What's new
What's new

Tsugami C300 CNC Lathe accuracy

sukumaran

Plastic
Joined
Apr 24, 2018
What is the dimensional accuracy of Tsugami C300 CNC lathe Machine. My requirement is to maintain 0.002 mm dimensional accuracy in hart turning process.
 
Last edited:
Kind sir, this machine you are referencing has most best dimensional accuracy and provides number one quality in all applications.
 
What does Tsugami say?

I would want a guarantee of performance/ turn-key on a deal like that.

How are you going to measure it?
 
Choice of machine is just one piece of the puzzle.

2 microns is difficult to hold even in a grinding application.

Holding the part in your hand can warm it enough to grow the diameter by more than 2 microns.
 
Extremely difficult if possible in all on lathe. In the past I coped with task like this Comparative Measurment but the demand was 0.01 mm only. And this was hell of the task.
Pay attention:
My requirement is to maintain 0.002 mm dimensional accuracy in hart turning process.
As I understand it is not turning accuracy only demand, but the demand of the whole process, including production, QC etc.
Anyway - good luck.

Stefan
 
Gonna be hard or impossible.
Imho.
Or expensive.
Depends.

I think hard turning needs a hefty cut, no ?
And with the very hot cuts and chips, the thermal issues are likely to be above 10 microns, 0.01 mm, ..
In D and length and feature positioning.

And I suspect that roller burnishing is not going to work, or is it ?
I have no idea if Your material can accept roller burnishing to finish with.

So your process would need to pre-account for the thermal issues and cut to gcode with the final part ending up to spec after it cools.
This expects a lot of stability from your machine, material, everything.
Challenging.

It might be possible to let the part cool, and then finish with a diamond tool custom made, perhaps cutting front and back simultaneously to even out the stress.
2 inserts a given size apart, doing the final cut in a fixed way, might achieve this.

A lot depends on the speed size volume and features required.
Cutting 2 bearing seats on a shaft seems feasible.

Especially if manual QC and adjustment is acceptable.

Complex geometry with defined locations to 2 microns seems not easily feasible.
At least in a production setting.
Doable, yes, easy, no.

I might / would expect to incorporate digital dtis w. spc and feedback loops, testing the features for size, and automatic program adjustments via macros using the spc data.
I would expect to spend about 2-3-4 weeks building this, with 1000$ worth of kit (6k$+ if using name brand stuff).

Your original post seems to indicate some level of production, and volume, and this makes it both harder and easier.
Easier because significant costs can be spread over lots of parts.

Harder because it is likely to be hard to very hard to make some features to 2 microns, depending.
If it´s just 2 bearing fits on a shaft,
I don´t see a problem,
if they are expensive parts and don´t need to be made in high volume.

But hard turning would tend to indicate high volume or hard (or expensive) material.

I don´t see the Tsugami lathe itself as a limiting factor at all.
Any competent machinist can use it to make a couple of cylindrical features to 2 microns, on a shaft for example.

But making eg. complex cones for rocket impellers, perhaps in inconel, $$$ material, ... ?
Or very high speed turbine housings, like for nuclear centrifuges ?
Or space rocket fuel pumps running at 200.000 rpm ?

My "feel" for above is that "sure ! I´ll do it" followed by a very hefty NRE charge to *test* capability, with customer supplied testing equipment and go/nogo gages and material provided by customer.
My "feel" is that I would expect at least 100k in NRE charges paid up front to test/make any of the mentioned stuff.
The same applies to automotive.
They expect endless liability and perfect delivery .. so let them pay for it up front.

Lots of auto parts are made to 2 micron tolerances, cheaply, in volume of millions per year.
Diesel injectors are an example.
Gearboxes are another.
But getting there cost 2 decades and lots of millions.
 
Do not index your turret.
Do not use any rapids.
Do not make any unnecessary moves. Do not home the x axis after part is done.
Use flood coolant.
Make sure the final pass on the od is consistent.
Make sure shop is climate controlled.
Keep rpms on the low side.

maybe, big maybe you can hold .002mm.
 

Some of you here might get a kick out of this ^^^.

