Current Limiting for a DC Motor

Thread Starter

NM2008

Joined Feb 9, 2008
135
Hi,

I am wondering if somebody could help with a current limiting circuit I am trying to implement.
What I have is a small 12v dc motor with gearbox.
This will be operating a mechanical valve.
As the valve reaches the end of its travel the motor will load and begin to stall.
This valve will be operated at most once a day.
The motor/gearbox does not have enough torque to cause damage to the valve, just enough to close it tightly. Also the gearbox is a worm type and will hold the valve closed without the need for constant power.
Due to is envoirnment limit switches cannot be used.
The unloaded/normal operating current is 50mA.

What I am looking for is a way to shut this motor off when is reached max travel/ beginning to stall.

For example when the motor current rises to 100mA current limiting will act and reduce the current to near 0mA or have a relay act and break the circuit.
The circuit can remain in an open or current limited state until it is manually reset i.e. to open the valve.

I do not have experience with building current limiting circuits.
What I have been looking at online are circuits using LM317 and LM339 (as I have both in storage) as current limitings circuits, but not sure if im on the correct path for the above application.

Thanks in advance,
Nigel
 

Sensacell

Joined Jun 19, 2012
3,420
Hmmm...

The problem here is the mechanical energy stored in the rotating parts- the armature of the motor.
The inertia will cause the valve to close tighter than the motor can un-do, even with full torque- it will get permanently stuck.

The only way you have a prayer of getting this to work is to have a speed control system that runs the motor slow enough, such that the motor cannot get stuck. If you creep into position, the energy stored in the rotating armature can be minimized, so the breakaway torque can be lower than the motor stall current rating.
 

Alec_t

Joined Sep 17, 2013
14,263
Another way of preventing the over-tightening would be to use a slipping clutch mechanism, which would begin to slip as soon as the load torque exceeded some limit less than the motor's starting torque. A lack-of-rotation sensor would then switch off the motor.
 

Tonyr1084

Joined Sep 24, 2015
7,829
Why not just build a timer circuit. If under normal operation the motor takes - oh, lets say 5 seconds to close the valve, then a drop-out timer could supply power for 7 seconds. Or 10 seconds.

Use a comparator and an RC circuit to set up the time delay. When motor power comes on to close the valve the RC circuit begins to climb until it surpasses the reference voltage. When it times out it can shut off motor current.

With a little thought you can further expand the circuit so that it holds itself off until you manually reset the system so as to open the valve electronically as well. Again using the timer circuit to control operation.
 

Thread Starter

NM2008

Joined Feb 9, 2008
135
Sensacell and Alec_t thanks for the input.
This type of valve is similiar to a butterfly valve. Its purpose is to divert air flow. I have tested this to the point of stalling the motor. I do not seem to have the problem which you are speaking of. The motor and gearbox are be capable of opening the valve without issue from a closed position i.e. closed with hard stall at 12v.
The valve rotates through 90 degrees. The gearbox output rotation is 3rpm. It takes roughly 5seconds to go from open to close.

Tonyr1084
When you say comparator and RC circuit are you referring to something similiar to this? http://home.cogeco.ca/~rpaisley4/ComparatorTimeDelay.GIF
Do you think a time delay circuit will be accurate enough? Will it lead to overshooting or undershooting? as its based on time.

I found this on youtube which seems fairly similiar to what I want to do,

No information is given with this circuit, but from the looks of it he seems to be passing the load current through a wire wound resistor then possibly using the voltage drop across it to switch on/off a MOSFET inturn latching and unlatching the relay cutting power to the motor?
Any ideas on this type of circuit?

Thanks
 

Tonyr1084

Joined Sep 24, 2015
7,829
Without knowing more about your valve system I can't say for sure. I assume you have a motor that closes the valve (as you said) but you don't say whether there's a shutoff once the valve is closed. You DID say you can't use a limit switch. So assuming when the valve is closed the motor remains powered until some human interaction changes things.

Your video is interesting. I wondered if you could use a relay closed through the motor, and with resistance from the motor the excess current would cause the relay to drop out - thus, removing power once the valve is fully closed.

As for the RC Comparator circuit - yeah, something like that. Only, instead of waiting to turn the power on you want to hold power for a period of time. Overshoot? Undershoot? Both are possible. But assuming you don't mind running the motor to fully closed, overshoot shouldn't be a problem.

Keep in mind I don't fully understand your project. Only you can determine what's best for your requirements. The video you provided is interesting nonetheless. It appears to cut power once the motor stalls enough to cause the relay to drop out. In order for me to accomplish that task I'd have to do a lot of experimenting with various relays and motors. But I'm fairly confident there's a magic spot where when the motor stalls the current it draws is sufficient to drop out the relay.

Let me bang a picture out.
 

Thread Starter

NM2008

Joined Feb 9, 2008
135
Tonyr1084,

Thanks again for the input.
Where the valve is nobody can see it or will be near it. It will be controlled by a toggle switch. So say the valve is open, somedoby flips the switch and the valve begin to slowly close. The only input the operator has is the toggle switch controlling opening and closing, so the motor needs to shut off automatically when at open or close.

With regard to the video im thinking the same, that the circuit he is using seems to be fine tuned to the components he has on the bench, whch for my application is not really practical either, as when the system ages how do we know that the relay will still drop out at the same current.

At the moment I am considering your solution with the time delay. As the rotation speed is low, a time delay circuit might be the simpliest solution for this.
 

