MechMate CNC Router Forum

Go Back   MechMate CNC Router Forum > After Building the Beast - Operating , Troubleshooting and Maintenance > Troubleshooting
Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
  #1  
Old Sun 23 August 2009, 07:16
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Rough Cut on Diagonal

A potential client asked if I could cut a grid out of 1/8" aluminum in a diamond shape for the openings in the doors for kitchen cabinets. I told him that I could and proceeded to cut the shape out sentra using 1/8" end mill. He was concerned with how the corners would look. He didn't want them to be round off. He was happy with the results but I was unhappy with the quality of the cut.

The diamonds were cut with the longer points of the diamond on the X axis and the short points on the Y axis. The sides were not smooth but had chatter. I was concerned that I was having a problem with my pinon and set screws again.


I created some larger diamond shapes and cut them out of solid surface material using 1/8", 1/4" and 1/5" bits at 35, 70 and 100 IPM. In every case, there was a cut problem. I looked over some of the other things that I have been cutting and I found a slight indication of the same problem.

I started to look for things that might be loose. I even took the motors off to check to see if I had a loose pinon and set screw but they were fine. I talked with Sean about the problem also.

Router Bracket

I decided that I might have a problem with the way my router bracket is mounted to the Z Slide. I have attached a diagram. Because I am using preground rails, I could not attach the router bracket directly to the Z Slide. I had to allow for the rails and the screws that hold the rails to the Z Slide. I used most likely to long of spacers or set offs - they were .43" wide X .75" long. I think they might be flexing or had the potential for flexing.

I decided that I wanted the eleminate the set offs completely and add the Quick Release Plate. I picked up a piece of .5" aluminum and created a plate 3.25" x 4.5", the same size as the Router Bracket Plate. I then milled the back side of the plate so that it would fit between the rails and allow for the screws. I have not added register pins because the fit is so thight.

Though I have no experience using a Bridgeport milling machine, I am very pleased with how the plate turned out and how tight it fits.

Spider and V Wheels

Some builders are reversing the eccentric bushings to set the V Wheels off of the Spider instead of using a Bearing Support Base in Drawing M120220T. I used a bushing that was 5/8" Wide x 1/2" Length. This gave me about .30" space between the back of the Z Slide Tube and the Spider. Not much space but it was enough not to cause a problem. Additional info: I am using a 3/16" bolt to hold the V Wheels.

When I added the Quick Release Plate, I created another problem. I need more space between the Z Slide Tube and the Spider to allow for the nut and washer that attaches to the 3/8" bolt that is used to secure the QR Plate. I got the additional space by adding a 1/4" X 1" X 1" piece of aluminum between the bushing and the Spider.

I don't know why I have not presented this problem before this but it is never to late.

I have the piece put together but I have not tightened the V Wheels and checked for squareness and tried cutting any thing yet.

I am looking for a reality check.

Gerald, Do I need to create a 20mm Bearing Support Base and replace my bushing and 1/4" aluminum piece? Would that be better than what I have created?

DXF 2000 file with I think the right measurements.


Pictures for viewing. Sorry that the Bridgeport is sideways. Fixed



I hope this makes sense to others besides just me!

I also created another problem for my self. With the changes to the position of the router, I have to redo my dust collection foot. Which is fine if my cut problem goes away.

Thanks
Reply With Quote
  #2  
Old Sun 23 August 2009, 08:13
Alan_c
Just call me: Alan (#11)
 
Cape Town (Western Cape)
South Africa
Send a message via Skype™ to Alan_c
Nils

Do you get the same result if you cut a square with the sides parallel to the X and Y axis? If you are only getting it on the "diagonal" then it is probably because of resonance between the X and Y motors. I get the same result when I cut a circle, there are 4 areas of roughness around the perimeter interspersed with 4 perfectly smooth areas. I am hoping that the reduction belt drives will reduce this problem.
Reply With Quote
  #3  
Old Sun 23 August 2009, 08:34
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Nils, what you are seeing is a very regular pattern:


. . . . which us guys with direct-drive motors have been battling with for years. I think you have geared motors? Anyway, we found that we could stiffen the structure of the machine, but that the marks would remain for diagonal cuts.

