Author: admin

I scrapped a treadmill and salvaged the motor for a belt grinder project. My goal is to chuck the original control panel and just use the motor and it’s controller… but unlike the previous treadmill I did this to, interfacing with this one was not as simple as feeding it a PWM signal. The motor is a Johnson 90v DC motor, model JM01-013, and the control board is labeled MLH0910PC. I’m hoping the following information helps at least one other person – I spent hours searching and wasn’t able to come up with anything useful.

The control panel (I’ll refer to this as the panel going forward) has all of the displays and connects to the rest of the interface buttons and the safety key switch. The motor controller contains the AC connection, runs the motor and incline, as well as supplies power for the panel.

There is a single cable connecting the controller and panel with 8 wires: two are the safety key switch and pass straight through to the controller, two for power, two for ground, and the last two are for serial data. On both ends of the connector there is a MAX3085 RS-485 transceiver chip. Initially I used a MAX485 on a breakout board to listen in, but I was struggling to extract useful data with a logic analyzer and PulseView. It was clear that there was a message and response from one unit to the other, but for whatever reason I wasn’t able to reliably pick up the start and end bits for the response. I had much better luck connecting the logic analyzer directly to pins 1 and 4 of the MAX3085 (on either board works)… this let me see the RX and TX data on separate channels, then it could be decoded easily.

From what I can tell the panel is the master, it always initiates communication (either a command or a query), and the controller only responds. Messages are sent roughly every 70-100 ms at 9600 baud and consist of 5-7 bytes. The first two bytes look like an identifier: the panel prefix is 0-255, and the controller is 0-127. I looked at all of this in decimal because why not, also it made it easy to work with in Excel. I was able to isolate the packet from the panel that seemed to control the speed, then I recorded the values for every speed available from .5 to 12 MPH in tenth of a MPH increments.

I used an arduino to replay all of the messages through the MAX485 from power-up to running and was pleased to see the motor start to run (I kept the safety switch shorted). Working backwards, I started eliminating the different messages until all I had left was the speed command and the motor continued to run – showing me that I can ignore pretty much everything else – great! Did some more experimentation and here is what I found:

• Only speed commands are needed
• All acknowledgements from the controller can be ignored
• Any new speed lower than the current one will be accepted by the controller and it will coast down to that setting
• A new speed higher than the current one will be accepted if it is not too much higher. For example, the motor will accelerate from 5 to 5.1 mph
• A new speed too much higher than the current one will be rejected and the motor will coast to a stop. For example, commanding it to accelerate from 5 to 12 mph.
• Speed changes are very gradual (probably to keep you from falling on your face and/or butt)
• It won’t exceed 4000 RPM (12 mph), or perhaps I picked bogus values when I tried

This means that interfacing with this is going to be pretty easy. My approach was just to create a lookup table for the known speeds that I mapped out. When increasing the speed I ramp through each of the steps on the way to the final target speed, this seems to keep the controller happy and it doesn’t panic and coast to a stop. There is no need to ramp down to a slower speed, only to a higher one.

The 7-byte packet that controls speed is pictured above. I mapped all of the speeds from zero to twelve MPH, see the table below. RPM was measured with no load, but I expect it to be similar under load – the motor has a toothed wheel (maybe optical or probably hall-effect) and I’m guessing it uses closed-loop speed control.

Mission accomplished! I’ve actually had this motor sitting around for over a year but never made any headway on figuring out how to control it, now I can move on and make something useful with it.

