unsure how to drive a 320V DC motor for lathe

Thread Starter

toadboy65

Joined Mar 16, 2018
23
I am repowering a lathe, and I already have a motor from a GE stress test machine. I have the whole GE unit, actually. But driving the motor is a little above my current skill level. The electronics from the Stress test machine are extensive, as it includes an EKG and is normally controlled from a remote computer.
The motor data is at the bottom of the post
It would be nice to be able to vary the speed between 500 and 1500 rpm, but I would be satisfied to be able to just drive it around 1000 rpm. There is plenty of gearing selection on the lathe. In a simpler world, I could just hook the motor up to a power source with a fixed voltage, and it would turn at a fixed rate. The system as installed in the GE machine used 5 hall effect leads that seem to allow the driver board to apply voltage to the different phases of the motor to maintain constant torque, and to use powered braking.
Anyway, unless there is a cost-efficient off the shelf system available to drive the motor, I will at minimum need to reuse the motor driver box, and figure out how to give it the signals it expects from the main control board via a serial cable.
So, any help is appreciated.
T
If the motor plate image did not come through, the data is:
HP-3
RPM-4000
Frame-50HZ Ser.F-1
Volts-320 Amps-21.2
Insul F Amb-40C
Duty- Cont Encl-Tenv
Brg/De-6205 Brg/Ode-6205
 

MisterBill2

Joined Jan 23, 2018
18,167
I am repowering a lathe, and I already have a motor from a GE stress test machine. I have the whole GE unit, actually. But driving the motor is a little above my current skill level. The electronics from the Stress test machine are extensive, as it includes an EKG and is normally controlled from a remote computer.
The motor data is at the bottom of the post
It would be nice to be able to vary the speed between 500 and 1500 rpm, but I would be satisfied to be able to just drive it around 1000 rpm. There is plenty of gearing selection on the lathe. In a simpler world, I could just hook the motor up to a power source with a fixed voltage, and it would turn at a fixed rate. The system as installed in the GE machine used 5 hall effect leads that seem to allow the driver board to apply voltage to the different phases of the motor to maintain constant torque, and to use powered braking.
Anyway, unless there is a cost-efficient off the shelf system available to drive the motor, I will at minimum need to reuse the motor driver box, and figure out how to give it the signals it expects from the main control board via a serial cable.
So, any help is appreciated.
T
If the motor plate image did not come through, the data is:
HP-3
RPM-4000
Frame-50HZ Ser.F-1
Volts-320 Amps-21.2
Insul F Amb-40C
Duty- Cont Encl-Tenv
Brg/De-6205 Brg/Ode-6205

You have an interesting project, that is certain! For starters, I would not call the motor a DC motor, although the driver supply is DC. The first step will be to locate the circuit diagram of the driver portion of the system. If you are fortunate the driver portion will be a separate assembly, which will allow you to provide inputs to make it run at the speeds that you choose. I am thinking that the original machine was some sort of treadmill package, and the commands from the computer selected a speed of the treadmill. So at some point the motor drive package gets a speed command, which in the lathe application you will provide. If you are very fortunate there will be a voltage input for setting the speed, if not, then possibly a digital word, and suddenly the project is a lot more complex. But the first thing that you need is to locate the circuit drawing.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
You have an interesting project, that is certain! For starters, I would not call the motor a DC motor, although the driver supply is DC. The first step will be to locate the circuit diagram of the driver portion of the system. If you are fortunate the driver portion will be a separate assembly, which will allow you to provide inputs to make it run at the speeds that you choose. I am thinking that the original machine was some sort of treadmill package, and the commands from the computer selected a speed of the treadmill. So at some point the motor drive package gets a speed command, which in the lathe application you will provide. If you are very fortunate there will be a voltage input for setting the speed, if not, then possibly a digital word, and suddenly the project is a lot more complex. But the first thing that you need is to locate the circuit drawing.
I do have all the diagrams and full service manual. But the circuitry that is there is way beyond what a regular treadmill would have. As I mentioned, this is a little beyond my comfort level, but I can probably work it out, given enough time. I sort of hope that someone sells a driver box that I could just hook up to the motor.
My other alternative is to install the motor on the lathe, and hook up the power and control boxes as they were on the original stress test machine. I do also have the computer with working software that drives the unit, but of course it is a lot to go through just to get the motor to run. And feedback from the elevation and other systems would need to be spoofed. As I mentioned, the unit is designed to combine treadmill functions with an EKG and other physical data, and log all that in the software. There is also a printer attached.
I am also a little apprehensive about pre-start status checks. It is not clear to me yet whether the unit will run without positive inputs from all the extra functions I will not need, or whether it will just register faults and shut off.
Last year I completed sort of a similar project, putting the motor and instruments from a newish car into a 1963 corvair. I spent months working out how to fool the CPU into thinking that missing components were installed and operating normally. But in that case, it was mostly something simple like a resistance value or a ground that was required. There were just lots of them.
This project is a little different, as I really do not have a comprehensive understanding of these motors.
 

