Hi there everyone,
A few years back I built a CNC router capable of 63" x 100". If any of you need anything done please let me know. Your design, my machine. I am only looking to help the community so NO CHARGE other than material and shipping, or if you have something that takes a rediculous amount of time to design/cut.
I am located in 91344 (SoCal).
Happy holidays,
Joel
A few years back I built a CNC router capable of 63" x 100". If any of you need anything done please let me know. Your design, my machine. I am only looking to help the community so NO CHARGE other than material and shipping, or if you have something that takes a rediculous amount of time to design/cut.
I am located in 91344 (SoCal).
Happy holidays,
Joel
my experience in CNCing a synergy horn
Can we talk about the Sketchup to CNC path and handoff interface? Importing someone else's design might be more work than you think! (to say nothing about shipping parts across country) Here is my experience with the process so far.
I drew a design in Sketchup and worked with a local cabinetmaker with CNC to make a prototype. That prototype cost me way too much - about 10 hours of his time. I think we can do much better on the design import next time but its a learning curve and we need to learn to trust each other and the tools to be comfortable.
I drew each piece of wood is a separate component in Sketchup. I exported each component as a separate (3D) DXF file. At the CNC guy's insistence, I manually added dimensions to each DXF file. (they apparently don't export from Sketchup) He imported each file, converted it to 2D and removed the dimensions after verifying them. He had a lot of trouble with lines not being joined which I think can be attributed to the 3D to 2D converions process and removing of dimension lines.
This is what I think would work much better:
*** strip each component down to a 2D outline dwg in Sketchup,
*** cut and paste it onto a cut sheet drawing leaving adequate (1") cleareance
*** export the Sketchup cutsheet dwg to DXF or import directly into VCARVE or ASPIRE (both seem to work based on my playing with Aspire trial edition)
Cutouts and holes in a component would be covered by the 2D projection but I haven't verified that yet.
The bevels in a horn are a special problem. The design intent is best conveyed with a 3D Sketchup drawing. Theoretically, bevels can be done by CNC but it again requires manual import and tool path programming steps. My guy simply cut the bevels on his Unisaw, which worked, although somehow we got slightly off on one piece. That was a disappointment to me because the horn flares is where I wanted the accuracy of CNC. I've since bought a V groove bit to do the 45 degree bevels that are common in my corner horn design but the horn has some non-standard bevel angles in it as well.
If this design import process can be smoothed out to where it takes only about 2 hours for a complete design, then I think its well worth the cost. If it takes 4 hours, I'll probably still do it but I would grumble. I don't mind my time in Sketchup doing extra work; that is play for me, but standing around watching someone else work is excruciating
!
So what are your thoughts on this process?
Jack
Can we talk about the Sketchup to CNC path and handoff interface? Importing someone else's design might be more work than you think! (to say nothing about shipping parts across country) Here is my experience with the process so far.
I drew a design in Sketchup and worked with a local cabinetmaker with CNC to make a prototype. That prototype cost me way too much - about 10 hours of his time. I think we can do much better on the design import next time but its a learning curve and we need to learn to trust each other and the tools to be comfortable.
I drew each piece of wood is a separate component in Sketchup. I exported each component as a separate (3D) DXF file. At the CNC guy's insistence, I manually added dimensions to each DXF file. (they apparently don't export from Sketchup) He imported each file, converted it to 2D and removed the dimensions after verifying them. He had a lot of trouble with lines not being joined which I think can be attributed to the 3D to 2D converions process and removing of dimension lines.
This is what I think would work much better:
*** strip each component down to a 2D outline dwg in Sketchup,
*** cut and paste it onto a cut sheet drawing leaving adequate (1") cleareance
*** export the Sketchup cutsheet dwg to DXF or import directly into VCARVE or ASPIRE (both seem to work based on my playing with Aspire trial edition)
Cutouts and holes in a component would be covered by the 2D projection but I haven't verified that yet.
The bevels in a horn are a special problem. The design intent is best conveyed with a 3D Sketchup drawing. Theoretically, bevels can be done by CNC but it again requires manual import and tool path programming steps. My guy simply cut the bevels on his Unisaw, which worked, although somehow we got slightly off on one piece. That was a disappointment to me because the horn flares is where I wanted the accuracy of CNC. I've since bought a V groove bit to do the 45 degree bevels that are common in my corner horn design but the horn has some non-standard bevel angles in it as well.
If this design import process can be smoothed out to where it takes only about 2 hours for a complete design, then I think its well worth the cost. If it takes 4 hours, I'll probably still do it but I would grumble. I don't mind my time in Sketchup doing extra work; that is play for me, but standing around watching someone else work is excruciating
!So what are your thoughts on this process?
Jack
One place that might really benefit from CNC 3D machining is the round to rectangular conversion where the CD mounting plate mates to the horn. I started with a .25" CD mounting plate but ended up with one .75" thick to optimize bandpass change for the mids (long story). That long a hole through the plate would optimally have a conical or exponential taper. If you could figure out how to do that on a CNC it would definitely be an advancement. OTOH this might be better suited for 3D printing. In any case, we'd need to standardize a mating surface at the back end of the horn to take advantage of whatever we arrived at.