If pushed for time scroll to 17 mins in and you see them using the "LOL" method (Back in the day; Little Old Lady) - method of statistical process control for sizing and more critically groupings of sized individual rolling elements for tapered roller bearing (Gabet(sp)) assemblies.

It's a scream (although quite didactic) to see these automated light bulbs lit at various bins and trays to direct the QC specialist / technician place a particular sized roller to a particular bin.

The statement to me "2 micron dimensional accuracy " ~ seems almost meaningless - I know it means more to
most of you guys.

What's kinda cute in this video - [I think it's Circa 1962 - ish (I may be wrong) ] - is the Moore-like metrology processes,

So even for a manual lathe 30 millionths roundness (on a test artifact) as plotted and tested for the Colchester lathes seemed routine / doable.

Today newer Renishaw tricked out (3d strain) probes and other seem to be creeping in on "In process" gauging so at least (theoretically) you can have a smaller "Band" of parts within tough/difficult form tolerances. I know there are older made in Switzerland sensors and gizmos for in-process gauging that have been used for many years on round parts.
 


Today newer Renishaw tricked out (3d strain) probes and other seem to be creeping in on "In process" gauging so at least (theoretically) you can have a smaller "Band" of parts within tough/difficult form tolerances. I know there are older made in Switzerland sensors and gizmos for in-process gauging that have been used for many years on round parts.
It has nothing to do with the probe itself. All of them are repeatable enough. The measurement tool are the axes of the machine itself, and this where the process control should be aimed.
 
It is all about repeatability. For hardturning on a lathe inprocess gauging does not really work. It isn't ocilating back and forth like a cylindrical grinder taking small amounts of material per pass.
 
It is all about repeatability. For hardturning on a lathe inprocess gauging does not really work. It isn't ocilating back and forth like a cylindrical grinder taking small amounts of material per pass.

For hard turning that's a very good point,

I.e. depth of cut and heat / annealing at the surface means you can't have a corrective finish cut.

but, at least the next part can be "Comped" somewhat. For lesser value parts in number.
 
It has nothing to do with the probe itself. All of them are repeatable enough. The measurement tool are the axes of the machine itself, and this where the process control should be aimed.

Indeed !

But increasingly I'm seeing higher "Accuracy" "precision" and "repeatability" probes being specc's as standard with various machines.

For example Rengauge / RENGAGE™ like OMP 600

OMP600 high-accuracy machine probe

RENGAGE™ technology

Rengage diag 1.jpg <--- Click to blow-up.

@PROBE you'd be in a good position to de-bunk or partially endorse their utility ? (perhaps). [Perhaps beyond a more newer shinny more expensive thing to sell people on devised by Renishaw.].

This become interesting / almost controversial RE: Mazak's AG (All gear machine) (integrex) B axis gear oriented machine + 3d mapping of parts in-situ using 3d -ish (strain gauge) based probes + easier interface to drive or "Tame" all these processes or aid in the building of a stable process for a particular part family.

I.e. the basic question; is it possible to iteratively converge on an accurate process that is statistically meaningful, controllable and useful - ON the machine (Studer Grinders are exempt from this ;-) ish) OR should everything be ultimately controlled by separate inspection ? And perhaps slightly more on-topic as per OP's general opening statement / post ?



i-200 Integrex AG (GEAR machine) does anyone know anything about this machine ?

Somewhat cringe making ^^^ from me / my end. But good overall debate.

More useful details have emerged about the AG machine and process and capability (since then).

Hardinge have pushed for a long time hard turning but I have to admit I'm more a fan of cylindrical grinding and other grinding based processes. Hard turning can be useful.

Concentricity, roundness, ovality, (trilobarity ? (sp)?), tapericity , cylindricity, etc. blah etc. blah VS.

<snip> My requirement is to maintain 0.002 mm dimensional accuracy in hart turning process.

OP seems rather enigmatic on what is really wanted.
 
Indeed !

But increasingly I'm seeing higher "Accuracy" "precision" and "repeatability" probes being specc's as standard with various machines.

