forum.cncdrive.com

Greetings-

I am looking for advice on upgrading the spindle motor drive on my old mid-80's vintage mill. OEM design uses a brushed DC motor (see nameplate below) driven by a Servo Dynamics servo driver. Reasons for upgrading: both the driver and the tachometer mounted on the motor are damaged, and rather than trying to fix everything I'd prefer to modernize and just dump the old stuff. I have components for repairing the driver but the tachometer is needed for the driver and that could be more problematic for repairing. Also the driver takes up a huge amount of space in the cabinet. The motor is in good condition--I just replaced the bearings and it runs very well, so I'd prefer to keep it. Advantages to keeping the motor are mounting hardware would not need to be retrofitted, the motor includes a built in brake which I like having, and if I remove the tachometer there is a ready made space for an encoder, if needed.

I have been looking at CNCDrive's DG4S-16035 as a replacement but have questions:

1. Is this a good way to go or is there a better alternative to driving the motor, using UCCNC for program control? I know next to nothing about servo drives and just wonder whether this is overkill for driving the motor or is this a preferred method? What would be involved in configuring--is it fairly straightforward to set up for a spindle motor?

2. Recommended PSU?

3. Looking through the manual for the DG4S-16035 it appears that it needs an encoder for operation, what encoder is recommended?

4. Not sure I would need a braking board since the motor has a built in brake and stops the motor fast enough to prevent excess back voltage, but on the occasion that the motor speed needs to drop in the middle of a program it could be a problem. Since a braking board is relatively cheap in the scheme of things I would probably include it anyway.

Trying to keep this brief initially but if more info is needed let me know.

Thanks,
Dan

That's a rather large brushed DC motor. Unless I'm reading it wrong, it looks like 11Nm, 180VDC, and 17.0A (not sure what 'pulsed amps' means)

First step is to see if you can find a DC motor drive capable of 180V and 17A. That's about 4hp.

You can either find a 'dumb' speed control for brushed motors, or turn it in to a servo with feedback from an encoder. And just because it's a servo doesn't mean you have to use it in positioning mode - they work fine in speed mode.

If you go the servo route the second step is to pull the back cover and see if the shaft driving the tach can accept an encoder. There are plenty of those available, and CNCDrive sells some, too.

The DG4S-16035 should work, but as you see it requres an AC-DC power supply. Finding one of those in the kW range you need isn't easy - Antek power supplies are fairly inexpensive but not much over 1.5kW. I'm not an expert on those sorts of power supplies, but maybe CNCDrive or someone else can help there.

An alternative is to find a used (ebay) AC-input servo drive capable of driving DC brushed motors at the amps you need. You might be able to find an AMC (American Motion Controls) drive, or a Copley Xenus XTL-230-40 can do 20A continuous and drive a DC brush motor.

spumco wrote:That's a rather large brushed DC motor. Unless I'm reading it wrong, it looks like 11Nm, 180VDC, and 17.0A (not sure what 'pulsed amps' means)

Yes that's correct. Not just large in power but in size as well. It's on a rather small mill with a 30 taper spindle (see photo). I don't know what pulsed amps means either, in fact there's not a lot that I know about motors in general (except steppers) which is why I need the help, LOL.

spumco wrote:First step is to see if you can find a DC motor drive capable of 180V and 17A. That's about 4hp.

You can either find a 'dumb' speed control for brushed motors, or turn it in to a servo with feedback from an encoder. And just because it's a servo doesn't mean you have to use it in positioning mode - they work fine in speed mode.

At the moment I'm using a variac and a bridge rectifier to operate it which is fine as I'm still in the process of refurbishing and don't need a lot of power, just enough for some light cutting of aluminum to make adapter plates for new axis drive motors.

spumco wrote:If you go the servo route the second step is to pull the back cover and see if the shaft driving the tach can accept an encoder. There are plenty of those available, and CNCDrive sells some, too.

The DG4S-16035 should work, but as you see it requres an AC-DC power supply. Finding one of those in the kW range you need isn't easy - Antek power supplies are fairly inexpensive but not much over 1.5kW. I'm not an expert on those sorts of power supplies, but maybe CNCDrive or someone else can help there.

An alternative is to find a used (ebay) AC-input servo drive capable of driving DC brushed motors at the amps you need. You might be able to find an AMC (American Motion Controls) drive, or a Copley Xenus XTL-230-40 can do 20A continuous and drive a DC brush motor.

A big question for me is whether I really need as much power as the motor is capable of supplying with a matched power supply. I have no expertise in determining just how much power I really want and translating that into meaningful numbers for sizing. I know that when the original servo drive worked there was more than enough power to match a 30 taper spindle tooling.

