harryball
07-14-2007, 01:00 PM
Let me ask my questions first, is there a way to eliminate the small hesitation/pause when calling a file with FP? I'm calling a file 32 times with offset each time to produce grooves. Between each call is a minor pause causing ramping.
If not... is there away to call a block of code internally with offset? i.e. instead of FP with offset create a loop with offset based on the final bit position at the end of each loop.
Now to share...
I've cut 28 sheets in a little less than 2 weeks. That's big production for me, remember I'm the guy that cuts a small groove every 1/2" over practically entire sheets, some on both sides. Anyway, the experience has sent me experimenting today since I didn't really want to work. I've been further tuning and getting braver with messing with ramp values. I thought I'd share.
I was able to cut more than 7 minutes off of a production file just with the ramping adjustments. Mine were all defaulted to .4. I was playing with slow corner speed which was set to 35 and I did move a couple of items slightly but never much. I decided to get brave... or dumb one. Let me say, I only changed values on the VR fill in sheet for this adventure, all my other settings and jog speeds remained the same.
I chose a file component that represented mostly pre-drilling and ran it. It was taking 2m31s to run. I then started altering jog ramping since it accounted for most of the moving. After a few minutes tweaking and repeated runs I settled on settings that runs it in 1m38s. Thats huge when tossed into the mix.
I then started playing with the move ramps. Slow changes but continued tweaking. I ran a grooving file this time, one that runs a zigzag of grooves every 1/2". The first time I ran the file for baseline it took 2m7s to run. After my tweaks I'm at 1m45s. Not as impressive but considering I run this pattern 6 times in one production file it adds up.
My actual numbers are not important, it's the fact I was willing to tweak to see what would happen. What is important is my methodology which I'll share.
It's important to have a baseline from a representative file to begin tweaking. This is so you'll know if you are making real gains. I suggest tweaking jog ramping and move ramping separately. This calls for at least 2 different files, I used 3 myself. I used a predrilling file for jog tunning since it's mostly jogging hole to hole. The other file does continuous cutting so that was for move ramping.
I figured out my jog ramping can be tuned and it's probably safe to leave them as my default settings. I MUST be aware at my speeds I need to always eject the network card when running files. I don't want any polling going on while I'm jogging. I tuned past my current values but noticed diminished returns after changes and much harsher tool moves. I settled on numbers that for me are comfortable and the tool is not excessively banging around.
My move ramping is a different story. I found some optimal settings for doing roosting grooves (a .015" deep groove cut with a 5/32" bit every 1/2") but those same settings seemed too aggressive when I ran my cutting files. This makes sense, there is practically no resistance on the bit and coming up to speed quickly is easy. But when I've got a 1/4" compression bit cutting into 1/2" material in a single pass... I better take it a little easier.
In the end I determined good VR values for roosting grooves and added them to the file so I could change them, then change back for regular cutting. I also found some tweaks for my regular cutting as well (my 3rd test file). Though not as aggressive it saved about 10 seconds on my sample file. I made these my default settings. Even so, I may find they are more aggressive than I'd like when I cut my 1.25" cedar sides or cabinet jobs.
For the final test I ran my full production files in air cut and compared before and after tuning... WOW! Before total elapsed was 27m35s and I was proud to get those numbers. I've done all I can do with minimized jogging, grid cutting etc... all the toolpath tweaking that can be done was done. After really shocked me at 20m8s!!! I had just cut 10 sheets with this production file, that's a savings of more than an hour of run time. I was pleased to find at least modest savings on most and a few dramatic savings like this when I air cut my other production files.
Spending a little time tweaking values can pay off. Adding VR tuning to specific files that can handle it pays off as well in production type files.
Now I know someone is going to ask me for my numbers, but I've thought about it and I'm not going to tell you... my numbers are not your numbers because we are not cutting the same thing. I think it's important you not plug in someone elses numbers and run with it thinking it's safe. I can tell you my numbers for cutting roosting grooves is not safe for cutting anything else. While my bot, which may be different than yours, runs smooth and quietly with my numbers cutting my file yours may not.
I found that in no case changing a number slightly caused disaster. I could easily see the trend after just a couple of changes as well as see and hear the tools reaction. I also made my changes with a leading 0 i.e. default is .4 and I entered 0.4, this helped me identify if I had actually changed the number or if I'd forgot to change it.
