Voice coil motor drivers

[NpR]

Hello!

DISCLAIMER!: This was, in my opinion, the closest matching forum for this sort of halfway experimental question/musing. If I am mistaken, please correct and point a more appropriate forum for this topic.

And then, to the topic:
While trying to update BOM for an existing design, I ran into an interesting deficiency in recently introduced dedicated VCM (voice coil motor) drivers.
The VCMs I'm using have a DC coil resistance of about 35 ohms and rated nominal current of 120mA. To achieve this, the minimum supply voltage would have to be abour 4.5V (also considering the driver voltage drop). However, the more recent dedicated VCM drivers are only rated up to about 3.6V, which means that they are physically unable to reach the full 120mA drive current with 35 ohm motors  . In other words, there doesn't seem to be a suitable device with sufficient voltage compliance currently active on the market.

Another disclaimer: There most certainly are current sink type DACs such as the AD5821 or AD5398, but they lack any anti-ringing or motion control related functionality, which means that this would have to be implemented in the host process (not necessarily optimal or even possible).

Okay, so to the question. Are you aware of a VCM driver (essentially a current sink DAC with some form of ramping / slev rate control functionality) that would fulfil following requirements:

• Supply voltage 5V tolerant
• 120mA maximum drive current
• I2C control (1.8V logic compatible)
• Built in damping / anti ringing functionality

The best solution I've found thus far can offer supply voltage tolerance up to 4.3V. This, however, is just barely enough and would also require a non-standard, separated power supply.

And from curiosity's point of view, Do you have any knowledge / educated guesses on why the more recent driver ICs seem to only comply with a meager 3.6V supply voltage. The discrepancy is interesting, considering there are quite a lot of VCMs with DC resistances falling in the 35...40ohm range on the market.

Thanks in advance!
Nick

[jmelson]

Computer disk drives have voice coil motors for head positioning, and they at least used to run off 12 V.  Of course, I suspect those coils have MUCH lower resistance than 35 Ohms.  Maybe the drivers used in CD/DVD drives might be what you are looking for.
Jon

[moffy]

Three possibilities but 100ma seems to be the common limit. If not compatible with 1.8V you can always use level shifters.
1. https://fscdn.rohm.com/en/products/databook/datasheet/ic/motor/mobile_module/bu64291gwz-e.pdf
2. https://www.ti.com/lit/ds/symlink/drv201a.pdf?ts=1711055114746&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FDRV201A
3. https://www.allegromicro.com/en/products/motor-drivers/bldc-drivers/a3907

[MarkT]

And then, to the topic:
While trying to update BOM for an existing design, I ran into an interesting deficiency in recently introduced dedicated VCM (voice coil motor) drivers.
The VCMs I'm using have a DC coil resistance of about 35 ohms and rated nominal current of 120mA. To achieve this, the minimum supply voltage would have to be abour 4.5V (also considering the driver voltage drop). However, the more recent dedicated VCM drivers are only rated up to about 3.6V, which means that they are physically unable to reach the full 120mA drive current with 35 ohm motors  . In other words, there doesn't seem to be a suitable device with sufficient voltage compliance currently active on the market.

What sort of performance are you looking for in the time-domain?  Do you have an inductance figure for the VCM too?  I wonder if it might be possible to use standard switched MOSFET H-bridge?  If not there are plenty of audio amplifier chips capable of driving such loads linearly, you just need to remove any DC-blocking capacitors that may be present - or perhaps a class D audio chip?  An opamp driving a current-buffer would also seem a reasonable approach for the linear approach.

I'd beware of assume 4.5V is enough BTW, for instance self-inductance and motional back-EMF might be significant.  Need to know more about the setup to figure this out.

[NpR]

Hi!

The driver would need to be unidirectional, since the voicecoil motor being driven is a coil-and-spring type. If I could affect the digital side of things, I could simply use a FET+freewheel diode totem pole driven by a PWM output of the ISP (with maybe a gate driver inbetween). Alas, this is not in the range of possibilities due to two reasons: The ISP doesn't have an easily available PWM output and it might not have enough processing overhead to handle anything but the simplest de-ringing functionality (which is typically offloaded to the VCM driver in most applications).

The VCM inherent inductance only really comes into play when changing the drive current rapidly (which is typically avoided due to ringing). Most of the VCM drivers are linear current sink type and thus operate in DC range most of the time.

Most audio amplifier chips that would fit the form-factor requirements are analog only, so an additional DAC would be needed. Please consider that most VCM drivers talked about are roughly the size of poppinseed rather than a cockroach, so there isn't too much room for additional circuitry, and also not much cost overhead for BOM inflation.

The original question wasn't so much on how to drive a VCM but rather to source suitable ICs that do so. There are some chips available through Asian manufacturers, but information on those is rather scarce. Also, the 102mA drivers suggested by Moffy don't provide enough drive current for the VCM we are using and are thus not applicable.

Here's a brief list of possibilities I've found thus far:
DW9714V (would allow supply up to just 4,3V. More detailed info hard to come by)
AT5510 (would allow 5,5V supply, but doesn't seem to have any anti-ringing functionality)
AD5821 / AD5398 (5,5V supply again but no anti-ringing control)
CXMD3291 (5V supply, other information hard to come by)

Others may exist, but are beyond my capabilities of searching online.

Any thoughts?
Nick