Maker Pro
Maker Pro

Pulse Width Modulation

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
I'm attempting to use a pulse width modulation control unit to operate windshield wipers on an old automobile. The little package unit I purchase works fine on the bench but because the PWM power input ground and the wiper motor ground cannot be separated on the car, the grounds are commoned and the unit will not control speed.

Any advice on a way to get around this? Do I need to build an isolated 12DC to 12DC power supply? If so, can someone suggest a DC-DC converter that would handle the 3 amp load. I haven't had much luck.

Thanks - Walt
 

Minder

Apr 24, 2015
3,478
Joined
Apr 24, 2015
Messages
3,478
There should be no reason why it would not work, post the schematic, why is the PWM unit Not being powered from the same supply as the wiper motor?
M.
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
I don't have a schematic of the PWM unit, but here's the hook up. On the bench, if the two grounds are jumped together the motor runs at full speed with no speed control.

Waltpwm.jpg pwm.jpg pwm.jpg
 

Minder

Apr 24, 2015
3,478
Joined
Apr 24, 2015
Messages
3,478
There seems as though there is a conflict with the unit because the motor frame is at ground potential.
You may have to see if you can isolate the -ve terminal on the motor from frame GND, or use a regular isolated motor,
It may prove difficult to to do either.
Unfortunately, generally auto wiper motors were not designed to operate with isolated -ve supply.
The PWM unit is switching, (or trying to) switch the -ve side of the motor.
Another alternative is modify to a P Mosfet output of the control.
M.
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
Isolating motor -ve would be a can of worms if at all possible and a regular motor is not an option due to originality. Do you see any issues with an isolated power supply feeding the PWM unit?

Walt
 

Minder

Apr 24, 2015
3,478
Joined
Apr 24, 2015
Messages
3,478
The Mosfet in the PWM unit still has to be referenced to motor common. Which is also connected to auto 12v com.
M.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Windscreen wipers are generally controlled via scr , the speed being delays between wipes rather than constant speed control of the motor.
 

Kiwi

Jan 28, 2013
471
Joined
Jan 28, 2013
Messages
471
What make and model of vehicle?

What brand wiper motor?

Single speed or two speed wiper motor?

What are you actually trying to do???
 

Minder

Apr 24, 2015
3,478
Joined
Apr 24, 2015
Messages
3,478
The OP mentions an OLD automobile, so I would hazard a guess there is no solid state control back then, ;)
I also assume from the OP that wiper control method is sought.
Many of the really old models, either did it mechanically or even not at all.
M.
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
The vehicle is a 1957 Mercedes 190SL and the objective is speed control rather than intermitment operation. It's a two speed motor but generally the slow speed is to fast.

I did have success with a dc/dc isolation power supply borrowed from another project ,see attached, but it's a kinda expensive route to take. It seems isolated dc/dc converters that handle 3 amps are pricey.

Walt
 

Attachments

  • pwm2.jpg
    pwm2.jpg
    26.3 KB · Views: 90

Minder

Apr 24, 2015
3,478
Joined
Apr 24, 2015
Messages
3,478
Still doesn't make sense as posted, if the motor -ve is still connected to the motor frame.?
Motor connections do not look right.
M.
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
No being as adept with electronic circuitry as most of you, it makes even less sense to me. Attached is a photo of bench testing, you'll note -ve is properly connected to the zinc die cast housing and +ve is connected to the slow speed terminal. The other three are high speed, maintain, and unused.

If input and output negative on the PWM unit are jumpered, as would be the case with the unit mounted in the car, the PMW goes to 100% duty. An isolated dc/dc converter feeding the PWM eliminates that.pwm3.jpg

Thanks - Walt
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
The OP mentions an OLD automobile, so I would hazard a guess there is no solid state control back then, ;)
Yes, but I was referring to add-on devices.

Earth will be for the park switch, not the motor wiring directly.
Although there are many ways to achieve the 2 speed arrangement in original equipment, I have chased up a 2 brush motor diagram to save drawing out again.
Yours may or may not be the same but it gives an inside into how perhaps your iso supply is operating.

To the right are motor and park switch components.
To the left is the "Off"..."Low"..."Hi"....speed selector switch.

Cheers ..Jorgo
 

Attachments

  • wiper_2Speed.jpg
    wiper_2Speed.jpg
    33 KB · Views: 98

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
No being as adept with electronic circuitry as most of you, it makes even less sense to me.
The problem is that the motor negative is connected to ground though the motor's houing.

The PWM controller you are trying to use has to go between the ground of the motor and the circuit ground, which you cannot since the motor's ground is already connected to the chassis through it's housing.

There are two ways to fix this.

1. Isolate the motor ground from the housing. If the motor has a ground wire that just happens to be connected to the housing, detach that and then check for continuity between the housing and the wire. If there is no connectivity, then you can use the PWM unit by connecting to plus output to the newly separated wire, and the - output to the chassis ground.

2. Get a PWM controller that switches the +12V instead of the ground wire. You may be able to do this by adding a P-channel MOSFET to the output of your existing PWM CONTROLLER. However, if you do that, the duty cycle will be reversed, i.e. it will be at full power when you turn the control all the way counter-clockwise, and at no power when you turn the control all the way clockwise. This you can fix by swapping the 2 outer connections to the potentiometer. If you want to go this route, we can try to talk you through it.

Bob
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
The motor internals are packed so tightly that they're a bear to work on. I have a non-functional spare that I'll inspect for the possibility of isolating the ground.

In the interim (and probably an easier solution) a rough sketch of a P-channel MOFET diagram would be greatly appreciated. No problem with switching the potentiometer around.

Walt
 

WaltP

Jun 16, 2017
10
Joined
Jun 16, 2017
Messages
10
Another question for the group. I've experimented with both an isolated power supply and building the circuit proveded by Jorgo , also attached below except with 1 p-Mosfet rather than two. I like this approach as it's been fun building the circuit and I'd like to make my first PCB but the voltage supplied to the motor is only 8.5 to 9 and with cold components, it's not enough to start the motor. I get the full 12v with the same power supply and Chinese PWM unit.

If I procure addtional P-Mosfets, will the output voltage tend to increase?


Thanks - Walt
 

Attachments

  • P-Mos3.png
    P-Mos3.png
    12.8 KB · Views: 172

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
With 2 output mosfets I would imagine it would be half the "on" resistance, so half the voltage drop on the mosfets.
Just what this figure actually is depends on how hard the mosfets are turned on.
Spec sheet says around 0.018 ohm.
 
Top