So take some time and give it a try.
Robert
If not... is there away to call a block of code internally with offset? i.e. instead of FP with offset create a loop with offset based on the final bit position at the end of each loop.
Now to share...
I've cut 28 sheets in a little less than 2 weeks. That's big production for me, remember I'm the guy that cuts a small groove every 1/2" over practically entire sheets, some on both sides. Anyway, the experience has sent me experimenting today since I didn't really want to work. I've been further tuning and getting braver with messing with ramp values. I thought I'd share.
I was able to cut more than 7 minutes off of a production file just with the ramping adjustments. Mine were all defaulted to .4. I was playing with slow corner speed which was set to 35 and I did move a couple of items slightly but never much. I decided to get brave... or dumb one. Let me say, I only changed values on the VR fill in sheet for this adventure, all my other settings and jog speeds remained the same.
I chose a file component that represented mostly pre-drilling and ran it. It was taking 2m31s to run. I then started altering jog ramping since it accounted for most of the moving. After a few minutes tweaking and repeated runs I settled on settings that runs it in 1m38s. Thats huge when tossed into the mix.
I then started playing with the move ramps. Slow changes but continued tweaking. I ran a grooving file this time, one that runs a zigzag of grooves every 1/2". The first time I ran the file for baseline it took 2m7s to run. After my tweaks I'm at 1m45s. Not as impressive but considering I run this pattern 6 times in one production file it adds up.
My actual numbers are not important, it's the fact I was willing to tweak to see what would happen. What is important is my methodology which I'll share.
It's important to have a baseline from a representative file to begin tweaking. This is so you'll know if you are making real gains. I suggest tweaking jog ramping and move ramping separately. This calls for at least 2 different files, I used 3 myself. I used a predrilling file for jog tunning since it's mostly jogging hole to hole. The other file does continuous cutting so that was for move ramping.
I figured out my jog ramping can be tuned and it's probably safe to leave them as my default settings. I MUST be aware at my speeds I need to always eject the network card when running files. I don't want any polling going on while I'm jogging. I tuned past my current values but noticed diminished returns after changes and much harsher tool moves. I settled on numbers that for me are comfortable and the tool is not excessively banging around.
My move ramping is a different story. I found some optimal settings for doing roosting grooves (a .015" deep groove cut with a 5/32" bit every 1/2") but those same settings seemed too aggressive when I ran my cutting files. This makes sense, there is practically no resistance on the bit and coming up to speed quickly is easy. But when I've got a 1/4" compression bit cutting into 1/2" material in a single pass... I better take it a little easier.
In the end I determined good VR values for roosting grooves and added them to the file so I could change them, then change back for regular cutting. I also found some tweaks for my regular cutting as well (my 3rd test file). Though not as aggressive it saved about 10 seconds on my sample file. I made these my default settings. Even so, I may find they are more aggressive than I'd like when I cut my 1.25" cedar sides or cabinet jobs.
For the final test I ran my full production files in air cut and compared before and after tuning... WOW! Before total elapsed was 27m35s and I was proud to get those numbers. I've done all I can do with minimized jogging, grid cutting etc... all the toolpath tweaking that can be done was done. After really shocked me at 20m8s!!! I had just cut 10 sheets with this production file, that's a savings of more than an hour of run time. I was pleased to find at least modest savings on most and a few dramatic savings like this when I air cut my other production files.
Spending a little time tweaking values can pay off. Adding VR tuning to specific files that can handle it pays off as well in production type files.
Now I know someone is going to ask me for my numbers, but I've thought about it and I'm not going to tell you... my numbers are not your numbers because we are not cutting the same thing. I think it's important you not plug in someone elses numbers and run with it thinking it's safe. I can tell you my numbers for cutting roosting grooves is not safe for cutting anything else. While my bot, which may be different than yours, runs smooth and quietly with my numbers cutting my file yours may not.
I found that in no case changing a number slightly caused disaster. I could easily see the trend after just a couple of changes as well as see and hear the tools reaction. I also made my changes with a leading 0 i.e. default is .4 and I entered 0.4, this helped me identify if I had actually changed the number or if I'd forgot to change it.
So take some time and give it a try.
Robert