Hello, I want to control a washing machine universal motor with a microcontroller and MOSFET, but by connecting the GND of the motor and the microcontroller, the micro resets, please guide me, thanks

Moderator edit: New thread created from this.

 

The method that was traditionally used for this was with an IC device developed especially for the purpose by Motorola,
TDA1085.
There may still be the APP note that was published with it.

 

but by connecting the GND of the motor and the microcontroller, the micro resets

That sounds as if you could have a 'ground bounce' issue, where the high motor current creates a significant voltage drop across a grounding conductor shared by the micro. A 'star ground' arrangement (Google that) may help.
Welcome to AAC.

 

How are you using the Mosfet? DC Motor Voltage?
Either way, generally with a Universal motor, you will need some kind of RPM control !

 

If you end up using a Triac, there is also Fairchild APP AN-3006 fig 3

 

What sort of washing machine has a UNIVERSAL motor, that is, a motor with a wound field and also a commutator with brushes?? I have never seen such a washing machine. Perhaps if was a front load machine?

 

Hello, I want to control a washing machine universal motor with a microcontroller and MOSFET, but by connecting the GND of the motor and the microcontroller, the micro resets, please guide me, thanks

Moderator edit: New thread created from this.

Hi,

This could be a ground problem like Alec said, or it could be just draining the power supply for a moment, long enough to reset the uC which could happen either by a complete loss of voltage or a significant drop in voltage which the uC sees as a brownout.
In either of these cases you could separate the motor power from the uC power, with an extra filter cap for the uC power rails. The separation component would ideally be an inductor, but that's not always easy to work with so a resistor or diode could be enough. A diode allows the cap to maintain the voltage level before the motor turns on and so the uC sees no change or very little change. A resistor could work but the value has to be right, and a linear regulator after the resistor would work even better. In fact, a different regulator for both the motor and the uC chip.

There is one more possibility though and that is EMI. EMI can also couple to the uC chip or lines running to the chip and cause problems with the uC chip. If this is the cause, then you may be able to detect it within the code and force an interrupt to code that still allows it to work normally. This assumes that it's a one-time occurrence, or a multiple occurrence that is manageable through code in a similar way. This involves being able to restore the state of the uC once the EMI spike is over.
This can actually happen even if the motor is not directly tied to the uC circuit with any direct wiring and it is turned on (or even off) at some point.

 

What sort of washing machine has a UNIVERSAL motor, that is, a motor with a wound field and also a commutator with brushes?? I have never seen such a washing machine. Perhaps if was a front load machine?

Like I mentioned in post #2, Some time ago, a few manuf wanted to use a Universal motor for WM usage, Motorola came up with an IC to control these motors using a simple pulse wheel sensor for feedback using a Triac, due to the Universal motor operating in a runaway condition.
TDA1085.
They worked quite slick as I used one for my band saw, also there was a treadmill manuf. used it for their product.

 

Like I mentioned in post #2, Some time ago, a few manuf wanted to use a Universal motor for WM usage, Motorola came up with an IC to control these motors using a simple pulse wheel sensor for feedback using a Triac, due to the Universal motor operating in a runaway condition.
TDA1085.
They worked quite slick as I used one for my band saw, also there was a treadmill manuf. used it for their product.

OK, and certainly a brush-type motor would be good for a treadmill since they are not expected to be a long-lasting appliance .
But why would the "GND" connection from a large motor connect directly to " a microcontroller and MOSFET "? That seems like control connections will need an opto-isolator, because of both motor noise and also inductive spikes. What power source is the micro using in this application?? That may be part of the noise connection.

 

OK, and certainly a brush-type motor would be good for a treadmill since they are not expected to be a long-lasting appliance .

Don't know why you would say that ?
It belies my experiences with brushed motors?
One down side is that the series field motor normally operates in a runaway condition and should always posses a load.
The main characteristic is Very high torque down to zero RPM, one reason series field motor was used for starter motor for many decades.
Where they are used often in domestic situations, they rarely run for more than a few minutes or at least less than an hour at a time.
e.g. The one in my central VAC ran for close to 40yrs before I had to change the bearings and brushes.

As to sharing a common GND, it is routinely done in CNC control systems, where low voltage computer control is wed with servo motors, spindles and solenoids.
The recommended method is to carry out equi-potential bonding of all ground systems together with the service earth GND, and terminated in a star point ground.
Incidentally it is also now recommended to earth GND both ends of shielded cables which can be successfully done where equi-potential GND bonding of all machine systems is done.
{Reference:- Siemens Earth Ground bonding techniques}

In wiring control enclosures, I always earth grounded all system to a central Star point where possible.

 

For shielding, tying both ends of the shield to whatever passes as "ground" can work with some installations. For industrial test machines, which sometimes use load cells and bridge type pressure and flow transducers, every machine is a custom product and the standard scheme did not always work. Some products are noisy and some DC motors have been VERY NOISY and required unique shield connection arrangements. Measuring fluid pressure ripple noise in a vane pump-drive motor combination was an especially noisy application. Ultimately even the pressure transducer had to be un-grounded and isolated, and the sensor connection cable shields all lifted from frame ground connections all the way back to the amplifier module.
So just because some scheme for shield grounding works quite often is no assurance that it will be the solution every time.

 

Some products are noisy and some DC motors have been VERY NOISY and required unique shield connection arrangements

DC motors and AC variable speed (VFD) spindles were used for many decades in the earlier years of motion control, Each Servo's was generally equipped with a high resolution 5v quadrature encoder.
We never used shielded cable for the motors, usually tightly twisted conductors for noise cancellation.
The encoders however, were wired using specialized shielded cables.
The controller itself was operated at 5vdc.
Although there was opto isolation used extensively, all systems were Earth GND bonded.

 

DC motors and AC variable speed (VFD) spindles were used for many decades in the earlier years of motion control, Each Servo's was generally equipped with a high resolution 5v quadrature encoder.
We never used shielded cable for the motors, usually tightly twisted conductors for noise cancellation.
The encoders however, were wired using specialized shielded cables.
The controller itself was operated at 5vdc.
Although there was opto isolation used extensively, all systems were Earth GND bonded.

Certainly 5 volt logic signals from a low impedance driver are much less likely to be affected by external noise pickup, while an inexpensive processor module with higher impedance inputs and lower threshold voltages is much more likely to suffer from noise pickup.
And I am certain that those high resolution encoders were quite well designed , as were the receivers for those signals. And some of those encoders even used complementary outputs, for additional noise immunity. Not much comparison with a wash machine motor.

 

The encoders used were mainly RS485. i.e. Complementary pair per signal.
.

 

IN applications that do not use complementary signals from low impedance drivers, plus optical isolation, adequate shielding is a better choice. In addition, all of the power circuits should be separate from the control circuits.
Now I am asking about what sort of control the TS wants to achieve. Is it speed control only, or also directional control, or both speed and direction. There are different schemes to achieve either or both.

 

From the OP's statement " control a washing machine universal motor with a microcontroller and MOSFET, ".
One avenue would be to incorporate the TDA1085 into the design or simulate what the IC does with a micro.
If reversing is required, extra circuitry is required in order to reverse the fields WRT the armature.
A Triac control rather than Mosfet will be much sturdier for this application, IMO.

 

One example out there for washing machine.

 

One note of caution here is that every bit of that circuit, except for the tachometer output, is tied to the 0 volt AC mains.
Certainly the triac is a better control device than a mosfet.