Printable View
What depth of cut, speed, feed, stepover was that? Looks like you took it very gently. Seems a bit of a random place to choose to cut it within the aluminium sheet, but I guess you had your reasons.
I see you're very generous with the cutting fluid!
Finished part looks nice, are there any marks in the corners due to backlash. Were you using G64? I think so, but it's difficult to tell in the videos.
Jonathan,
I did take it very easy with that cut. Depth of cut was only 0.5mm, step over was 20% which would be just under 1mm and feed speed was 10mm/sec for that cutter was 10mm/sec. The only cutter I was was not new so I did go slowly with it. The alu also has some hard sections in it so that was why the cutting fluid was generously applied. The starting point was a mistake, I for got to set my dxf to 0,0 in cad and I think cut2d picked that up.
I have had a look at the program file and cant see any G64 codes in it. Youll need to excuse my ignorance in G codes at the moment, I havent got round to learning them yet!!
Some photos are attched of the cut edge. Quality looks good for an old cutter.
Attachment 4567Attachment 4565Attachment 4566
what would you look for or what would it look like . for backlash??
James
James,
backlash would look like an uneven cut. The finish edge at the corners would be like a wavey line. This happens when there is play in the mechanics of the machine like a set of cogs not meshing together correctly so that when your turn one cog it moves slighlty before the other one moves. In a profiling machine that drives atart to wobble a bit at a change of direction until the machine sort of balances itself out.
t I have found that the best way the check for backlash is to run your machine at a 45 degree cut so that both x and y axis are travelling at the same time ie cut a square but rotate it to 45 degrees on the bed. The corners should be nice a smooth without any wobble in the cut surface. Cutting at 45 also tells if your speeds are the same in both axis at the same time. If they are not you would have an uneven cut surface all the way along the cut.
Ian
Whether or not you get a wavy line/arc depends on the direction of cut (climb/conventional), the cutting force and the friction of the slides. Either way you will get a 'blip' on the corners, or wherever any axis changes direction. Here's an example:
Attachment 4569
Attachment 4570
The corners should be smooth, but as you can see each corner has a line/offset portion where the axis changed direction. For that part it doesn't really matter - it's still shiny :smile:
That was on my milling machine which currently has more than it's fair share of backlash.
This part was made on my router which has very little backlash, so the corners are smooth with no 'blips':
Attachment 4579
Hi Ian,Quote:
Originally Posted by Web Goblin
Congrats on your first Hot Chip's. .:toot: . . . . . (Now the BUT.:lol:)
I think your single spindle mount is causing chatter/resonance looking at the edges I can see chatter lines.!
I know you say the cutter was old but that does look more like flex chatter..? Also when cutting it sounded ringy more like a resonate ring not a blunt cutter.
Too little DOC, too slow spend n feed or more importantly wrong chip load can be just as bad for cutter than too much, esp in sticky Ali. Don't be shy give the cutter some work to do as your going far too conservative. . . .Heat is the enemy so send it away with the chip.!!
Just what I was thinking - the vertical lines are a bit of a give-away.Quote:
Originally Posted by JAZZCNC
Definately too low feed and stepover - with only 20% stepover you need to feed much faster to attain the same chipload, look up chip thinning. This explains some of it, though it is aimed at making you buy the program...
http://www.cnccookbook.com/GWCalcFeedsSpeeds.htm
I was joking when I said that was generous!Quote:
Originally Posted by Web Goblin
Mach3 may have defaulted to it anyway, it depends.Quote:
Originally Posted by Web Goblin
G64 puts mach3 in constant velocity mode, so that within reason the feedrate remains constant. This means that external corners will become slightly rounded - more if the acceleration is very low.
G61 puts mach3 in exact stop mode, so it stops once it has got to the point specified by each line of code, then continues. This makes the motion very 'jerky', though depending on the situation it will make the final part more accurate.
Using constant velocity makes the motion a lot smoother, and a constant feedrate is clearly faster overall and kinder on the cutter. To try it just type G64 at the begining of the program. You'll notice the most difference when a program uses lots of small lines to approximate a curve.
Hi There,
How are you finding the High power kinco drives? I can't decide between those and the normal DC powered variety, any reason you chose those?
Regards
Andrew
The 2M2280 drivers run on 240VAC they run the motors at a high voltage, not sure if it is actually 230*2^0.5 as that seems a bit much. This get much much higher speed from the motors than the 'normal' 70V (or so) drives, however it will make the motors run hotter. That's not necessarily a problem - as long as the case temperature is not above about 80-90C it's fine.Quote:
Originally Posted by andrewbond
I'm thinking of getting a couple to use on either my lathe/mill or router. They're only suitable for Nema 34 (or 42) motors, not 23/24. A Nema 23 motor with high inductance might be ok... but with how much those drivers cost that's probably not the best option.