The facts that the marks only occur on diagonals and are regular, shows that there is an interaction between the steps/drives of the 2 axes.

The things we learnt to do:

1. Clean and grease the racks. Easy to do and has a signicant influence.

2. Check pinion gears for wear. If the pinion gear tips touch the bottom of the V in the rack, there will be backlash at that point. Is that regular pattern perhaps matching the rack profiles of X and Y racks? (tricky to determine on the diagonal - cannot just hold a rack off-cut against it)

3. Experiment with faster or slower cutting speeds. Sometimes it helps

4. Increase the resolution. We learnt that geared motors were much better than ungeared motors.

5. And then get to wondering if the backlash in the gearbox is causing it . . . . . . .

(I have moved the issue about z-slide roller support bases to its own thread here)
Reply With Quote
  #4  
Old Sun 23 August 2009, 09:43
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Gerald,

Yes, I have geared motors. That being the case should I be seeing the same issue?

Do you think that I might have had some movement because of the extended stand offs that I used on the router bracket or am I just wishing that was my problem?

One of the things I thought of doing for the grid pattern that I have to cut is to turn the pattern and cut it so that the lines run more on a horizontal / vertical direction.
Reply With Quote
  #5  
Old Sun 23 August 2009, 10:00
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Geared motors cures that problem while cutting wood, but cutting very hard stuff makes the cut quality suffer. The unknown is the backlash of the geared motor . . . .

I don't think that the stand off's can generate such a regular pattern.

Test your cutting directions in some scrap with a test pattern (star?) - there will probably be some better things to do.
Reply With Quote
  #6  
Old Sun 23 August 2009, 10:35
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Gerald,

I think the problem is still in the wood but you don't see it because the grain is more giving. When I was testing, I cut the same pattern in wood and could hardly detect it but it was there. It will be interesting to see if the problem still exist when I finish changing things.

Thanks for your input and knowledge. I will post sooner next time.
Reply With Quote
  #7  
Old Mon 24 August 2009, 00:14
hennie
Just call me: Hennie #23
 
Roodepoort JHB
South Africa
Nils the sharpness of your cutter is also a factor.
Reply With Quote
  #8  
Old Mon 24 August 2009, 02:39
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Hennie,

Yes, the cutter needs to be sharp but that was not a factor. I used new bits.
Reply With Quote
  #9  
Old Mon 24 August 2009, 04:57
MattyZee
Just call me: Matt
 
Adelaide
Australia
Does the pattern change with different angles?
I reckon the repeatable pattern points to the drivetrain. But given the short pattern (i'm guessing <10mm?) i don't reckon its the rack or pinion. The pinion would produce a pattern closer to 70-80mm depending on the gear (i doubt such a repeatable pattern is occuring around the pinion in a discrete integer multiple). It may be something internal to the gear motor. If the motors are 7.2:1, then a mis-shaped internal pinion may result in a pattern closer to the 10mm you are seeing.
It could be either the X ot Y motors. The easiest thing to try might be to swap the Y and Z motors. and see if that improves things. If its not the Y, then maybe swap with one of the X motors.
Reply With Quote
  #10  
Old Mon 24 August 2009, 06:08
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Here is a straight line drawn diagonally across my PC monitor:

My pixel size is about 0.5mm, but I can create any "roughness step" length by changing the angle slightly. The pattern is created when the line has to decide which pixel is the nearest.
Reply With Quote
  #11  
Old Mon 24 August 2009, 06:19
MattyZee
Just call me: Matt
 
Adelaide
Australia
Yes, very good Gerald. But the steps in Nils photo looks to be more than the 'resolution' of his motors. I was interested in seeing if an angle closer to the X or Y axis was better. If it is the Y motor having an internal eccentricity for example, a cut almost parallel to Y axis should be worse than one almost parallel to the X axis.
Reply With Quote
  #12  
Old Mon 24 August 2009, 06:58
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Matt,

I can not do any cutting until I get my modifications complete which I would like to finish this week if my welder has the time.
Reply With Quote
  #13  
Old Mon 24 August 2009, 09:56
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
I don't think that backlash would be a factor because the motors do not reverse directions on that particular cut. Also, because both motors are energized (moving) on a diagonal cut, neither axis would go into its reduced current mode, so neither axis would be pushed by the other axis.

