Using VFD for speed control on a 1236T Precision Matthews Lathe

[Firebrick43]

Thanks for the graph. I did not consider losing torque at higher RPMs. I need to learn more about the vector-rated motor.

A vector rated motor isn't going to help you at higher rpm. The drive maintains the rated motor hp on the nameplate when its programmed in. HP is a (rpmxtorque)/5252. SO if hp will remain the same(on any motor) as you pass 60hz, the torque has to fall off to maintain a constant HP. If the drive didn't the motor would have to much amperage going into the motor for the wire to handle.

Vector or inverter duty motors really only help in slow speed, of very high speeds. Slow speed because of their higher rated H class insulation. Many small sub 2 hp inverter motors also are constructed differently in the rotor to go above 3600 rpm, many of the inverter duty motors can go 5500 or 6000 rpm, which is not going to help you.

I agree with mksj on newer motors not having issues with insulation, if they are name brand. You will see some motors, like baldor super E say "inverter ready" which is not the same as inverter duty. Just means they have upgraded insulation. A lot of imported non branded motors use crappy materials.. I really don't know on that lathe what is installed. I wouldn't trust it however below 20hz but your gear train should be able to get you there even if you have a 1:1 ratio on your belt.

As for shielding, Yes, you should have your vfd in a box(sized sufficiently for the thermal load) and use either shielded cable rated for VFD or use metal conduit/flextite that is shielded

A brake can help with all threading. Some mount a small switch to the way that can be reposition. The switch is moved and wired so when its actuated by the carriage it stops the VFD and comes to a stop. The tool can be retracted(hopefully by a quick reacting threading tool, VFD reversed back to the stop, tool extended and have a few thou added to the cross slide and vfd started again. If done in a relatively slow manner the tool point will end up in the same place each and every pass, with one big chip building up. I have never seen lathes that need a braking resistor added as threading should still be done relatively slow and the internal resistors are sufficient to do so. External resistors can be benificial if turning heavy objects or at high speed to bring things to a stop in emergency situations. Without the external resistor the drive will still stop things MUCH faster than before unless the lathe is equipped with some kind of mechanical/electric emergency brake.

 

[B2]

@sewaldrep

All that has been said here is valuable. But let me add to it.

I have the PM1440GT with the factory shipped 3HP motor. I drive it via the Hitachi WJ200 VFD, but do it via solid state control electronics rather than relays. I am NOT proposing that you rebuild your VFD conversion. But you may find something useful at the following link where I provided a "complete" description of my conversion in a document called .... Part2... See:

VFD conversion via solid state electronic components. PM1440GT, VFD, 3-phase

VFD conversion using solid state electronic components. I finally finished documenting my PM1440GT lathe VFD conversion and want share it. It took me longer to write this up than it did to actually build the new electronics! (Are these sorts of things every really finished?) I apologize...

www.hobby-machinist.com

The pot, and all of the other circuits, plus parts lists, for controlling the VFD is also part of this description. (The pictures are also posted separately at the link, but are not in the Part2 report as it might have made the file too large to upload.) In going from the VFD control electronics to the front panel, where the pot is located, I used a multi wire shielded cable (see the parts list, "8 core 20 AWG PVC Tinned Cu Marine" cable.. There are lots of types of wire available. Just make sure that the individual wires are stranded, not solid, so that they will not break when bent multiple times and get a decent covering which is oil resistant and not likely to tear open. There is little current flow so the gauge is not critical. As you can see I went with 20 AWG which works fine and is not so large that it is difficult to attach to the pot terminals. You will want a strain relief clamp on the cable.) where the shield is grounded at the control electronics at the VFD end. While I did not find that I needed to do so you can also add a capacitor at the wiper of the pot and at the wiper input to the VFD... and/or a short distance from the VFD input. This will suppress noise spikes. Even a small capacitor like a 1uF will help considerably. Depending upon the noise spikes duration one may need a larger capacitor. The relay noise can last several milliseconds. The input resistance of the VFD pot input times the capacitance pretty much sets the RC time constant of the circuit. Assuming that it is greater than 1KOhm (it needs to be, and appears to be, in order to use a 1K Pot. as an input) Then 1k*1uF~1msec. It will also, of course slow down the response time for your setting the speed, but unless you use an extremely large capacitor you will never notice this as the response time of the motor to the pot setting is slow anyway. Getting rid of the relays, as in my conversion, pretty much eliminates the long time constant spikes. Hence, most of the noise in my conversion is due to the switching inside the VFD and these spikes are quite short (microseconds). While I do not know the VFD internal working code, my guess would be that input signals are taken when the VFD spikes are not occurring as the VFD as complete control of all of this. (I were building and selling VFDs, I would design the VFD workings to do this. )

