[acannell]

I just ordered it and I'm guessing I'll get it in about two weeks.

It can tilt up to 90 degrees, controlled by a handwheel instead of just a clamp that you adjust by sliding it.

The part number is supposedly TSK160 but I've seen other TSK160's that dont have a tilt-by-turnable-handle feature.

Which makes it perfect to put a motor on!! So then two motors, one for rotation, one for tilt, and now you have a fairly legitimate 5 axis setup..although only 90 degrees instead of 180 youd get with a trunnion.

I will be sure to report all my experiences with it. It weighs something like 79 to 97 lbs (kinda not clear on the specs..big surprise)

It should fit into my mill with room enough to still be able to do work.

How can I put it through the MILL so to speak (pun intended) and analyze how bad/good it is?

Comments? Thoughts? IDEAS???

It goes for $449 + $85 shipping on ebay, but I found that ebay sellers website and ordered it there for only $425 + $70 shipping, and its in the USA!

its utoole.com Home Page

I didnt get the chuck or plate thats in the picture..

6" Prcision Tilting Rotary Table

[dwalsh62]

Size and power of drive motors are an issue when converting this type of table to CNC not to mention removing as much backlash as possible before the table has any real accuracy in a CNC application.

You could essentially lap the gears to get almost unmeasurable backlash and all it really takes is polishing compound, a reversible electric drill and time.

When you are done you should disassemble the entire table and wash every part well to ensure you have no polishing compound remaining which would cause premature wear and only use enough compound to lightly coat the worm gear during laping to prevent bearings from being messed up.

Of course if yours is a bushing type (not bearings) then you'll have to lap the bears outside of the case to avoid damaging bushings and the best you'll get on output is 0.15deg error minimum due to the slop in the bushings.

Installing a 1.0KW 90mm or 110mm servo motor requires a lot of room, I've found that a 600W 60mm servo motor with a 3:1 reducer will give the equivalent power of the 1.8KW motor but do it in a significantly smaller foot-print and for gear type tables you'll need as much power as possible due to the drag induced by the gears (not very efficient).


Of course the best solution would have been to buy 2 harmonic drives and build the table yourself, it's much easier than you think.

Two harmonic drives with the perfect ratio (72:1) such as 14-72 and 17-72 would give you a very accurate table and you can get even one digit resolutions, somehting you cna't get with 90:1 that most of the gear types are built with.

I've build many 4th axis using 14-72 harmonic drives with 4in and 5in 3-jaw and 4-jaw chucks, attached is a picture of a 4th axis with a 4in independant 4-jaw chuck using a 14-72 harmonic drive with a 100W 2500ppr quadrature encoder (10,000 steps per revolution, 4,000RPM rated motor metal was polished including the hardware and black zinc'd for that glossy black look), 100W at 72:1 gives 7,200W (7.2KW) at the output of the harmonic drive and yes, you can achieve great results with small motors using harmonic drives due to the efficiency of the reduction system.

Harmonic drives come in direct drive and belt drive configurations, a belt drive is useful if you want to have a through-hole (29mm for 17-72, 35mm for 20-72, 42mm for 25--72) but the through hole- types are usually more expensive but a 20 or larger size can be outfitted with an 5C collet system as well as a chuck giving more flexibility to the work piece and it's length and also easier to implement a hydraulic clamp.

If anyone tells you that 72:1 ratio is not available, stop talking to them, they clearly don't have enough knowledge about harmonic drives to help you.

72:1 is much better because it's 5deg per revolution of the motor, using a 2500ppr quadrature encoder (10,000pulses per revolution) gives 0.0005deg per step so it's easy to get 0.001deg (two steps).

90:1 gives 4deg per turn and 0.0004deg per step makes it impossible to get 0.001deg, 0.01deg is about the best you can get (25 steps) and hard to find in a harmonic drive

100:1 gives 3.6deg per turn and never resolves to a 1.

80:1 gives 4.5deg per turn and never resolves to a 1.

50:1 gives 7.2deg per turn and never resolves to a 1.

Using harmonic drives has many advantages, they can be ratio corrected by employing a timing belt and pulleys, 50:1 ratio can be corrected to 72:1 ratio by using a 50 tooth timing pulley (on the motor) and a 72 tooth timing pulley (on the harmonic drive), 25 tooth and 36 tooth give the same results and are smaller in size.

Since the belt is on the input, you don't have to worry so much about backlash or belt-stretch affecting precision and if you do the math you can see that a max error of 0.5deg on the input is about 1.25 steps and highly unlikely you could measure any error on the output using a timing belt (belts would be maybe 0.015deg of stretch under excessive load) and this would be about 0.00375 steps.

The output shaft on the 14-72 is about 40mm, this is good for up to a 4in chuck, to tilt a table it is more than adequate.

For a table with a 6in chuck I would probably recommend a 14-72 and 20-72 or 14-72 and 25-72 depending on the style needed.