Tonyr1084

Joined Sep 24, 2015
7,829
a time delay circuit might be the simpliest
I agree.

Here's that picture I promised:

[edit]

In retrospect, based on the greater information, the relay circuit drawn won't give you the reversible feature you seek. However, using the time delay method, you still have to energize it. Meaning the operator will switch the switch to "Open" or to "Close" and then push a "Start" button.

I'm also assuming your system is DC powered. AC power won't lend itself well to reversible motors or time delays.
 

Attachments

MaxHeadRoom

Joined Jul 18, 2013
28,576
You could use a time-off delay timer, the switch energizes the timer and once the switch opens the timer continues until the set time is up and switches the motor off.
The other is a current detect circuit using a comparitor across a very small value series resistor, <.1Ω.
Max.
 

Tonyr1084

Joined Sep 24, 2015
7,829
Why not go with a Momentary DPDT switch with center off? The operator will have to hold the switch in the "Open" or "Close" position for the duration of the valve movement. Then when released the switch snaps back to center off. The only down side to that is 'what if the operator doesn't hold the switch long enough?'. HEY! It's only for five seconds. NO?
 

shortbus

Joined Sep 30, 2009
10,045
Showing a photo of the valve and where it goes would be a big help to get answers. You say a micro switch can't be used, is this because the motor/gearbox is directly coupled to it? Or due to other reasons? A "flag" or "stud" can be added to the coupling to trigger the micro switch in some cases. Or there are many different types of micro switch actuators out there, maybe adapt one to the pipe wall and trigger it by contact with the valves butterfly. There are many other ways if the nature of the problem is shown.
 

Thread Starter

NM2008

Joined Feb 9, 2008
135
I spent a good bit of time experimenting with this today.
I replicated the circuit in the youtube video I found earlier.

This circuit is very tedious, as for its operation it uses a resistor in series with the motor. In my case this was 10ohm. As the motor stalls the voltage accross the resistor falls. This voltage drop can be used to deactivate a transistor through two resistors totaling approx 290kohm connected to its base.
The transistor in turn operates a relay. I had this fed to a second relay wired so as it latches once power is applied.
This gave effect that as power is applied the motor comes to life and the relay latches, as I put load on the motor the voltage across the 10ohm resistor would drop enough stop the transistor conducting and cause the latched relay to trip, thus stopping the motor. It would then remain off until the latch was reset.
The issue with this circuit is, the cutoff current depends greatly on power supply voltage, the resistance of the series resistor and the balance between the pull down resistor and input resistors to the base of the transistor and all of the parameters are in turn linked to that exact motor (for example a change of motor with a few ohms out in the motor windings and everything else is out). Any fluctuation between any of these and the trip current can increase or decrease by 100mA easily.
Overall it does seem to be a very accurate circuit, even after alot of fine tuning/trial and error.

What I think what I'll decide upon is a combination of the lm317 constant current circuit and a delay circuit possibly based off a 555.
When power is applied a delay to power off circuit is triggered. This circuit will drive the lm317 constant current circuit set to 125mA.
For example:
The valve takes 5 seconds to go from open to close.
The delay to power off circuit can be set to say 10 seconds once power is applied. The motor should start from stop with 125mA(still need to check this fully). As the valve closes the operationg current should rise from 50mA and be capped at 125mA. It will stay stalled at 125mA for the next 5 seconds until the delay times out cutting power to the circuit after 10 seconds.
Likewise for opening.

Thanks to everybody for the input and ideas.
I will work on completing this over the next few weeks and post pictures of the completed unit.
 
Last edited:

MrSoftware

Joined Oct 29, 2013
2,186
I skimmed, so forgive me if this has been covered. But look at the way auto-windows in your car work. You press the button once, the relay latches and the window automatically rolls up or down until it hits its limit, then a moment later the motor turns off. The window in my car uses what is basically an auto-resetting circuit breaker, I forget the real name for it. Under normal operating current load the device is fine. When the motor stalls and the current goes very high, the breaker trips turning off the circuit and unlatching the relay. After cooling for a few seconds the breaker automatically resets. Are you looking for basically the same thing for your valve?
 

hp1729

Joined Nov 23, 2015
2,304
Hi,

I am wondering if somebody could help with a current limiting circuit I am trying to implement.
What I have is a small 12v dc motor with gearbox.
This will be operating a mechanical valve.
As the valve reaches the end of its travel the motor will load and begin to stall.
This valve will be operated at most once a day.
The motor/gearbox does not have enough torque to cause damage to the valve, just enough to close it tightly. Also the gearbox is a worm type and will hold the valve closed without the need for constant power.
Due to is envoirnment limit switches cannot be used.
The unloaded/normal operating current is 50mA.

What I am looking for is a way to shut this motor off when is reached max travel/ beginning to stall.

For example when the motor current rises to 100mA current limiting will act and reduce the current to near 0mA or have a relay act and break the circuit.
The circuit can remain in an open or current limited state until it is manually reset i.e. to open the valve.

I do not have experience with building current limiting circuits.
What I have been looking at online are circuits using LM317 and LM339 (as I have both in storage) as current limitings circuits, but not sure if im on the correct path for the above application.

Thanks in advance,
Nigel
Just a rough draft ...
R1 monitors motor current. When it gets up around 100 mA the voltage across R1 reaches 600 mV turning on Q2 which turns off Q3 setting the latch. This turns on Q5 which turns off the motor. A reset buttons turns off the latch which starts the process over again.
How does the valve get put back into the other position?
 

Attachments

Top