MPH	Byte 1	Byte 2	Byte 3	Byte 4	Byte 5	Byte 6	Byte 7	Measured RPM
0.0	0	255	241	2	0	0	221	0
0.5	0	255	241	2	0	161	16
0.6	0	255	241	2	0	195	201
0.7	0	255	241	2	0	230	186
0.8	0	255	241	2	1	8	144
0.9	0	255	241	2	1	42	116
1.0	0	255	241	2	1	76	105	332
1.1	0	255	241	2	1	111	188
1.2	0	255	241	2	1	145	33
1.3	0	255	241	2	1	179	197
1.4	0	255	241	2	1	213	216
1.5	0	255	241	2	1	248	18
1.6	0	255	241	2	2	26	156
1.7	0	255	241	2	2	60	188
1.8	0	255	241	2	2	94	101
1.9	0	255	241	2	2	129	79
2.0	0	255	241	2	2	163	171	670
2.1	0	255	241	2	2	197	182
2.2	0	255	241	2	2	231	82
2.3	0	255	241	2	3	10	43
2.4	0	255	241	2	3	44	11
2.5	0	255	241	2	3	78	210
2.6	0	255	241	2	3	112	8
2.7	0	255	241	2	3	147	154
2.8	0	255	241	2	3	181	186
2.9	0	255	241	2	3	215	99
3.0	0	255	241	2	3	249	250	1016
3.1	0	255	241	2	4	28	96
3.2	0	255	241	2	4	62	132
3.3	0	255	241	2	4	96	229
3.4	0	255	241	2	4	130	70
3.5	0	255	241	2	4	165	87
3.6	0	255	241	2	4	199	142
3.7	0	255	241	2	4	233	23
3.8	0	255	241	2	5	11	64
3.9	0	255	241	2	5	46	51
4.0	0	255	241	2	5	80	212	1358
4.1	0	255	241	2	5	114	48
4.2	0	255	241	2	5	148	87
4.3	0	255	241	2	5	183	130
4.4	0	255	241	2	5	217	38
4.5	0	255	241	2	5	251	194
4.6	0	255	241	2	6	29	136
4.7	0	255	241	2	6	64	186
4.8	0	255	241	2	6	98	94
4.9	0	255	241	2	6	132	57
5.0	0	255	241	2	6	166	221	1700
5.1	0	255	241	2	6	200	121
5.2	0	255	241	2	6	235	172
5.3	0	255	241	2	7	13	63
5.4	0	255	241	2	7	47	219
5.5	0	255	241	2	7	81	60
5.6	0	255	241	2	7	116	79
5.7	0	255	241	2	7	150	236
5.8	0	255	241	2	7	184	117
5.9	0	255	241	2	7	218	172
6.0	0	255	241	2	7	253	189	2042
6.1	0	255	241	2	8	31	135
6.2	0	255	241	2	8	65	230
6.3	0	255	241	2	8	99	2
6.4	0	255	241	2	8	134	54
6.5	0	255	241	2	8	164	175
6.6	0	255	241	2	8	202	118
6.7	0	255	241	2	8	236	86
6.8	0	255	241	2	9	15	112
6.9	0	255	241	2	9	49	234
7.0	0	255	241	2	9	83	51	2385
7.1	0	255	241	2	9	117	19
7.2	0	255	241	2	9	152	158
7.3	0	255	241	2	9	186	122
7.4	0	255	241	2	9	220	103
7.5	0	255	241	2	9	254	131
7.6	0	255	241	2	10	33	132
7.7	0	255	241	2	10	67	93
7.8	0	255	241	2	10	101	125
7.9	0	255	241	2	10	135	222
8.0	0	255	241	2	10	170	20	2725
8.1	0	255	241	2	10	204	9
8.2	0	255	241	2	10	238	237
8.3	0	255	241	2	11	16	132
8.4	0	255	241	2	11	51	81
8.5	0	255	241	2	11	85	76
8.6	0	255	241	2	11	119	168
8.7	0	255	241	2	11	153	118
8.8	0	255	241	2	11	188	5
8.9	0	255	241	2	11	222	220
9.0	0	255	241	2	12	0	105	3070
9.1	0	255	241	2	12	34	141
9.2	0	255	241	2	12	101	161
9.3	0	255	241	2	12	103	69
9.4	0	255	241	2	12	137	139
9.5	0	255	241	2	12	171	127
9.6	0	255	241	2	12	206	49
9.7	0	255	241	2	12	240	235
9.8	0	255	241	2	13	18	188
9.9	0	255	241	2	13	52	156
10.0	0	255	241	2	13	87	116	3414
10.1	0	255	241	2	13	121	237
10.2	0	255	241	2	13	155	78
10.3	0	255	241	2	13	173	110
10.4	0	255	241	2	13	224	92
10.5	0	255	241	2	14	2	210
10.6	0	255	241	2	14	36	242
10.7	0	255	241	2	14	70	43
10.8	0	255	241	2	14	105	131
10.9	0	255	241	2	14	139	32
11.0	0	255	241	2	14	173	0	3752
11.1	0	255	241	2	14	207	217
11.2	0	255	241	2	14	242	80
11.3	0	255	241	2	15	20	195
11.4	0	255	241	2	15	54	39
11.5	0	255	241	2	15	88	131
11.6	0	255	241	2	15	123	86
11.7	0	255	241	2	15	157	49
11.8	0	255	241	2	15	191	213
11.9	0	255	241	2	15	225	180
12.0	0	255	241	2	16	4	119	3996