The 4th axis in the picture (including motor, driver, chuck, material, harmonic drive, hardware, anodizing and zinc plating) was made for just under $400.00, adding another axis to it would basically only increase by the cost of the harmonic drive and motor and a little material, your $500.00 item needs work and by the time you're done will be around $800.00 and still not have a chuck on it and most likely be stepper motor driven.

Don't get me wrong, I don't hate stepper motors, I preffer something with significantly more power, higher RPM and error protection so don't start this stuff that you can get 3,000RPM from your nema 23 stepper motor as everyone knows that after 2,000RPM (assuming you can get it to run at this RPM without losing steps) the motor has no usable power and you can stop it with your finger and thumb.

I've also attached some additional pictures of tables which may give readers some motivation in building their own, none of the tables are using stepper motors so don't be fooled by appearance, the one with the blue fixture plate is using 100W 60mm servo motors with 1000PPR encoders (nema 23 stepper motors are 57mm).

[acannell]

Size and power of drive motors are an issue when converting this type of table to CNC not to mention removing as much backlash as possible before the table has any real accuracy in a CNC application.

"Real accuracy" is a relative term to the task at hand. Otherwise its just hobbyist talk to justify working on the machinery instead of the machining, IMO, which seems to be a favorite theme on cnczone.

100W at 72:1 gives 7,200W (7.2KW) at the output of the harmonic drive and yes, you can achieve great results with small motors using harmonic drives due to the efficiency of the reduction system.

How did you come up with this gem? Power goes down in any system not up. Try hooking up 7200W load to that harmonic drive and powering it successfully with 100W.

[dwalsh62]

How did you come up with this gem? Power goes down in any system not up. Try hooking up 7200W load to that harmonic drive and powering it successfully with 100W.

It would be impossible for power to go down and you don't have to be a rocket scientist to figure this out.

Gear reduction lowers RPM and increases torque, this is a fact, you obviously need a little more education on the subject, SEE: "HOW IT WORKS"

[acannell]

It would be impossible for power to go down and you don't have to be a rocket scientist to figure this out.

Gear reduction lowers RPM and increases torque, this is a fact, you obviously need a little more education on the subject, SEE: "HOW IT WORKS"

You certainly do not put 100 watts of power into something and get 7200 watts out of it as you describe here:

"100W at 72:1 gives 7,200W (7.2KW) at the output of the harmonic drive "

That is completely wrong and I am curious how you can be designing or assembling anything if you are working with ideas like that.

In any case, power certainly does go down in any system. Every system has loss of some kind and in a gear train most of that loss is friction. So with your 100W in you are doing pretty good if you get 98W out of it.

So you are 0 for 2 on your fundamentals about energy conservation. Maybe you'd like to go back and edit your posts before you continue to tell people about your motors and 4th axis machines designed with that kind of thinking behind them. I wont say anything if you do. Everyone screws up as a beginner!

[daniellyall]

pull it to bits and clean it before you use it, it will have crap in it

[acannell]

pull it to bits and clean it before you use it, it will have crap in it

id really like to..Im concerned that doing so may be more damaging then not doing it though..

btw I got a good deal on DTE medium which is supposed to be the oil fill for it..arriving today!! the fun begins sooN!!

[keebler303]

A 100W motor at 72:1 puts out as much TORQUE as a 7200W motor directly coupled to a load (ideally).

The 100W motor is still only giving you 100W of POWER (again, ideally), regardless of the gear ratio.

acannell has it right...

[acannell]

My new Asian friend...a scrap book

clocks in at 79 lbs

I measure less than about a tenth 0.0001" flatness across the table while not moving, and also rotating the table 360 degrees!!

[kb0nly]

Thats really neat looking! At least they seem to do a good job of crating it.

[acannell]

Thats really neat looking! At least they seem to do a good job of crating it.

they did except that they left the handle on the main wheel which then hits the wall..they could have just unscrewed it like they did with the tilt wheel

[daniellyall]

it does not look to bad you might be lucky with it

[aarggh]

How on earth are you holding that piece of aluminium while machining it?

cheers, Ian

[acannell]

How on earth are you holding that piece of aluminium while machining it?

cheers, Ian

theres a hole through the table and I drilled and tapped a hole in the block and used a threaded rod and nut..def not an approved method but for the test I just wanted to get something going

[acannell]

okay I took it apart to see what I could see.

I wasnt going to take it apart, but I tried filling it with oil through the little spring loaded ball "zerk" fittings, and one of them got stuck..the ball wouldnt re-seat..so I figured that was a good excuse to take it apart and fix that at least..meanwhile check for grit and see how it all works.

Long story short:

Wow there really isnt any oil in it! They say fill it before use and I guess thats for real. Things seemed greasy at least.

There are separate oil fittings and they all actually do work. They have little tunnels through the casting and go to various places, I tested them and the oil gets to where it should.

There are no ball bearings anywhere. Looks like just ground surfaces for everything.