By the way, I used a 1K, 2Watt, 10 turn pot. This was a mistake as it is far too many turns and so is slow to adjust.

I sometimes use my lathe for unconventional projects. One of these is winding electrical coils in which case it best to have very slow speeds and it takes a long time to winded several hundred turns and get all of the wire turns neatly and tightly placed.. The load is small, mostly the lathe gears and 3 or 4-jaw chuck. So I have run my lathe very slow. The motor seems to run even down to a few (2-3) Hertz. In fact, the WJ200, designed jog speed is only 6Hz, which I use a lot. In winding the coils I have to start and stop the turning a lot, sometimes reversing the process, so sometimes I just use the forward and reverse Jog feature as it is easier to reach and operate than the lathe switch handle.

Also, wrt to the hole that you have where you can put the pot, you can buy 7/8" metal hole plugs for the currrent holes and you can then cut a hole in them to insert the pot shaft. They are not perfect, but work. When I first started my conversion I was going to do that, so I purchased some, but then I added lots of other features to my front panel so I just remade it. The description and a spread sheet to generate the milling Gcode is here: Click Here . Search "The Hillman Group 58147 7/8-Inch Metal Hole Plug, 8-Pack" on Amazon for the hole plugs.

You should not have problems. Go for it!

Dave L.

 

[Ischgl99]

I have a 1236T on a vfd and never installed a speed pot, rather installed a selector switch for 30hz operation that gives me all the nameplate speeds without changing the belt. The belt is on the higher speed pulley groove, so Position 1 on the switch gives me 60hz from the motor and corresponds to the high speed settings for each gear, position 2 is 30hz and is the same as the low speed gear settings. I never experienced a situation where 30hz did not have enough torque or HP for what I was doing, nor did I really need speeds in between those. That may change at some point, but for now it works. This is also on the factory motor, I haven’t found the need to upgrade it for this style of operation.

On switch setting 2 for 30 hz, I also programmed a fast stop for threading, otherwise I use coast to stop. I mostly do metric threading, so need to stop fast before reversing. Since threading is relatively slow, there is very little inertia that needs to be braked, even with the 4 jaw chuck, my vfd can handle a braking time of 0.2 sec without an external braking resistor. I didn’t try any slower than that, so it might stop even faster.

 

[mksj]

With an inverter/vector motor, Hp remains flat, and torque declines in a somewhat non-linear fashion above the base speed, but by going with a smaller motor pulley and increase the RPM above the base speed you gain the mechanical advantage of the applied force (torque/Hp) on the spindle. When I have installed vector motors on lathes they maintained tighter speed regulation with load regardless of RPM. The advantage of higher speed is significant going with a smaller motor pulley, and the motor have prodigious short term (peak) torque almost down to 0 speed. Most vector motors are TEBC or TENV so cooling is not an issue over the motor speed range. If you look at most factory installed vector motors on mills and lathes they need to cover a 10 fold speed range (typically 20-200 Hz) and are used with a single speed or 2 speed mechanical gearbox. They are usually oversized to accommodate for the loss of mechanical power when running them below their base speed. When used in a multi-speed gearbox like a standard lathe, this is less of an issue. The attached document describes use of variable speed operation motors as well as the the mechanical operational limits for Marathon electric motors to address the original thread discussion.

