Current limiting a DC gear motor

"Lars Bentley"

** The only way to limit the torque *available* from a DC motor is by limiting the current available and that automatically means the voltage to the motor and hence rpm will drop when more load is applied.

Your post seems to falsely imply that applied voltage, rpm and available torque are independent.

Taking your 200mA figure as a given, try connecting a 24 volt, 5 or 6 watt lamp in series with the motor. Even with the motor stalled, this will limit the current to about 200mA.

..... Phil

t

it

Phil, thanks for the response. To clarify a little, i dont care if the rpm drops somewhat because this circuit will only engage if the motor is pulling the spool too fast. The tape is fed at a costant rate to the spool, but as the diameter grows the spool will not have to turn as fast to keep up. i need the full torque at times to accelerate the spool quickly from a stop, but the full torque will harm the feeding mechanism if allowed to continue so the current limiting circuit will engage when tension on the tape has reached a certain level and disengage when there is to much slack. i suppose i could just have the circuit cut completely under tension, but then the switch will be contantly bouncing around and rapidly switching on and off. Btw i can use a lamp because itthe machine is used for photographic purposes and must remain in the dark.

"Lars Bentley"

Phil, thanks for the response. To clarify a little, i dont care if the rpm drops somewhat because this circuit will only engage if the motor is pulling the spool too fast.

** A DC motor will simply stall if ever the load requires more torque to turn than is available.

The tape is fed at a costant rate to the spool, but as the diameter grows the spool will not have to turn as fast to keep up. i need the full torque at times to accelerate the spool quickly from a stop, but the full torque will harm the feeding mechanism if allowed to continue so the current limiting circuit will engage when tension on the tape has reached a certain level and disengage when there is to much slack. i suppose i could just have the circuit cut completely under tension, but then the switch will be contantly bouncing around and rapidly switching on and off. Btw i can use a lamp because itthe machine is used for photographic purposes and must remain in the dark.

** Huh ??

You can or cannot use a lamp ???

Stick the lamp inside a box - you fool.

... Phil

** =A0Huh ??

sorry for the typo, i can NOT use a lamp. This is a complicated mechanism that is inside a complexx camera which can have no light at all, there isnt much room and its not worth risking a light leak which could ruin thousands of dollars worth of film and processing. I'm sure there is another way to limit the current. ANyone else have a suggestion?

You could put the lamp inside a box. But that may not solve your whole problem. The problem with a lamp (or a resistor) is that to limit the current to no more than 200mA would mean that the motor only _gets_ 200mA when it's stopped and has minimal voltage at its terminals -- at any higher speed it'd drop some voltage itself, and so the resistor would limit the current further.

What you really need is active control of the tape tension, but that requires considerable work; given the level of expertise that you're demonstrating in this post you'd need to flog things pretty hard before you got what you wanted.

--
www.wescottdesign.com

"Tim Wescott"

** A lamp behaves very differently from a resistor - that's why I suggested it.

Brief surges of current, up to 10 times the steady value, are possible.

..... Phil

reducing the current to the motor will reduce the torque proportionately

but with more tape on the spool it more torque will be required to maintain the same tension in the tape.

you don't have to use a lamp, there are other current-limit circuits. depending on how tightly you need to control torque.

It isn't clear whether you want to use the motor to both accelerate and deccelerate your take-up spool.

For decceleration you'd want to reverse the sign of the current being fed into the motor, which eliminates the simplest solution,

I'd think in terms of a National Semiconductor LM12CLK power op-amp, which costs some $50 and comes in a TO-3 can, but can dissipate 80W. If you configure it as a Howland current source, you can control the output current with an easily switched low-current voltage signal, and usee an alanog switch to swap between two sources - one limited to to generating currents in the range +/-200mA whhere the other can force curents up to +/-1A.

df

Phil Allison's incandescent lamp is ingenious, but tricky to match to your particular load.

Phil's vituperation has to be ignored - he seems to suffer from some variant of Tourette's syndrome.

-- Bill Sloman, Nijmegen

Very interesting suggested use of the lamps characteristics. Although I'm not sure that the duty cycle of the lamps characterists will match the needed motor drive duty cycle. But, cool idea. Lot of lamps to choose from. Mike

On a sunny day (Sun, 6 Sep 2009 06:45:30 -0500) it happened "amdx" wrote in : Very interesting suggested use of the lamps characteristics.

You hope, soon all incandescent will be banned :-)

You need to have a catenary arm with roller that the tape can ride under, providing a constant pull on the arm. This arm will have either a mechanical spring or air cylinder to adjust tension.. This arm will also have a POT on it that will adjust a simple drive circuit that regulates the motor.

I think constant tension through out is what you're looking for..

Low power DC drives are already made to do this for you, with just a little extra work..

If you don't want to buy a drive, for this simply app, you could construct a simple circuit that will do this..

Something to think about..

Lars Bentley wrote in message ...