MisterBill2

Joined Jan 23, 2018
18,167
I do have all the diagrams and full service manual. But the circuitry that is there is way beyond what a regular treadmill would have. As I mentioned, this is a little beyond my comfort level, but I can probably work it out, given enough time. I sort of hope that someone sells a driver box that I could just hook up to the motor.
My other alternative is to install the motor on the lathe, and hook up the power and control boxes as they were on the original stress test machine. I do also have the computer with working software that drives the unit, but of course it is a lot to go through just to get the motor to run. And feedback from the elevation and other systems would need to be spoofed. As I mentioned, the unit is designed to combine treadmill functions with an EKG and other physical data, and log all that in the software. There is also a printer attached.
I am also a little apprehensive about pre-start status checks. It is not clear to me yet whether the unit will run without positive inputs from all the extra functions I will not need, or whether it will just register faults and shut off.
Last year I completed sort of a similar project, putting the motor and instruments from a newish car into a 1963 corvair. I spent months working out how to fool the CPU into thinking that missing components were installed and operating normally. But in that case, it was mostly something simple like a resistance value or a ground that was required. There were just lots of them.
This project is a little different, as I really do not have a comprehensive understanding of these motors.
OK, since you have all of the schematics this can be a learning experience if it is possible to post copies of at least some portion of them. If you can locate the motor and motor driver circuit portions that will be a good start.
The question now is about the hardware: Is all of the circuitry on a single circuit board? Or are there separate boards for some of the functions? If the motor driver board is separate then the project can be a lot less effort, but if all of the circuit is on one single board then it gets a bit more complex. So now more questions, about the motor. Does the motor have brushes and a comutator, and second, how many wires that are not part of the hall effect sensors section. That can provide a bit of information as to just what kind of motor it actually is, since the nameplate data does not help in that aspect. One more mechanical question is about the linkage between the motor and the treadmill portion: is there a lot of speed reduction, so that in actual use the motor is turning a lot faster than the big belt driving rollers? That would be the best situation, because for a lathe there is already a great deal of reduction, so that the motor can run at a higher speed.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
OK, since you have all of the schematics this can be a learning experience if it is possible to post copies of at least some portion of them. If you can locate the motor and motor driver circuit portions that will be a good start.
The question now is about the hardware: Is all of the circuitry on a single circuit board? Or are there separate boards for some of the functions? If the motor driver board is separate then the project can be a lot less effort, but if all of the circuit is on one single board then it gets a bit more complex. So now more questions, about the motor. Does the motor have brushes and a comutator, and second, how many wires that are not part of the hall effect sensors section. That can provide a bit of information as to just what kind of motor it actually is, since the nameplate data does not help in that aspect. One more mechanical question is about the linkage between the motor and the treadmill portion: is there a lot of speed reduction, so that in actual use the motor is turning a lot faster than the big belt driving rollers? That would be the best situation, because for a lathe there is already a great deal of reduction, so that the motor can run at a higher speed.
I will try to post some of the diagrams this evening.
Here is a link to a pdf of the service manual for the treadmill part- http://www.medteh.info/_fr/93/T2000-field-ser.pdf
For motor wiring, there are three wires: orange, white, and blue. Also a blue temp sensor wire wrapped around those in the junction box.
The ribbon cable with the hall effect wires has six wires.
As far as the the mechanical part goes, I can probably work with whatever I can get. The original motor ran at two speeds, low was about 865, and high was 1724. But I could either mount the original pulley system on the new motor, or fabricate intermediate reduction gearing. The new motor is supposed to provide constant torque at all speeds, but that might entail the hall sensor feedback system working properly.
There are two main boxes in the unit. One is a control for the drive motor with two PC boards, the other box is the general control system, which has 220v AC in, and to which the three wires from the motor connect. The drive motor control box is only connected to the motor by the hall effect ribbon cable. A serial cable connects the two boxes. An additional serial cable connects the general control box to the computer console.
also, thanks for helping!
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
That is a 3 phase brushless motor so it requires the drive from the Stress Machine, typically the control input is either 0-10vdc analogue, or a 5v PWM signal.
If you can locate the separate motor board it may be possible to figure it out the control entry by tracing the likely connector pins to a identifiable IC.
Also the pinouts are also labeled.
It sounds like the P.S. may also be separate.
Max.
 

