Starting values for the PRTalpha

[richards]

Would someone who has a PRTalpha with a Colombo spindle check my assumptions?

(1). To cut plywood (baltic birch) at 10-inches per second, I'll need to ramp the cutter into the work. The PDS Colombo site says "When programming your work-piece, entry angles should be kept from 0º - 20º off the table. This will keep axial loading low and allow feed rates at 100%." My math shows a ramp of 2.06-inches for .75 inch plywood meets the minimum requirement. I plan on using 2.0 to 2.5 inch ramps on a straight-line portion of each piece and a 2.0 to 2.5 lead-in cut tangent to parts that don't have a straight-line portion long enough for a ramp. (All ramps will be hand-edited using tool paths generated with PartWizard.)

(2). Using that same chart from PDS, drilling at 90-degrees, shelf-holes for example, the feed rate should be reduced to 5-percent of normal, or .25-inch per second. Is that the optimum speed for maximum spindle bearing life?

(3). Assuming a chip load of 0.012 inch (using 120 inch-per-second, 10,000 rpm and a single flute cutter for the older PRT machine) would I run the spindle at full speed with a two flute cutter to approximate the chip load if I cut at 10-inches per second ((600 inches per minute / 24,000 rpm) / 2 flutes = 0.0125 chip load)? BTW, the Onsrud catalog suggests 18,000 rpm at 400-1000 inches-per-second with a 1/2-inch 2-flute cutter in plywood and 100-600 ips at 18,000 rpm for their 1-flute cutter in plywood.

(3). Which Onsrud cutter should I use? 60-121 1-flute 7/8-CEL, 60-123 2-flute 7/8-CEL , 60-123C 2-flute 7/8-CEL chip breaker, 60-123MW 2-flute 7/8-CEL maximum wear? All other things being equal, I would like the cutter that gives the longest life at the highest feed rate and the smoothest cut. (I would also like to have the eye sight that I had when I was twenty, the body that I had when I was twenty-five, and the mind that I had sometime in the past - but I've forgotton just when.)

(4). What is the expected cutter life at the higher speeds? Assuming 10-12 hours cutter life at 2-ips feed rate and 9,000-12,000 rpm in plywood, should I expect roughly 2 hours tool life at 10-ips?

(5). Is the 3-hp Colombo adequate for the job, or do I really need the 5-hp spindle? My business plan is centered around 3/4-inch sheet material (baltic birch, MDF, and melamine coated particle board). Shopbot said that they have done extensive tests with the 3-hp spindle at 10-ips in plywood with no problems. Normally I would accept that without question, but I really don't want to buy the 3-hp spindle and end up buying the 5-hp a few months down the road. However, the problem is that I am limited in power. I have 240VAC single phase at 50-amps available without paying $2,000 to run another 100-amp line. Because I'll be moving into a shop with plenty of power in 6-8 months, I would prefer to use the existing power without modification and I shudder at the thought of going house to house in my neighborhood with an extension cord in hand asking if it would be possible to borrow an hours worth of electricity. I have very good neighbors but, at that point, some might think that it would be better to help me find my lost marbles.

Any other suggestions/comments about the PRTalpha would be appreciated.

[richards]

Is it lazy fingers or a numb mind? Question 3 should have been 120 inch-per-minute
(3). . . using 120 inch-per-second, . . .

Also note that 3 must be my favorite number, thus having two questions labeled (3).

[elcruisr]

Michael,
I run a PRT with an Alpha upgrade. I've run the Alpha upgrade for about two months now and you can run at 10" per second but remember a few things. That figure is straight line speed, it's going to slow down for each corner so don't plan on cutting the entire perimeter at that speed. This is true of all CNC machines. On complex parts (and I run 'em by the thousands) the real speed measured in time vs. perimeter can drop to more like 7"/sec.

I run a 5 hp spindle. Never had any experience with the 3 hp spindle. I'm sold on it enough that I just bought a second one as a back up.

As for drilling I do peck drilling at 2"/sec. My spindle has around 5,000 hours on it and is about due for new bearings. Maybe slower drilling would have made it go longer but I find that is acceptable. I also didn't do ramped entries for about the first 1500 hours and that probably didn't help the bearing life. (Artcam Insignia will set up ramps for you) I'm told by the techs they usually rebuild at 5,000 to 7,000 hours. Remember the warm up and cool down cycle!

I use 60-121s from Onsrud since I've used up several hundred of them and have them pretty well dialed in. At those speeds I find 16,000 gives me a good result. The 2 flute version is OK as well and can be run at LOWER RPMs. As for life span I find I can cut 5 to 8 hours in some species of plywood, less in the nasty stuff like virola and luan and MDF is hard on edges as well. The MW series of tools (long life) have never seemed to run very smooth for me, one of their experts is looking into it as he thought the geometry should be the same, no answer yet though. Resharpened bits also seem to have about 75% of the life of a new tool. We resharpen two to three times.

