PDA

View Full Version : Holding full sheets when cutting many small parts



mikejohn
02-25-2005, 03:45 AM
I will be cutting many different small parts from 4mm and 6mm Birch ply (B/B).
In the past, I had these cut for me by both laser and CNC router (the router produced the better finish). In both cases the companies asked for a maximum sheet dimension of 300mm x 600mm. So at the moment my files fit these sizes.
However, I would cut about 3 full size (2440mm x 1220mm) sheets for a single production run from these files.
In other threads on vacuum hold down, it is normally stated that cutting small parts can be a problem. It would obviously be easier to lay down a single sheet, and let the ShopBot cut away for 2 hours (my estimate at 25mmps). I have been advised to use a downcut, clamp the 0,0 end, and use the bread slicing method of cutting. That is, start furthest away from the clamps and cut towards them.
My alternative suggestion was to make holding jigs approx. 300 x 500 and use these, setting up one whilst the other is cutting.
Does anybody cut similar small parts (maybe 1400 from a single full sheet)? How do you hold down?
.........Mike

gerald_d
02-25-2005, 04:15 AM
Mike, could you mail me a dxf of your 300 x 600 layout? (Is the grain direction important?) what size cutter do you plan to use?

srwtlc
02-25-2005, 01:02 PM
Mike,

Hers's what I have done numerous times. I don't have a vacuum table so I've made some dedicated jigs that I use a 6hp ShopVac with. Here's a shot of one that has been used to cut 1000's of canoe paddle keytags. They're about 3" long x 3/4" wide.


5115
5116
5117


It's made from two layers of 3/4" MDF with one layer having the plenum and the other the holding frame with vacuum openings cut to just under the finished part size. I cut my 1/8" baltic birch into blanks that fit into the jig. The paddles have a tab on each end to keep intact a web that keeps everything held together because there's not a lot of vacuum surface. I use a 1/8" downcut spiral bit at about 0.8"/sec. I'll often make another holding frame for a different part (matching canoes) that I can just drop into the base jig.

Scott

gerald_d
02-25-2005, 02:26 PM
Mike kindly mailed me his .dxf file and his problem is not going to be easy. Many of the parts are tiny with a maximum dimension of only 0.5". The part at the top right is only 0.1" wide for the narrow section:


5118

mikejohn
02-25-2005, 03:16 PM
Gerald
the file was cut succesfully for me by someone in England on a small machine.
It was this experiment that finally got me to order my ShopBot.
We do a lot of work on the finished pieces.
This is an example of one automaton' showing the cat with the very thin parts.
I had to save it at very low resolution, so its a bit naff.

5119
5120

His head follows up and down as the washer woman pounds the clothes.
The washing blows about on the line, and chickens peck in the background.
This is only a prototype, 'real' boxes will be made out of hardwood (using a stepped joint, cut on the ShopBot,I hope !)
Will I need more than one tab on thin parts?
How about long, thin strips of plywood, the full length of the Shopbot (96") held down along either side of the long length by some type of T track jig?
If the strips are only 200mm wide, surely they wont lift?
Or are there other problems I am unaware of?
..........Mike

mikejohn
02-25-2005, 03:38 PM
even by splitting a picture into three, the result is not great.
This is a contemplative automaton, although he has just been disturbed by a phantom door knocker!
Again, a lot of small parts, like the slippers.
........Mike

5121
5122
5123

zeykr
02-25-2005, 03:43 PM
We're cutting some pretty small parts out of 12" x 24" 1/16th ply on a shopvac powered vacuum jig.


5124

The smaller parts are .118 inch think by .785 inch long with a .035 dia hole. We leave two half height tabs per part and cut slow with a 1/32 circuit board endmill.

Our vac jig is a little more generic than Scotts (I like his design!) and consists of a plenum cut from 3/4 mdf and sealed with PU and an upper bleeder board of 3/4 mdf faced on both sides and glued to the plenum.


5125


5126

As long as we can keep our 1/16th ply from warping too bad before we get around to cutting, it works very well. We store the ply in the house (shop is not very climate controlled) and keep a heavy board on it to keep it flat.

mikejohn
02-26-2005, 01:01 AM
Ken
That is seriously small parts! Much smaller than I need to do. I do notice your material size 12" x 24" (30mm x 600mm) is the same size as my original thinking. What speed do you run at with such a small bit?
Somebody needs to explain to this knowledge limited Englishman how this vacuum hold down works!
I understand that the material can be sucked flat when placed on the vacuum jig, but as soon as you start cutting you create a hole, the area of which will soon be the same as the area of the vacuum tube, when no vacuum will exist. I understand about onion skinning (I think) but this isn't how everyone works.
Vacuum obviously works. I just don't understand how.