I'd prefer to stay away from used stuff at this point, otherwise I'd probably concentrate on just doing what I need to get the original drive up and running again. The drive is an 'all in one' unit with onboard PSU and braking module, so I've considered as an interim step using the PSU to power a new servo drive until I can find a proper PSU for it. Only drawback there is another reason I want to ditch the whole thing--it requires 3 phase 115VAC input. I do have a rotary phase converter to use my garage 240VAC, but that also means retaining the original 3 phase transformer to step down the voltage. That would be the only reason I would need the transformer now, which also adds significant weight and space requirement. Looking to slim down as much as I can.

Thanks,
Dan

If you wish to stick with 'new' stuff, have you considered replacing the DC motor with a 3-phase induction motor and VFD?

You can get a 2-3hp motor pretty cheap, and there are tons of single-phase in, 3-phase out VFDs available. And hookup is usually very simple, including connecting to UCCNC via whatever BOB you're using. And if you have a 3-phase RPC, that opens up quite a few options for VFDs.

Depending on the VFD, you can control spindle speed (through UCCNC) via 0-10v analog signal, PWM input, or through MODBUS directly from the computer.

I'd measure the DC motor and see what the frame size is (if standard) and start looking for 3-phase induction motors that will fit the real estate on the head.

Induction motors tend to be heavier than equivalent HP permanent magnet motors, but there are a few fairly inexpensive aluminum-framed motors I've seen on ebay that look like they'd be worth a try.

https://www.ebay.com/itm/332426926065

And I believe those motors are 'encoder-ready', meaning they have a shaft extension beyond the cooling fan for a rotary encoder. Add one of those and UCCNC will let you do synchronized tapping.

The other benefit of a new, quality induction motor is that they tend to have max speeds well in excess of the nameplate synchronous speed. I have an 1800rpm Marathon motor that has a max speed of 5400rpm. Meaning you could either gain spindle speed, or fiddle the belt ratios so you have more torque at the same max RPM as your DC motor.

The setup would be: 120vac 1ph powers the steppers & electronics via AC-DC power supplies, and the VFD is powered by the RPC through a contactor. Contactor coil is controlled by UCCNC via a 24VDC relay as well as through the e-stop chain. The e-stop relay is set up to be a 'holding' relay that requires a momentary reset button to be pushed before the contactor is energized. That way both you and UCCNC can stop the spindle (kill the VFD), but only you can re-energize it.

As for motor sizing, that is a deep rabbit hole on its own. Two suggestions:
1. Look at other mills of about the same size and 30-taper spindle and see what they're using. Tormach 1100 has 2hp. Mine (same-ish size) has 3hp. A Robodrill or Brother Speedio have around 5-10hp at that size. It really depends on the rigidity of the machine - more rigid means you can remove more material per cutting edge... which requires more torque/HP.

2. Download a free/trial copy of HSM Advisor. Input some approximate values for machining cuts you would like to make. i.e. 1/2" 4Fl carbid end mill, 1" DOC and 30% WOC in aluminum and see what the software reports the required HP is. You'll be surprised at how little HP you need to do 'hobby-level' machining.

Although this thread is pretty old now I thought I would post an update, along with a question.

So after a lot of research and head scratching I decided to abandon the DC motor and go with a new 2HP 3 phase motor and VFD--overall I'm very happy with the decision. I chose the 2HP only because it was the largest motor that would fit in the available space, along with some minor modifications. I was really expecting I would have to lower my expectations of how much power I would end up with but to my surprise I ended up with at least as much power as I would ever need to use on this machine, with power to spare. So overall I'm glad I went this route. Thanks Spumco and others for your advice.

Now for the question:
With the success on the home machine, about the same time the machine I sold to my employer decided to blow out the spindle driver as well so time to do the same conversion to it. On the first machine I found a reasonably priced Chinese VFD (not Huanyang) and found it relatively easy to setup and configure (considering it was my first go at it). By the time I got to the second machine the same VFD was out of stock, and besides it was only single phase input (good for my garage) but the work machine is 3 phase input only. So this time I bought a Huanyang from Ebay, but with this one I have a problem that I can't figure out, even after a lot of research. The problem is that no matter how I configure it I get a much higher speed out than commanded. I've looked at a lot of parameters online but so far haven't found anything that fixes the problem. I definitely have the VFD configured for 0-10V input (I haven't done the math but the speed seems to be roughly double what it should be so I even changed that parameter to 0-5V but it didn't help at all). I've checked the actual 0-10V output from the BOB terminals and also at the VFD end and I'm getting exactly the voltage I would expect for the speed commanded, so I can only conclude that the problem is at the VFD end. The frequency display on the VFD shows the wrong value given the voltage input but the value does match the actual motor speed measured with a digital tach, in other words the displayed frequency matches the motor speed correctly but not the commanded speed/voltage input.

So does anybody have any clue which parameter I need to change?

Thanks,
Dan

I seem to have finally stumbled on the right parameter: PD072 does the trick, although the math doesn't correlate exactly (unless I'm figuring it wrong, which is as likely as anything) so I just kept fiddling with the value until I got what I wanted.