Piezoelectric focuser on eBay

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: Pau, rjlittlefield, ChrisR, Chris S.

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

Ray,

Yes the Piezo elements require almost no current, just a charge to satisfy the capacitance. So the only current is just dq/dt, and we don't need speed, so dt is large, thus current is small. However this does require a large voltage for reasonable displacement so what I've designed basically is a high voltage digitally controlled source with low current capability. This utilizes a DC to DC converter to create the HV from the normal motor supply (~12V) and a 12 bit DAC to control a HV linear regulator with a 0 to 200 volt output.

I have the piezo stage now (just arrived) and it has 4 strain gauges arranged in a way to utilize a precision analog bridge as position indication, as I had figured. I've already got a position feedback design in process (parts were ordered this morning) for this as I suspected it would be arranged this way, so got lucky here :D It is capable of fully differential operation for a bridge configuration and with 16 bit resolution.

I don't think the regulator will support enough current for the voice coil, certainly the DC to DC converter won't. The feedback components are all designed for a 0~200 V output, so these will need changing.

Another design would be required for the voice coil with a lower voltage range and much higher current capability. The components I've ordered won't support the higher currents, so new components will need to be ordered (I'm trying to keep the cost down by using readily available components that are inexpensive).

Maybe can we look into this at a later date, but now I've got enough on my plate and a PCB to layout :roll:

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Can you make the 12 bit DAC available as a separate output? If so, then an external DC buffer stage would give the current sourcing capability needed for the voice coil.

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

Ray,

Sure I can make it available. Keep reminding me so I don't forget though :roll:

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

The initial circuit design is completed as is the PCB.

This directly interfaces with a Raspberry Pi via. the 40 pin standard GPIO connnector with a cable (couldn't get to fit the RPi case).

If things work as planned here' the features.

1) Interface with RPi
2) Base circuits power from motor supply (~12-24V)
2) External Power supply input from ~20 to ~250VDC
3) Has filtered output for external DC to DC Converter
4) Has built in DC to DC Converter on PCB which creates ~100~200VDC from motor supply
5) Multiple safety LED indicating HV Input, HV Output and commanding HV Output
6) High Current LED Indicator
7) 12 Bit DAC Controlled output from 0 to ~200VDC
8) Negative feedback désign to drive high capacitive loads (Piezo devices) and stable with such (output could handle ~100ma or more at low input supply voltages).
9) Single external potentiometer to manually control output for setup and calibration
10) DAC Output available (Ray), 0-5V with 12 but resolution
11) 16 Bit Differential Input with Low Pass Filter for Strain Gauge in Bridge configuration
12) Mostly uses readily available parts and only 1 surface mount part (DC to DC Controller)
13) Modest overall cost, things can be left off like the Strain Gauge ADC and DC to DC Converter to save $

Anyway, here's what the initial V1 PCB looks like.


Image

I'm getting ready to order additional components for the entire setup including on boards DC to DC Converter, let me know ASAP if you want the parts and PCB.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Thanks for including the DAC output Mike!

viktor j nilsson
Posts: 253
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Post by viktor j nilsson »

Looking great!

I'm interested in both the parts and PCB.

I have a question regarding the the manual control pot. Will this be useful only for calibration and setup? I'm thinking that in many cases, it would be great to have real-time manual control of focus. Would this pot be useful for this?

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

viktor j nilsson wrote:Looking great!

I'm interested in both the parts and PCB.

I have a question regarding the the manual control pot. Will this be useful only for calibration and setup? I'm thinking that in many cases, it would be great to have real-time manual control of focus. Would this pot be useful for this?
Yes, the external pot will allow positioning by way for controlling the output. This should allow you to control the output from say 0 to 200V, or whatever high voltage is determined as the max voltage.

Get a 10 turn 10K pot with a Dial type knob. This will allow you to check the piezo linearity vs. voltage and compare with the DAC generated linearity. There's a 4 pin header (CTL-IN) that has 5V, DAC Out, CTL-Input and Ground. Just connect the pot to Ground (4) and +5V (1) and wiper to CTL-Input (3).

Best,
Last edited by mawyatt on Sun Jul 14, 2019 8:52 pm, edited 1 time in total.
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

I've done some simulations and happy with the results so far, so getting ready to release the PCB soon for the 1st generation piezoelectric driver. Need inputs from those interested as I'm only going to fabricate a few boards, so when these are gone they are gone!!

The design concept is to deliver current from a high voltage source to the ceramic piezo elements quickly without overshoot, but have a slower delay when moving backwards. I've limited the command position voltage from the DAC with a LPF with a time constant of 1ms, this limits the slew rate on the command voltage for the linear control system amplifier/regulator response and thus the required transient currents in the output transistor. The design is for driving a heavy capacitive loads like seen with these piezo elements, conventional amplifiers don't like heavy capacitive loads, connect one up to these piezo elements and you'll likely have a power oscialltor and fracture the elements. :shock:

This controller is designed to control these elements, and architected as such. I don't have exact models for the critical parts, so used similar devices based upon engineering judgment.