An external braking resistor is still recommended for fast stops, not necessarily for threading but when turning heavy loads at speed. The consequences of rapid stops can result in a buss over voltage error and the VFD going into a free run mode with no braking. This has the potential for a crash into the spinning chuck/work. Some VFD's will adjust the braking rate to minimize this risk, but then it also will alter the stopping time/distance. Most VFD's have an internal braking resistor, but the ability to dissipate high buss voltage is limited and varies by size/model of VFD. I did a 5 Hp motor install on a 1640 lathe, and we forgot to connect the braking resistor, stopping with an 8" chuck was around 3+ seconds. We then hooked up the external braking resistor and the braking time went down to 1 second. We also encountered buss over voltage errors with too rapid speed changes w/o the resistor. I thread at up to 600+ RPM, so having reliable (linear) braking is a must.

A main concern with not using an external braking resistor on a lathe, is that if you are spinning at high speed and hit the E-Stop, you want the lathe to reliably stop as fast as possible when engaged. Some VFD's, have a specific input programming parameter for this function and can be wired into the VFD input controls.

With regard to variable speed on the fly, my experience is that certain operations like boring and large drilling have a sweet spot which is usually over a narrow speed range, too fast and the tool with chatter, too slow cutting performance diminishes as well as chip formation. Being able to adjust the spindle speed allows one to optimize the speed for the material/operation. My larger MT drills and annular cutters the cutting the ideal speed range can be +/-10%, there is an advantage to speed adjustments during operation.

Another major advantage of going 3 phase for many machines is smoother motor power delivery which can show up in surface finish and also allowing frequent start/stop cycles and fast stopping which is a limiting factor with capacitor start single phase motors.

 

[sewaldrep]

Thanks Mark, I have been threading standard threads but not metric. Most of my threading had been done at 90RPM. Experimented with different threading tools and settled on HSS made by Robert Warner. I use the compound at 29.5 deg. I tried using the cross slide and as long as you take like cuts it works, otherwise too much pressure on tool. I need to research what size brakes I need. Can you refer me to that? Looks like they may need to be mounted in a separate box.

 

[sewaldrep]

@sewaldrep

All that has been said here is valuable. But let me add to it.

I have the PM1440GT with the factory shipped 3HP motor. I drive it via the Hitachi WJ200 VFD, but do it via solid state control electronics rather than relays. I am NOT proposing that you rebuild your VFD conversion. But you may find something useful at the following link where I provided a "complete" description of my conversion in a document called .... Part2... See:

VFD conversion via solid state electronic components. PM1440GT, VFD, 3-phase

VFD conversion using solid state electronic components. I finally finished documenting my PM1440GT lathe VFD conversion and want share it. It took me longer to write this up than it did to actually build the new electronics! (Are these sorts of things every really finished?) I apologize...

www.hobby-machinist.com

The pot, and all of the other circuits, plus parts lists, for controlling the VFD is also part of this description. (The pictures are also posted separately at the link, but are not in the Part2 report as it might have made the file too large to upload.) In going from the VFD control electronics to the front panel, where the pot is located, I used a multi wire shielded cable (see the parts list, "8 core 20 AWG PVC Tinned Cu Marine" cable.. There are lots of types of wire available. Just make sure that the individual wires are stranded, not solid, so that they will not break when bent multiple times and get a decent covering which is oil resistant and not likely to tear open. There is little current flow so the gauge is not critical. As you can see I went with 20 AWG which works fine and is not so large that it is difficult to attach to the pot terminals. You will want a strain relief clamp on the cable.) where the shield is grounded at the control electronics at the VFD end. While I did not find that I needed to do so you can also add a capacitor at the wiper of the pot and at the wiper input to the VFD... and/or a short distance from the VFD input. This will suppress noise spikes. Even a small capacitor like a 1uF will help considerably. Depending upon the noise spikes duration one may need a larger capacitor. The relay noise can last several milliseconds. The input resistance of the VFD pot input times the capacitance pretty much sets the RC time constant of the circuit. Assuming that it is greater than 1KOhm (it needs to be, and appears to be, in order to use a 1K Pot. as an input) Then 1k*1uF~1msec. It will also, of course slow down the response time for your setting the speed, but unless you use an extremely large capacitor you will never notice this as the response time of the motor to the pot setting is slow anyway. Getting rid of the relays, as in my conversion, pretty much eliminates the long time constant spikes. Hence, most of the noise in my conversion is due to the switching inside the VFD and these spikes are quite short (microseconds). While I do not know the VFD internal working code, my guess would be that input signals are taken when the VFD spikes are not occurring as the VFD as complete control of all of this. (I were building and selling VFDs, I would design the VFD workings to do this. )

By the way, I used a 1K, 2Watt, 10 turn pot. This was a mistake as it is far too many turns and so is slow to adjust.

