Flip Milling Process

[groucho]

I laminated several sheets of rectangular 1/2" marine plywood sheets together to form a 3 1/2" thick slab. The first goal is to clean up the vertical edges with a 1/2" carbide end mill with two flutes. I want to create a clean block with true vertical faces on all four sides so I can flip mill it with accuracy. This is my first experience using more than the first 7/8" of a bit and it didn't go so well. I oversized the stock about a 3/16" +/-on all four sides and thought the 1/2" bit would have no trouble trimming out the block as it dropped 1/2" per pass and ate about 3/16" into the wood as it cut rectangular toolpaths around the perimeter of the stock. I ran the router at 18000 rpm using my 2.5 HP spindled PRS Standard machine. Here is a picture of the block after the procedure and another block before being trimmed:

2949629497

The bit only had 2 1/2" of actual blade so I didn't cut all the way down the side of the block and cleaned up the remaining material with a flush trim bit in my router. In the attached picture you can see the rippled face where the shopbot milled it and the smooth face where the router bit surfaced it.

Maybe this procedure is inherently loud because the bit is eccentrically loaded in a trimming operation but from the very beginning of the tool path the bit was loud trimming out the material and by the time it was almost done the bit was really loud. It finished the routine but the bit wound up sitting on the part at the end of the process. It was still in the collet but it turns out it had wiggled loose a bit. I'd like my next results to be more like that router bit and a lot safer :). Maybe the answer is self explanatory. I need to tighten up the bit more when I play with a 1/2" bit instead of my usual 1/4" fair?! Maybe longer router bits also just produce a less quality cut because they have to extend so far out from the spindle? I'm really hoping this kind of performance is not something I should get used to. The large bits are expensive and I'd like to produce something with more accuracy. I wound up tossing the collet and the bit. The bit was chipped on the end.

2949429495

I've purchased another bit. Can anyone give any advice on how to do this right? Maybe make toolpath passes in smaller vertical increments? 1/4"? Maybe I just need to make sure that collet nut is tight? Any help is greatly appreciated.

[steve_g]

Michael…
This is asking a lot from a machine that cost less than 100K... But it can be done using the proper techniques! First, the part must be held down absolutely solid! Second, because of the tremendous mechanical advantage the wood has when using a long bit, take smaller bites. Third, use a “onion skin” cutting strategy. By “onion skin” I mean cut over size about .01” with a climb cut, followed by a final finish pass in the conventional direction.
I don’t know how old your machine is… but it must be “tight” for success with this operation. Try these suggestions and let us know how it went! Oh, and your bit must be TIGHT in the collet!
SG

[jerry_stanek]

Also a spiral bit may be a better choice

[steve_g]

Michael…
I’ve been thinking about your question and re-read your post… somehow it got past me that this was cut with a ¼” bit. I’m not sure you will ever be happy with the cut quality using this bit! Was your flush trim bit a ½” one?
The adage “when you’re a hammer, everything looks like a nail” may apply here. If you have a table saw, it’s the better tool for this job! If you’re determined to do the job on the bot, a half inch spiral bit would be a good investment!

https://www.magnate.net/ProductDetails.asp?ProductCode=2011

https://www.centuriontools.com/router-bits-toolcase/solid-carbide-wood-router-bits.html?sku=12US22.1254&pcn=Upshear&pid=779&sat=1

Also, a down cut bit would be a good investment…
SG

[groucho]

Sorry I wasn't clear. I AM using a 1/2" bit.

[groucho]

29498

I do have a tablesaw and I was worried about the imprecision of the tall vertical cut but I take your point. Much more expedient and I can always measure things with the calipers and take my time to get it close. Yes, I definitely have my "shopbot goggles" on these days. :)

Attached is an image of a cup design (only one half of it actually). This is what I'm ultimately heading toward producing in Fusion 360. The machine hasn't been used a lot and I think it is in relatively good shape (with the possible exception of that bit incident :)). Do you think flip milling a cup part like the one I'm showing is possible if I onion skin as you say? It measures about 3" tall.

Thank you so much for your response.

[steve_g]

Michael…
I’m not a fan of flip ops… I’ve used steel dowel locater pins, and other “precise” locating methods, sometimes it works and other times with the same jig, the result is less than stellar. I just finished a chair design that I spent “extra” time designing to avoid cutting from two sides. I’m convinced that my poor success rate has nothing to do with my locating method and everything to do with the nature of wood… it moves when stress points are cut! Sometimes you have to think outside the box… here’s a 10” deep “Vase” I made by stacking rings and cutting the exterior on my indexer…
SG
http://www.talkshopbot.com/forum/attachment.php?attachmentid=29499&stc=1

[Burkhardt]

Right now I am working on a set of very thin walled lamp shades 3.5" diameter and 3.7" tall. In a way similar to your project but with a kind of tulip shape, i.e. the wall have varying slope (not vertical).
For such things (or similar deep bowls) I usually cut the inside cavity of all the parts first without much regard to positioning. Then I machine a plug from scrap wood which is exactly the negative shape of the cavity but only 1 or 2 inches tall and drill a hole in the middle for a vacuum connection. Now I can machine the outside of all parts one after the other by seating the cavity onto the plug, if necessary with a thin gasket on the bottom. This guarantees precise position of the outside relative to the inside, even if the wall is less than 1/10" like my lamp shades.