shortbus

Joined Sep 30, 2009
10,045
For motor wiring, there are three wires: orange, white, and blue. Also a blue temp sensor wire wrapped around those in the junction box.
The ribbon cable with the hall effect wires has six wires.
As far as the the mechanical part goes, I can probably work with whatever I can get. The original motor ran at two speeds, low was about 865, and high was 1724. But I could either mount the original pulley system on the new motor, or fabricate intermediate reduction gearing. The new motor is supposed to provide constant torque at all speeds, but that might entail the hall sensor feedback system working properly.
There are two main boxes in the unit. One is a control for the drive motor with two PC boards, the other box is the general control system, which has 220v AC in, and to which the three wires from the motor connect. The drive motor control box is only connected to the motor by the hall effect ribbon cable. A serial cable connects the two boxes. An additional serial cable connects the general control box to the computer console.

From your link the motor, while DC is actually a BLDC (brush less DC)motor, All you should need is contained in Section 6 of the manual. Have you disconnected any of the cables yet? If not be sure to unplug them not cut them, unplugging them will make life easier in the long run.

A quick look at the manual it seems like the main power board will contain the AC to DC part of the circuit. Then from there a cable goes to the motor controller, and another cable goes to the actual motor.

One caution on doing something like this, taking a tread mill type motor to power something else, is to not deviate too much from the original RPMs that the motor was intended to be used at. Most motors like this are fan cooled, either internally or by one mounted on the end of the shaft. Slowing them down below the 'low' speed from original use will cause them to over heat. And being that the motor is BLDC, the motor may not be able to reverse direction. Since your controller was from a tread mill type use reverse may not be possible with out a new controller, because of the way BLDC motors are made. Don't know what kind of lathe you have, but in my lathes the motor has to be reversed to reverse spindle directions.

Pictures of the main control board may get you more help to figure out what parts are necessary to take for the power supply.

Max types faster. :)
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
It also appears that the output of the μp is PWM and then it is converted to analogue via the LM358 and the 4053 I.C.'s.
You may be able to intercept the analogue and feed it with a separate control that will give you a wider RPM range, you may lose the closed loop differential encoder feed back from the two extra halls that are fitted for this purpose.
It is going to take a little reverse engineering and trial and error.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,167
It also appears that the output of the μp is PWM and then it is converted to analogue via the LM358 and the 4053 I.C.'s.
You may be able to intercept the analogue and feed it with a separate control that will give you a wider RPM range, you may lose the closed loop differential encoder feed back from the two extra halls that are fitted for this purpose.
It is going to take a little reverse engineering and trial and error.
Max.