Edit 10/25/21: Here are the codes blinked out by the control board:

And here is a simple arduino sketch. You can probably ignore the 7-segment and rotary encoder parts, but it is successfully controlling the motor drive through a MAX485 chip on a breakout board.

/*
 * Encoder on A2, A3 
 * Button on A1
 * MAX 485: 5v to DE, D8 to DI.  A and B to A and B of the treadmill controller
 * TM1637 display clock on 2, data on 3
 * Short out the middle "safety key" pins 
 * Take power from one of the treadmill controller pins (which one?)
 * Treadmill header has 8 pins (left to right, latch at the bottom):
 * 1 - 12v
 * 2 - 12v
 * 3 - A or B?
 * 4 - Safety switch
 * 5 - Safety switch
 * 6 - A or B?
 * 7 - Ground
 * 8 - Ground
*/




#include <AltSoftSerial.h>
//https://www.pjrc.com/teensy/td_libs_AltSoftSerial.html

#include <SimpleTimer.h>
//https://github.com/jfturcot/SimpleTimer

#include <RotaryEncoder.h>
//http://www.mathertel.de/Arduino/RotaryEncoderLibrary.aspx
//https://github.com/mathertel/RotaryEncoder

#include "OneButton.h"
//https://github.com/mathertel/OneButton

#include <TM1637Display.h>
//https://github.com/avishorp/TM1637

#include <EEPROMex.h>
//https://github.com/thijse/Arduino-EEPROMEx

const int encoderButton = A1;
const int encoderA = A2;
const int encoderB = A3;
const int dispClock = 2; //D2
const int dispData = 3; //D3


AltSoftSerial ss;
SimpleTimer timer;
OneButton encoderButt(encoderButton, true);
RotaryEncoder encoder(A2, A3);
TM1637Display display1(dispClock, dispData);

bool halt = true;
int setSpeed = 0;
int curSpeed = 0; //ramp this up until it reaches the set speed
int savedSpeed = 0;
const int speedCount = 123;

const byte b1[] = {0,0,0,0,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,6,6,6,6,6,6,6,7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,9,9,9,9,9,9,9,9,10,10,10,10,10,10,10,11,11,11,11,11,11,11,12,12,12,12,12,12,12,12,13,13,13,13,13,13,13,14,14,14,14,14,14,14,14,15,15,15,15,15,15,15,16,16,16,16,16,16,16,16};
const byte b2[] = {0,161,195,230,8,42,76,111,145,179,213,248,26,60,94,129,163,197,231,10,44,78,112,147,181,215,249,28,62,96,130,165,199,233,11,46,80,114,148,183,217,251,29,64,98,132,166,200,235,13,47,81,116,150,184,218,253,31,65,99,134,164,202,236,15,49,83,117,152,186,220,254,33,67,101,135,170,204,238,16,51,85,119,153,188,222,0,34,101,103,137,171,206,240,18,52,87,121,155,173,224,2,36,70,105,139,173,207,242,20,54,88,123,157,191,225,4,20,54,88,123,157,191,225};
const byte b3[] = {221,16,201,186,144,116,105,188,33,197,216,18,156,188,101,79,171,182,82,43,11,210,8,154,186,99,250,96,132,229,70,87,142,23,64,51,212,48,87,130,38,194,136,186,94,57,221,121,172,63,219,60,79,236,117,172,189,135,230,2,54,175,118,86,112,234,51,19,158,122,103,131,132,93,125,222,20,9,237,132,81,76,168,118,5,220,105,141,161,69,139,127,49,235,188,156,116,237,78,110,92,210,242,43,131,32,0,217,80,195,39,131,86,49,213,180,119,195,39,131,86,49,213,180};
const int sfm[] = {0,262,314,366,418,471,523,575,628,680,732,785,837,889,941,994,1046,1098,1151,1203,1255,1308,1360,1412,1464,1517,1569,1621,1674,1726,1778,1831,1883,1935,1987,2040,2092,2144,2197,2249,2301,2355,2406,2458,2510,2563,2615,2667,2721,2772,2824,2878,2929,2981,3033,3087,3138,3190,3244,3295,3347,3401,3453,3504,3556,3610,3661,3713,3767,3818,3870,3924,3976,4027,4079,4133,4184,4236,4290,4342,4393,4447,4499,4550,4602,4656,4708,4759,4813,4865,4916,4970,5022,5074,5125,5179,5231,5282,5336,5388,5440,5493,5545,5597,5648,5702,5754,5805,5859,5911,5963,6016,6068,6120,6171,6225,6277,6329,6382,6434,6486,6539,6591,6643};