Gerald's post #10, showing a straight line, displayed on a diagonal shows how the problem could be caused. We're dealing with digital electronics. If there is a 10:1 ratio between the X-axis and the Y-axis, one of the axes is going to get a step pulse 1/10th of the rate of the other axis. A step is a finite movement, not a partial movement, so the slower axis may make a step at somewhat irregular intervals.

The other factor, which is probably the major factor, is the hardness of the material. The cutter bounces and giggles and wiggles as it chews through the material, after all, aluminum is not butter. It resists the cutter. That bouncing is transferred to the structure of the machine. If the harmonics are right, the structure can cause the bouncing to be amplified.

On my Shopbot, with no gearing, I had a severe chatter problem, even worse than the photo of the aluminum. With a 3:1 belt-drive and then with upgraded 7.2:1 geared motors, the problem was reduced, but not eliminated. I get chatter every time I cut aluminum or plastic. Most of the time when I cut MDF, I get chatter that can be seen but hardly felt with the fingertips.

When I complimented the owner of a 'big iron' machine that cost over $250,000 about the smoothness of his cuts, he told me that he had spent hours and hours making various trial cuts at different speeds before he finally found the proper feed speed and the proper spindle speed to eliminate most of the chatter. After listening to him, I tried various speeds on my machine and found that feed speed is a large part of the chatter problem. Large circles require a different speed than small circles. Diagonal cuts require speed compensation depending on the angle. Cuts with a straight-line portion and a curve usually require two speeds.

Because almost all of my work is with wood, I found that sanding the edge was the fastest solution on one-of-a-kind parts. When I cut aluminum or plastic for customers, I show them some worse-case samples so that they know what to expect. Usually the results are better than the samples, but at least the customer didn't expect better than he got.
Reply With Quote
  #14  
Old Mon 24 August 2009, 10:34
isladelobos
Just call me: Ros
 
Canary Islands
Spain
Send a message via MSN to isladelobos Send a message via Yahoo to isladelobos
We can test
it is possible two printer ports and two PMDX.?
1- X motors.
2- Y & Z motors.
I do not think that solves a lot.
Reply With Quote
  #15  
Old Mon 24 August 2009, 16:11
Alan_c
Just call me: Alan (#11)
 
Cape Town (Western Cape)
South Africa
Send a message via Skype™ to Alan_c
What might also help if it is being caused by the cutter flexing, is to do a rough cut at a larger size and then a finishing cut on size (you CAM program should be able to do that, I use Enroute3 with that capability) that way the cutter is only taking a small "bite" and will flex a lot less. This obviously increases the cutting time but if it is a critical part or there is going to be lots of them it could save a lot of sanding.
Reply With Quote
  #16  
Old Mon 24 August 2009, 21:03
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
This thread brings back lots of memories More factors:

- Small diameter bits magnify the problem, bigger bits give smoother cuts.

- V-rollers of the gantry/car/slide must be seating down firmly. The y-car in particular must have all 4 rollers seating with equal pressure. In general, everything on the mechanicals must be firm.

- Low power PCs and Mach3 together have a problem in generating a train of smoothly spaced steps for multiple simultaneous axes.......(hence the SmoothStepper)
Reply With Quote
  #17  
Old Tue 25 August 2009, 03:28
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Some additional information:

I have a drawing of what I cut.

I made a number of test cuts using 35, 70 and 100 IPM, I used 1/8", 1/4" and 1/2" end mills. The bits were a new or almost new. I cut the pattern in solid surface material.

