Designing a new router called Brevis-HD

[mactec54]

Hi All- Back on line at last. 16 days no connection thanks Telstra. Systems seem to be normal now. So update... I'm going with 1610 screws, now I have to size motors. I've looked at whirling of the screws and my FE and other equations seem to be in conflict. So that's an interesting area to sort. My development model has developed a few bad habits so its time to rebuild into correct sub assemblies from scratch and build a production grade model. Been sorting some wiring and I have a meeting with the sheet metal company to sort some preferred metal thicknesses and bending data. Then its onward to the build. I have decided to solder some of the bracing on the machine vs welding or bolting... minimal distortion, very compact and easy to do...Peter

Do you have a line on the whirled Ballscrews

[pippin88]

Hi MD - Thanks for the feedback, lock 10mm in Danno. So the 5mm lead gives heaps of force but not enough speed? Is that the reasoning? Peter

Stepper torque drops off quickly with increase in RPM and screw whip becomes an issue.

10mm is definitely a better choice.

For wood cutting the mechanical advantage from 5mm vs 10mm is not an issue. Neither is the resolution.

[joeavaerage]

Hi,
my mini-mill and my new build mill both have 5mm pitch screws.

I like the extra mechanical advantage (12 for 5mm pitch 20mm diameter verses only 6 for 10mm pitch 20mm diameter) of the 5mm screws. Remember however
that it was always my intention to mill steel and therefore thrust was more important than speed.

I suspect that peteeng will want to direct couple the steppers to the screws being the cheapest, no fuss way to do it, therefore the combination of ballscrew pitch and
stepper torque are linked. If an otherwise good quality stepper is available at best price but with limited torque then 5mm pitch is indicated. If however a very 'torquey'
stepper can be had at the right price then 10mm pitch is indicated.

Another consideration is availability. It would appear that the Chinese manufacturers have adopted 5mm pitch as a quasi standard and therefore there is a huge selection
of models to choose from and therefore likely to be had at best price, particularly for a volume sale.

The success or otherwise of peteeng's design will be cost. The market is full of cheap Chinese made routers and trying to sell in that market is difficult proposition.
Good design, good construction and good components could and would generate good market appeal but that appeal will not withstand a significantly higher price
than your Chinese competitors.

Finding suppliers that offer exactly matched screws and/or steppers at the right price is a critical goal. Its been my experience, admittedly with C5 ground ballscrews
rather than C7 rolled screws, that the cost and availability of the ballscrews themselves is the determining factor in the overall performance and price of the machine
that can be made with them.

I wish you luck Peter, one thing that must be said about Chinese manufacturers is that they make stuff at a price that is so low that it's hard to believe. I'm sure
with research you will find the right combination.

Craig

[peteeng]

Hi Mactec - I don't understand your question re: whirling ballscrews? Whirling is called whip in USA.

Joe and others - Balancing cost and performance has been a major juggle. Some items are cheap and some are dear. Trying to make something that has commensurate parts with the performance target is tough. Once you upgrade something this cost and "over performance" flows thru the machine and its hard to dial it back. So cost escalates. Takes several design rounds to figure out various design/cost spaces. Chinese machines are the competition but this machine is bigger then these and second generation buyers understand the value vs what the cheap Chinese machines offer. I'll keep plugging at it. I've started the drawing stack for HD so must be close.

Re: Whip or Whirling. There are three approaches to calculating this issue. 1) A simple equation but you need to know the eccentricity of the screw. This eccentricity is difficult to predict 2) The first vibration mode of the screw is the same vibration as its free mode vibration ie its like a guitar string plucked in the middle. This is a simple eqn and does not require a known eccentricity 3) Let the FE do it in a modal analysis It should be the same result as 2) . But so far 2) and 3) have given me different results. But the results have been that the 12mm dia screw would be fine up to >500rpm. Now I'm at 16mm it will be better so I have not revisited the numbers. But I shall to be complete. Now I'm back on line I can pick the motors and then complete the torque and whip/whirl numbers....Peter

Here's a bowl I'm working on at the moment as a side interest, the new chinese tool was getting a bit blunt by this stage!!