The good news is that at least for a metal working lathe the drive motor usually does not need to provide a wide range of speeds because there is a lot of gear combinations available. So the variable speed function is a luxury, unless some specific application requires a particular speed.
One interesting thing is that a permanent magnet brushless DC motor is really a synchronous motor. So if three phase power is available then it could be run as a standard synchronous motor at some input voltage. IF the speed that resulted was satisfactory then it would be a simple solution. So at least the theory predicts one simple option, while the reality may get in the way. And it may be that the whole purpose of the motor change was to work around the lack of three phase power.
So now for one more question: Does the stress-testing treadmill work? If all of the circuit is working then it is a matter of finding what part is needed, but if the existing system does not work then the project becomes a lot more complex.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
From your link the motor, while DC is actually a BLDC (brush less DC)motor, All you should need is contained in Section 6 of the manual. Have you disconnected any of the cables yet? If not be sure to unplug them not cut them, unplugging them will make life easier in the long run.

A quick look at the manual it seems like the main power board will contain the AC to DC part of the circuit. Then from there a cable goes to the motor controller, and another cable goes to the actual motor.

One caution on doing something like this, taking a tread mill type motor to power something else, is to not deviate too much from the original RPMs that the motor was intended to be used at. Most motors like this are fan cooled, either internally or by one mounted on the end of the shaft. Slowing them down below the 'low' speed from original use will cause them to over heat. And being that the motor is BLDC, the motor may not be able to reverse direction. Since your controller was from a tread mill type use reverse may not be possible with out a new controller, because of the way BLDC motors are made. Don't know what kind of lathe you have, but in my lathes the motor has to be reversed to reverse spindle directions.

Pictures of the main control board may get you more help to figure out what parts are necessary to take for the power supply.

Max types faster. :)
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
Everything is disconnected, but I assumed that I might have to put it back together, so it is all labeled and undamaged. The lathe (a 16x40) apparently switches directions through gearing, so that is one less thing to worry about.
I do not know what rpm range the motor normally ran at on the treadmill. I guess I can do some math and figure that one out.
I have so far been unable to post images here, but I will be in the shop tomorrow, and will try again.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
The good news is that at least for a metal working lathe the drive motor usually does not need to provide a wide range of speeds because there is a lot of gear combinations available. So the variable speed function is a luxury, unless some specific application requires a particular speed.
One interesting thing is that a permanent magnet brushless DC motor is really a synchronous motor. So if three phase power is available then it could be run as a standard synchronous motor at some input voltage. IF the speed that resulted was satisfactory then it would be a simple solution. So at least the theory predicts one simple option, while the reality may get in the way. And it may be that the whole purpose of the motor change was to work around the lack of three phase power.
So now for one more question: Does the stress-testing treadmill work? If all of the circuit is working then it is a matter of finding what part is needed, but if the existing system does not work then the project becomes a lot more complex.
The stress testing system works perfectly, or did before I took the motor out. I even have all the passwords to the original software. And you are right about three phase power being the issue. As far as I know, the original lathe motor is functional, as it at least passed the tests I was able to perform on it. I am in a fairly remote location, so the expense of adding three phase power is prohibitive. I was in the process of investigating phase converters when I thought about the stress test machine just sitting there on the other side of the shop.
I am more mechanical than electronic in knowledge orientation and skillset, so if I get the motor to run at some fixed speed or range of speeds, I can almost certainly manufacture a gearing solution to deliver the desired RPMs to the lathe.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
the expense of adding three phase power is prohibitive. I was in the process of investigating phase converters when I thought about the stress test machine just sitting there on the other side of the shop.
.
My test on using a BLDC on 3ph has not produced any decent results, I have tried it on a VFD with the same result, some manufacturers such as Hitachi mention theirs with run a BLDC but when I contacted them on the issue I was told it would have to be custom for each motor.
If you can use the existing BLDC drive, this will be the best scenario.
Max.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
I guess I am going to start by reconnecting all the components temporarily, with the motor on a stand and a tach installed, and just start subtracting unneeded elements one by one, like the elevation motor. Then, the real work of figuring out from the prints and from experimentation of how to give it start/stop/speed request inputs begins.
I can also see if motor heat is an issue. The lathe already has a coolant system running though it, I could build a cooling jacket for the motor, if needed. Or a fluid cooled heat sink.
 

