[gio666]

It certainly reduces the problem, but does not cure it.

To the opening poster: have a very good look at the costs. In a dual screw you need the two screws and two motors. However in a single screw you have the cost of the beam at the bottom joining, a moving cable system if you implement, often wider bearing spacing is required. My point is that these little things often add up, probably not to as much as dual screws but you won't save a full screw and motor cost.

With centre screw there is a big sacrifice in rigidity. You can't support your table in the middle.

Perhaps I sound too negative on single screw, but I've built a machine with round supported rails with a centre screw, with a gantry width of 2 foot (cutting width 400mm) and it's just not good enough. Racking leads to inaccurate cuts, and the table sags in the centre. So I like to help others avoid my mistakes / issues.

Your machine was raking because on using supported rails, not from single ballscrew. Supported rails have no supported at all if any forces are toward the opening in the bearing. All that's there is plastic. If mounted on the side, it will rack left and right. If mounted on top , gantry will move back and forth. I use single ballscrew on everything and not had problem with racking.
Here is single screw machine cutting 540ipm at 1 inch pass .

540 ipm - YouTube

[maly-miller]

It certainly reduces the problem, but does not cure it.

To the opening poster: have a very good look at the costs. In a dual screw you need the two screws and two motors. However in a single screw you have the cost of the beam at the bottom joining, a moving cable system if you implement, often wider bearing spacing is required. My point is that these little things often add up, probably not to as much as dual screws but you won't save a full screw and motor cost.

With centre screw there is a big sacrifice in rigidity. You can't support your table in the middle.

Perhaps I sound too negative on single screw, but I've built a machine with round supported rails with a centre screw, with a gantry width of 2 foot (cutting width 400mm) and it's just not good enough. Racking leads to inaccurate cuts, and the table sags in the centre. So I like to help others avoid my mistakes / issues.

From the isometric photo down with the bed hidden you see that I can support the frame with any number of extrusion needed. so the problem of racking I guess is in the rigidity of the frame itself and not the single ball screw. I have no doubt that the dual screw is better in rigid frame or not, but if you calculate the bend factor and the inertia of the anti vibration extrusion that I ordered from 8020, you will find that this design is an over kill. However, I trust the testing. Although my calculations should be ok, but I am known of missing numbers and decimal points

The extrusions will be delivered 2moro and I will start the build soon. I will give you all step by step feedback about the build in this thread

[maly-miller]

Here's one link that describes it with a drawing - cockrum.net It's called a moving knot cable/pulley system. Search the web and on CNC Zone to find more examples of it.

Thanks CarveOne I will check this out for more knowledge

[maly-miller]

Your machine was raking because on using supported rails, not from single ballscrew. Supported rails have no supported at all if any forces are toward the opening in the bearing. All that's there is plastic. If mounted on the side, it will rack left and right. If mounted on top , gantry will move back and forth. I use single ballscrew on everything and not had problem with racking.
Here is single screw machine cutting 540ipm at 1 inch pass .

540 ipm - YouTube

Thanks for sharing this video, gave me more confidence. What do you think about the design and the extrusion used?

[maly-miller]

I am now stuck in a very stupid donkey work, calculating the number and size of screws. Do you guys have a a way to just buy screws in bulks. Is there like a standard or assorted collection that I can buy and it will help me? Or I have to go through hell calculating them group by group?

[CarveOne]

Just for the record, I'm in favor of dual screws or dual r&p over single screw any day. The single versus dual drive systems both have their benefits and detractions. The moving knot suggestion is just a way to help minimize racking in the event that you go with single screw and have some racking issues.

Slaving X/A or Y/A is how I built my current two machines and I don't regret it at all. I did have problems with the first attempt but that is history now. It was an incorrect Mach3 related setting. Paying more for a second motor and screw are just what I would do without a thought about cost. To me it's just a necessary expense.

[LeeWay]

I am pro dual drive as well. Not everyone experiences racking with a single screw though. Totally dependent on the machines design and the materials used.
Your design is kinda nice in that if you do get racking, you can slide the screw assembly over and add another. Your initial prototype would cost more in that event, but the subsequent machines could be made cheaper.
I do see the benefit of proving to yourself one way or the other. Like many of us posting in this thread, we have been there and done that and have formed an opinion based on the result we had.

I will be interested to see how the build goes for you.

[maly-miller]

As the title, screws mmmmmm I didn't think that I will be buying screws for $100. Please guys tell me that I didn't get screwed and it is normal when you buy 10 or 15 packages of 50 and 100 different sizes screws you pay such an amount. I never bought this amount of screws before.