//eeprom memory locations
const int memAddress = 20;
const int memBase = 350;


void setup() {
  //interrupts for encoder
  PCICR |= (1 << PCIE1);    // This enables Pin Change Interrupt 1 that covers the Analog input pins or Port C.
  PCMSK1 |= (1 << PCINT10) | (1 << PCINT11);  // This enables the interrupt for pin 2 and 3 of Port C.
  //Serial.begin(9600);
  ss.begin(9600);
  timer.setInterval(100, sendSpeedPacket);
  encoderButt.attachClick(startStop);

  // Set Up EEPROM
  EEPROM.setMemPool(memBase, EEPROMSizeNano);

  //Load the stored speed value
  setSpeed = EEPROM.readInt(memAddress);

  display1.setBrightness(2);
  updateDisplay();
}

void loop() {
  // put your main code here, to run repeatedly:
  timer.run();
  encoderButt.tick();
  readEncoder();
  updateDisplay();
}

void sendSpeedPacket() {

  if(halt == true) { //send the stop packet
    ss.write((byte)0);
    ss.write((byte)255);
    ss.write((byte)241);
    ss.write((byte)2);
    ss.write((byte)0);
    ss.write((byte)0);
    ss.write((byte)221);
  }
  else { //send the current speed packet
    ss.write((byte)0);
    ss.write((byte)255);
    ss.write((byte)241);
    ss.write((byte)2);
    ss.write((byte)b1[curSpeed]);
    ss.write((byte)b2[curSpeed]);
    ss.write((byte)b3[curSpeed]);   
  }

  if (halt == true){
    curSpeed = 0;
  }

  if (curSpeed < setSpeed && halt == false) { //this ramps the sent speed up until it hits the set speed
    curSpeed++;
  }
  if (curSpeed > setSpeed && halt == false) { //this immediately reduces the sent speed to the set speed
    curSpeed = setSpeed;
  }

  //Serial.print(halt);
  //Serial.print(" ");
  //Serial.println(curSpeed);
  
}

void startStop () {
  halt = !halt;
  EEPROM.writeInt(memAddress, setSpeed); //save the speed every time the button is pushed... so that on next power up it is not zero
}

void readEncoder() {
  static int pos = 0;

  int newPos = encoder.getPosition();
      if (pos > newPos) {
        setSpeed = setSpeed - 1;
      }
      else if (pos < newPos) {
        setSpeed = setSpeed + 1;
      }
      pos = newPos; //--keep

      setSpeed = constrain(setSpeed, 0, speedCount);


}

ISR(PCINT1_vect) {
  encoder.tick(); // just call tick() to check the state.
}


void updateDisplay() {
  display1.showNumberDecEx(sfm[setSpeed], 0, true);  //Set time should always show on display 1
//  Serial.print("Minutes: ");
//  Serial.print(seconds / 60);
//  Serial.print("Seconds: ");
//  Serial.println(seconds & 60);
}