Ye Try G61 with an high feed rate/accelleration and watch it turn into a "Darlick on Whiz" as it try's to break dance thru the workshop wall after it's destroyed all in it's path. . .:rofl:Quote:
Originally Posted by Jonathan
Andrew, the kinco drives are good. I have Nema 34 motors on them and they are fine. The only problem is that the drives generate alot of heat but a couple of cooling fans will take care of that.
Guys, thanks for info on cutter loading/feed speed. I willl have to play with it to get a feel for the right speed and loading. The step over and feed speed was what cut2d generated when I picked that cutter. They can be modified but I decided to go with the stock option as a first timer.
I do have plans for two spindle mounts but havent gotten round to them yet. I have ordered some of the single flute cutters off ebay so that could be a good test when they arrive.
Jonathan I am using USBCNC not Mach3. It has something called look ahead feed built in to it. I think this looks ahead of the block you are running and looks at the program to see what is coming up. It then tries to calculate the optimum speed for those upcomming blocks to keep the machine moving smoothly, well thats what I got out of reading the manual anyway!
Mach also use's look ahead.Quote:
Originally Posted by Web Goblin
CV (constant velocity) and Exact stop are Gcode commands which basicly tell the control software how to react.
If your code outputs G61 then the look ahead won't apply because it's been forced into exact stop mode and look ahead only works in CV mode.
If your code breaks arc's down into tiny lines(like lots of software does) then this will have the affect of jerky movements, it will also have the affect of stopping and starting between arcs and straight lines as exact stop accelerates and deaccelerates point to point, this affects cut quality badly. (Plus it's bloody annoying)
Exact stop will not round tight corners like CV does at high feeds but it can cause the machine to run very slow and quite rough as it accellerates then deaccelrates between each point.
As you can imagine at high feeds this will hammer the machine and motors as it flys upto the corner at high speed then jumps on the brakes for the tight bend.!!
I think if you check there will be settings within USBCNC to help control CV's angular distance where it changes for tight radius and controls rounding at high feeds.
But I think even USBCNC can be put into exact stop mode so G61 will apply basicly cancelling look ahead and CV mode. . . . . Worth checking out.!!
Thanks for the info,
I'll probably go for these as I won't have to buy seperate power supplies (I'm going to be using the 8nm nema 34 motors)
Did you mean the drives themselves generate alot of heat, or that they cause the motors to get hot?
Andrew
It's probably both - running the motors at a high voltage tends to make them hot, but as long as it's below about 80C it's not a problem.Quote:
Originally Posted by andrewbond
Andrew,
the motors get warm to touch but nothing bad. The amplifiers do get quite warm so if you are going to mount them in a cabinet forced cooling would be required.
Ian
Had a shot at cutting some new spindle motor mounts today. The first one is cut and it came out well. I used one of the single flute cutters from ebay.
Attachment 4662Attachment 4660Attachment 4661Attachment 4663
There is a litte bit of marking on the third photo at one of the corners but its nothing to worry about. This is cut from 25mm thick alu. Inner size was programmed to 80.5mm and it came out 80.4mm so thats fine as well. Just need some more time to get the second one done and get them drilled and tapped out for fitting.
I also did some testing on my tool height setter. Fitted a new 6mm single flute cutter and ran 10 measurement cycles of the tool height routine. The results are below.
1, 30.5087
2, 30.5037
3, 30.5012
4, 30.5012
5, 30.4963
6, 30.5012
7, 30.4963
8, 30.4988
9, 30.4988
10, 30.4988
From min to max the difference is only 0.0124 so thats good enough for me.
Regards
Ian
Hi Ian,
The corner marking is chatter probably due to radius being the same size as cutter, if you can make your radius just slightly larger than the cutter this won't happen.
Nice job as well by the way.! great aint it watching the machine build it's own bits. .:toot:
Jazzcnc,
thanks for the info on the cutter radius. It was a 3mm rad and a 6mm cutter. I will alter the drawing for the inner radius corners for the next one. I have alot to learn about milling, this is not something I would have to worry about in profiling.
Your right, its great watching the machine cut its own parts. I wonder if I can get it to build a replica of itself:lol:
Ian
I have been cutting two new spindle mounts this week to reduce the vibration of the spindle motor. After finishing the second one I had sent the machine to the home position which is where I usually leave it when switching it off. On travelling with the X axis I noticed a grinding sound. On further investigation I found this:
Attachment 4716
Looks like the shaft I used to join the motor to the pulley had been working its way through the bearings and had the coupling rubbing against the motor mount.
I need to strip this out and make two locking collers for the shaft to stop it moving.
At least it hasnt done any real damage.
Ian
Ian there's a big gap between the worn coupling and the bearing mount has it been floating.?
Sorry but I'm not familiar with your setup and can't tell from this close up pic but that looks like it's an FK bearing in the mount so it shouldn't float at all.?
