In Search of A Better G0704 Spindle

[MontanaAardvark]

If there's one area of my mill that seems to make sense to upgrade, it's the spindle. At least, if I keep working mostly in aluminum. I use G Wizard Speeds and Feeds calculator, and I find that it virtually always tells me to max out my spindle at its 2200 RPM. On steel (and I don't have a ton of experience here) the RPM is usually much lower and the spindle seems adequate.

I'm really not that determined to increase the horsepower. From what I see in G Wizard, most of the time I'm limited by cutter deflection and not the spindle HP. Plus, I think these are systems, and if the motor is too powerful for the rest of the mill, it's more likely to cause the mill or the drive components (ballscrews, etc.) to deform too much. I'm not going to throw out a good deal on a 1-1/2 HP motor, but I'm not looking for a 2 or 3 HP option, either.

I have the option of a conventional 220 outlet near my mill. If I stick with 120V, the computer, lights, control box, and everything are running on one 15A circuit. I don't think adding more power on 120 is a good idea.

What I'm looking for is something in the 1 to 2 HP range, that will do 5000 RPM (as a guess). 220 AC is preferred. I'd like to control it from inside Mach3, so I think that means an encoder and I suppose that means replacing the control box on the side of the head stock. I expect I'll have to rebuild the headstock, at least partially. Probably improve the bearings. It would be nice if got quieter.

I really don't know what I'm getting into or how involved it's going to be. Can the whole thing be done for $500?

 

[JimDawson]

Can the whole thing be done for $500?

Probably not using new hardware, using used hardware....maybe.

220 (240) volt is the only way to go. Not sure that Mach3 requires an encoder for spindle control, I don't think so. You will need a BoB that has a provision for spindle control if you don't already have one. You will want a 3 phase, 4 pole (1725 RPM) motor and a single phase input VFD rated for the motor. Most of the small (<5hp) Baldor motors are rated at 6000 RPM max, I wouldn't try that with an import motor. Carefully check the motor specs before running one up that high. I would recommend going direct drive and use a sensorless vector VFD. A 2 HP would be my choice for your machine, you don't have to use all of it, but it will give you better low speed performance. I can run mine from about 10 to 5000 RPM. My mill is quite a bit bigger than yours and I'm running a 3 HP Baldor, has plenty of power.

For the bearings, you're on your own there. Can't offer any advice.

 

[dave2176]

For spindle bearings look to what Hoss did (Daniel last name fell out of my brain at the moment). He has G0704 dot com. I would also go with a 2 HP 3 phase motor plus VFD. Running a 3 phase motor on single phase 220 only produces 2/3 of the rated horsepower. Starting with 1-1/2 HP would only yield 1 HP output and the ponies seem smaller than they used to be years ago.
Dave

 

[spumco]

Yes, you can - but just barely and depending on components selected.

Basic components:
220v 3-phase Ebay motor, look for "vector duty" (I'm partial to Marathon BlackMax motors)
VFD with 220 1p input (as Jim mentioned) and 3p output. Automation direct's GS-3 series is quite nice and has optional high-speed encoder inputs. I went with a Hitachi VFD, and it's worked well.
BOB (or sub-board) with 0-10v output to control the VFD. Mach3 outputs a PWM signal that the BOB converts to the 0-10v

Optional upgrades:
- Belt drive conversion - highly recommended. You won't need the gears if you get the right motor and pulley ratio. Determine lowest reasonable speed you'll be cutting anything (for me it's tapping at 500RPM), and the max spindle speed you want. When you're shopping for a motor, check the name plate to see what the max safe RPM of the motor is and multiply that by your pulley ratio. There's your top end.

Belt drive also lowers the head weight, and is much quieter. No more oil in the head - no more dribbling. No gear backlash so rigid tapping is possible.

Your spindle bearings will eventually get wrecked, but just run the stockers for a while with some good grease and when they give up the ghost replace them with as nice a set as you can afford. Don't worry about them right now.

- Encoder for spindle and/or motor. This will permit rigid tapping and much more accurate speed control if the VFD gets information back from the actual motor RPM. No more 'bogging' down on heavier cuts unless you overload the VFD completely. It also permits significantly increased torque at low RPM (like 200% over name plate in some cases), so your 'loss' of low-end torque with a 1.5:1 belt drive is no longer an issue.

Check your control software (Mach3?) and see if it is capable of dealing with a pulley ratio on the motor while still permitting rigid tapping. If not, you may be stuck deciding if you want rigid tapping or feedback to the VFD - you can't have both in this case.

- External braking resistor for VFD. Permits stopping the spindle RIGHT NOW.

Here's a really slick idea...

Buy a big 1.8kw (2.5hp) servo and DYN4 drive from DMM Technology and install it as the spindle motor. No separate VFD needed as you'll be using their servo drive to power it. The drive takes 220v 1phase. Already comes with an encoder, and Mach3 is capable of driving a spindle with step & direction pulses. You get:

1. Smaller, lighter motor than the equivalent power induction motor
2. Motor is new, not used.
3. Motor and drive are a set from the same company - no tearing your hair out with incompatible signal voltages or encoder signal types. Trust me, this is huge.
4. Built-in extremely nice encoder, with - and this is important - encoder signal pass-through features. The encoder signal can be sent to Mach3 or other controller so you can have a real-time RPM display.
5. Factory made shielded cables so you aren't fiddling with sizing big 8ga wires from the VFD to the induction motor and chasing down RFI/EMI. VFD's are noisy.
6. Spindle positioning capability. Make a single point tool and have the spindle index it - instant broach for square corners or internal splines!

