I have an Ox with a TinyG controller and I’ve been trying to mill some aluminum. When the router is engaged in the Aluminium the motors become over loaded and either stall or lose steps. According to calculations I should have my spindle set below 5000 rpm, but slowing the spin rate only makes it worse. What is the best way to solve this? If I reduce the feed would that help?
Dropping feed rate is often a good way to weld your end mill to the aluminium, breaking them. You need good chips to take the heat away from the cut.
Need a bit more information to help you with the issue, namely depth of cut, end-mill size and type and if and what cutting fluid you are using.
I’ve had good luck with cheap 3.125mm (1/8’) single flute HSS end-mills at >760mm/min, est. >10kprm, 0.2mm per pass and WD40 as cutting fluid. They are probably conservative settings, but for the few small jobs I’ve done in Aluminium they have worked.
The calculations I have seen would take into account bit speed (rpm) and movement speed, I would think. They seem to be optimizing the chip side taken with each bite. So I would think the calcs would be a function of movement velocity.
I would also assume the calcs would have to make assumptions on the torque available in the spindle. My impression is that the 400W class variable speed spindles don’t have a whole lot of torque capability, and even less when running slow…
You might find that increasing S speed and decreasing F velocity will help. Taking less material with each pass would help as well (reduce your Z down increment)
Note as well that some aluminum alloys are much harder to cut than others.
Have you searched this forum for Aluminum suggestions?
I recall seeing some good writeups.
Are you using any lubricants?
I will try some lubricants and the suggestions. Thanks for the tips!
To mill aluminum you need (IMHO) at least an 800W spindle, mist coolant, single flute bit, and a controller that will pump out at least 3A @24V to your stepper motors. A Tiny G is junk AFAIAC. You also need a lot of luck and a lot of practice to get it right.
MG
Plenty of people have great results with the TinyG.
If you don’t have bulk WD-40 on hand, kerosene is cheaper (and the main component by far of WD-40, which is why it works as a convenient substitute), and denatured alcohol is cleanest. Keep in mind the flammability of all of these before you start spraying them around anything that could be a problem (VERY hot or any sparks). Proper spindles are fine, but routers with their brushed universal motors are not a good idea.
@Jon_Miller I find that WD40 makes a mess, because it is oil based. I have had good luck with just plain old water, or a 50/50 water alcohol mix. Note that alcohol may produce flammable vapors, ensure adequate ventilation (not a shop vac!) Also, just rubbing beeswax based cutting paste on the aluminum before cutting is sometimes enough.
MG
400w brushless, but manually controller speed and no tach connected. @Darrell_n I’m only pushing 2.4A at 25volts and I get there. Shallow fast cuts seem to be the solution, its not fast, I wouldn’t race to do commercial work on it, but it works.
Ok, thanks Jeremy
@Jon_Miller Just because lots of people use something doesn’t necessarily make it good. And, I did say it was my opinion, you are entitled to your own. I think the TinyG has a boatload of issues:
- It has everything on one board. If you blow a stepper driver, it’s junk. $150 down the drain.
- It doesn’t have the current drive capability to drive a NEMA23 stepper motor properly. You need 3A to get max torque out of a NEMA23 stepper motor, even more for a high torque motor.
- It’s expensive, $130-180, depending on where you buy it. I can buy an arduino and 4 stepper drivers for less than half of what a TinyG costs. And, I can replace a $15 stepper module if I blow one.
- GRBL is open source and constantly updated. Software on a TinyG is closed source, and you are completely dependent on whether Synthetos wants to update it or not.
- It only offers 1/8 microstepping max. Other drivers have 1/16 and 1/32 as standard options. Also, most stepper driver modules have settings for idle current and decay, which the TinyG does not.
The TinyG was originally developed for 3D printers, and is perfectly fine for NEMA17 motors, but it was not made for use on a CNC machine. There are other all in one boards that are much better, like the Panucatt Gradus M1 for half the price. And, the Gradus has replaceable stepper driver modules, with a wide range of available models.
Having said all that, I don’t really like ANY all in one board solution for most of the reasons outlined. As I said, they are fine for 3D printers and laser engravers, but anything like a CNC that requires maximum performance you are much better off with an Arduino and individual stepper driver modules. You can also upgrade to a 32 bit driver like an ESS or UC300, which is , IMHO, much better bang for the buck.
Disclaimer: These are my opinions based on extensive building and testing. I do not have any affiliation with any of the brands discussed.
MG
at 16k and a single flute dont cut slower then 800mm/min. You need a air blast if cutting beyond 1* tool Diameter. Your DOC depends on teh machine, but it will probably be 0.3mm or less…
@George_Allen another consideration is whether your speed is actually what you think its set to?
If you are measuring the spindle speed ignore the rest of this post. If not you might want to verify it is running the speed you want.
…
I am running a tinyG, RioRand motor driver and 400W spindle.
I have completed extensive tests and have proven to my own satisfaction that the RioRand driver has some problems that result in the spindle running substantially faster than the Gcode “S” settings expect.
i.e. The driver does not shut off properly during the PWM off cycle (when using this spindle motor).
…
My point being that if your not measuring the spindle speed you may actually be running much faster than you want for the bit and material you are cutting.
I am working on a new driver design and will post when I get it proven to work properly.
@SirGeekALot using an optical tac with a reflector on the shaft. I found out the best place for the reflector is down on the collet nut.
https://www.amazon.com/Neiko-20713A-Digital-Tachometer-Non-contact/dp/B000I5LDVC/ref=sr_1_1?s=industrial&ie=UTF8&qid=1509927683&sr=1-1&keywords=optical+tachometer&dpID=41pFO83qSaL&preST=SY300_QL70&dpSrc=srch
The speeds may not be right. After my PSU blew a replacement PSU blew my TinyG. I have a CNC xPRO V3 on order. I’ll give y’all an update when/if things get up and running
@George_Allen once you are on GRBL run the piece-wise function. You can get spindle speed through Gcode very close.
1.7A is all that is needed to cut 1/4” aluminum with a 800w on an ACME or GT belt drive system. A 400w DC spindle will cut 5000/6000 series Al. Use a single flute upcut. Blast with air for surface finish. Go slow, speed up till chatter is gone, light DOC.
I just received my xpro v3 today and hooked it up on the machine. TinyG & xPro seem pretty similar. I think I may get to liking this controller a little better, but the functions are a little different. It’s almost like learning a new language. Im used to getting all the machine configurations in code I understand like $1PO=0 meaning motor 1 is set to normal and $xvm=XXXX for the feedrate max for X. I will have to learn grbl a little to figure it out.
I don’t have an 800w spindle, only a 400w or 500w, I can’t remember. What diameter of bit do you use?