Hey all,

The brief summary for this project is that I want to make a 2 position encoder for my industrial sewing machine. I want my machine to stop with the needle in the up or down position, have a button that will advance the needle up or down and a button that will produce 3 stitches for backstitching. I plan on using an arduino based circuit for this project.

Here is a link to the machine I have. http://www.westchestersewing.com/wsm/5550/

I have a degree in physics and a some experience in electronics, but I don't have a lot of experience in building power supplies or circuit design in general. I own an outerwear company and I want to modify my machine to make it easier to use.


Here is my main problem right now.

1. From what I can tell my machine seems to be controlled by a magnetic proximity sensor. The proximity sensor is a small 3 pin device and there is a link to a picture of it at the bottom of this page. There is a small square neodymium magnet that gets moved closer to the sensor when the foot pedal is pressed down. However, when I measure the voltage across the pins to the sensor I measure ~8V AC for all pins. When I measure the resistance across the pins of the sensor they don't change with the introduction of a magnetic field. If we look at the picture from left to right (so upside down from what we might normally consider pins 1, 2 and 3), 1 and 3 measure 147.5 k K Ohm, 1 and 2 give 142.1 k Ohm and 2 and 3 2.35 k Ohm.

Can any one tell me what this sensor is and propose a method for sensing and then controlling this circuit with an arduino?

Here is an image of it http://hubbleouterwear.com/Backups/IMAG0074.jpg

 

That appears to be a hall effect sensor. If it's controlled by the position of the foot pedal, it's for motor control. If it were set up near the needle bar, it could be used as a needle position sensor which is what you need.

A 3-stitch back stitch is quite short; are you sure you don't want 5 or more?

I don't consider this a feasible project because even if you could attach the needle bar position sensor, simply controlling the motor will not give you precise control of the needle bar position. For that you would need some mechanical braking designed into the head and probably some kind of stepped advance for the motor. You would also need to sense the position of the foot so you could lock out the 3-stitch function when the foot was raised.

I know the machines with these features built in are more money but you really can't expect to just tack on (pun accepted) a programmable back stitch and needle up/needle down feature.

 

How easy is it to manually control the motor speed and stop the needle in the up position? Does it have the servo motor or clutch motor?

 

I thought it might be a hall effect sensor, but my machine can be run very slowly with the foot pedal, and my understanding is that hall effect sensors are on or off not variable .The motor has a break in it that engages when the foot pedal is not pressed down so the motor stops quickly, and but it can still spin if the the motor gets power and the break is engaged . My plan was to send a signal to the motor when the foot pedal was released after sewing so that it would progress slowly until the needle was in the desired position.

 

I thought it might be a hall effect sensor, but my machine can be run very slowly with the foot pedal, and my understanding is that hall effect sensors are on or off not variable.

Hall effect sensors are available in both bi-polar and linear versions. For example, the A3144 is bi-polar, but the SS49E is linear.

 

Hall effect sensors are available in both bi-polar and linear versions. For example, the A3144 is bi-polar, but the SS49E is linear.

Thank you, I looked around for a while and couldn't find a linear sensor.

I'm probably going to sense the output of the hall sensor replacing the input with the 5v output of the arduino, but what is the best way to drive the 8v that my machine needs to be signaled to start sewing? I was thinking of a transistor where I would hook up the original 8v input to the hall effect sensor to it.

 

Is this 8V signal different from the Hall sensor output? What normally generates that?

 

Is this 8V signal different from the Hall sensor output? What normally generates that?

I have done some more investigating into how my machine is controlled and here is what I have found.

This analysis is based on the idea that the center pin is the ground pin. Pin 3 is the left pin, pin 2 is the middle ground pin and pin 1 is the right output pin. The pin numbers are referenced from the picture I provided earlier. Here is the link again.

http://hubbleouterwear.com/Backups/IMAG0074.jpg

Note* Earlier I mentioned that the voltages were AC. This is not the case. They are DC. I measured the voltage before with the Hall effect sensor disconnected and that must have affected my measurement.

The power for this circuit comes from a rectifier circuit. This circuit is not directly powered by the 110V AC wall voltage.

In the absence of a magnetic field:
Pin 1 and the ground pin 2 are at 12.3V
Pin 1 and pin 3 are at 5.7V

In the presence of a strong magnetic field:
Pin 1 and and 2 are at 12.3V still
Pin 1 and 3 are separated by a few milivoltes (so basically zero resistance).

In the presence of a weak magnetic field:
Pin 1 and 2 are at 12.3V still
Pin 1 and 3 are at a couple volts 1.8-2.4V.

Conclusions

To me it looks like pin 1 is the source voltage, pin 2 is the ground and pin 3 is the output of the sensor. When there is no magnetic field, the sensor does not open the circuit but imposes a high enough resistance to keep pin 3 at 5.7V which is enough to keep the motor off.

When a magnetic field is applied the sensor reduces the resistance between pin 1 and 3 even further which signals the motor to start.

There seems to be a small range of voltages that allows the motor to operate. The range seems to be from ~10-12V.


Questions

How might I control this circuit?

My Idea
I was thinking about calling pin 1 the ground to the arduino and then sensing the voltage by the arduino on pin 3. If the voltage was below the "off voltage" the machine would be "on" and when the voltage went above the "off voltage" the arduino would send an analog signal to pin 3 of the sensor that would lower the voltage to an appropriate voltage that would progress the needle slowly until the down hall effect sensor was engaged.

I know that the arduino has an analog input range from 0 to 5 voltes. Would the 5.7V overload the arduino, or would it just clip off the extra .7V?

 

If you believe that pin 2 is the ground pin and pin 1 is the positive voltage supply pin, you should be making voltage measurements between pin 2 and pin 3, shouldn't you?

 

If you believe that pin 2 is the ground pin and pin 1 is the positive voltage supply pin, you should be making voltage measurements between pin 2 and pin 3, shouldn't you?

I'm measuring between pin 1 and 3 because the voltage difference between those two dictates the applied impedance of the hall effect sensor. I suppose I could measure the voltage between the ground and pin 3 but I would just measure 12V when the sensor is engaged (I think).

 

I don't think so. My experience with a linear Hall effect sensor is that the voltage at the output pin is positive with reference to ground, varies from near 0V to some fairly large percentage of the supply voltage, and is directly proportional to the strength of the magnetic field. It is that voltage with reference to ground that will provide the motor speed control. Depending upon whether the motor is AC, or DC, an appropriate interface is provided.