Ken are you willing to share what those tiny parts are for?


.............Mike

paco
02-26-2005, 07:01 AM
Ken,
do you have to sand 'em?!

bleeth
02-26-2005, 07:40 AM
Mike:
"Somebody needs to explain to this knowledge limited Englishman how this vacuum hold down works!"
Vacuums suck.
"I understand that the material can be sucked flat when placed on the vacuum jig, but as soon as you start cutting you create a hole, the area of which will soon be the same as the area of the vacuum tube, when no vacuum will exist."
Using a small diameter cutting bit, Working with a fairly small overall area and correspondingly powerful vacuum force means overall piece won't lose grip. Note that tabs are still being used and the nature of a shop vac (drawing it's required air thru the hose rather than from an outside source means that as a bit of space is opened up by the cutting bit you are actually helping your machine. If you are using a shop vac to hold parts that are not being cut thru then it is best to have a "bleeder" valve to let air into the system.
Notice that our fellow botters generally speak of two distinct types of vacuum. Large volume/low pressure (such as shop vacs, feins, etc.) or small volume/larger pressure (operated from venturi controlled valve systems by electric or compressor driven pumps).
Without degenerating into a discussion of horsepower/amp ratings the large volume systems are best for systems such as Kens as his layout means that the voluum being pulled by his vac is greater than what can be pulled thru the cut gaps and the bleeder board.
If I were cutting your file with all the area of the board not being used in your parts as show in the file you sent Gerald I would probably screw it down, use down cutters, leave tabs and not even bother about a vacuum.

Dave

"There is more than one way to skin a fox"

zeykr
02-28-2005, 10:01 AM
Mike,
The parts are accessories such as control horns, motor mounts, servo mounts etc for Radio Control aircraft.

Running the 1/32 bit at .4 ips with 2 passes to avoid breakage in the 1/16th ply.

We use the same or slightly larger bits (.054")on the vac jig to cut Depron foam at much higher speeds for the rest of the kits. Depron is a 3mm or 5mm foam sold for insulation there in Europe, but it's imported to the U.S. just for R/C aircraft.


5127

zeykr
02-28-2005, 10:04 AM
Forgot to answer Paco!

Paco - no we don't have to sand at least for our application. We just divide the sheets and ship complete with the tabbed parts and the modeler cuts them out with their exacto knife.

zeykr
02-28-2005, 10:28 AM
oops - double posted.

mikejohn
02-28-2005, 12:07 PM
Interesting stuff Ken.
How precise are your small cuts?
your .118" is our 3mm. I will be delighted if I can get within 0.02".
2 passes in 1/16" ply (1/32" per pass) is .03" in Z, which appears accurate to me.
I wonder if any one cuts out male and female pieces, marquetry style, and how close a fit they get.
...........Mike

beacon14
02-28-2005, 01:57 PM
My experience is that you will acheive better results with a vaccuum, especially with the thin plywoods. The screws don't keep the material from vibrating at the point of cutting the way a vaccuum hold-down does. I think the vaccuum will give you cleaner cuts and longer tool life.

I have done male letters out of 3/4" solid wood and female pockets out of veneered MDF, and the fit was so close you couldn't see a gap

paul_n
02-28-2005, 02:41 PM
If you scroll around on this guys page, you will see the claim to cut veneer....
www.scrollsaws.com/images/CNC%20Router/Vacum3.jpg (http://www.scrollsaws.com/images/CNC%20Router/Vacum3.jpg)

Don't know, have not tried it !!

Paul

mikejohn
02-28-2005, 03:05 PM
Thanks Paul.
I don't want to try marquetry, I just use it as an example. I was really trying to find out the precision I can expect when my ShopBot (still somewhere in transit !) is up amd running.
..........Mike

zeykr
02-28-2005, 03:17 PM
Mike,
None of the parts we are currently cutting require close tolerance. These control horns slide together fairly tightly, though probably not as close as you'd be wanting, but I'd think .02 could be held.


5128

I suspect main thing will be taking your cutter into consideration so the parts match up - on software that does male/female pocketing the software considers this. Depending on your software and since you're not really doing pocketing you'll probably have to account for bit size in your toolpathing. Saw a thread on it here recently how someone was doing manualy in a cad package - there are other experts here that would be much better at addressing that than I.