On the other hands the extrusions arrived YAAAAAAY 83 Lbs of fine aluminium extrusions . Will start on labeling them for cutting during this week and start cutting them this weekend.

Stay tuned.

[LeeWay]

I have never bought that many Ball screws, but I did buy 15 pounds of coated decking screws yesterday for another project. I pretty much got screwed on the price for those at about $80.00.

[CarveOne]

I am pro dual drive as well. Not everyone experiences racking with a single screw though. Totally dependent on the machines design and the materials used.
Your design is kinda nice in that if you do get racking, you can slide the screw assembly over and add another. Your initial prototype would cost more in that event, but the subsequent machines could be made cheaper.
I do see the benefit of proving to yourself one way or the other. Like many of us posting in this thread, we have been there and done that and have formed an opinion based on the result we had.

I will be interested to see how the build goes for you.

Leeway,

I haven't built a machine with a single screw in the middle, but have the notion that the single point of weakness is in the way the screw is attached to the gantry under the table through the single anti-backlash nut. If there were two anti-backlash nuts mounted on a wider bar or plate under the table it would help reduce the racking dramatically. Same principle as is used on other parts of the machine where two bearing blocks are used on a ball screw. Instead of two anti-backlash nuts, one of them could even be a regular ACME nut to save cost.

The other reason I don't like single screw under the table is that it gets in the way of supporting the table with more than four legs. That has been discussed and improved with better table designs.

[maly-miller]

I have never bought that many Ball screws, but I did buy 15 pounds of coated decking screws yesterday for another project. I pretty much got screwed on the price for those at about $80.00.

That's a relief. I have a lot of different sizes too. M8, M6, M5, M4 with flat heads, socket and Button with different length too. So I guess am good Time to start on collecting info about the aluminium plates and angles that I will buy next

[maly-miller]

Leeway,

I haven't built a machine with a single screw in the middle, but have the notion that the single point of weakness is in the way the screw is attached to the gantry under the table through the single anti-backlash nut. If there were two anti-backlash nuts mounted on a wider bar or plate under the table it would help reduce the racking dramatically. Same principle as is used on other parts of the machine where two bearing blocks are used on a ball screw. Instead of two anti-backlash nuts, one of them could even be a regular ACME nut to save cost.

The other reason I don't like single screw under the table is that it gets in the way of supporting the table with more than four legs. That has been discussed and improved with better table designs.

I was thinking the same but in a different way may be if I widened the adapter that connects the nut to the gantry it will make a difference. I will be trying different ideas during the build if I have racking problems.

[CarveOne]

Since an anti-backlash nut is fairly short in length and is usually made of Delrin there is little resistance to twisting due to off center leveraging. Using two nuts spread apart on the screw tends to reduce the twisting. If severe enough, the screw itself may start to deflect when near the center of travel.

[maly-miller]

Long time since I posted anything in the thread. I was fetching the materials from here and there. Got the extrusion safe at home, The MDF sheets that will be used as templates before working with Aluminium sheets.

Shortly I will start on working on the Z-axis in MDF first. Will post photos soon

[maly-miller]

Started on cutting the extrusion yesterday, Very good experience actually. I thought it will be so much pain since I am cutting with my Bosch table saw. However, this Diablo Blade is very good that it cuts through aluminium like butter. I also felt the weight of these extrusion so I decided on making a table for the machine that will be also the assembly table. Once it is assembled I don't think I will be able to move it. Will post photos later for the table. Here is a photo of my first cut.

Attachment 235240

[maly-miller]

I started assembling the machine yesterday evening. I did the main frame. thought about the way to properly put the extrusions together. I came up with a way that probably right. Some scrap wood cut the same size and double sided tape. Squaring the corners to square the whole machine. Photo below shows the placing of scrap wood on each corner. However, since wood and double sided tape is not rigid enough, I got all 3 corners square except for the 4th, it is off by 0.5 mm (about 0.019685 inches) I am thinking of spending the time to square it. What do you think guys? Do I spend the time or this won't make much difference and I just stop being so picky?

Attachment 235280

[maly-miller]

Finally After a year of ups and downs, work and money issue my CNC was born. Here is a photo

And there is no racking with a middle screw

Cheers

[RicknBeachcrest]

Congratulations. You built a machine to be real proud of. I'm sure you will agree, building a cnc router is not for the faint of heart.

I'm using my machine about once a week. It still amazes me what I can do with it.

Now the fun can start.
Rick