I saw the same results no mater what size or speed. I had the router set at 20,000.

For any one that has the time, I would be interested in see if you get the same results or what kind of results you do get.

I will be checking to see what my Mach Kernel setting is set at but I believe it is at 25,000.
Attached Files
File Type: dxf Test Diamond.dxf (15.0 KB, 45 views)

Last edited by sailfl; Tue 25 August 2009 at 03:34..
Reply With Quote
  #18  
Old Sat 05 September 2009, 02:25
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Gerald,

I thought my problems were resolved until I cut the Fleur De Lys model from my Aspire - Vectric Art 3D collection.

I sized the Model to .5" X 5" X 5.75" and cut it out of MDF. I used a .25 O Flute End Mill for Roughing out, .25 Ball Nose and .125 Ball Nose for finish cuts and a .125 End Mill to cut the model out of the material. I used an .125 End Mill to cut the model out to retain the detail.

The machine is doing a fantastic job of cutting the surface as you can see from the photo. When the .25 End Mill was roughing out, the surface cuts were absolutely smooth - the machine is cutting flat. Even the Ball Nose bits did a fantastic job except for one area that does not show up on the photo where there is a slight indent.

The problem is chatter, when I cut the model out, which appears on both sides of two leafs - there must be a name for them but I don't know it. I did use a smaller bit which I ran at 50 IPM. Maybe for this small model, that was even too fast.

I am going to recut my diamond model in solid surface material and MDF to see what the results are there.

An area of concern are the washers that I used for the Gantry V Wheels. May be there is play there. I will take a picture of that when I have light. I might have to make some bushings for that area.

I realize that the Milwaukee has some run out.

Does cutting slower on a small surface help.

These are the things I want to cut so I want to resolve this problem.

I am open to additional suggestions. I do think that the Z Slide is not the problem.
Attached Images
File Type: jpg DSCN0403.jpg (169.0 KB, 830 views)
File Type: jpg DSCN0406.jpg (166.6 KB, 825 views)
Reply With Quote
  #19  
Old Sat 05 September 2009, 05:05
hennie
Just call me: Hennie #23
 
Roodepoort JHB
South Africa
Nils did you cut it out in one pass and at what speed?.I have noticed on my MM when I do fast cutting with the smaller dia bits it does the same so maybe try cutting at a slower rate.Also when the bits start getting blunt it does the same.
Reply With Quote
  #20  
Old Sat 05 September 2009, 05:06
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Pictures of the washers that I had to insert for the V Wheels on the Gantry / X Axis.

I think that I might have a problem here and that I need to make some bushings like I did for the spider. What do you think?
Attached Images
File Type: jpg DSCN0411.jpg (171.4 KB, 801 views)
File Type: jpg DSCN0414.jpg (167.1 KB, 795 views)
Reply With Quote
  #21  
Old Sat 05 September 2009, 05:09
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Hennie,

The material that is being cut at this point is .20" and it was cut with one pass at 50 IPM. I did another cut this morning to test slower speed and I cut it at 28 IPM and the chatter was still there.
Reply With Quote
  #22  
Old Sat 05 September 2009, 05:36
javeria
Just call me: Irfan #33
 
Bangalore
India
Ah nils I have the same problem! I cut some mdf with a 8 mm straight edge router bit nd there were chatter marks all over. the spped was 55 ipm @ 0.4 inch depth.

i suspect my spindle mounting now cause I have only one clamp holding the spindle!

Last edited by javeria; Sat 05 September 2009 at 05:38..
Reply With Quote
  #23  
Old Sat 05 September 2009, 06:22
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Nils, every time I think of a reply for you, I realise that I have said it before. I have reviewed my previous posts in this thread, and they all still apply.

Changing washers behind the gantry rollers won't change anything because the rollers are pushed up by the gantry weight (on the z-axis the load oscillates).
Reply With Quote
  #24  
Old Sat 05 September 2009, 06:39
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Gerald,

I hear you. I will recheck that the Gantry and the Y-Car to verify that they are sitting firmly on the rails. The Gantry is not so difficult, the Y-Car is a pain. I will check back with you.

