Hi Jeffeb,
Isn’t 1/32 settings on ramps for steppers resulting in 6400 steps for revolution? All 3 jumpers in on drv8825.
Going in I thought it would be like 3d printing too where you hit the green button and walk away. I was wrong…a lot of people on here look down in shame on that. Unless your machine is fully enclosed in like bulletproof chamber or it is rolled outside I wouldn’t trust mine for too long.
Once, I did leave the machine running alone. Everything went fine but the router had to be left on not cutting about 2 hours. For some reason there was some sort of ash as if the wood underneath burned a little. Made no sense to me at all. I wasn’t cutting wood and the bit was retracted above my aluminum by 2mm. Nothing broke but I was confused.
Ah, I see what you’re saying. Yeah, that’s a mistake. It’s set to 200 steps/mm, but without the jumpers they are basically set to 6400 1/32 steps/mm… Yeah, that would break some stuff. Also, you’d end up with gigantic parts…
Thank you all for support!
I am learning and adjusting the settings. Cutting through wood and aluminum like trough butter now :).
I found good settings for aluminum using trochoidal milling, it is excellent and the cuts looks fantastic. Done quite some letters, squares and models without any problem. Started milling a small frame, all perfect at beggining and somewhere in the middle the bit started losing revs. I paused, raised the Z and cleaned the bit as small aluminum parts were fused to the bit. At that point continued the job but material fused again, I couldn`t “rewind” back the gcode for few seconds in repetier as it always returns when pause was activated.
Why would aluminum stick to the bit? Feed rate to high/low or bit is already damaged? How long your bits last usually?
Aluminum is a soft material in respect to other metals. It also has a gumminess property. It isn’t just shown when milling either. Try cutting or filing it. It just likes to stick. Not to mention it makes big chips naturally.
Either you’re feed is slow, your chip evacuation is not nearly good enough, or you’re cutting the wrong alloy. Go for 6061 t6. Or t6511 for flatbar is what I order.
1 Like
20 some odd hours! I did leave a couple gallon ziplock bags full of water sitting outside the cutting area. It’s also mdf. I have a hard enough time catching the stuff on fire on purpose! I think that was also one of the times I collapsed a 5 gallon bucket that sits under my dust deputy.
Thanks Kevin,
Seems like the alloy is not homogenous and on some areas the material has different properties (probably not). I don`t know exactly which alloy are the plates I got but I will ask in future for specific alloy and the one you mentioned. Will also play again with settings, decrease the step and increase the feed. Yesterday at one point was excellent and I thought nothing can stop me anymore 
.
Attached video of trochoidal milling letter “B” , 27 mm/s feed rate , 3mm Z step , 3mm/s plunge , trochoidal width 50%. I started with trochoidal step 4% and each next “B” had 1% troch step more. At 9% (not on video) the aluminum start sticking to the bit and I stopped.
I chose 8% trochoidal step as last best and started milling large piece when the alu sticked again on the bit somwhere at the middle. I saw on Kristian video, he had settings of 5% troch step and much higher feedrate(35 mm/s) and I will test those settings today.
Barry mentioned above lower feed rate but higher troch step (10%) with less doc. I guess will have to cut more and test what works for me.
Barry,
Nice picture. If you have time let me know how you made it?
Oh no, if your aluminum plate has different material properties in different patches I would be really nervous lol. It would have to have been fabricated by wild rednecks!
Wait so what end mill is that exactly? It better be a single flute!! Don’t even try double it doesn’t really make any sense. With how fast we are spinning those sharp edges around there is only a need for 1, any more is just asking for a clog barely, if any of us, have a proper mist lubricant system.
Your chip evacuation looks pretty good but not great. If you plan on doing a lot of aluminum get some loc line and a source of air. Way better than policing the machine with your vacuum.
If you’re running the ramps 1.4/arduino mega, you will find there is a speed limit. It could just be be especially since I have oscillation on estlcam on, but I max out at a certain speed. Probably my accel limit. After like 800mm/m for me I cant tell it circles any faster
Hi Kevin,
I am sure the problem is in the plate 
, hehe. Yes, I am using double flute 
as my single flute broke and saw few videos where guys succeded and cutting nicely on MPCNC. I do understand the single flute is way better for aluminum and chips evacuation thus I am already on the way buying them.
Noted related the evacuation system. Will get or print some loc line and have to find better air compressor as mine is quite loud.
Currently I am trying to make a 3D printer frame - mk2.In case I start building 3D printers than I guess cutting them with laser or other ways will be much cheaper and easier.
Wil have alook at the limits and accelerations. I could only measure if estimated job time corresponds to actual one.
Can’t print loc line, just get it off ebay for the el cheapo price of $10. So worth it. Injected molded acetal, their tolerances are really tight. They literally snap on.
If you are big thin plates then slotting is generally what you will do. I like to reserve trochoidal for thicker parts. Otherwise it just takes waaay too long, like your 18 hour gcode lol. Nothin more fun than doing an all nighter with a vacuum right??
1 Like
I opened the picture in estlcam, it will open a new wizard for this.