[peteeng]

Hi All - I've done some reading on whirling and "oil whip" and other transverse vibration effects of shafts. There's quite a few. But its clear that "whip" and whirling are separate phenomena. Oil whip is due to the clearance in journal bearings. This clearance allows the shaft to precess and at certain speeds vibrate badly. This is unlikely to occur in the bearings used in our CNC machines. Whirling is the effect that is caused by out of balance transverse forces on shafts. eg shafts with rotors in which the rotor is not uniform so uneven centripetal forces make the shaft vibrate. In the case of horizontal shafts the force of gravity is enough to deflect the shaft downward then the shaft rotates and has to spring back the other way (when that deflected bit of shaft is up and gravity sucks down) and this small springy action excites the shaft at its natural frequency. This is called Whirl. So learnt something today. Peter

[machinedude]

Hi MD - Thanks for the feedback, lock 10mm in Danno. So the 5mm lead gives heaps of force but not enough speed? Is that the reasoning? Peter

in my case that would be the main reason since i have a 4' x 8' work space. small machines can get away with slower feed rates and not have as much trouble this way. since plasma was an interest acceleration was another reason. right now i'm out in left field with the plasma and seeing what can be done with a milling head spindle i had laying around. plasma is still an interest but because the plasma cutters are so expensive it is last on the list. i'm trying to see if i can make different head for different processes so my build is all over the map i actually have the X and Y axis running at a 4 to 1 ratio to keep the speed in a better place on the torque speed curves of the stepper motors i am using. will see how that works out with the different process of these different heads im' building but is right where i need to be for plasma.

[peteeng]

Hi All - I have found the correct equation that agrees with the FE work. This example is a 30mm diameter shaft 4m long. I use this as it was a worked example in an article I read. In the Dia12 by 1.4m long case it will produce a critical speed of 1660rpm which is unlikely from the stepper. The new 16mm dia would be 2213rpm which is even higher. The eqn is for effectively fixed ends or described as long bearings. Short bearings are mentioned in the pdf. The FE calculates the whirl as slightly under the restrained end calculation even if the ends are free.

I've run Scoot at 25m/min. Scoot uses a 24mm diameter pulley and a 16mm belt, so this means 1 rev = 75.4mm so 25000/75.4= 331rpm (using a gecko 540). Looking at big N23 and N24 motors seems 600rpm is about half speed. So using the 1610 means that 10x600= 6m/min rapid or maybe 12m/min absolute max. But 6m/min for a small router is fine. Brevis no1 uses a stepper motor with 400steps to the rev and a 10mm belt I can get it to 15m/min no worries. But belt drives are easy to get moving fast. They don't like going slow.

Now I know the math I can use the FE in confidence, prefer that to using spreadsheets.

So now to size the motors... Peter

[mactec54]

Hi Mactec - I don't understand your question re: whirling ballscrews? Whirling is called whip in USA.

Joe and others - Balancing cost and performance has been a major juggle. Some items are cheap and some are dear. Trying to make something that has commensurate parts with the performance target is tough. Once you upgrade something this cost and "over performance" flows thru the machine and its hard to dial it back. So cost escalates. Takes several design rounds to figure out various design/cost spaces. Chinese machines are the competition but this machine is bigger then these and second generation buyers understand the value vs what the cheap Chinese machines offer. I'll keep plugging at it. I've started the drawing stack for HD so must be close.

Re: Whip or Whirling. There are three approaches to calculating this issue. 1) A simple equation but you need to know the eccentricity of the screw. This eccentricity is difficult to predict 2) The first vibration mode of the screw is the same vibration as its free mode vibration ie its like a guitar string plucked in the middle. This is a simple eqn and does not require a known eccentricity 3) Let the FE do it in a modal analysis It should be the same result as 2) . But so far 2) and 3) have given me different results. But the results have been that the 12mm dia screw would be fine up to >500rpm. Now I'm at 16mm it will be better so I have not revisited the numbers. But I shall to be complete. Now I'm back on line I can pick the motors and then complete the torque and whip/whirl numbers....Peter

Here's a bowl I'm working on at the moment as a side interest, the new chinese tool was getting a bit blunt by this stage!!

