Lathe: Infinitely variable vs Geometric speeds

[GLCarlson]

A recent thread- and several in the past- have discussed the use of the VFD on the lathe. There are good reasons to use one- it's the cheapest, and often only, way to run three phase motors in the home shop.

Beyond that, though, there seems to be a tacit view that infinitely variable speed is in and of itself a good thing, worth having on all machines. I wonder if that's so, particularly for the modern gearhead lathe. A well designed lathe comes with geometrically spaced speeds on the pulley or gearhead. Those speeds were carefully chosen to optimize performance and tuned over decades of experience. Most of us here, I suspect, cut aluminum, steel, or brass in sizes ranging from a quarter inch or so up to about six to ten inches. Available speeds are a bit slow for small diameters at least on my 13" Standard Modern, but still usable, and just fine for work in the 1-10 inch range with HSS bits and in most cases carbide. Tooling, technique, and skill affect my results; within reasonable limits, sfm does not.

So, a question for the experienced hands here. Is it really worth the effort to convert a well-designed lathe with geometric speeds to a VFD, absent any other reason to do so than to get infinitely variable speed? What objective evidence exists -if any- that this is worthwhile?

I understand the theoretical argument that one can run at exactly the recommended sfm, but in practice one is rarely at that speed even with a VFD- and in any case, the recommended sfm must be a range, not a unique single number. And, no doubt, if one is building or converting to direct drivewith a couple of speed ranges, VFD is the technology to use. The question is whether there's practical value in changing a well designed, fully functional lathe to gain nothing but variable speed.

My vote is "no", as evidenced by the simple fact that this remains a project I've never gotten around to doing, despite thinking about it with some regularity.

The discussion should be interesting.

 

[kvom]

The gears are chosen to cover a given range without too many gaps. If the motor runs at constant speed then the torque depends on the gear reduction and is constant for a given gear setting. With a VFD the motor speed varies, so does torque.

Unless you leave the gears/pulley fixed, then it will be difficult to determine the actual speed as you need to program the ratio based on the current gear. I suspect the real advantage of a VFD is the ability to quickly adjust speeds to eliminate chatter.

A major problem with VFDs is that the on/off/rev function needs to be under VFD control rather than with the lathe's normal switch. This is not too difficult with a mill, but is likely to be so with a lathe. For a 3-phase motor a rotary phase converter may be a better choice.

 

[velocette]

Hi
One big disadvantage with a gear head lathe and a single phase motor is that it does not run smoothly as a three phase or DC motor.
On my six speed gearhead lathe the uneven running was evident by chattering noises from the gears in all ratios.
I posted a question to the boffins on HMEM and got loads of valuable information why this was so.

http://www.homemodelenginemachinist.com/showthread.php?t=17383

Fitted a DC motor and controller and the difference was a very quiet running lathe with much smoother accelerate and decelerate with braking on the motor
Thread cutting is a breeze with slow forward and fast reverse set from the control panel.

Keep the gear box and select the approximate ratio and as you reduce the diameter of the work you can tweak up, the spindle speed.
A basic rev counter on the spindle will help.
One more addition that will help is an amp meter on the motor leads.

Eric

 

[WOB]

It is obvious that you have never used a lathe with an infinitely variable speed spindle long enough to appreciate the advantages. And I'm not going to start enumerating them all now. My lathe has a 2 HP motor/VFD that has a 24:1 speed range available at my finger tips and on the fly. At 54 ( 5 hz) rpm , I can ream and power tap as well as find the high spots of a workpiece in the 4-jaw with a test indicator in the toolpost. At 1300 (120 Hz) rpm, the chips come off fast, hot and smoking blue. If I turn a large disc, I can start at the outer rim at speed and increase the speed at the tool approaches center, keeping the sfpm rate fairly constant and obtain a good finish all the way in. The 3-phase motor runs smooth and quiet with none of the annoying torsional vibration that all single-phase motors produce.

Finally, look at the trouble and expense that the tool room lathe makers like Monarch and Hardinge went to to provide variable speed. Their customers demanded it and for good reason. Quit thinking like a model maker and try to think like a real machinist. You simply don't know what you are missing.

WOB

 

[IanN]

Hi WOB,

Thank you for adding some humour to the thread with your hilarious comments about continuously variable speeds:

<SNIP>
look at the trouble and expense that the tool room lathe makers like Monarch and Hardinge went to to provide variable speed. Their customers demanded it and for good reason. Quit thinking like a model maker and try to think like a real machinist. You simply don't know what you are missing
</SNIP>

As I'm sure you are aware, the introduction of variable speed was a major cost reduction exercise - initially replacing expensive gear wheels with cheap variable diameter pulley systems and then replacing these with even cheaper electronics.