If you want to send a file of a few of your parts, I'd give it a try just to see.

mikejohn
02-28-2005, 03:34 PM
Ken
In fact most of what I'm planning for the ShopBot doesn't require fantastic accuracy. With the rocking horses, it's only the cutting out and joining of some of the parts, where 0.1" is fine for this. Its all hand finished, and its only in a few areas I need even this accuracy. For the automata, I also have some 'slack'. It would be nice, though, to have some very intricate mechanisms, with very sutle movements. Maybe here 0.02" would be nice!
I realise that the right combination and choice of bit, rpm and ips are very important.
..........Mike

gerald_d
03-01-2005, 12:58 AM
Mike, we have done "marquetry" (didn't know there is fancy word for it) Overall size probably 100mm with narrow sections about 4mm wide. (There are 2 boardrooms in London for which the furniture was made in Cape Town - the company logos are in the chairbacks and table). For these our client (the joiner) supplied the 2mm thick veneer glued to melamine faced chipboard. He lifted the pattern off the melamine, not us. The fit was totally perfect. As the first sample, to give him confidence, we cut formica inlays for him - also totally perfect fits.

mikejohn
03-01-2005, 03:16 AM
Gerald
That is what I wanted to hear.
In case anyone else following this thread is as slow thinking as I am, let me tell you what has only just occured to me. It is obvious once thought through, and certainly obvious to anyone already operating a ShopBot.
When I look at the remains of the 300mm x 600m plywood that was CNC cut for me, I have a very weak piece of wood, with more holes than Swiss cheese, so I have been wondering how I keep this flat to the table whilst cutting.
I now realise, of course, this is not what I am doing!
Using the 'farthest away' sliced-bread method of cutting, the cutting is always happening adjacent to the remaining solid piece of Plywood.
It's similar thinking to tabbing. Cutting, say, a circle with a 50mm (2") circumference, after 50% of the cut you still have a 'tab' of 25mm. Even after 90% of the cut your 'tab' is 5mm.
So is it only in the final part of any cut you might get lateral movement of the piece? And if this is likely, is it solved by placing a small tab opposite the finishing tab?
I don't mind if you RAOTFL (I'm learning
) about my slow uptake. I'm posting this to help other new or soon to be owners who may also need to 'get it'.
-20C more snow
...............Mike

gerald_d
03-01-2005, 04:30 AM
Mike, I thought that I was preaching my bread-slicing sermon (http://www.talkshopbot.com/forum/messages/401/1469.html) to the dead (the connection to Vector probably slayed them). You seem to be the first one that actually understands the logic! However, you must realise that a heck of a lot of shopbotters don't know how to control the starting/ending point of their cut. (<------ Let's see if anyone bites on this provocative statement
)

"Using the 'farthest away' sliced-bread method of cutting, the cutting is always happening adjacent to the remaining solid piece of Plywood." This was also our major relevation - complete with light-bulbs, harps, horns and drums!
"It's similar thinking to tabbing. Cutting, say, a circle with a 50mm (2") circumference, after 50% of the cut you still have a 'tab' of 25mm. Even after 90% of the cut your 'tab' is 5mm." That is right, but only if the tab is attached to something solid. No sense in having a tab to a bit of flimsy scrap.
"So is it only in the final part of any cut you might get lateral movement of the piece?" Yes! "And if this is likely, is it solved by placing a small tab opposite the finishing tab?" Normally no. With tight nesting, an opposite tab will be to flimsy scrap. The solution is to leave a tab at the start/end point - simply stop cutting before you have gone right around.

This "final tab" can be refined so that you take the load off the cutter as you near the end. A trick that Sean came up with, is to plunge(z) the cutter over a distance(x/y) so that an inclined entry is made. Then when the cutter has gone right around and completes the profile, the thickness of material being cut progressively gets less and less. Now that I think of it, this is his trick to avoid leaving a final tab - to make the part release gently. Anyway, the absolute key is to realise that your offcut is your anchor.

If you do leave a final tab, you will learn to pick a position relative to the grain of the wood so that you can "snap" the tab (grain parallel to tangent on profile). Also, you will learn to put this tab on a "outside" part of the profile that is accessible to a sanding machine.

mikejohn
03-01-2005, 06:33 AM
Gerald
I completely ovelooked the fact that, if you start and end the cut at the point closest to the clamps, then a tab diametrically opposite is simply hanging onto a sliver of waste.
I like Seans plunging cut idea.
Over what x,y distance would he make his plunge? (depends on thickness of material, I guess). Does this mean his end point is beyond his start point?
How does he achieve this? Or is this a case of having one line above another line acceptable?