MisterBill2

Joined Jan 23, 2018
18,167
My test on using a BLDC on 3ph has not produced any decent results, I have tried it on a VFD with the same result, some manufacturers such as Hitachi mention theirs with run a BLDC but when I contacted them on the issue I was told it would have to be custom for each motor.
If you can use the existing BLDC drive, this will be the best scenario.
Max.
Thanks for the response, Max. I had no experience in that area, but it seemed like it was worth at least a few minutes of thought. The re-purposing of a motor like this is an interesting challenge indeed. The only treadmill motor that I have had experience with was a brush-type device that primarily served as a loading device. While it could have become a useful generator it had enough physical problems to make that use not worthwhile.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
I guess I am going to start by reconnecting all the components temporarily, with the motor on a stand and a tach installed, and just start subtracting unneeded elements one by one, like the elevation motor. Then, the real work of figuring out from the prints and from experimentation of how to give it start/stop/speed request inputs begins.
I can also see if motor heat is an issue. The lathe already has a coolant system running though it, I could build a cooling jacket for the motor, if needed. Or a fluid cooled heat sink.
Sounds like a good approach, one suggestion is if the motor has a flywheel like many do, I would remove it for a lathe application.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,167
I guess I am going to start by reconnecting all the components temporarily, with the motor on a stand and a tach installed, and just start subtracting unneeded elements one by one, like the elevation motor. Then, the real work of figuring out from the prints and from experimentation of how to give it start/stop/speed request inputs begins.
I can also see if motor heat is an issue. The lathe already has a coolant system running though it, I could build a cooling jacket for the motor, if needed. Or a fluid cooled heat sink.
Worst case the motor might need a fan blowing on it. The nameplate description includes 40 degrees C and TENV, (Totally Enclosed Non-Vented), which means that it can survive a bit of heat. Depending on the amount of heat sinking the power electronics may need a fan or even a blower.
One suggestion is to research other circuits for brushless DC motors so that you can be familiar with the arrangement of the drive circuitry. That could be useful in understanding the portions that you would need to preserve.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
The removal of any flywheel will also also allow the motor and electronics to run cooler on the likes of a lathe where the speed is often constantly changing/stop/starting etc.;)
Max.
 

Thread Starter

toadboy65

Joined Mar 16, 2018
23
So today I hooked everything up as planned, and ran the system. I left the flywheel on the motor, otherwise it had no load. The motor would not start until I had the software running, and the software would not let me start until I plugged in the EKG system. I left the treadmill elevation motor disconnected with no problems.
The console has buttons for "motor start","motor stop", "increase speed", and "decrease speed". Each momentary press of the increase or decrease buttons changes the treadmill speed by .1mph.
The motor is nearly silent, except when changing speed. I compiled a graph of the indicated treadmill speed vs. the rpms obtained from a temporarily installed digital tach. I ran it for an hour at the highest rpm that I am likely to need for this application, and there was essentially no heating of the motor. Of course there was no load.
"motor start" and "motor stop" seem to use programmed gradual acceleration and deceleration. The "emergency stop" button used magnetic braking and stopped it almost instantly. Since there was nobody hooked up to the EKG leads, it would give regular alarms indicating that the patient was dead.
I suppose the next step is to study the schematics (as best I can understand them), and tap into the serial connection between the computer and the drive system box, and see if I can figure out what is going on between them. There is also a service port on the drive system box. I suspect that there are manual speed settings there somewhere.
test rig.jpg
 
Top