Whirling is a thread cutting process that they use for high precision thread cutting and mass production of medical screws Etc

https://leistritzcorp.com/machine-to...crew-whirling/

Can be set up to be done on most lathes also

[peteeng]

Hi Mactec - Seems whirling means different things to different people. Oil whip is different to whirling. See video or do search for whirling vibration. Whirling as a cutting process seems to be in reference to preventing whirling using rests. Thanks for the link



I've picked 3Nm motors with 10mm diameter shafts. The Z axis will have a brake on it

https://www.omc-stepperonline.com/ne...ter-mm%5B10%5D

This is the motor for the X and Y drives. At 600rpm, half torque and 90% efficency it will produce 86kgf of thrust using the 1610 screw.

https://www.omc-stepperonline.com/st...ter-mm%5B10%5D

This one for the z axis. At 90% efficiency and half torque (600rpm) will provide 200kgf thrust from the 1204 ballscrew. I have to check buckling loads now.

Cheers Peter

heres a link to oil whirl and oil whip stuff
https://www.machinerylubrication.com...oil-whirl-whip

[peteeng]

Hi All - Now I have picked the motors its time to check buckling of the screws. Using Eulers buckling eqn or Hiwins eqns we can check that the screws wont buckle under expected loads. The Z axis motor is 3Nm holding torque, this on a 1204 screw will produce 432kgf. The screw from bearing to bearing (being conservative, Hiwin goes bearing to centre of nut) is 326mm long. This buckles at a load of 609kgf so we are good here. I had to look up the root diameter of the screws for this as well vs the outside diameter. The gantry screw from bearing to bearing is 1400mm and its a 1610 (with 12.9mm root). The motor/screw can produce 173kgf of load. The screw buckles at 104kgf so we have an issue here. So I changed the position of the drive nut to "shorten" the screw. Its now 1225mm long and buckles at 136kgf so still no good The X dirn (dual drive) is 700mm long and buckles at 417kgf and remember it has two motors so its 2*173=346kgf so its good as well. These are quick checks not detailed checks by the way. So I may downsize the X drives a bit and I'll have to think through the Y drive (gantry width) or reduce its motor size.

If I use the OD of 16mm I get 1320kgf buckling load so flies in. I may model the screw to see what the actual buckling load of a screw is verses using its root diameter.....otherwise back off the motor size. Cheers Peter S

Some 30 years ago I was using a large lathe that had been converted into a metal spinning machine. It was used for tapering yacht masts and flagpoles. The drive screw was maybe 4m long. I did something wrong and the screw which was about 40mm diameter buckled up and banana'd if that's a word. I had to go to the owner of the company and explain myself. He was good about it and said he'd done the same thing himself and that the foreman had straightened the screw with oxy last time it happened. He said give that a go, So I did and was quite pleased that I got it workable again... shouldn't use long screws in compression....

[peteeng]

Hi All - I looked at the Hiwin formulas and they have some factors in them which are a little different to the first principle equations. So they have accommodated various things in them from theoir experience I expect. So I applied the Hiwin equations to this case and it looks good. The cross gantry motor with 1610 can push 173kgf and the Hiwin formula allows 287kgf, all good. The critical speed is 1228 rpm which is higher then I expect to go. These are also based on a span of 1400mm which is the bearing spacing. It's more like 1200mm to the nut so these are conservative. So I'm happy to proceed with the current motors and screws.

Excellent start, more detailing.,... Peter

[machinedude]

all this information is also referred to as shaft critical speed. there are all kinds of calculators for this online some more detailed than others. way to scientific for me i always hated the theory in school and enjoyed the lab day where you jumped into things and got your hands dirty it's all important stuff but i never get to wrapped up in the details myself. i have a hard enough time finding time to make things as it is so i tend to fly by the seat of my pants on details like this

[mactec54]

Hi Mactec - Seems whirling means different things to different people. Oil whip is different to whirling. See video or do search for whirling vibration. Whirling as a cutting process seems to be in reference to preventing whirling using rests. Thanks for the link
Cheers Peter

I guess anything that referred to a rotating shaft could be called whirling it has nothing to do with the rests or shaft supports it's is the cutting process where the cutter and shaft both rotate that is called the thread whirling some more and a thread whirling cutter

Long Shafts being support is required for all forms of precision machining Grinding Thread Whirling Etc..