Speed selection defines the rate of metal removal (time takes to machine an item) and tool wear (the number of items that can be machined before replacing tooling inserts or regrinding tooling and resetting tools, with the inherent machine downtime associated with these activities). This equation can make the difference between profit and loss on a job for a commercial operation, but is of minimal interest to the hobby user.

In my case, I estimate I spend 5% of my workshop time machining - the rest of the time is spent in thinking about how I will tackle a job, setting up the machine, finding the material and tooling, setting up the work, etc. Often I will spend ten minutes or so just enjoying a cup of coffee and admiring a particularly fine finish I've just produced on an item.... "Efficient machining" for me means getting the maximum life from my milling cutters, reamers, taps, dies, drills, etc. I do not have a tool and cutter grinder and so cannot easily sharpen milling cutters, reamers, taps, etc so taking more time over the machining and getting longer tool life is more important that maximising product throughput.

Based on the above, fine tuning the machine speed is of no importance, and it is worth remembering that the quality of a machined surface has almost no dependency on machining speed: It is almost impossible to cut too slowly (although at very low speeds chatter tends to become an issue), slow speed just means the job takes more time to machine. It is almost impossible to cut too fast, you just end up destroying the edge of your tools unnecessarily quickly.

This means there is a real benefit for me to cut at surface speeds much lower than those "recommended". My lathe, a Drummond made in 1910, has three speeds (plus a back gear which gives me three more, very slow, speeds) so I can choose a cutting speed of "fast", "medium" or "slow". This limited number of speed options has never been an obstacle to the production of high quality work.

Electronic speed control is of enormous benefit in CNC applications because it allows simplified computer control of speed, but for manual operation electronic speed control is used for the simple reason that good quality gears are expensive, and electronic circuits are very cheap.

All the best,
Ian

 

[kvom]

My Monarch 10EE lathe has continuous variable speeds 0-2400 and uses neither cone pulleys or cheap electronics.

That said, a quick change gearbox is quite sufficient for most uses.

 

[GLCarlson]

The original question was whether to replace a functional constant speed drive with VFD/DC drive. Implicitly, in a home shop (or prototype development) environment.

To summarize the comments (and my thoughs about them):

Reasons in favor:

Change to 3 phase to lessen vibration
The 1 vs 3 phase vibration argument is theoretically sound, but doesn’t seem to be an issue in practice in most cases. A 1 phase motor that cogs or wiggles is probably a bad motor.

Threading is easier
A valid argument if one has not mastered threading at speed, the lathe doesn’t have a threading dial, and the lathe is the only available tool. None of those apply in my shop. Mostly, I threadmill on a CNC mill these days.


Better sfm control on large facing cuts, if one adjusts during the cut
Theoretically true. Is it supported by any examples of broad utility in practice? None were given, nor have I seen any elsewhere.


Chatter management
As above, with the additional observation that one has feed options as well. A good machinist can get great finishes with a gearbox.


Reasons against (mostly from my own list):

IanN’s discussion of the realities of the prototype shop environment is spot-on.
Even though I do have full tool grinding capabilities including carbide, tool life is important. As is machining strategy, fixturing, and so on. Throughput is secondary to fit, finish, and accuracy.


Major rewiring job even if 3 phase motor already in place.
VFDs require a different contol design and are (judgementally) more fragile than other phase converters. Older motors -even three phase- may not work well with VFDs.

The speed-torque curve depends on the drive chosen (VFD or DC)
VFDs have a speed-power optimum. Constant speed gearhead multiplies torque- high power at low speeds. Torque at low speed is sometimes irreplaceable.

Other:

Run 3 phase motor: no need to change motor or shop power
VFD is the acknowledged answer here, not the question at hand



It must be a good idea, Monarch/Hardinge did it.
My first thought was that Ford made Edsels, too. IanN nailed it- gears are expensive, silicon is cheap - which probably explains why tiny machines are all DC motor drive. More to the point, even today very few OEMs offer a VFD/DC drive except in CNC lathes (Kent is an exception); gearhead seems to be the default and most common choice for machines with a swing over a few inches. Though I admit I’ll look at a Kent if my Standard Modern ever dies, and a 10EE with modernized drive would be interesting too- as a replacement. The question was conversion.


Try it, you’ll like it.
Not an answer. The question was, why will I like it? Too many reasons to list isn’t a reason.