Edit: Ok I stopped being lazy and went back and looked thru the thread I understand now, they are FK bearings but also fixed at the other end.!. . . . Still looks like float thou.?
Ither a FK or FKA depending on what model mount hes go!
I read it as the shaft has moved through the bearing block, which in turn has pulled the lovejoy into contact with the housing...
Ye but what shaft.? Motor shaft.? or Screw shaft.? Either way neither should be floating.? . . .something seems strange.!Quote:
Originally Posted by m_c
This pic from earlier in the thread reveals all.
Attachment 4012
The motor coupling would have enough movement in it to allow the pulley shaft to hit the mount without losing drive.
The shaft that connects the motor to the x axis pulley is the one which has managed to move through the bearings so that the coupling was rubbing against the motor mount.
Attachment 4717
A picture of the complete motor mount. I didnt think that the shaft would travel through the bearings so I didnt fit any locking collars. I will make them and fit them on the motor side of the last BK mount before the pulley. This will stop any further movement.
I suspect this has been caused by the pull on the shaft from the tension on the pulley belt but this is a guess. Seeing as I had to hammer the shaft through the bearings I really didnt think it would move.
Ian
m_c,
thanks, I really need to learn to type with more than two fingers.
Ian
I feel sorry for those bearings. Hopefully you just hammered on the inner ring. Often heating up the bearing first does the trick if it's that close.Quote:
Originally Posted by Web Goblin
Then learn with this, you won't regret it. I have been using this layout for about 5 years now:Quote:
Originally Posted by Web Goblin
http://www.dvzine.org/
Jonathan,
I dont like heating up bearings. I just normally put the shaft in the freezer for a while and yes I used a tube the same diameter as the bearing inner to hammer the shaft in. With a nylon hammer I might add.
Ian
Argh I see now but I would call this the Y axis and the earlier post said X axis thats what's thrown me.!!Quote:
Originally Posted by Web Goblin
Quote:
Originally Posted by Web Goblin
You don't have to heat bearings up that much. 100degC is more than enough for installation, and it won't do any damage to the bearing or whatever grease is inside it.
Often sitting them on top of a hot radiator for an hour or so is enough to give that bit vital clearance for easy installation.
jazzcnc,
I would normally call this the Y axis as well but USBCNC calls it the X axis. For some reason the designer has the axis the wrong way round from what have seen is normal. I have tried to change it in the software but there does not seem to be an option for it.
Ian
Better still do both that way they just slip on sweet.Quote:
Originally Posted by m_c
I find oven on low works good for large items like crankcase's etc . . .We race motoX so with rebuilding engines and forks I get plenty of practice.!!
Dont know much about USBCNC but surely it's upto you which Axis moves by how you connect the drives to the board. Hisn't it simply a case of connecting the X axis motors to the USBCNC designated X axis output.?Quote:
Originally Posted by Web Goblin
Then when USBCNC see's G-code telling the X axis to move it outputs pulses to it's X axis output which are connected to your X axis drives. . Simplizz's.!!
Like Mach it shouldn't and probably doesn't care which Axis is which, it just outputs pulses to it's designated pins or in your case connector on the board.
You could call them Jim, spock and Mc Coy it doesn't care (Can you tell I'm watching Star trek.! Lol)
I did that to start with but it gave me some problems with homing the machine so I changed it. Also its really annoying to watch the machine go one way and the graphics on the pc go the other! I did ask the designer if it could be changed but I didnt get a reply to my email so I will take that as a no.
Ian
Here is a screen shot of the jog screen. You will see what I mean.
Attachment 4718
The direction of the axis can be reversed but not the x and the y.
Ian
Ian that looks right to me? orientation should be side on to the machine not head on as you may be thinking? It's not a mil
Yep I agree thats how I view it. . . . Mach works like this as well.Quote:
Originally Posted by 2e0poz
Ian I have downloaded the USBCNC software so I'll go have a look.
I see what you mean. It might just be me. I am used to profiling machinery and we always use the rail as the x axis and it is always head on on the screen as well. I am getting used to it slowly. I have plans to mount the pc under the frame of the machine and to mount the screen and keyboard on an arm at the front left corner of the bed so that the pc screen will be at 45 degrees to the machine so then I will be looking at the machine and the screen in the same direction so to speak so the machine movement will follow the screen movement.
Ian
Think how a 2D cad screen is laid out, most routers are the same. A mill is head on and centered (point 0) so your machine will be xy axis positioned bottom left (point 0). Easy mistake :-)
I've just loaded USBCNC with some of my G-code and it shows perfectly just like it would in Mach.
Like Poz says just view it as the side is your front view and it will make sense.
I'm now going to have a good rumage around in USBCNC software, I may just invest in one for a lathe project I have in mind.?