I am slloooooooooooowly catching on!
...........Mike

gerald_d
03-01-2005, 07:21 AM
Sometimes, a tab onto a good sliver of waste is planned to stop movement in the x,y direction, but waste is seldom any good to hold down in the z-direction.

The plunging cut is drawn in AutoCad, before the file is converted. I think he just draws a straight line in at about 45 degrees, in 3D space, starting from slightly outside the profile. The cut then finishes on the starting point.

There are no hard and fast rules about anything - if Nick, his pal is here for the day, Sean doesn't bother with the lead-in. Nick gets told to push the part down with a stick just before it releases. Also, if packed sawdust in the groove holds the part, then tabbing tricks are not needed. With this flexible approach, .sbp files are never saved - we only save the AutoCad files.

But, I think that we are drifting away from how other guys hold really small parts........
(We have relatively little experience in this area)

ron brown
03-01-2005, 07:38 AM
Gerald,

You aren't the only one using the ramp into and from a cut to leave a tab or, even cut the thing clean off. I use Rhino. There are several ways to achieve the same tabbing effect. It is also a good way to enter and leave a cut without having a plunge point in the file. That is an important consideration on the oldest framed machines that were not terribly rigid to begin with.

Ron

gerald_d
03-01-2005, 01:40 PM
Ron, I didn't think that we are unique with that, and ramping into cuts has often been mentioned here. But I do think it is pretty smart of a 20-yr-old to figure out something to do at the beginning of a cut that will make his life easier at the end of the cut - maybe it is just misplaced fatherly pride.

mikejohn
03-01-2005, 02:14 PM
Gerald
Can you change the cutting speed within a file?
If you have your entry ramp offset even just a couple of degrees in Autocad, you can create the .sbp file in Vector. It works, I just tried it in simulation. I wondered if, as you approach the last part of the cut, that is cutting away the ramp, you could slow the shopbot right down, you could cut leaving the smallest possible tab by stopping just short of the end of the ramp.
Do you think this is possible?

Ron
I don't think Gerald was claiming to have invented ramping
. I think he was just illustrating a novel purpose for it, (novel to me anyhow).
...........Mike

gerald_d
03-01-2005, 03:08 PM
I am sure you can insert a command to change speeds, or all sorts of other fancy things. But we never do. Also, I don't think there is a way of using code to say "stop just before the end".

Remember that the part comes loose when you are still one tool diameter away from the start point. You want to take the load off about 2 diameters from the start point. Dropping the speed will change the cut quality, maybe even leave a burn mark?

mikejohn
03-01-2005, 03:33 PM
Gerald, thanks
Incidentaly, you covered this 'cutting past the ramp' technique over a year ago. Maybe Ron learned about it from there

........Mike

gerald_d
03-01-2005, 03:40 PM
Nah, don't pull his leg like that - he is nearly as old as you!

fleinbach
03-01-2005, 05:05 PM
You can change speed on the fly by hitting the shift key and tapping the > or < to speed up or slow down.

mikejohn
03-02-2005, 06:33 AM
Gerald
I'm not that old.
Please keep my email request for ShopBot plans for a mobile commode for the workshop, with inclusive zimmer frame(walker), just between ourselves
..............

sales@fretworks.co.uk
03-02-2005, 03:21 PM
Mike

When you get your Shopbot up and running useing it in a metric setting be extremely carefully when you increase the feed speed
with the shift key and tapping the > to speed up, above the limit of the ShopbotPRT (90mm) it will stall and then disappear across the table
with dire results....... you will need to alter the speed settings in the "ShopBotW.prm" file.


Paul

mrdovey
03-03-2005, 01:09 AM
Paul...

I've seen Rick's vacuum hold-down in action and have been planning to build one for myself from his DXF.

Rick has used his little CNC router to build /tiny/ box-jointed boxes with sliding tops that need a magnifying glass to look at.

Fun stuff!

mikejohn
03-03-2005, 02:36 AM
Morris
Your reference to Rick's work is intriguing, and I would love to see examples of his work.
But I can't find any reference to Rick in this thread

Any chance on pointing us towards his vacuum hold-down?
..........Mike

fleinbach
03-03-2005, 06:40 AM
Mike

Here is Rick's site http://www.scrollsaws.com/

When ever someone is pointing to a picture from a site just drop everything past the .com to get to the home page.