Has a 3000RPM max, so a 2:1 pulley ratio would be the ticket for your aluminum milling. And you'll have gobs more low-end torque, even with the pulleys, than your current set-up.

$378 for the motor
$267 for the drive
$60 for two cables (signal and power)

OR...

Go with the .75kw (1hp) 5000RPM servo (half the price) and do a 1.25 or 1.5:1 pulley ratio. Less money, still plenty of torque down low for steel cutting and bigger drills, and potentially a higher top end.


-S

 

[Boswell]

Here's a really slick idea...

I have not had a big desire to modify my gearhead PM-45M CNC mill, but this is a very compelling plan.

 

[MontanaAardvark]

Here's a really slick idea...

Buy a big 1.8kw (2.5hp) servo and DYN4 drive from DMM Technology and install it as the spindle motor. No separate VFD needed as you'll be using their servo drive to power it. The drive takes 220v 1phase. Already comes with an encoder, and Mach3 is capable of driving a spindle with step & direction pulses. You get:

1. Smaller, lighter motor than the equivalent power induction motor
2. Motor is new, not used.
3. Motor and drive are a set from the same company - no tearing your hair out with incompatible signal voltages or encoder signal types. Trust me, this is huge.
4. Built-in extremely nice encoder, with - and this is important - encoder signal pass-through features. The encoder signal can be sent to Mach3 or other controller so you can have a real-time RPM display.
5. Factory made shielded cables so you aren't fiddling with sizing big 8ga wires from the VFD to the induction motor and chasing down RFI/EMI. VFD's are noisy.
6. Spindle positioning capability. Make a single point tool and have the spindle index it - instant broach for square corners or internal splines!

Has a 3000RPM max, so a 2:1 pulley ratio would be the ticket for your aluminum milling. And you'll have gobs more low-end torque, even with the pulleys, than your current set-up.

$378 for the motor
$267 for the drive
$60 for two cables (signal and power)

OR...

Go with the .75kw (1hp) 5000RPM servo (half the price) and do a 1.25 or 1.5:1 pulley ratio. Less money, still plenty of torque down low for steel cutting and bigger drills, and potentially a higher top end.

This is a really cool sounding idea. Seems like it hits all the features I'm looking for.

My whole thing here was "what are other people doing, or wish they should have done?"

I need to look around a bit.

 

[spumco]

When my motor, VFD, or anything else spindle-related breaks, this is what I'm going to do. I may not go with the DMM servos, but I'm absolutely going to have a spindle servo. All the 'grown-up' VMC's I've seen or investigated use servos because you need the indexing/homing capability to align the drive dogs on CAT30/40/50 taper tool holders.

I researched used servo motors and drives for a while before punting when I stumbled across a really nice, really cheap Marathon motor. I was pretty overwhelmed & discouraged because new servo/drive packages from brand-name sources are unbelievably expensive, and trying to find just the right motor-drive match in a sea of used ebay stuff takes a degree in servo tech.

If I ever start making things for profit and not just fun, the first $1k my mill makes me is going towards servos for the x/y/z axis. The second is going to a higher-speed servo spindle setup I outlined above.

 

[MontanaAardvark]

... and trying to find just the right motor-drive match in a sea of used ebay stuff takes a degree in servo tech.

Ain't that the truth. I could use a primer on spindles, motors and controlling them. Applications notes from the manufacturers are usually a good place to learn. I need to learn what techniques are common, who makes what parts, the trades between different approaches and all that practical knowledge that comes from working in a specialty.

I understand servo controls in general. I'm a EE and I designed feedback loops for a living for 30 or 40 years. In very different context. No motors were involved, everything was at frequencies motors can't work at. I should be able to understand this stuff, but it's a whole different field from anything I've had to study.

 

[cjtoombs]

For spindle bearings look to what Hoss did (Daniel last name fell out of my brain at the moment). He has G0704 dot com. I would also go with a 2 HP 3 phase motor plus VFD. Running a 3 phase motor on single phase 220 only produces 2/3 of the rated horsepower. Starting with 1-1/2 HP would only yield 1 HP output and the ponies seem smaller than they used to be years ago.
Dave

The loss in horsepower is a feature of solid state phase converters, not VFDs. VFDs will provide the horsepower they are rated for at the machines design RPM. Lower RPM than the design RPM decreases horsepower linearly (half the design RPM, half the rated horsepower). This is why folks above are recommending 2 to 3 hp motors, because of the HP loss at lower RPM. Speeds higher than the design RPM are at the rated horsepower.

 

[cs900]

The loss in horsepower is a feature of solid state phase converters, not VFDs. VFDs will provide the horsepower they are rated for at the machines design RPM. Lower RPM than the design RPM decreases horsepower linearly (half the design RPM, half the rated horsepower). This is why folks above are recommending 2 to 3 hp motors, because of the HP loss at lower RPM. Speeds higher than the design RPM are at the rated horsepower.

I'm no VFD expert, but I do believe TORQUE is constant up to the rated RPM of the motor (usually 60Hz on the VFD). So while while I agree with what you say bout HP, torque is what you actually care about as it is the force you can deliver to the cutting edge.

That in mind a 1.5 HP motor rated for 3600 RPM (2.19ft-lbs) will need to be geared up 39% to get up to 5000RPM. You'll get an equal loss in torque, so at full speed you're only at 1.33ft-lbs. Compared to a 3 Hp you're at 2.66 ft-lbs. I believe this to be the reason people choose larger motors.

Also interesting, and to compliment what cjtoombs pointed out, after 60hz HP stays constant and torque drops off non-linearly. So running your motor right at 60Hz is the peak of HP and torque! That's good to know if you plan on doing lots of a particular material, you can plan your gearing around what HP/torque your application will require.