Update 11/2022 – here’s a list of some of the unique commands I saw. I didn’t test or try to capture the incline setting, but if we’re lucky maybe it’s in here:

These are the unique commands and responses I recorded just after powering up:
Source	Unique Message		Count	Comment
Panel	0 255 14 0 14			1	
Motor	128 0 14 0 5			1	
Panel	0 255 9 64 0 95 241		2	
Motor	128 0 9 0 107			2	
Panel	0 255 159 0 9			30	Common while idle and running
Motor	128 0 159 64 0 0 160	30	Probably reply to 0 255 159 09
Panel	0 255 143 64 0 0 187	81	Speed command? 143
Motor	128 0 143 0 238			81	


And here are the unique commands while it was running at a steady .5 mph:
Source	Message						Count	Comment
Panel	0 255 143 64 0 133 8		77	Speed command? 143
Motor	128 48 143 0 48				77	Same acknowledgement to speed command as 1 mph
Panel	0 255 111 64 0 20 14		76	Same while running at 1 mph
Motor	128 48 111 0 69				76	same response as running at 1 mph
Panel	0 255 159 0 9				23	Common while idle or running
Motor	128 48 159 64 0 37 56		11	
Motor	128 48 159 64 0 165 180		10	

Part 2 for the lathe power feed. I’ll make at least one more to go over the wiring and controls.

This mount puts the motor on the left side of the lathe and replaces the original banjo for the changegears. This is the easiest/cheapest way I could come up with to attach the motor to the leadscrew. Originally I intended to extend the leadscrew and drive it from the tail end, but that is a bit more challenging and I thought this would be easier for the home-shop. The motor will interfere with the cover, so you may need to cut a hole in it or leave it off. I put some thought into this so that it could be made with basic tools (no slots, welding, etc).

Bill of Materials

Quantity	Description
1	Steel or aluminum plate 4″ x 5″ (or larger). 3/16″ or 1/4″ thick
4	M5 x .8 Screw – 35mm long
1	M5 x .8 Screw – 15mm long
4	Spacer 5mm ID – .700″ long (see drawing) (make from round stock)
1	NEMA 23 Stepper (notes below)
1	GT2 pulley – 60 tooth 12mm bore
1	GT2 pulley – 20 tooth (bore to match stepper shaft)
1	GT2 belt – 200 tooth, 6mm wide

Tools required

• M5 x .8 tap
• 4.2mm or #19 drill for M5
• 8mm or O drill (.3160″)
• 5mm or #7 or #8 drill
• Tools for cutting the plate
  • Hacksaw and a drill is the basic requirement
  • Bandsaw, scroll saw, mill, CNC, LASER… these are optional. Use whatever you want
• File or sandpaper
• Lathe (for making spacers)

The only critical dimensions on this part are the four tapped M5 holes for mounting the motor. It is also important that the four spacers be the same length so the motor is square to the plate. Other than that, the outside contour of the part only needs to clear any obstructions. The circular part with the offset hole is used to adjust the belt tension, it rotates and presses against the leadscrew bushing/bearing – if that doesn’t make sense it should be more clear in the photo. The OD of the spacers is only important for the one that is closest to the leadscrew pulley, it may be necessary to relieve this for clearance.

I recommend printing the drawing out at 1:1 scale and gluing it to your plate, center punch the hole locations for drilling and use a saw and files to get the right outside shape.

Hopefully it is clear how this mounts up to the lathe in the photos. Take off all of the change gears including the key on the leadscrew and set them aside (or probably lose them forever, which is what I will probably do). The banjo also needs to be removed, but the nut and washer will be reused. Don’t tighten the set screws on the pulleys until you have the motor mounted, this way you can shift them around for alignment. Remember to put the belt on!

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

Regarding the motor – I chose a 1.9 nm stepper. I have not tested this thoroughly yet, but so far I have this running at 2 amps and it doesn’t seem to have any trouble moving the carriage while cutting in aluminum. I think a smaller motor will probably be sufficient – but I don’t know the limits yet.

That’s it! I’ll follow up with the details for the control in another post.