Hi there Jack,
I haven't found the secret suace that makes this a "just hit cut" operation yet. My experience is that I can import DXF's and DWG's, but sometimes, they need a lot of extra work to join corners and perform general cleanup on them. My preference, is to simply start with a detailed sketch from whatever program, sketchup is nice, and then I can look at each part and basically rebuild it manually in vcarve. By building it in vcarve, I know that it "works" and is true to size. tedious? Yes, but it is reliable and it means I can add tool paths easily. So working from a projection in 2D is generally just fine.
Angles, bevels, compound curves are getting into 3D which are something that I am glad you brought up. I am happy to add any angle/bevel that I can do with a standard router bit (90*, 60*, 45*, etc.). Odd angles though are a problem for me. A) I dont know how to program them in an efficient manner, and B) because I dont know ho to program them efficiently, they end up taking hours to cut
, and C) I cant make any promises on them if I do them. I would rather use the CNC to rough cut something and then finish by hand.
For right now, lets stick to basic 2D shapes. I am offering this for free, so lets start simple I am open to ideas though. If we can figure something out...great...we all benefit!
I haven't found the secret suace that makes this a "just hit cut" operation yet. My experience is that I can import DXF's and DWG's, but sometimes, they need a lot of extra work to join corners and perform general cleanup on them. My preference, is to simply start with a detailed sketch from whatever program, sketchup is nice, and then I can look at each part and basically rebuild it manually in vcarve. By building it in vcarve, I know that it "works" and is true to size. tedious? Yes, but it is reliable and it means I can add tool paths easily. So working from a projection in 2D is generally just fine.
Angles, bevels, compound curves are getting into 3D which are something that I am glad you brought up. I am happy to add any angle/bevel that I can do with a standard router bit (90*, 60*, 45*, etc.). Odd angles though are a problem for me. A) I dont know how to program them in an efficient manner, and B) because I dont know ho to program them efficiently, they end up taking hours to cut
, and C) I cant make any promises on them if I do them. I would rather use the CNC to rough cut something and then finish by hand. For right now, lets stick to basic 2D shapes. I am offering this for free, so lets start simple
I am open to ideas though. If we can figure something out...great...we all benefit!No $$...just material+postage. If we need an occasional cutter we can add that too. But no $$ for my time or the machines time. The end goal is to get peeps out in the garage building and then listening.
You are describing a tunnel that blends a square opening at one end to a round at the other? This would be a small enough part to try 3d work on.
Not exactly although just that much might be an improvement.
We want the blending to take place over a longer distance than just the thickness of the CD mounting plate. What we would have to do is cut off the back perhaps 1.5" of the horn and replace it with a block that has been milled to the desired shape. The CD would mount to the back of this block. The front would mate with the further truncated horn and the milled insides would form the last 1.5" (or so) of the horn.
If you search for posts by HulkSS in this thread about 16 months ago you will find a definition of the profile of that tunnel. Hulk was going to use 3D printing.
http://www.diyaudio.com/forums/multi-way/88237-suitable-midrange-cone-bandpass-mid-unity-horn.html
We want the blending to take place over a longer distance than just the thickness of the CD mounting plate. What we would have to do is cut off the back perhaps 1.5" of the horn and replace it with a block that has been milled to the desired shape. The CD would mount to the back of this block. The front would mate with the further truncated horn and the milled insides would form the last 1.5" (or so) of the horn.
If you search for posts by HulkSS in this thread about 16 months ago you will find a definition of the profile of that tunnel. Hulk was going to use 3D printing.
http://www.diyaudio.com/forums/multi-way/88237-suitable-midrange-cone-bandpass-mid-unity-horn.html
2015 Plans
The horn designs I am working on all incorporate salami-slice construction. As such, their components are intended to be manufactured on a CNC panel router like the one you have built. These will be ready for prototyping sometime during the 1st. quarter of 2015.
What tolerances can you maintain when:
1) MDF Cutting?
2) HDF Cutting?
3) Aluminum Cutting?
4) Hole Drilling (location)?
N.B. Parts are nested. Do you have a vacuum table?
Regards,
WHG
The horn designs I am working on all incorporate salami-slice construction. As such, their components are intended to be manufactured on a CNC panel router like the one you have built. These will be ready for prototyping sometime during the 1st. quarter of 2015.
What tolerances can you maintain when:
1) MDF Cutting?
2) HDF Cutting?
3) Aluminum Cutting?
4) Hole Drilling (location)?
N.B. Parts are nested. Do you have a vacuum table?
Regards,
WHG
export/import process
I hadn't seen this post when I responded to the other one. Your process is almost exactly the same as that of the guy I'm working with and I understand the reason for it. Although tedious, it works.
I think the 2D projection import will be a distinct improvement over the 3D import I tried the first time. I will do whatever cleanup I can in the process of making the 2D drawings and I can do a test import into Aspire or Vcarve demos, if you tell me what to look for. I saw my local partner struggle with unjoined corners and extra lines. They creep into the drawing when you make too many changes....so the export/import process should include the original designer doing the cleanup.