1 Like
Hi,
I found the problem was in my 3mm double flute bit which got dull after 5 hours of cutting??!! Perhaps I pushed too much when searching for settings but definitely will search for cheaper bits around. Also aluminum could be t4 and will ask next time for T6.
I tried with 4mm double flute and is cutting well using repetier. Only Ramps crashed during milling after 1 and half hours.
Currently I tried to set up ESTLCAM controll. All went well and I prefer it against Repetier but trochoidal movements xy axis are really odd, slow, rough and not fluid.
Should I start a new post? Here the link to video :
Currently I set 3200 step/rev on all axis but tried also all other stepping settings.
I set max feedrate to 50mm/s and used Ryan`s settings found in the forum.
Linear milling is ok and manual movements as well.
What could be the issue as the XY movements during trochoidal milling are extremely slow, rough and not fluid?
Here video for better understanding :
I am trying to mill a frame few days already and learning hard way/having plenty issues when I depart in few days and I am in the rush.
During tests today trying to set up ESTLCAM controller on my MPCNC with RAMPS 1.4 , drv8825 :
- settings drv8825 1/32 6400/rev estlcam won`t allow it
- settings drv8825 1/16 3200/rev estlcam accepts but max trochoidal feed rate is 6 mm/s
- settings drv8825 1/8 1600/rev estlcam accepts but max trochoidal feed rate is 6 mm/s
- linear milling/ trochoidal off - feed rate is high and no problems on that side
Returned back to Repetier and found with 1/32 steps the max trochoidal feed rate is 10-12 mm/s. Above the 10-12 mm/s starts to jitter a little. When I set 30 or 50 mm/s there is no visible increase in speed, only more jitter.
I can use repetier only for trochoidal milling with similar settings as Barry mentioned - lower feed rate 10mm/s and higher trochoidal step as 9%.
With ESTLCAM as controller which I prefer, can do only slotting.
So, either this machine cant go above the mentioned feed rate for trochoidal milling, the software cant handle it or I did something wrong. If you have any advises, what can be checked, dont hesitate to suggest as I dont know where else the problem resides. It is originally build as per ryan tutorial (GT2, pitch2mm , pulleys 16T, steppers Nema17 in parallel with current around 1A on drv8825). The steppers are cold, drv8825 heatsink has around 50-60 degrees C during ops. Loaded MArlin RC7 with additional LCD display library and laser add on on AUX2.
Additionally I measured the straight movements XY axis and feed rates can be easily acchieved up to 50 mm/s and they also correspond to actuall speed.
I don’t use EstlCAM’s firmware. Are you sure you have the steps/mm set right? Why are you saying steps/rev?
Can you just try to move the bit by 10mm in software and see if it moved 10mm in real life?
Why did you switch from Marlin?
Hi Jeff,
I use estlcam as cam software to generate gcode. It has a possibility to control mpcnc and is versatile with probing etc.
Which software you use?
When in 1/32 mode per instance it takes 6400 microsteps for one revolution of stepper which is 32mm with gt2 pulley og 16teeth . One step is 0.005 mm.
The problem is trochoidal milling speed. Not slotting and linear movements.
Most people here use Marlin (I use grbl, but there’s not much help here for that). I use EstlCAM for CAM, it’s awesome.
Referring to things in steps per rev is just confusing to me since I am used to the Marlin setting of steps/mm. So I would call that 200 steps/mm. Maybe I’m the only one that is getting confused.
I was asking “Why use the EstlCAM firmware, instead of Marlin?” EstlCAM firmware is fine, but there are “proven” techniques that work here, including trochoidal milling in Marlin.
I agree you should stick with marlin, proven and very easy to diagnose.
You have to set steps per revolution as well as distance per revolution, if you want to stick with estlcam control, from there you also need to tune the motion and acceleration settings. If there is a specific feature you think marlin doesn’t have I will look into it, but as far as I know we have everything but estlcam’s built in surface mapping, which I don’t actually think works with the mega anyway.
Hi Ryan,
I set up all settings, searched in forums and applied same settings as others. Each time I changed the jumpers I changed the settings in EstlCam, reprogramm, etc. Seems the max feedrate for troch mill is allways 6mm/s which is too slow. When it comes to regular movements the feed rates are high and all ok.
Back to Marlin.
- Using Marlin and repetier found max trochoidal feed rate is app 10-12 mm/s. I was wondering as some people reported higher feed rates for trochoidal milling weather they really acchived them or something wrong with my MPCNC. I can set 30mm/s in estlcam and controlling with repetier the speed visually doesn`t change or increase compared to 12 mm/s.
- Which software for controlling mpcnc you use? Repetier, Fusion, others?
test. Same could be probably acchieved on RAMPS but I dont need it currently. They can come handy with longer jobs. Perhaps same can be done with Repetier but I am not aware of it. Minirambo has atmega 2560 and can perform mesh level calibration which could be similar to surface scan. I was thinking even adding an SPI FRAM to store position and last line of gcode in order to resume job even if loss of power occurs or reset needed due to loss of communications,…maybe in the future.
Hi Jeff,
Actually is the same if you use 200 steps per 1mm or 6400 microsteps/revolution, while I rather use the latter since the same is also used when settings ESTLCAM.