Rainy day today, got the prototype firmware working. For whatever reason using a display and sending pulses to a stepper motor with the Uno was a real struggle. I tried several different libraries for the I2C LCD and stepper motor and wasn’t able to get a single update done quickly enough to fit between steps. This was easy to hear, sometimes you can see it, and depending on the pulse frequency it would even stall the stepper or cause it to reverse direction if it was changing speed at the same time. At this point I am just not updating the display in the “Jog” mode until you stop or change direction – the numbers are really meaningless anyways and if you’re cutting you should adjust it by looking at the tool not the screen. My 3D printer can do it without a problem, wonder how?

This idea has been floating around in my brain for awhile. I am finally getting started on it, and hopefully writing it down here will help me remember to finish it.

The “problem”
The mini lathe uses change gears for the leadscrew; for anything other than threading they are a pain in the butt (in my opinion). Changing the feed rate is annoying, and for regular turning I think it would be great if it were variable, just like a power feed on a milling machine.

This isn’t a novel idea, but I won’t let that stop me. Here are some others who have done this already:

• http://www.mini-lathe.com/Mini_mill/Modifications/Power_feed/power_feed.htm
• https://www.homemadetools.net/forum/mini-lathe-carriage-feed-70736
• http://www.varmintal.com/alath.htm#Ultra_Fine

I’m aiming for overkill, so here’s what I’m going to try to make this do:

• Use a stepper or DC motor coupled to the tailstock end of the leadscrew. Ideally permanently attached and can coast if you are using the change gears, or with some kind of clutch otherwise
• Microcontroller and rotary encoder for closed loop (if we use a DC motor) speed and position control. Knob to set the speed.
• Several “modes”
  • Jog: feed in either direction at the set speed, like a normal power feed
  • Set: after jogging to a position, save that position to memory for later
  • Auto: feed in selected direction up to the saved position.
• LCD to display the current speed, saved speed and saved positions

“Auto” mode (might rename that) is the neat part. I want to save two feed rates, one for cutting and another for reversing back to the beginning of the cut. With the half nuts engaged the whole time, first you would enter “set” mode, jog to the end of your cut and zero this “left” position. Then jog back past the start of your cut and zero this “right” position. A feed rate for each direction can also be saved. Now return to “Auto” mode. Setting the switch to move left will feed in at the cutting speed you set and will stop when it reaches that position. Retract your tool, then jog right and it will move back to the start of your cut at a much higher feed rate, stopping at the right position. I think this will be neat for taking repeated cuts and getting the feed rate just right every time. Another way to say all that is “I want a power feed with virtual stops and fast reverse”.

Using a rotary encoder or stepper on the leadscrew means that the positions are always relative and will be lost if you disengage the half nuts. Backlash will still exist, but if you approach your set points from the right direction I don’t think it will be a problem. On the upside, since we’re not measuring real units, (just pulses and steps) there should be no calibration required and it should be easy to use a variety of encoders and steppers regardless of their resolution. We can display numbers on the LCD for position and speed but we don’t have to care if they are inches, rotations, minutes, whatever. If this works out I’ll probably add it to my larger lathe as well.

I tried to measure the torque required by fixing a rod perpendicular to the leadscrew, hanging a weight on it, and measuring the radius where it started to move (I’ll add a photo of how I did this if I can find one). This was all done with the half nuts engaged

• .40 Nm to overcome friction only
• .93 Nm to do a very light cut
• 1.16 Nm is my guess for a minimum

It’s probably best to double that or more – the required force will increase depending on the cut you’re taking and how much friction there is between all the moving parts. There is probably a big difference between dirty and lubricated ways. I ordered what was advertised as a 1.9 Nm NEMA 23 stepper, and I’m going to use a timing belt with 20 and 60 tooth pulleys, so that should reduce the maximum speed and triple that torque number. My guess at speed was that 0-200 RPM would be appropriate, and it may be possible to get away with a smaller motor and larger reduction.

I’ve got some parts on order and I’ll work on the electronics while I wait. If this works out I’ll try to document it so that it’s reproduceable. It’s going to be a challenge to condense all these switches and buttons into a small package that doesn’t look humongous compared to a mini-lathe!