Here's a plot of the simulated controller response to a (Blue) 8 volt to 80 volt change, with a gain of 40, this is an input command (Red) of 0.2 to 2 volts. 5 volt input creates 200 volt output, thus 2 volts commands 80 volts output. Note the output device (Q2) peak collector current (Grey) limited to 700ma and transient power dissipation (Green) to under 50 watts within the ~1.5ms time frame. Also note the output voltage (Blue @ 79.99468V), current (Grey) and Power (Green) at 2.5ms!

Best,

Image
Image

Updated circuit for symmetrical fast output capability, or fast rise, slow decay as shown above. Green is input command, Blue (79.9926V @ 4.4ms, 4.00V @ 53.4ms) is output, Pink is PNP current (762ma peak) and Grey is NPN current (337ma peak).

Image
Last edited by mawyatt on Mon Jul 15, 2019 10:27 am, edited 1 time in total.
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

dmillard
Posts: 595
Joined: Tue Oct 24, 2006 7:37 pm
Location: Austin, Texas

Post by dmillard »

mawyatt wrote: I'm getting ready to order additional components for the entire setup including on boards DC to DC Converter, let me know ASAP if you want the parts and PCB. Best,
I'm very interested in the parts and the PCB. Thank you for investing your energy and insight into this project.

Smokedaddy
Posts: 1710
Joined: Sat Oct 07, 2006 10:16 am
Location: Phoenix, Arizona
Contact:

Post by Smokedaddy »

... count me in.

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

Getting close to releasing the PCB. The simulations for the Piezo Controller with heavy (10uF) capacitive load looks beautiful and is configured to allow a fast rise and decay, or a fast rise and slow decay system, or slow rise and decay. The slew rates are controllable by part selection. So the system should be easily configurable for different uses and profiles.

The initial voltage scale factor is now 0 to 4 volts produces 0-160 volts with 12 bit resolution. I'm using a precision reference design at 4.095V for the DAC, so each bit is 1.000mv or 40.00mv at the output.

PM me for more details if you want.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Assuming the voltage range and full-scale deflection are aligned, with 100um full scale and 4096 steps, you'll get ~24.4nm steps. Should be nice and smooth.

pbraub
Posts: 76
Joined: Fri Feb 02, 2018 1:23 pm

Post by pbraub »

I also would be interested in the PCB (2 if possible). I would like to adapt one to micro-manager via their arduino device adapter.

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

I will be releasing the PCB board today, so PM me with your requirements and if you want the parts and I'll give you the costs. Here's some the Version 3 preliminary specs.

PCB size 90mm by 80mm

Controller Input voltage is standard 12V connector.
Controller Input voltage nominal +12VDC @ 2A max, range TBD
+12VDC Filtered Output for external DC to DC Converter, 2 pin XH connector
Controller External High Voltage Input (~200VDC) is XH 5 pin connector with remote shutdown
High Voltage Piezo Output is 0 to +163.8VDC, 12 bit resolution, connector is XH 3 pin
High Voltage Piezo Output capable of driving 10uF Piezo capacitance, max value TBD
Max Piezo Output Range set pot from 60~187VDC
Internal Ref of 4.095VDC
External manual piezo control by way of external precision lab multi-turn 10K potentiometer
External header pins for Vref (4.095VDC @ 1ma max current), DAC Output (0-4.095VDC @ 1K Source Impedance), Piezo Controller Analog Input (0-5VDC for 0-200VDC Output, has LPF with 1ms TC)
Strain Gauge Input, 6 pin XH (Vref, In1 +-, In2+-, Gnd), also on header pins
Differential 16 Bit ADC with LPF & PGA
Header pin for ADC and DAC address
LED Indicators for +12V, RPi +5 Power (RPI connected), HV IN Present, HV Output (2 LEDS), High Current

12V to 200VDC Low Noise Step Up converter, set pot output range from 91~224VDC

Raspberry Pi Interface
RPi Interface connector is standard RPi 40 pin type
Standard RPi 40 wire Ribbon Cable to RPi

Two surface mount components

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2479
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater

Post by mawyatt »

The Printed Circuit Boards are on order. This is Version 4 of the Design. Here's what the Layout looks like.

Image

As you can see this is a pretty busy board, includes Raspberry Pi interface, a 12 to 200 volt DC to DC Step Up converter, high voltage amplifier/regulator capable of driving high capacitive piezoelectric loads, regulator, voltage reference, 12 Bit DAC, 16 Bit ADC for strain gauge position sensing and bidirectional drive and sensing capability (this needs to be verified, and may need piezo element modification, the piezo element I have works and this is what the design is based upon).

BTW I've run the piezo element forward and reversed and used the strain gauge as feedback. With a +- 30V drive the strain gauge shows a +- 1 part in 1000 change (0.1%), so this requires some serious precision to resolve accurately which has been incorporated into the PCB.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

Post Reply Previous topicNext topic