You'll need a good hold down system. We upgraded to a 15 hp FPZ vacuum blower and I consider this to be just adequite. A bigger blower or better yet a vane pump would be great.

Leave yourself plenty of time to learn the machine and software. If you think you will be in full production in a few days, you're in for a suprise. There's alot more to running good production numbers than it seems untill you try it. Just remember this is a great machine but it will only do what YOU tell it to do. I've found over the last two years that it's been me that had to climb the learning curve! Not the machine...

Eric

[gerald_d]

Eric, you mentioned that you can run at 10" per second, but what is now your preferred speed with the Alpha compared to the Classic?

[richards]

Eric,

Thanks for the quick reply and for the information. I have a couple of 60-121 that I bought for testing when I assumed that I'd buy the pre-alpha machine, so I'll try them first. 12,000 to 18,000 rpm is the highest torque output on the spindle (according to the charts) so 16,000 rpm is great.

When I realized that .25-inches per second was the "ideal" drilling speed, I got really concerned. That means 1.5 seconds per shelf-pin hole plus jog time to raise the cutter and move to the next hole; however, I hadn't even thought about peck drilling. What is your depth per peck? Do you raise the cutter 1/4-inch or so above the material with each peck to clear the chips?

I wish I had enough power available to use a good vacuum hold down, but I'll have to settle for tabbing until I move into the new shop. Tabs will cripple production. In fact, cutting the material free with a jig saw and cleaning off the tabs at a router table will probably take longer than cutting most of the parts on the Shopbot in the first place; but, I'm hoping some of the customers will perfer tabs to ease the transport of parts from my shop to theirs (full skeletonized sheets would be a lot easier to handle than lots of fairly small parts).

When you slow the feed speed down for a curve or corner, do you ramp down in several steps, or can you go from 10-ips to 5-ips or 6-ips in one speed change without losing steps? I guess the same question is also valid for increasing the feed speed as you leave a curve or corner - one big leap from slow to fast or several incremental steps? My experience with stepper-motors in designing and programming process control computers for the photo industry showed that I could accurately stop a motor in three decremented steps, but ramping up to full speed required significant care. Also, I'm guessing that your curve/corner speed must be somewhere around 1/2 of your straight-line speed, given the average speed of about 7-inches per second.

I'm sure the first few months with the machine are going to be "interesting"; however, given my thirty years in the computer industry, I have learned to keep accurate records of all tests and to increment the difficulty of each test in small steps. Learning the Shopbot will probably be much like learning a new computer language, at least I'm hoping that will be the case - old men and new tricks, well you've already heard that story.

Mike

[elcruisr]

Gerald,
I run 10"/sec on most sheet material with simple rectangular parts. I slow down for plastics and sheets with very complex parts.

For instance I have one production run I do on a regular basis. This is a 48" x 96" sheet of 3/4" ply that gets cut into 40 parts in a very tight nest with little waste. Each part is composed of short arcs and short straight segments. Each also has six sloped tabs to keep the nest together. I can set any speed I want but working through all the short runs and tabs the machine will hit about 6" per sec max before ramping into the next corner. The faster jog speeds do save a very large amount of time. Total cut time has gone from around 18 minutes to just under 9. That's money in my pocket!

Cutting solid hardwoods, which we do occasionaly, I'm still learning the limits of. It's more of a hold down issue with the board than anything. But I have so far been cutting solid 3/4" oak, maple and lyptus in two passes at 6" to 7"/sec with a single or double flute compression tool.

Eric

[elcruisr]

Michael,
We set up our peck drilling in Artcam, just click the option in the drilling routine. I usually will drill in .25 to .375 increments and clear .05 between pecks. Keep the rpms way down for drilling, we typically use 6500 to 7500 to prevent burning. Also look at the cnc drills offered by onsrud, cheaper than a 1/4" router type bit and designed for the job.

Ramping threshhold, speed and distance is all controlled in the Shopbot software. I'm finding that tuning these values has been my biggest learning curve on the new Alpha. I seem to use one set of values for large simple parts without tabs and a slightly different set for my small complex ones. A few conversations with the tech department once you have a machine will let you figure out which stuff to experiment with to fine tune for optimum performance for your parts.

About tabbing slowing you down, it dosn't need to. Here's how we run. We have an outfeed table at the end of the 'bot. As soon as a cutting cycle is done we pull the sheet on to this table. Vacuum off the spoil board, reload and start another cutting cycle. We then turn the cut sheet over on the outfeed table (use enough tabs to hold up to this) and cut all the tabs with a laminate trimmer with a 1/4" solid trimmer bit while the next sheet is cutting. No lost cycle time and minimal handling time. Parts go from there to pallets and out the door. I think our record day was a little over 60 sheets of larger parts.