But even with a 1/2" bit it becomes difficult to go deeper than maybe 4 inches because the long leverage often makes it chatter and pit the surface. As mentioned previously the straight flute bit that you have been using also contributes to noise and vibration since it smashes the entire length of the flute into the material in one impact.

[groucho]

29505

Interesting. That is actually very similar to what I'm working on (see fusion rendering above) and would be very interested in any wisdom you have to share. That plug idea is great! Wish I had a vacuum table but it sounds like you might even consider it with a shopvac. Just started my second pass on the cup tonight with a much less noising bit. I think the trick with this vertical milling work is to just do much less ambitious passes than one usually would when one is cutting panels and working flatter. This small pass approach just seems to traumatize the bit less as the mechanical advantage of the work imposes itself on the tip of the bit. The wall of this cup is a 1/4". Glad to hear you're able to get the wall as little as 1/10" of an inch.

[Keith Larrett]

A different approach to this project would be to slice the model into seven 1/2" sections. Just cut the seven cross sections out of a 1/2" sheet of your marine plywood, glue them together and all you will need is a bit of light sanding inside and outside of the vessel.

I'm not familiar with Fusion 360, but I've used the slice function in Aspire on a project when I didn't have a long bit available.

That's a nice looking vessel.

[Burkhardt]

The plug fixture makes very thin walled cups possible (as long as they don't have any cracks or knots..). When I cut the plug on the machine, which must be tight fitting but not a press fit, I drill a bit size hole in the middle as future x/y zero-reference (also serves as vacuum conduit). You must register inside cut and outside cut very precisely or the wall will be uneven and the cup look cock-eyed.

The other advantage of this method is the part comes off the machine as a finished shape (no tabs of onion skin to cut off) and you can use 90+% of the wood without much loss.
I am not sure if the shopvac will do it, since the holding surface area is already small. For my lamp shades it is only 7 square inches which comes out to 60-70 pounds vertical holding force at 20 inches of vacuum. A Shopvac has much lower vacuum capability which may be O.K. for spoilboard/sheet good duty but probably not enough for such a pod. But a small oil-less vacuum pump is only about $150 on eBay.

[Burkhardt]

Maybe a few pics should explain it better:

1. Roughing the inside cavity with an end mill
https://lh3.googleusercontent.com/nKHdHkCH-fx04KB79ac-ehMg4LklXNp7L60loeSe7FWghicTAtkUf-qh0O9oTIbLGetF8h23ObRfHE6WntM33DUCU-_m9ltRSHe42a2RMMRcbslahl1psWYp2O6PTPEkRtoDeSvvg8Gu Zh3I_9LJZuINkwi3aSFGasMRaxJZvBhY7fNvqMhOpKNUHkQa_J 6HcR3YNd2ht5eymIp2bG7mWoT5BbjGdPwVNMALE6kMsRLiPk1k dq1BAQUH5VdNPyKdeS7DRN2hhkdTPc-595DDQz4vEiBEVQizjcpgAyiJ_X6TZL1pCH6WCCr994ikkA4yc rVvHhe4w-3u598w_XTFjz9Adl6gxkStH2FBp_bmoypPdA1vhgu4LSdHx9mZ 1wjp8S9du5qiEXFiQvBjVVP-4YTuByqgzH1kSRxrOxNoNnzZradhJ7Mz-1Bz_xLz0zSY5923tnd9Az9OeH8ATTrbbzO9KGVky11rDXk6-cgHzh6H9xks0g_TmYk1DS3JcIX-EDAVgxpXegIjj3iKuenVQNWLOPDxR8k-jgKqAwA3y8zs3QBf0OgYxdBzvH8LCQOLO66Km9sCXKiy8dsmrN d2xY8E9lyxVd_5JiIV6Cv4wCMrEUWx7rp6=w1172-h950-no

