After the long thread on drilling and peck drilling I thought I'ld post this video for you as an alternative way to drill holes that seems to be faster than peck drilling for many of my applications. Yes, I'm doing this dry. The darn day job keeps interfering with my hobbies and I still have not finished my enclosure. It ramps in at about 5IPM and finishes the pass at
20IPM. The cutter is turning about 15000 RPM.
Yes, they are slightly out of round due to backlash and run out, but not enough to matter for most applications. If I needed them perfectly round I think I'ld pocket them .005 to .010 undersize and then drill or ream as needed.
Yeah, I'm sure there are a million reasons why this is a bad idea, but for a ton of applications it works great.
At 15,000 rpm and 5 IPM, you are annoying the metal out of the way, not cutting it. You want to cut the chips, not rub it to death. Maybe less rpm? Or maybe clean the hole out under size at a faster feed rate, and then a slower finish pass to clean up the hole to size. Speed might not affect your backlash anyway. But the servo tuning might affect it's ability to keep up to the feed rate.
Here is an example of a (small) full scale machine. A Robodrill. Just for grins... In aluminum, I was running 24,000 rpm, and pocketing at 180 ipm with a
3 flute Zirconium cutter 3/16" Diam .28" deep. Spiral plunge into the bottom, and interpolate the hole. Holes from .500", .345" .280" .214"
30 holes in 28 seconds including tool changes and rapids all over the part. Within a few 10ths over Thousand of parts.
You have an AHHA control? I used it for years. Great control for its time.
I made a Mach parallel port to RC500 AHHA control box cable and was able to switch from Mach to AHHA with just a cable switch. At the time Mach (1) was not as good as AHHA. I tried again when Mach2 came out but very shortly later I upgraded to servos and a closed loop control.
From what little I've read Mach has come a long way. This would be a quick simple upgrade.
Actually, this is done a lot now, just peruse some videos on Youtube. I'd do it myself quite often, but one of the flaws in Ahha is that thread milling (the only option for this sort of approach) does not work well. When I have to threadmill, I drop back to an earlier version that sorta works...
Just one of the reasons I'm seriously looking to switch to either EMC or Mach3 in the near future...
Yeah, it's been a mostly dead reliable system. But it's getting hard to find cheap computers without any AGP slots and finding video cards that don't cause problems.
I'm sorta pondering building a dual boot system and trying both Mach3 and EMC. But it takes time to get used to any new system, and both will require modifying all my programs (probably close to a hundred), just not sure I really have the time to invest in that comparison. I am quite sure I'd be happy with either, just have to make the call...
I did a side-by-side shootout when I started futzing with CNC. Two identical surplus PCs, one with Linux/EMC and the other with Windows/Mach3 just switching the cable from the machine between the PCs. At that time it was EMC vs. Mach3 and Mach3 won by a healthy margin for "refinement", both worked properly, but Mach3 was far cleaner and easier to configure and manage. Now that EMC2 is available I expect it would end up largely in a draw with either solution working equally well.
I do this all the time. I make many equipment panels that have a wide range of cutouts (slots, rectangles and round holes) of varying sizes. So, I just put in a 1/8" solid carbide
4-flute end mill and cut everything but 4-40 and 6-32 holes with one tool. I have some routines that generate the G-code for generic rectangles and round holes posted on my web pages at
You just fill in the answers and it generates pretty efficient G-code for the operation. I then stitch the separate G-code files together to do the whole panel.
Perhaps your router can only handle these feedrates, I generally run about 10 IPM with only 2800 RPM. For the panels, I use a routine that "trepans" the hole, cutting just inside the final dimension, the slug drops out and then it makes the finish pass. I generally plunge about .050" each step down with the 1/8" end mill, so I do a 1/8" panel all the way through in 3 passes.
Say, i went looking for how I built the Mach to AHHA converter cable. I couldn't find it. But I did find when I chopped that cable up to try Galil. So, the notes below have columns that explain all the AHHA function. I had used a 37 pin rainbow color ribbon connector from digikey.
This should be worth what you paid for it.
DB37 connector from 1900 to Ahha step driver (brown tape on first 10 wires, red tape around 2nd 10, Orange around 3rd 10, Yellow around last seven)
Wire color db37pin 1900 pin Function Ready to run Logic state Output on (GRN=LOW RED=HI) limitswitch On brn - brn 1 17 Grnd brn - red 20 34 PWMX brn - org 2 33 SIGNX brn - yel 21 31 PWMY brn - grn 3 30 SIGNY brn - blu 22 28 PWMZ brn - vlt 4 27 SIGNX brn - gry 23 25 PWMW brn - wht 5 24 SIGNW brn - blk 24 74 GRND _ OUTS GRN
red - brn 6 66 OUT1 Auxout1 - Drive enable on RC500 GRN RED E stop on pendant powers out1 drive enable pin
6 GRN RED red - red 25 67 OUT2 Auxout2 - Mister GRN RED red - org 7 68 OUT3 Auxout3 - Spindle CW GRN RED red - yel 26 69 OUT4 Auxout4 - Spindle CCW GRN RED red - grn 8 70 OUT5 Auxout5 - not used now GRN RED red - blu 27 reserved by Ahha - don't connect GRN red - vlt 9 reserved by Ahha - don't connect GRN red - gry 28 reserved by Ahha - don't connect GRN
red - wht 10 55 GRND limit switch ground GRN red - blk 29 52 RLSX Axis 1 minus limit switch GRN RED org - brn 11 53 FLSX Axis 1 plus limit switch GRN RED org - red 30 49 RLSY Axis 2 minus limit switch GRN RED org - org 12 50 FLSY Axis 2 plus limit switch GRN RED org - yel 31 46 RLSZ Axis 3 minus limit switch GRN RED org - grn 13 47 FLSZ Axis 3 plus limit switch GRN RED Relays in cablinet all NC to pin 10 LS GRND
org - blu 32 Axis 4 minus limit switch RED org - vlt 14 Axis 4 plus limit switch RED org - gry 33 55 GRND Grnd for Auxins GRN org - wht 15 65 ABORT Auxin1 - ALLWAYS E-STOP GRN *need to change state above Galil says abort input active org - blk 34 Auxin2 RED
yel - brn 16 Auxin3 RED yel - red 35 Auxin4 RED yel - org 17 reserved by Ahha - don't connect RED yel - yel 36 reserved by Ahha - don't connect RED yel - grn 18 reserved by Ahha - don't connect RED yel - blu 37 AUxin8 Nothing yel - vlt 19 56 INCOM +5 volt DC output Nothing
Hmm, maybe, but I've gotten used to what my own routines do, and I can modify them when I feel like. I now have a rectangular routine that ramps down, I have to add that to my circular one, too. But, sometime I have to check out some of these routines and see if I like them.
usually .062 to .125" thick. No, not loud at all. There's a slight hum when the cutter is in the work.
I know the C code for it, no question you could do a G code sub, but it would be just a little complicated.
But, a while ago I ran into some SERIOUS problems with subroutines. I had a slot that was to be repeated many times, so I programmed it once and offset the work coords to repeat it. EMC would cut it OK, although the Axis preview was not right. But, EMC2 was NOT able to restart the job in the middle at all. You could set the correct parameters to the right state for a restart, but when you did the "run from line" it would just go to the end of the program, following all the subroutines. I had to give up and do it without subs.
I don't know for sure if this would behave the same way in your use, but it put me off subroutines in a big way. This was just last year, or maybe even early this year, so I don't think this has been fixed.
John Kasunich's comment on this problem was "subroutines are evil, don't use them!"