I sometimes use my lathe for unconventional projects. One of these is winding electrical coils in which case it best to have very slow speeds and it takes a long time to winded several hundred turns and get all of the wire turns neatly and tightly placed.. The load is small, mostly the lathe gears and 3 or 4-jaw chuck. So I have run my lathe very slow. The motor seems to run even down to a few (2-3) Hertz. In fact, the WJ200, designed jog speed is only 6Hz, which I use a lot. In winding the coils I have to start and stop the turning a lot, sometimes reversing the process, so sometimes I just use the forward and reverse Jog feature as it is easier to reach and operate than the lathe switch handle.

Also, wrt to the hole that you have where you can put the pot, you can buy 7/8" metal hole plugs for the currrent holes and you can then cut a hole in them to insert the pot shaft. They are not perfect, but work. When I first started my conversion I was going to do that, so I purchased some, but then I added lots of other features to my front panel so I just remade it. The description and a spread sheet to generate the milling Gcode is here: Click Here . Search "The Hillman Group 58147 7/8-Inch Metal Hole Plug, 8-Pack" on Amazon for the hole plugs.

You should not have problems. Go for it!

Dave L.

Thanks Dave, I can see you put lots of effort into the solid state control. Very good documentation too. My relay logic is very primitive compared to this solid state design. Certainly cuts down on space requirement and should reduce noise issues. I certainly would need to better understand the electronics to troubleshoot. This looks to be something that could be built and sold with the machine....very impressive. When I get better at understanding the machining process might take this on as a project. I have an electronic solid state background but its been a while. I am dated somewhat, we used wire wrap planes and sockets. MOSFET technology was the solid state design....been a few years.

 

[llamatrails]

One point I have not seen addressed on this and other threads has to do with where to place the gear selectors?

I have my VFD controlled 1236T pulleys both on the larger side, and basically find that each gearbox setting gives me the same spindle rpm range with the pot.

Is there an advantage to one gearbox setting over another, i.e. just use the lowest for a mechanical advantage?

 

[Just for fun]

Good question.

 

[B2]

Thanks Dave, I can see you put lots of effort into the solid state control. Very good documentation too. My relay logic is very primitive compared to this solid state design. Certainly cuts down on space requirement and should reduce noise issues.

Thanks. @sewaldrep . I am pretty old school or is it just plain old! Anyway, I know the wire wrap technology all too well. In my design I just stuck to proto board and solder to keep it simple and to not create a space issue with conducting wire wrap pins sticking out the back side. If I ever find the time I will, or maybe someone else will, make printed circuit boards (PCB) for the control electronics so that it easier for those who have a limited electronics background to use the board. The circuits are really pretty simple and I used bi-polar transistors as they are very robust.
If you decide to try the solid state approach let me know. Maybe I can answer questions or support you somehow.

Good luck with your projects.

Dave L.

 

[sewaldrep]

Thanks. @sewaldrep . I am pretty old school or is it just plain old! Anyway, I know the wire wrap technology all too well. In my design I just stuck to proto board and solder to keep it simple and to not create a space issue with conducting wire wrap pins sticking out the back side. If I ever find the time I will, or maybe someone else will, make printed circuit boards (PCB) for the control electronics so that it easier for those who have a limited electronics background to use the board. The circuits are really pretty simple and I used bi-polar transistors as they are very robust.
If you decide to try the solid state approach let me know. Maybe I can answer questions or support you somehow.

Good luck with your projects.

Dave L.

Thanks Dave, glad to see someone knew what wire wrap technology is.........so I am not the only old one........