Any horn will have non-standard angles where the horizontal and vertical horn flares meet even if the major angles are standard. They are easy enough to do on a table saw, with jigs. Its probably best to leave this step until assembling the horn and then cut to fit.
I'm set for this design and I'm on the east coast, so I won't be asking you to cut parts for me in the immediate future. But I would like to get that export import process optimized. Others may use it and it will help with my local CNC guy.
I can send you a Sketchup design of my 90x45 2way horn if you want something to play with for yourself. I should have the fat lady singing through it by the end of the holidays.
And there is that 3D challenge waiting to be taken on. Doing that blending I spoke about in my other response manually is my own next challenge. Results will be more predictable if we can get the computer/CNC to do it for us.
Jack
Hi there Jack,
I haven't found the secret suace that makes this a "just hit cut" operation yet. My experience is that I can import DXF's and DWG's, but sometimes, they need a lot of extra work to join corners and perform general cleanup on them. My preference, is to simply start with a detailed sketch from whatever program, sketchup is nice, and then I can look at each part and basically rebuild it manually in vcarve. By building it in vcarve, I know that it "works" and is true to size. tedious? Yes, but it is reliable and it means I can add tool paths easily. So working from a projection in 2D is generally just fine.
Angles, bevels, compound curves are getting into 3D which are something that I am glad you brought up. I am happy to add any angle/bevel that I can do with a standard router bit (90*, 60*, 45*, etc.). Odd angles though are a problem for me. A) I dont know how to program them in an efficient manner, and B) because I dont know ho to program them efficiently, they end up taking hours to cut
For right now, lets stick to basic 2D shapes. I am offering this for free, so lets start simple
I hadn't seen this post when I responded to the other one. Your process is almost exactly the same as that of the guy I'm working with and I understand the reason for it. Although tedious, it works.
I think the 2D projection import will be a distinct improvement over the 3D import I tried the first time. I will do whatever cleanup I can in the process of making the 2D drawings and I can do a test import into Aspire or Vcarve demos, if you tell me what to look for. I saw my local partner struggle with unjoined corners and extra lines. They creep into the drawing when you make too many changes....so the export/import process should include the original designer doing the cleanup.
Any horn will have non-standard angles where the horizontal and vertical horn flares meet even if the major angles are standard. They are easy enough to do on a table saw, with jigs. Its probably best to leave this step until assembling the horn and then cut to fit.
I'm set for this design and I'm on the east coast, so I won't be asking you to cut parts for me in the immediate future. But I would like to get that export import process optimized. Others may use it and it will help with my local CNC guy.
I can send you a Sketchup design of my 90x45 2way horn if you want something to play with for yourself. I should have the fat lady singing through it by the end of the holidays.
And there is that 3D challenge waiting to be taken on. Doing that blending I spoke about in my other response manually is my own next challenge. Results will be more predictable if we can get the computer/CNC to do it for us.
Jack
hmmm, interesting idea. I'll have to think about that.
that does seem like a good way to do the first couple of inches of horn to get the profile right. Why not the whole thing? In fact salami slices is how 3D printing works, although the slices are a good deal thinner in 3D printing.
So you set the horn down on the bench with the mouth pointing at the ceiling (or vice versa) and make slices with horizontal cuts spaced so that the each slice overlaps the next say 50%.
Then you make some small number of nests of slices, interleaving slices among the nests so as to leave enough clearance for the cutting tool and perhaps the tabs that are needed if the CNC doesn't have vacuum hold down.
But when you stack them up, you end up with a lot of smoothing to do. Perhaps you can do the slicing with a V groove bit whose angle matches your horn angle. Of course then you need more separation between parts because the V bit cuts a wide swathe.... which may just mean you need more nests.
Sounds like a fun design!
I've got an 11 liter 2-way monitor clone project that is stalled out at the cutting stage. Drivers and crossovers and veneer ready to go. Would be awesome to take advantage of your skills/offer and get to listening! Is this the type of project you would be interested in? How do we proceed?
BK
What tolerances can you maintain when:
1) MDF Cutting? -- usually around+/- 0.05 - but it depends on
- speed...if I cut fast the accuracy drops
- cutter...smaller bits flex more, and some cutters are trying to pull the piece off the table more than others.
- material...some mdf is different than others
- Part layout...if the webbing that connects all the parts on the sheet is too flimsy, the machine will simply bend the material as it cuts
3) Aluminum Cutting? -- No...waaaaaay to slow and messy...Plexiglas is okay though.
4) Hole Drilling (location)? -- I use a v-bit and have it locate points.
Honestly, the accuracy is very good, but not $100,000 machine great...not good enough for anything that would go in a nuclear reactor, outer space, or the human body
Everything that comes off the machine needs a little sanding. It is wood after all, and wood isn't perfect.I have had more issues with material not being uniform throughout, and material changes due to humidity.
No Vac Table...just clamps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.