Eric

[richards]

Eric,

It's good to hear that the Shopbot software handles all of the ramping parameters. Manually adding ramping code to files is more than even an old programmer wants to deal with. (Missing even one turn/curve/corner could throw off the steps enough to ruin the part.)

Thanks also for the advice on removing tabs. I usually find the slowest way to handle every project. As you know, my method of using a jig saw to cut through the tabs really requires two people - one to cut and the other to move the sheet around so that the tab being cut hangs over the edge of the workbench. After that, using a table mounted router to smooth the tabbed areas takes more effort and space than your method.

I'll check the Onsrud site for cnc drills. Right now I'm using a Freud 5mm solid carbide router bit with a 1/4-inch shaft for shelf-pin holes (part number 04-512, I think. I'm not at the shop this morning).

Mike

[gerald_d]

I am having difficulty understanding why you want to jigsaw tabs afterwards? Why not use:
- a single tab per part,
- right at the end of the cut, i.e. don't cut right around
- tab running parallel to the board outer verneer grain
- tab length about 1/2 cutter diameter
And then manually break the tab?

If you had a ramped entry, the tab is less than 1/2 the thickness of the board - it will break easily without tearing the grain. We normally clean these broken tabs with an angle grinder, or a bench-mounted belt/disc sander

[johnny_s]

Eric L-
Do you have some pics of your upgraded prt? How much different, if any, is it than a brand new purchase? I noticed that the pics of a new machine have different springs on the z return. Also, the upgrade specs say hardware to stiffen the x carriage - not the y as well? Oh, it says to stiffen, but at the bottom of the sheet it says to "upgrade" x&y carriages, add $$$. So what is it exactly, and is it worth it - besides the speed?

[ron brown]

Johnny,

The best way to get the latest information is to ask ShopBot directly. The "new" ShopBot is different in several minor and yet major ways. "Is it worth it?" - that would depend on what one is doing. My "original ShopBot" with the small Unistrut and running on spring door rollers, with a 12V operating system will build a sign as well as a new "Alpha", although a bit slower and go through more bits doing it. But, at 60 miles an hour and 100 feet off the road, I doubt you could tell which sign was cut on which machine.

The person with superior skills and experience will beat the person with a better machine almost every trip unless the project demands the speed and accuracy of a "better" machine. I find knowledge the only thing that can't be purchased from a vender of bits, machinery or software.

I think I will upgrade my machinery when I find I need more production than can be done on one machine in two shifts. Right now, I doubt the machine runs more than 4 hours a day average. It is quite possible that having two slower machines might be a better choice than one fast machine. I don't know. I do know it will be a decision directed by a spreadsheet and not just the desire for the latest and fastest machine.

Ron

[johnny_s]

Ron-
I totally understand what you're saying. I originally bought the thing for 3-d work, but have only cut one file on it, due to my lack of cam programing knowledge. It gets a lot more 2-d use, but I have been slowly learning checking out different cam packages, for which, a better bot would be nice.

That's why I am seeing if the upgrade is worth it. I have PRT 120, and might consider just getting a 96 alpha and swapping motors and use the 120 with porter cable for oversize panels. I dunno, ideally, one solid - upgraded - machince would be best, but not if it only half the machine of the alpha.

I, like you don't use the machine all day long. But it was my wife that urged me to go 120 over 96, and I am glad I listened. At least for me, it pays to have the capacity when you need it.

[elcruisr]

We upgraded an existing 5' x 12' PRT. We used some steel channel iron as transverse stiffeneing for the legs, a worth while addition even if you don't upgrade. We also upgraded the carriages along with the new Alpha Steppers. The new carriages are much stiffer and have less vibration. With a 5' wide machine moving at 10"/sec. I feel that this was an important improvement.

The coil springs on the Z-axis are an improvement over the flat springs we had when we started two years ago. The Columbo balances better and this is important when your z speeds are hitting 7"/sec or more.

Sorry, don't have a picture at the moment. I might be able to get one next week.

Was it worth it? For us a resounding yes. We just started a contract for close to 50,000 parts on top of our other regular work load. For production work the upgrade was a no brainer. If time was not a critical issue than for me the closed loop steppers would still have me looking at it pretty hard. I don't think I would buy a new one to replace an old PRT but, me personally, I would upgrade the carriages along with the Alphas.

Eric

[johnny_s]

Okay, guess i'll be posing some more questions to SB directly, thanks for all the help guys.

[andrewm]

Eric,

How do you go about pricing with the new alpha? Shop time is obviously less if you figure that into your quote, but with the apha you can do more in less time? Just curious.

Andrew

[elcruisr]

We sort of split the difference. We went up a little on our hourly rate but are producing more parts per hour. This has resulted in a slight reduction in the cost per piece for our customers and we are still making good money.

Want a happy customer? Tell him you just saved him a little money....

Eric