For example Rengauge / RENGAGE™ like OMP 600

OMP600 high-accuracy machine probe

RENGAGE™ technology

View attachment 322595 <--- Click to blow-up.
The repeatability of both types (standard and so called Rengage) depends on the ability of the deflected stylus to return exact to it's initial position. In both types of probes it is done by mechanical system (invented in 19th century by Sir James Maxwell), called kinematic stage. The difference is that Rengage is generating the trigger signal using strength gage, while standard probes - using electrical normally closed contact. Renishaw claims, that Rengage assures the repeatability of 0.5 microns, comparing to 1 micron of standard probes. Assuming that this is true, this 0.5 micron "drawback", according to my personal experience, is negligible in metal cutting machining world. But Rengage are priced much higher than standard probes, so why shouldn't the producer try to convince us, that using it will let us get better results ? BS.

Stefan
 
Do you guys not realize that the two micron tolerance is only the "Indian" tolerance. The actual tolerance is probably +/- 10 microns but tell the Indians that,they'll end up making parts that are +/- 30 microns, and call it best quality. With a 2 micron tolerance, there's a "possibly" good chance that the finished parts will be within the actual size needed.
 
Do you guys not realize that the two micron tolerance is only the "Indian" tolerance. The actual tolerance is probably +/- 10 microns but tell the Indians that,they'll end up making parts that are +/- 30 microns, and call it best quality. With a 2 micron tolerance, there's a "possibly" good chance that the finished parts will be within the actual size needed.

That's what they said back in the day about the...

Japanese/ Japan,

Hong-Kong,

then Taiwan,

South Korea,

then Singapore,

Mainland China,

Vietnam,

etc.

etc.

I'm not descended from SE Asia nor the Indian sub continent but I have a lot friends who are (over the years); broad strokes - the Indian Government (recovering from British Imperial rule) have over the last five decades at least have been rather cagey (understandably) about foreign investment and development - [to the point of formally legislated restrictions and various moratoria to prevent the ingress of foreign companies and foreign $ ]. Historically it was a slippery slope for India trading with the Dutch East India Company and then later the "British Empire" through the 17th, 18th and 19th Centuries - through trade they lost their country and autonomy and were ruled by a foreign power. India has been through many upheavals to scrabble back to some semblance of independent power. Right now Covid-19 is really kicking their butts (unfortunately) ~ overwhelming.

I'm pretty sure you don't have to scratch too deep to find precision work being carried out in India.

So from an American perspective is India "Friend or Foe" ? or just another global manufacturing hub ?

_______________________________________

But for sure a distribution curve of many parts would yield 5% of the total parts being "In tolerance" (that could be improved upon) - Good argument for on floor inspection or separate inspection department to "guide" processes. And more forward looking to automation where this is ALL carried out by 'Robots" wherever they happen to be installed (globally).
 
Hard turning

Do not index your turret.
Do not use any rapids.
Do not make any unnecessary moves. Do not home the x axis after part is done.
Use flood coolant.
Make sure the final pass on the od is consistent.
Make sure shop is climate controlled.
Keep rpms on the low side.

maybe, big maybe you can hold .002mm.
Thanks for your suggestions. We are doing this kind of components in a Miyano lathe with lenear tool post plus minus 0.002 microns tolerence about 1200 Nos per day. Component weight is only 4 gms. This process we canot do on grinding machines. No coolant must be used. Only special carbide tools should be used which can machine components up to 62 HRC. If coolant is used tool will break. 6 bar air can be used for cooling the edge of tool with builtin airlines supply through the tool holders.
Sukumaran.
 
Thanks for your suggestions. We are doing this kind of components in a Miyano lathe with lenear tool post plus minus 0.002 microns tolerence about 1200 Nos per day. Component weight is only 4 gms. This process we canot do on grinding machines. No coolant must be used. Only special carbide tools should be used which can machine components up to 62 HRC. If coolant is used tool will break. 6 bar air can be used for cooling the edge of tool with builtin airlines supply through the tool holders.
Sukumaran.

That's great ! interesting about use of air.

I guess you mean +/- 0.002 mm

(not micron) + / - 0.002 μm <--- how many nano meters is that ?

or +/- 2 micron - 4 micron spread.

under the mythical "two tenths" "all day",

___________________________________________________

I believe that machine has some kind of thermal sensor and thermal compensation and something 'Special" about the way the X axis homes. I think the rest of the machine is as per 'Normal" what ever that is these days :-)
 








 
Back
Top