By the way it's a very interesting site

gerald_d
03-03-2005, 07:00 AM
(See Paul Nielsen's post above on Monday, February 28, 2005 - 2:41 pm)

fleinbach
03-03-2005, 07:14 AM
Here are some very interesting bits for scroll cutting.

http://www.precisecut.com/products/specifications/srdkerfspecs.htm

gerald_d
03-03-2005, 08:18 AM
Ouch! isn't that the type of thing a dentist uses for root canal work? I have an appointment next week.


Notice that they want very high rotation speeds.

mikejohn
03-03-2005, 08:20 AM
Frank
They sound very interesting.
Does anyone have any experience with these?

Gerald
How come I missed it, but you found it?

........Mike

mrdovey
03-03-2005, 02:38 PM
Mike...

Frank supplied the info you're after (Thanks, Frank!)

Rick is known worldwide as a scrollsaw guru - and he delights in working /small/. After get got into (wooden) clock design, he bought a tiny tabletop CNC router to increase precision on items like clock gears. I've visited a couple of times, and never managed to tear myself away in less than six hours!

He's visited my shop (to eyeball the ShopBot) and we swap ideas and methods when we get together. He gifted me with a DXF for a vacuum table (shown on his web site) for holding veneer while cutting inlays, and I showed him how my vacuum clamping setup works. If you ever travel in this part of the world, it'd be worth planning to visit his shop - his web site just "scratches the surface".

...Morris

mikejohn
03-03-2005, 03:08 PM
Morris
My main hope for my ShopBot is to produce more complex mechanisms for my automata.
Once you've linked more than 5 or 6 gears/cams/scotch yokes/geneva stops together, you really start coming across problems with precision.
On the scroll saw I can make the individual gear chains, but the small innacuracies sometimes add up, causing a very stiff, even inoperative, movement.
In the coming weeks, when my ShopBot arrives, I hope to report some successes.
Reading all on this, and similar, threads, I'm tending towards some sort of jig with something like 300mm x 500mm sheets of Birch ply.
..........Mike

mrdovey
03-03-2005, 06:09 PM
Mike...

Rick used 1/8" (3.175mm) plywood for his top - and I'm planning to use 1/4" acrylic to make the top just a bit more stiff (and I plan on setting the top on a 3" x 3" support grid) - which should be overkill for the veneer I plan to cut.

If you're a wooden clock builder, then you're going to have a lot of fun with your 'Bot - and I'd encourage you to introduce yourself to Rick in an e-mail (He seems to enjoy talking about the solutions he's found for making clock run smoothly and there aren't all that many clock makers in the neighborhood.)

Yes, I think you're going to have altogether /too/ much fun. (-:

...Morris

mikejohn
03-04-2005, 12:39 AM
Morris
Thanks again.
It's not clocks I make, but automata. Models that move when you turn a handle. I posted some rather poor pictures earlier in this thread here (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=29&post=21337#POST21337)
They do not show the mechanism however.
Once I'm up and running, I will show the mechanisms, maybe in a blog.
..........Mike

bill.young
03-06-2005, 08:32 PM
Hey Mike,

Have you seen flying pig (http://www.flying-pig.co.uk) yet?

mikejohn
03-07-2005, 12:48 AM
Bill
I have indeed seen the flying pig website.
We also make paper models, and its a real challenge to get the mechanism to work well.
Larger wooden automata (like those shown here (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=29&post=21337#POST21337) in the poor photos), take a long time to make, so are expensive. We exhibit all around England at large county shows (state fairs?) during the summer with the Rocking Horses and the automata. Rocking Horses are big ticket items, so we make some smaller toys, wooden puppets,small wooden automata kits, and now paper automata kits, which sell to cover the expenses. (We also sell bags of rare Rocking Horse manure)
.
If I can figure out how to post pictures on a blog, I'll show examples.
..........Mike

gerald_d
03-07-2005, 02:40 AM
Mike, you can post your pics on this site (http://www.scapenotes.com/notes/messages/2020/2020.html) of mine, and then link from here. The method of posting is the same, except that the size limits are bigger (50kB, 600w X 300h) - you don't need to register. If you register, I can give you much bigger limits.

mikejohn
03-07-2005, 06:04 AM
Gerald,
Thanks. I've done a test (it works)
.........Mike

bill.young
03-07-2005, 06:21 AM
Hey Mike,

I've posted some info on adding pictures to blogs here (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=6720&post=21749#POST21749)