I am not sure I want to move to a smooth stepper.....

Thanks

Irfan,

Yes, I would suspect that part of your problem is the way the spindle is mounted but now that I have reworded my bracket mount and the Z Slide, I don't think my problem is there.

Thanks
Reply With Quote
  #25  
Old Sat 05 September 2009, 07:20
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Gerald,

I have disengaged the motors for the Gantry and the Y-Car.

When I twist the Y-Car, I can not detect any movement. The V Wheels seem to be firmly sitting on the rails. I also visually looked at each of the V Wheels as I moved the Y-Car back and forth along the Gantry rail. I could not detect any gaps.

I did the same with the Gantry but it is a little more difficult to twist. I also visually looked at each of the V Wheels on the Gantry as I moved it along the X axis. No gaps.

Is there any thing I have missed? Is there any thing else I can try?

Thanks
Reply With Quote
  #26  
Old Sat 05 September 2009, 08:26
WFY
Just call me: Buzz #60
 
Pembroke, Ontario
Canada
Hi Nils,
Is there a possibility that the debris in the bottom of the "v" grove is possibly lifting the wheel off the rail slightly?
Reply With Quote
  #27  
Old Sat 05 September 2009, 08:55
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Quote:
Originally Posted by Gerald D View Post

The things we learnt to do:

1. Clean and grease the racks. Easy to do and has a signicant influence.

2. Check pinion gears for wear. If the pinion gear tips touch the bottom of the V in the rack, there will be backlash at that point. . . . .
Did you?
Reply With Quote
  #28  
Old Sat 05 September 2009, 09:00
Doug_Ford
Just call me: Doug #3
 
Conway (Arkansas)
United States of America
Nils,

Beautiful work.

I can see the facets in the edge of the workpiece but they appear to be tiny. I'm wondering if the runout in the router's spindle and the small amount of play in your V wheels are the cause. Can't the facets be easily sanded out? I mean, you're going to have to sand whatever you cut out anyway. Are you expecting to cut out objects and apply a finish without sanding them?
Reply With Quote
  #29  
Old Sat 05 September 2009, 09:13
Leko
Just call me: Leko
 
Kaukapakapa
New Zealand
Nils,

Have you tried disconnecting one of your x motors and trying a cut, then disable the other one & try a cut.

Realize that either your x axis or y axis are probably doing this pulsating all the time, but it only shows on diagonal cuts. When you are cutting in straight x or straight y directions the pulse is directly in line with the cut.

I can only think of 4 things that would result in such a regular pattern; loose grub screw, bent motor shaft, bad motor, or bad Gekko. I'd recheck your pinion to motor connections again.

2cents from somebody that only has a Mechmate in my head.
Reply With Quote
  #30  
Old Sat 05 September 2009, 10:43
sailfl
Just call me: Nils #12
 
Winter Park, FL
United States of America
Buzz, No debris. If the V Wheel lifted up I see a change in the surface but the wheels are clean.

Gerald, I will address those issues. Thanks for mentioning.

Doug, Thanks. The cut turned out very nice. I am concerned that there still is a mechanical problem and I want to eleminate those. The chatter is more than I want to sand. My goal is not to have to sand any thing....I realize that is not a realistic goal but I can still want it. I have checked the runout and it is not good and so some of the problem can be attributed to that.

I will do some more test cuts to see if my cut has gotten worse since I modified the Z Slide which had very little chatter.

Leko, I don't have any of the problem you mentioned but I will think about the motor test.

Thanks to all.
Reply With Quote
Reply

Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Potential problems with placing diagonal braces Gerald D 10. Base Table 0 Mon 20 April 2009 23:50
Foul between diagonal brace and tip of cross-support channel kn6398 10. Base Table 4 Thu 27 November 2008 18:38


All times are GMT -6. The time now is 14:58.


Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.