As the diameter of the film pack increases, you need more power applied to the motor to increase the torque. No constant current or voltage scheme will provide that. If you can live with the variation in tension from the inside diameter to the outside diameter, it is a simple matter of inserting a resistor in series with the motor to drop the voltage applied to the motor. Starting up from a standstill requires more current than while the motor is actually spinning. So at startup, there is a good chance that the takeup spool can't keep up, and a loop of film will spill out. To get around that problem, you can put a timer in the circuit that switches in a smaller resistor for however long the boost is needed. Being a DC motor, you can do this by putting a large capacitor at the output of the resistor. While the motor is disconnected, the cap charges up to a higher voltage, and this higher voltage will give the motor a boost at the instant the motor is connected. As the cap discharges through the motor, the resistor eventually takes over to supply a lessor voltage. This is a cheap and easy way to deal with your situation, but keep in mind that because there is no feedback, the motor can only do what you tell it to do. If you take a look at any tape deck, they deal with the same situation of keeping a constant tension on the takeup and supply reels. Many of the older cheaper decks do it as explained above. The tape tension easily drops in half by the time the reel is full. More sophisticated decks use a sensor to measure the tape tension. The sensor output feeds an amplifier which in turn controls the power supplied to the motor. A tension sensor will likely have to be a custom built part, but it sounds more complicated than it really is. It can be as simple as a potentiometer, photocells, LVDT's etc. Servo amplifiers are an off the shelf item. A complex one for a DC motor is not much different than a DC coupled audio amplifier.

o

ill

de

otor.

or is

t

do

e

ly

d

to

n

ps

or

n

ly

.

is

Thanks for the responses everyone. To answer some of the questions, No I don't need the motor to decelerate the spool, and I'm not worried about the take up spool not keeping up at start up because I can program the acceleration of the servo motor driving the feed mechanism. I am using a spring loaded tensioning arm/roller that the film rides on, and will trip a switch when the tension gets too high & switch to a lower torque while maintaining *some* tension on the film.

I suppose the other option would be to put a small potentiometer on the axis of this tensioning arm that controls a voltage regulator supplying the motor with power. I like this idea but this tensioning arm/roller can only turn a max of 45 degrees so the pot would have to have the resolution over a 45 degree turn to adjust the speed of the spool from 100 - 350rpm. I am also very cramped for space around this mechanism, so I would have to incorporate some very small trimpot, like the little pcb mounted ones that are usually turned with a tiny screwdriver and left at a certain setting. Can these tiny trimpots withstand frequent adjusting for hours on end without failing?

Let's say I use a small voltage regulator like this one here (never mind that it's 10v max output for now):

...and I removed the potentiometer that adjusts the voltage and wired in a pot that is located a few inches away on my tension arm. Would removing the pot from the pcb and locating it inches away cause problems?

Thanks,

Lars

ed

to

will

side

motor.

otor is

p

hat

n do

the

ally

ind

it to

ion

he

rops

nsor

in

kely

tc.

r is

Potentiometers are lousy angle sensors - amongst other things, they wear out - and a trimpot would be totally hopeless. The rotary version of the linear variabe differential transformer could do the job, but you might find it difficult to find one compact enough to retro-fit into the space you've got.

You could hang a neodynium-iron or samarium-cobalt magnet on your tension arm and measure the magnetic field as close to it as you can get - you'd probably want a linear Hall effect sensor, possibly two of them mounted so that their sensitive axes were both in the plane of the tension arm's movement, but at right angles to one another.

Allegro have a bunch of potentially suitable sensors

The traditional solution is a LED and a photodiode or two, but that won't suit you - through you might get away with an IR LED since most cine film isn't sensitive to 950nm radiation.

-- Bill Sloman, Nijmegen

On a sunny day (Mon, 7 Sep 2009 02:41:18 -0700 (PDT)) it happened Bill Sloman wrote in :

Are you aware that every model servo has one? never had a servo fail due to a worn out pot. But maybe I do not use them enough...

"Jan Panteltje" Bill Sloman

** Model servos move rather slowly. ** Models do not rack up enough hours before crashing or sinking to wear out a pot. ** Ditto.

..... Phil

On a sunny day (Mon, 7 Sep 2009 20:48:40 +1000) it happened "Phil Allison" wrote in :

But I use those for camera movement Phil :-) There is even auotpan, moves continuously from left to right and back.

But I really do not use it a lot, Have a spare servo in the drawer though.

Search in Digi-Key for "angle sensor" -- there's a pot they sell that's not much bigger than a trim pot that's specifically designed ?to measure the angle of a damper in an air duct.

I have no idea of the wear-out time, though. Pots do wear out, but pots that are made for continuous service wear out more slowly than el-cheapo volume controls.

--
www.wescottdesign.com

Studer used pots on their studio decks, and to my surprise, they lasted forever. Just buy a good quality sealed pot. The angular displacement would not have to be over the full range of the pot, because if your servo is set up correctly, the position of the sensor arm will not change. The concept is that the servo amplifier supplies whatever power is needed to keep the arm in the same positon, and the tension is determined by how hard the spring pulls on that arm. BTW a company called Electrosound used LVDTs in one of their models. Ampex used leds with photo transistors which would probably would'nt affect your film. Many others use visible light with photocells. You have alot of options.