2.Finish cut inside with a 1/2" ballnose:
https://lh3.googleusercontent.com/sqT00JyPMXiJMl5es25Uf-iAjPogfzMax5l9Zgehp4ypo9VvygMFxKOMyh184ozMH33BZyL-O5ZRS58Kk5lIYQsa0mnn8QeoXnFGu0nTWkEJZ2rHjD2iRvBUeo 3FIFpMwD7Ob7R9HkE9wzSh3pkzjTP5GG2UitDB6RPFMb58uo88 ab7No482VnlRXJAw77IxxE2iQ-zDtE5bRxQaG-EkQ9Td3lrlU49b9zttnS4pyB0GeGQ6Z40Bb0rx4sSHbkbGRchX k6SEYkOqWBB0qbnQ5p4HtSk7iBkLMLgIloh5TBSEgRrfPdSFLZ-0QZDUeGgfHX5Wb3pLvy4HCt4y3yh-LyIq0mjfFh1tLhLC6fEVLzRYxxejyUeZdsqBUBctAFk_asoiBr dbqlo2S0HJ8fSm1gm7nctIdtYrgNn2FoSoSDW3PcXgy70KwCM4 ojdMxljh86FfUSUX30L6TtV3GWRKb3ZtzCJPdJADU_fXYMYFlJ u6DRSw1IIk_6rwjSSpOSLfcE2YY3w7ForjV-D2Vvtw8SfpPe4LrpeAapOIsAaY3ZPa0WV62WN9HDVXv2y7qA6A do-KzpPklyw5pyLqZlpNu2UI6ipgOlWOFCVYTvf091w6lGxt=w925-h950-no

3. Vacuum plug fixture (the gash in the side is the result of an accident...). The gasket was originally a bit oversized and clamped in between the plug and the bottom plate. It was machined to size with the first part cut.
https://lh3.googleusercontent.com/zvElpNmjGqtNbGBHb669KRXhMJ_aaNxKZtaqVmzzibuBNZI0qZ AlruOel34Tfn1oVJN_z9-MxCH_b_eMXgaV6KDABRtHb-ezTuD4VsssMMOsXv2wd3iQw-C7FkxjQ-7VMfGfw5P0uAzwTF-t6X3tw64fYfTgdS34V8BCP6NOutLBV43JmmwPyPZvMXrjLhUqB EgftMsjXlEgycR3laPBBML-qkrfY7OPYz0WHfgz1zBFBSdknpgEcyykdOAW2glg7OHtTgCALJ nP6Ak4yVqULk9gLqs7p_eZ9J1vYUQiwKKHiQeoTtXILix10Ith 3kTEu9a4lNCYNCTR0cQUEHyKhiu1XOETTew5TzbP-loQ5_JsTzp-cuLgME-hPKsnQyp9QwZ0UGgaui2zg2nTorS8P-bTG5wQdgyV89oTl74pGY-Km7MlfEXNGPz93Je5_2PMk-7mAPS9l1pshJ_em2Ypd5iGVwdjA57JqZfHjuPV2YMG6Cl61Vdz EPNf71dKmwQI7-tnul3275Zf9x8D0NA3JzjbZIhR-3RIwBiEE_y5wVHfE7z-PqsuzPWg5UMQSZPc_Wcr3FZ7wnDOKNn8WCCYGwDLA9-WDAzGFzJAy5Ux1q3_7Moi=w1494-h950-no

4. Small lamp shade from redwood (bottom wall only 0.08")
https://lh3.googleusercontent.com/KY48KNf4IPms5bc8Dx62XSUv8_uquS7iNw1OU1Fg0EcbioQq9L CRPbWJyghKS4FmdkU4_IT1tsL7hOXrcRgew1MNV36Al4kyNtWH swz6yJoaL3ek1KeimBvaJJo70TVUckNCaeFVgNpa9AYeYv7iAh BrWk4_0cc5EAvhaU7MAnJJdsrdIkUtW4FPH_Hm5M6R2QBojuRn Fv4f6Xxkbnamo7j-qKJVpCHt6BaVpNClNLiczsX3lj_0GTW_bxBWytzWMlZuc7cGJV JyORgYw4Vq6rIhlE8VRhWmdAfw4XvlN7bF0RE-hyFsGvVHSXZ3Ca8ewWuGIESh8kd3qqh4w1XvGXz5bS3zwG1nsi f4AGKVvDoo8hV1ga6MH-rykzJNPOBU9zrGzEq_YZ8hp4W8dNXs0cXJqOwfvNxzFPCSK6rK RShPirMw5dwI8mPmchBFtwcJrRCZWX3T7LA3TIWIO_TRjlavTE Un2DefKcjqs6fxUM4yv00epyeH7mRtSqCQ-dPDfKZXTOEy6HzbzeMO8ypeXHLsh6hIPE3iR0momyBMk7GT7bp 17MUx3oM_dqyupyXw8FcLP0uXPNPmf0DLc2cxyjeFQ_58MZ8w1 ntZEjixufAGnAds=w1171-h950-no