNCing w/ my Calmotion no matter what. I have an 88HS controller without AFF or anything and I don't use a ton of HSM toolpaths but my machine seems to run smoother with line approximated arcs for the most part. You say "lazier post" and I'm certainly not claiming to know best on this subject, far from it but I figured the machine has to do the math on the radius then do the line approximation on its own because it can only move in straight lines in practice and all that takes processing time. I'm prob wrong, but all I know is it seem to be how my machine runs best for some reason. Thanks for the info.
I have the CNC88 control, it processes 1 block at a time, with small pauses between them. My machine has 156KB, which is just enough to do something useful. I'm planning to replace the control with LinuxCNC and Mesa hardware, retaining the stock drives and resolvers. Tormach based their control on LinuxCNC, with a different UI -- they also contributed a better trajectory planner, which really shows when watching these NYC CNC videos.
Awesome I hope your able to film some of it. I'd love to watch that! Cheers
That's brilliant!
Pressure pot on the clear John and it will look great!
You should collaborate with Precious plastics!
cool product BTW
so why don't you fix the tram on the machine?
Probably a lack of time. That's why I have not fixed mine.
I think I can see how you got that initial file over 7MB. But the machine I work on. That file will be reduced to 256Mb. Because I don't have a choice. :(
John, are you auditioning for the old school FedEx commercials? (I'm dating myself here...)
Hi john. Why not get a mold made by protolabs. Probably only $3 -4k for the mold and piece part price would be super low. Plus you can get 5000 odd pieces from a mold and if parts go out of tolerance they re cut the mold for free. Huge variety of materials
I think its because he is a machine shop and thats what he does. Thats like a mechanic taking his car to a mechanic.
we sometimes have Programs as long as 3 million lines ;)
Make a master positive then you can produce cheap negatives, then your product. I used to use Hapco products.
I am not a fan of this solution as it is going to take too much messing around cutting and fitting this stuff for each configuration. I have the aluminum plate, it is going to take me some time but I am going to make .5 x .3 steel slugs to fit each hole. They will sit flush in the holes, I will simply use a magnet to snag it out and put in a stud when I need to use a hole. That means even under a vise every single hole will be covered and below the surface of the table so it will never bother fixturing.
Please get rid of your 4 flute for aluminum go with two flute
now you need a robot to fill those molds with resin.
You should consider selling a silicone mold for those of us who already cast our own resins.
1st!