Does anyone have any good shields or circuits to accomplish the following:
INPUT:
0-5V PMW with s/w controlled frequency and duty cycle
(this would be the PMW analog output from an arduino)
DESIRED OUTPUT:
50V RMS (meaning around -70 to +70V peak) waveform
maintains the same frequency and duty cycle as input
sine would be ideal, but maintaining square wave would also work
able to source around 5W of power
POWER SOURCE for CIRCUIT
car battery: say 12-14.5V
So powering the arduino from 12V and writing the code to generate and control the PWM square wave is no problem. From there I need to shift the voltage from 0-5V to say -2.5 to +2.5V so that we have an alternating waveform. Then, I could use a transformer to step up the voltage around 28x to get -70V to +70V. Lastly, I would need to source the current through an H-bridge perhaps (?) to handle the of power.
Not sure. it seems like some shield out there should do some of the work for me, but shifting from +waveform to +/- waveform is problematic.
A simple transformer will do that for you, except you will have to have a circuit that can provide more than your required 5 watts? How will you measure the 5 watts?
Good catch on the RMS. The application generally uses an AC source and AC dimmer which chops the waveform. In order to automate, I was hoping to use a controlled duty-cycle, square wave instead. With this, let's assume we just want to stay with a square wave even though the manufacturer says a sine will work better. Since the manufacturer calls for 50V RMS, in square wave terms we're probably talking about -70V to +70V at the peaks.
This is for an electrochromaic film which requires power as a function of area. If we think of it as a light bulb, that's close enough (except that the film does need -V to +V and not 0 to +V). So something like an H-bridge or FET solution where the output matches the input but sources the current from the 12V battery would be best. Assuming my arduino s/w generates the correct waveform, and a Transformer steps it up, then I just need some type of buffer/amplifer/h-bridge to source the current.
Thank you for asking. I had asked the manufacturer for the spec sheet and what they sent me was almost useless. Now that you asked me, I did a random Google search and actually found something useful. I'll attach.
Interesting that the spec sheet lists 28 - 48V AC and the data they sent me states 28 - 60V AC. From other Googling, I understand the voltage range is pretty open.
But the stuff you have is definitely the low voltage version as opposed to the regular 110VAC version?
Is there any information online about the waveform from their controller as they state that a normal dimmer circuit will not work?
They expect users to use their wall-wart which is just 120VAC to 60VAC and a Home Depot dimmer. They have no advice about automating this outside of buying an automation controller from them for big $.
As the Home Depot wave-chopper will work, I'm assuming varying the duty-cycle on a square wave should work as well. When I emailed them, they said to only use a sine wave and not a square wave, but if you Google other manufacturers of what appears to be the same product, they say a square wave actually works better. I think these folks are just saying to use a sine to get customer to buy controllers from them.
Either you've misunderstood me, or I've misunderstood you. A square wave (at any duty cycle) that is -70V to +70V in peak amplitude will be 70V RMS amplitude.
Yes of course, I get that, but wasn't sure what that would mean in terms of what the load "sees" when using a square wave. Since you've made me think harder about it, I guess I should really be thinking about the area under the curve. Since a square wave has more energy delivered at each peak, we would need a lower amplitude to be similar to a sine. Aiming for a square wave amplitude closer to the sine RMS is probably a better bet. So for a square wave, maybe close to -50V to 50V full scale is probably more realistic, agree?
Exactly. I was trying to convey the topic in purely electrical terms to simplify things, but knowing the application would help I'm sure. So the film is driven by an AC source, and a typical residential AC dimmer can be used to vary the opacity from 0 to 100%. Since an AC dimmer merely chops the waveform, I was hoping by building a controller and varying the duty cycle that I could control the opacity similar to if we just used an AC residential dimmer (which of course would be manual).
It may be easier to transform the AC dow to 60 volts and then run that through an Arduino controlled dimmer circuit. Getting the power you need will be non-trivial with the approach you are considering.
Arduino output 0-5V PWM s/w controlled frequency and duty-cycle into...
PWM control board
This would be powered by car battery and would mirror the input signal from 0-5V to 0-12V with the current sourced through the FETs up to 400W which is more than enough.
I could then take the 0-12V and run it through a multi-tap transformer with a 10-1 turns ratio and a center tap that yields a 5-1 turns ratio. I would tap the 5-1 tap as the common for my load, the transformer common as the "-" for my load, and the 10-1 tap as the "+" for my load. My load should then see a square wave signal from -60 to +60V which is high but within the range of the spec. I'm thinking that will work, but I'll have to scope it.