I am trying to develop a driver to control a MEMS mirror by Mirrorcle. The specific component is A8L2.2-5000AU-TINY48.4. The data sheet only includes VdifferenceMax and and frequency information, but no where on it can I find the specifications for the required current and voltage needed to operate the device. Would I have to calculate that on my own? I've looked all over the web for this info, but got nothing.


They do give a bias voltage of 90v and the difference voltage implying that the two electrodes in each axis are relative to that bias voltage. So you will need two single-ended drivers per axis.

Assuming that is true your driver will need to provide about 0-180V for each axis.

The current required for electrostatic actuators is essentially zero and they act as a pure capacitive load. I would expect the actuator capacitance to be very low and dominated by the package and connections to the driver.

I designed a system with similar mirrors with >750 in one unit requiring ~3,000 drivers. In that case, the interconnect capacitance was hundreds of picofarads. But I only used 15uA in the output stage of each driver to keep total power consumption low.

  • \$\begingroup\$ Thanks! This was very helpful - I do have a few follow up questions. Having a bias voltage of 90V means I would most likely have to utilize a wall outlet or an equivalent high voltage source. Correct me if I am wrong, but that basically means it would be near impossible to miniaturize all this including the battery into a tiny portable enclosure? \$\endgroup\$ – joethemow May 13 '20 at 6:48
  • \$\begingroup\$ @joethemow - the power is very, very low, milliwatts. I don't see a problem in being battery powered. I had over 3000 drivers requiring a total of ~30W including the control processors. You will need a converter to create ~180V from the battery input. I wouldn't do it with a bias voltage, I would have 4 separate single ended drivers capable of doing 0-180V and control them differentially. \$\endgroup\$ – Kevin White May 13 '20 at 21:04
  • \$\begingroup\$ Gotcha. I don't think I follow when you say "I wouldn't do it with bias voltage". It is my understanding that bias voltage is the necessary voltage needed to power the actual device? Also if the mirror can only move in 1 direction why not just use 1 driver? I was thinking of using a motor driver to control the x and y axis. \$\endgroup\$ – joethemow May 14 '20 at 18:25
  • \$\begingroup\$ @joethemow - I don't believe so. It is just the mean voltage of the two electrodes for each axis. To meet their data sheet characteristics you need to keep the mean voltage at 90V. The MEMS won't know how you are driving it provide you meet that criteria. There is nothing active in it - just a couple of electrostatic actuators on each axis. I'm impressed that they angle vs voltage curve is so linear; most MEMS would have a square law which is a pain to drive with adequate accuracy. \$\endgroup\$ – Kevin White May 14 '20 at 19:50
  • \$\begingroup\$ @joethemow - most motor drivers would be PWM. That is not at all suitable. Usually they need two actuators per axis because in general electrostatic actuators can only pull, they can't push. so you either pull down on one side or the other and when they both have 90V the forces are balanced. \$\endgroup\$ – Kevin White May 14 '20 at 19:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.