I’ve been trying to “think like a machinist” since I first read “How to Run a Lathe” in 1961. I’ll keep trying.



I’m reassured that I haven’t really missed any major reasons in favor of such a project, and -until and unless a really good one comes along- I’m going to stick with my geometric gearbox and quick change feed.



I will admit to a bit of tool envy when I hear about a 1910 Drummond. I have fond memories of my first lathe, a 1919 13” South Bend. Old iron can do great work.

 

[Blogwitch]

Goodness me you lot, it isn't the machine that thinks it is cutting metal, but the person who is running it.
I keep going on about experience, and without it, I personally would be completely stumped with half the jobs I have done.
Putting something like this onto a machine will not make you a better machinist, just a better equipped one.

Now to the nitty gritty.

If you can use your machines as they are, then there is no use fitting variable speed, just carry on as you are but don't pull others to bits because of their thoughts and wishes.
I have an 18 speed gear head lathe, but at times, I could do with something between two gears, maybe to get a better surface finish, kill resonation or sometimes even be able to machine a certain material to a good standard, plus also the ability to cut threads using taps and dies on my machine at very low speeds.
Try a bit of ali bronze or certain stainless compounds if you think you know it all. You only have to be a few revs out and you can say goodbye to the job.

It is for that reason I am letting a friend fit a 3 phase motor and VFD to mine.

KVOM,

A major problem with VFDs is that the on/off/rev function needs to be under VFD control rather than with the lathe's normal switch.

I have been reliably informed that when mine is upgraded, everything will work on my machine as it does now, using the on/off, fwd/rev handle plus I will also be able to use the pendant controls, it is all to do with how the VFD is wired and programmed.
The only work I will need to do is put on a cheapo tacho to see what speed I will be doing.

Bogs

 

[hanermo3]

My opinion and experience.
I fitted my (12x, industrial, 1.5 kW, 12" lathe with the best possible option - a brushless ac servo drive with (very rigid and accurate) timing belts.

Technically, the lathe is == 5x more powerful.
It has the same practical torque as a modern cnc 11 kW industrial lathe, of over 50.000$.
A good industrial cnc lathe has 102 Nm Peak at 1300 rpm, 11 kW.

The 90 Nm is at all speeds 0-1000 rpm (belt drive wont support it under maybe 50-100 rpm or so).
I used 3:1 for 300 rpm servo : 1000 rpm spindle. 24:72 teeth.
Cons.
Drive kit == 1500&#8364;.
Belts == 200&#8364; (HTD 8 mm profile, 30 mm wide, heavy strong rigid mount. Lots of hours.).

I almost never used the 2-speed belt drive, and in 10 years only shifted it twice.
99% in 0-600 rpm.
Now, I get 0-1000 rpm.
All my stuff is in steel, mostly 10 mm plus D.
0-1000 rpm is ok to great for me.

Pros:
Perfect results.
2-4x more torque.
CSS.
C axis.
Never break anything major ! Servo faults and thats it, on impact/overload. (0.005 sec fault time.)
Limitless acceleration. Sanity keeps me to reasonable numbers.
10-30x accuracy on surface speed, perfect feeds, perfect rpm, perfect c axis tracking.

I´m not necessarily advocating this - especially since I import and sell servos etc.

Just pointing out that if $$ is not the issue, relatively modest investments may bring big benefits in some use.

I am changing out the mill (2.2 kW VFD) to servo.

 

[kvom]

I have been reliably informed that when mine is upgraded, everything will work on my machine as it does now, using the on/off, fwd/rev handle plus I will also be able to use the pendant controls, it is all to do with how the VFD is wired and programmed.

What needs to happen is that the lathe controls need to be detached from the motor and instead be wired to the VFD. That what I mean by under VFD control. What you can't do is just place the VFD between the wall socket and the motor and have the motor function operate as before.

If the lathe has a manual brake it will work, but if not adding a braking resistor to the VFD will be useful.

 

[Nick Hulme]

If the lathe has a manual brake it will work, but if not adding a braking resistor to the VFD will be useful.

Although not usually quite as effective as a braking resistor some better VFD's have the required circuitry and can be programmed for DC injection braking.

- Nick

 

[Nick Hulme]

Calculating speeds is very easy.

Your VFD will generally run from 0% - 150% of standard frequency/motor speed (although with a motor rated for the rpm 200% is do-able)

At 100% all gears give 1x rated speed at 110% it's 1.1x and at 200% it's 2x
Easy ;-)

- Nick