The teeth on the worm gear feel sharp, and there are marks on the table teeth..not sure if thats normal, significant, or because I turned it a bunch without a lot of oil in there.

I'm fairly impressed with how things look at this price.

There does seem to be some grit of some kind. I cleaned everything out with alcohol.

That yellow paint flakes off really easy..looks like it was applied in a rush with a paintbrush.

Doesnt seem like an oil sump can be maintained..there is no seal to the outside really..oil in the body can come out between the table and body when its vertical. So I think regular lubing is a requirement. I think it would be possible to add a simply rubber oring of some kind or wiper to make a seal..probably a good upgrade and will make it so I can freely use coolant during my 4th axis shenanigans.

I put it back together and adjusted things for pretty much no backlash and barely any stickiness to the handwheel. I'll see how it works and maybe loosen it up a tad.

I'm not sure what I'll do for the broken zerk fitting. I had to pop it out. Perhaps just a rubber plug would work. Or I could tap it and put a set screw in or something.

[dwalsh62]

You certainly do not put 100 watts of power into something and get 7200 watts out of it as you describe here:

"100W at 72:1 gives 7,200W (7.2KW) at the output of the harmonic drive "

That is completely wrong and I am curious how you can be designing or assembling anything if you are working with ideas like that.

In any case, power certainly does go down in any system. Every system has loss of some kind and in a gear train most of that loss is friction. So with your 100W in you are doing pretty good if you get 98W out of it.

You mean everyone else in the world has it wrong????

Torque multiplication is a byproduct of gear reduction, google confirms this.

We have a a machine that installs 10mm bolts, it's driven with a 40W 90VDC motor, I've removed the motor to service it a couple of times over the last 5 years, you can stop the shaft with your fingers or prevent it from starting by holding the shaft with your fingers, the machine has a gearbox with a ratio of 200:1, if you forget to engage the clutch it will twist the head off of a grade 12.9 8mm bolt and do it effortlessly, I guess my fingers are stronger than the grade 12.9 bolts or the machine has some magical powers if your opinion has any fact.

Information taken from a torque calculator located HERE
The output torque and speed of a gear reducer with 200 :1 ratio,
input torque of 7.35 in/lbs: (torque of 40W 90VDC motor)
1,470.0000 in/lbs (rounded to the nearest 10,000th)
input speed of 9840 RPM is: (RPM of 40W 90VDC motor)
49.2000 RPM (rounded to the nearest 10,000th)

Even if there is 25% loss (highly unlikely but I'll humor you with an unrealistic loss), there is still sufficient power to twist the head off.

[daniellyall]

be nice

[Richard_C]

Dwalsh62
You seem to be confusing power and torque.

A 200:1(in : out) reduction gearbox multiplies torque by a factor of 200, rotational speed by 1/200, and, in a frictionless world, the power output would remain the same.
Unfortunately, friction exists. You will always lose some power to heat and/or vibration in a gearbox.
Mechanical power is equal to the rotational velocity multiplied by the torque.
Lets continue the 200:1 example above. Input is 2 rpm and 2 ft.lb torque, with power of 2 x 2 = 4 units of power.
Speed is multiplied by 1/200, and torque by 200.
Therefore output is 1/100 rpm, and torque is 400ft.lbs, and power is 1/100 x 400 = 4 units of power.

[dwalsh62]

Dwalsh62
You seem to be confusing power and torque.

A 200:1(in : out) reduction gearbox multiplies torque by a factor of 200, rotational speed by 1/200, and, in a frictionless world, the power output would remain the same.
Unfortunately, friction exists. You will always lose some power to heat and/or vibration in a gearbox.
Mechanical power is equal to the rotational velocity multiplied by the torque.
Lets continue the 200:1 example above. Input is 2 rpm and 2 ft.lb torque, with power of 2 x 2 = 4 units of power.
Speed is multiplied by 1/200, and torque by 200.
Therefore output is 1/100 rpm, and torque is 400ft.lbs, and power is 1/100 x 400 = 4 units of power.

Torque is a measure of power, the only one confused is you, everyone else seems to understand that a gear reducer increases output power but you seem to be hell-bent on proving me wrong instead of clarifying my statements and this is your own stupidity, best to let this drop rather than continue and further corrupt the thread.

[acannell]

Torque is a measure of power, the only one confused is you, everyone else seems to understand that a gear reducer increases output power but you seem to be hell-bent on proving me wrong instead of clarifying my statements and this is your own stupidity, best to let this drop rather than continue and further corrupt the thread.

Wrong.

Torque is a measure of force, not power. Power is a measure of work done per unit time.

If you have more power then youve put in then you have a perpetual motion machine. I guess I should alert you to the fact that that is an impossibility.

Its pretty obvious that a reduction gear train reduces the rpm and increases the torque. And since the power is the product of those two, the power remains the same, ignoring frictional losses, which reduce it in the real world.