I've got a plan for a large electroluminescent display. EL panels are typically driven with a sine wave at around 1kHz, and I want to use mains power for this. What's the best way to convert mains power from 60Hz to 1kHz? I don't need a whole lot of current - 500mA RMS should be plenty.

I've been thinking converting to DC, then back to AC again would be the simplest way to do this, but I'm concerned about the efficiency of such a system; is there a better way? I also don't know how I'd produce a clean sine wave (EL is quite picky) at the output.

EDIT: I'm in the US, so this is 120V at 60Hz.

  • \$\begingroup\$ EL panels also require a higher voltage than 120/220V if I recall correctly. What voltage do your panels require? \$\endgroup\$ – JYelton Jul 31 '13 at 18:26
  • 1
    \$\begingroup\$ Everything I've read says about 100-120V, but if that's not the case then that's an additional consideration. I don't exactly have real panels - I'm planning on building my own by spraying phosphor and dielectric layers on a conductor, and using a conductive mesh on top. There are a lot of materials challenges to cover, but I wanted to tackle the electronics in the meantime. So I'm assuming my design will behave like conventional panels, but I don't know. \$\endgroup\$ – Trebawa Jul 31 '13 at 18:44
  • 2
    \$\begingroup\$ EL panels like about 100 V at higher frequency. I used to have an EL panel hooked to my stereo through a backwards 12V mains transformer. Different frequencies make different colors on the display! And it was a somewhat effective piezo tweeter. \$\endgroup\$ – markrages Jul 31 '13 at 19:23
  • \$\begingroup\$ What power level do you need for your display? \$\endgroup\$ – markrages Jul 31 '13 at 19:23
  • \$\begingroup\$ The higher the frequency you go (and I'm thinking 20kHz or more), the easier it is to design a "sinewave pure" step-up resonant converter for the EL display. \$\endgroup\$ – Andy aka Jul 31 '13 at 20:17

Just an idea and not the conventional approach I'll admit. How about using an off the shelf (cheap) audio amplifier driving a step up transformer (reversed audio output type 1:10 ratio - 10V in gives 100V out) instead of the speaker and a 1kHz sine wave generator as input.

enter image description here

The output would be isolated from the mains. The 'volume' control could be used to give a 'variable voltage' output and the sine wave generator frequency could be varied if required.

  • \$\begingroup\$ There are some reasonably cheap 100W RMS amplifiers, but their power curves mostly seem to depend highly on the load impedance (generally 4 or 8 Ohms for speakers). Will it be easy to find/construct a step up transformer that presents a similar impedance to the amp when the secondary is loaded with OP's panel? \$\endgroup\$ – us2012 Aug 1 '13 at 1:48
  • 2
    \$\begingroup\$ @us2012 It should be possible. My thinking would be that audio output transformers match to a speaker (4/8R) so using it in reverse should give the same loading on the amp. You could try an ordinary 'power' transformer, ferrite core type with a 10:1 ratio. (cheaper option) \$\endgroup\$ – JIm Dearden Aug 1 '13 at 7:25
  • \$\begingroup\$ Thanks! Is this something I can do with an amp I build myself? I've recently come into a large quantity of 2N3055 power transistors (and their PNP counterpart), so if I could use them for this that would be perfect. \$\endgroup\$ – Trebawa Aug 1 '13 at 15:29
  • \$\begingroup\$ @Trebawa Its certainly do-able. \$\endgroup\$ – JIm Dearden Aug 1 '13 at 16:55
  • 1
    \$\begingroup\$ I've done this with an old 25W Pioneer boat-anchor receiver. No instability was noted. I drove a small (<5 sq. in) panel through a backwards 12V mains transformer, using sine waves and normal music program. The load is mostly capacitive, so most amplifiers will be happier with some parallel resistance added. But it worked for me without any resistive load. \$\endgroup\$ – markrages Aug 1 '13 at 19:58

I would go to DC, then use a DAC with a lot of bits. This should give you a stepped sine wave that shouldn't be too hard to filter to a pure sine wave. And I would expect all this to be fairly efficient. And for safety I would use at least one transformer so the panels were isolated from the mains.

  • 4
    \$\begingroup\$ This doesn't really address the high voltage production side of things and using a DAC to generate a sinewave is a litle bit of overkill plus using a "lot of bits" is even more overkill as your filter (that you mentioned) could quite easily convert it to being decent waveshape. Also you say "go to DC" and I believe the OP meant convert 220Vac directly to DC - I'm not sure you understood that. \$\endgroup\$ – Andy aka Jul 31 '13 at 20:16
  • \$\begingroup\$ A 50% square wave has (generally) well-defined harmonics that can be filtered out. Anything beyond a 1-bit DAC is unnecessary. \$\endgroup\$ – Ignacio Vazquez-Abrams Jul 31 '13 at 22:30

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.