This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

amplifier/driver for miniature motor for low freq

Other Parts Discussed in Thread: LMV881, LM4675, LM4995, DRV8601, LM4673, TPA2005D1

Hi

we have design a motor with resistance of 100 ohm and inductance of 600uH.

The motor activate with signal of 5-15Hz (pwm or sinus between (4 to (-4)) volt).

For this I need current of around 40mA.

I need a chip with the smallest foot print.

I though of using LMV981 or  LMV881 and connect them to my analog (sinus or pwm) signal.

will this work?

do you have recommendation on something else?

thanks in advanced.

 

  • Hi Sam,

    Remember that high-current and small package may be mutually exclusive. It is hard to get the heat out of a small package. Even if you could get the heat out of the tiny package...you need an area for heat sinking.

    At ~40mA, the output voltage swing will be 1V less - so you may not have enough output drive. You may need more current than 40mA to overcome the "stall" current of a motor to get it started.

    What are the supply voltages? Will you be driving the load ground referenced or as a bridge (BTL)?

    Have you thought about audio amplifiers? They are just power op-amps.

    Audio power amps are designed to drive low impedance (4-16 ohms) inductive loads with much more current driving  capability than an op-amp. They also add better heat sinking and thermal limiting. You only need about 200mW. Perhaps class D to save power/heat?

    Your frequencies are in the low audio range...so should not be a problem...though PWM may be an issue depending on the period.

    The class D types (switching) are used in cellphones and are in small packages. Since they do not have to conduct linearly, they run much cooler and are in smaller packages than the class AB types..

    See the "dynamic" speaker drivers:

     http://www.ti.com/lsds/ti/audio-ic/dynamic-speaker-product.page

    For example, see the simple LM4675 class D or the LM4995 class AB, both in a tiny WSON or BGA packages.

    Regards,

     

  • Hi Paul

    thanks for your sweep answer.

    You are right maybe 4 to (-4) is a bit high but even 2.5 to (-2.5) are something I still cant achieve easily.

    the supply voltages I use are 3 and -3 volts.

    with those voltages I should get around 30mA (I dont understand why you wrote 1V swing it was 40mA*100ohm=4V the inductance in 10Hz is neglectable ). in a bridge do you mean like in DRV8601? this is also considerable option.

    I though of audio amps but they are for resistance of 2-8 ohm so I though it will be a bit problem because I have 100 ohm resistance dont you think? 

     from what I saw the dynamic speaker drivers are not rail to rail amps so I think I wont be able to get the negative part of the sinus, isnt it so?

    thanks again. 

  • Hi Sam,

    The audio amps should have no problem with driving lighter (100 ohm) loads...the swing should be even better with the lighter load. The audio amps should be able to easily swing close to the rails.

    The audio amp examples I provided are "BTL", or "Bridge Tied Load" (also called "Differential" drive) output. The speaker is connected between the two amplifier outputs.

    Each amplifiers output is biased at 1/2 supply voltage, so that the speaker sees 0V across it with no signal. The amps are driven 180° out of phase (one amp inverts). Examine figure 1 in the LM4995 datasheet. So that at full peak output (assuming single 3V supply), one speaker terminal is at 0V and the other is at 3V. This gives you full supply swing and 4X the power of a single-ended amplifier.

    The disadvantage of the BTL output is that it cannot drive a grounded load - the speaker/motor has to have two wires  to the amplifier.

    The output resistance of the op-amp causes a voltage drop as the load is increased. At 40mA, the LMV981 could only swing to within 1V of the rail (see the output swing graphs) so you would get less than a full output swing. Signal-level op-amps (like the LMV's) are not designed to drive heavy loads. Output drive ability also decreases as the supply voltage drops.

    So with the audio amps I mentioned, you would be able to drive the motor off the single 3V supply. The -3V would not be needed.

    I would investigate the motor drivers and the audio amplifiers...they are designed to do just what you are trying to do..and are in small packages.

    Regards,

  • Paul Hi

    Thank you very much for your detailed answer.

    from what you are saying I understand I need to connect the amp as a differential amp (fig 25 in the LM4995 datasheet). this is something I done before but this forced me to spend 2 DAC channels of my controller in order to generate "negative" voltage. 

    the other option was to use it as regular amp and add mux in the amp output to switch the motor voltage direction.

    because of that reason I though using drv8601 will be easier.

    Regarding the  LM4995, perhaps you have inside information and you know if there are spice model and EVL datasheet available for this component before ordering the EVL?

    thanks in advanced for all your help.

     

  • Hi Sam,

    You do not need two D/A's and you do not have to drive it differentially.

    The input can be single-ended. Just ground the "other" input. See Figure 1 or Figure 24 on the LM4995 datasheet, or Figure 24 on the LM4673 datasheet. The "second" amplifier internally inverts off the first amp  - so you do not have to provide the inverted signal (see the "AV=-1" lower-right amp in LM4995 figure 1). Differential inputs are used in noisy environments to reject common mode noise between the source and the amp.

    There are several TPA2xxxx class-D amps (TPA2005D1, similar to the LM4673) in TINA (in the "other" macros section). The models are encrypted, so they cannot be used in other SPICE tools....sorry...

    Regards,

  • Hi

    thanks for your answer.

    I think we are not align here (or that I dont understand what you mean).

    In my D/A I can create is only positive signals the signal I creat can be abs(sin) or something like this I can not generate negative signals.

    did you knew it?

    if you knew it, how can I get negative output from the amp without negative input? what is the difference in the input if I want to get 1v and -1v on my output.

    if you didnt know it, see my last post before this one.

    again, thank you very much for all your help.

     

  • Hi Sam,

    You do not need negative voltages if you use the audio amps in BTL mode. They are all single supply.

    Remember that with the BTL amps, the motor is not grounded. The motor does not have a "ground" reference. It's "ground" is the mid-point between the outputs.

    So for a single 5V supply, the motor "ground" will be +2.5V - since BOTH amplifiers will have +2.5V on their output and the net voltage across the motor is zero with no input signal.

    Similarly, the inputs are biased at mid-supply (2.5V). An input voltage above +2.5V will cause the motor to see a "positive" voltage (the non-inverting amp goes positive, and the inverting amp goes negative). Any input above 2.5V would cause the motor to see a positive voltage.

    So you would apply a sine wave with the zero point centered at mid-supply (+2.5V), swinging +1.5V to +3.5V to get a +/-2V (4Vpp) swing on the motor (assuming gain of two, where RF and RI set the overall gain).

    So you would apply your DAC output signal to the cap-coupled input (on Figure 1). The amplifier contains all the proper mid-supply biasing. Remember that the circuit shown is inverting - so you either have to reverse the expected polarity of your input signal - or just reverse the motor leads.

    Regards,

  • Hi

    Now I fully understand how it works. this is exactly the same way as the  DRV8601 works.

    the issue with this is that I am working with low voltages 2.8V top so my DAC can only give me around 2.5V and I dont have 5V for the supply.

    even if I would put the 2.8V to the supply I can only get +-1.4V swing.

    does this amp can work in regular mod like regular opamp? in that case I would connect a mux  to the amp output and switch the motor wires with the mux in a manner that very time I need to switch current direction, I switch the mux and the amp will give me abs(sin) according to my DAC signal.

    in that way I can get to 2X2.8vpp 

    can this amp work as regular amp and still have all it advantages?

    if not, is there another amp that do it in such small footprint?

    Regards,