So Whirling can be used for many different situations

v.intr.
1. To rotate rapidly about a center or an axis; spin.
2. To move while rotating or turning about: The dancer whirled across the stage. See Synonyms at turn.
3. To turn rapidly, changing direction; wheel: She whirled around to face him.
4. To have the sensation of spinning; reel: My head is whirling with data.

v.tr.
1. To cause to rotate or turn rapidly: whirl a baton.
2. To cause to move with a spinning motion: whirled the ball up into the air.
3. To drive at high speed: whirled the motorcycle around the corner.
4. Obsolete To hurl.

n.
1. The act of rotating or revolving rapidly.
2. Something, such as a cloud of dust, that whirls or is whirled.
3. A state of confusion; a tumult: The press room was in a whirl.
4. A swift succession or round of events: the social whirl.
5. A state of mental confusion or giddiness; dizziness: My head is in a whirl.
6. Informal A short trip or ride.
7. Informal A brief or experimental try: Let's give the plan a whirl.

[peteeng]

Hi Mactec - We need to add the whirling dervish to the list as well.. Peter

[peteeng]

Hi All - Back to business. Been looking at couplings and seems the disc coupler is the go. The flexible couplings only take <1.5Nm , the jaw couplings are good on torque 3Nm but are compliant so the disc couplers fit the bill with 5.8Nm rating. Anybody got couple stories? Peter

My story is to loctite everything together. The steppers have a history of shaking things apart....

[joeavaerage]

Hi Peter,
I have low lash (<2 arc mi) 10:1 planetaries on my mini-mill. It means that modest steppers give really great torque, at the expense of speed.

I used the spiral cut solid couplers because they were cheap and available. I realised not long after I got them that they have a certain amount of torsional twist
under load. Its not huge but as it turns out it dominates the sources of 'lost motion' in my machine. What I also discovered is that if I crash, and it has happen on a couple
of occasions that the coupler twists off like a carrot. In a sense it provides a mechanical fuse.....it was an unintended bonus, so much so that I have not replaced
them with more torsionally rigid couplers.

Craig

[mactec54]

Hi All - Back to business. Been looking at couplings and seems the disc coupler is the go. The flexible couplings only take <1.5Nm , the jaw couplings are good on torque 3Nm but are compliant so the disc couplers fit the bill with 5.8Nm rating. Anybody got couple stories? Peter

My story is to loctite everything together. The steppers have a history of shaking things apart....

Yes the Disc Couplers are superior to any other type if you have to use them

[joeavaerage]

Hi,

Yes the Disc Couplers are superior to any other type if you have to use them

Yes, I would agree but they are also $20 a piece whereas the spiral cut aluminum one are $20 for three.
Question...'does the extra expenditure result in a machine that is commensurately better...OR....is the cheaper alternative "good enough"?'

Craig

[mactec54]

Hi,



Yes, I would agree but they are also $20 a piece whereas the spiral cut aluminum one are $20 for three.
Question...'does the extra expenditure result in a machine that is commensurately better...OR....is the cheaper alternative "good enough"?'

Craig

That's a high price I have brought them at around $8.60 a piece here are some from this company which is a good brand

The cheap ones you have been using always have the breaking problem even with normal use

[peteeng]

Hi - Yes $20USD x 4 is a series chunk into the budget. Whereas I could just chew into the rating of the flexible type (1.5/3Nm) and get 3 for one. And a fuse sounds like a good idea on a starter machine. I myself in 40+ years of running machines haven't crashed one yet (wink wink) I remember being asked to run a subroutine once by a foreman on a welding robot. I advised its unwise to run an unfamiliar subroutine separate from its program (or programme) but he said lets do it. So the robot took off and it punched a hole in the steel safety screen. There were relative moves in the sub. The noise brought in everyone from the office and workshop plus the MD.

There are flexibles cut radially I think with bigger rating I'll have a dig for those. My current thought is that 3Nm is an extreme event and I like the idea of a fuse. There's a lot of potential thrusts in this system...and changing the coupling is easy vs whatever else decides to bend and its easy to upgrade if found to be an issue...Peter