Quiet EL wire inverter, based on a boost converter and a H bridge driver IC. Provides 360Vpp drive at 1.2kHz, up to 3M of typical wire. Device plugs into a USB port or wall plug.
The circuit has a basic boost converter controlled by a PIC10F204, that has a comparator and a voltage reference. The EL is driven by a microchip HV809, set to run at about 1.2kHz.
This is the third revision of the design, the first was based on a HV859/HV860. This basically functioned well, but the light output was way too low due to the lower voltage boost converter built into them. The second revision was based on a discrete device boost converter, (a oscillator, comparator, a AND gate, and a HV809). The second version had more light output, but still the boost converter was not able to reach 180V, got stuck at about 150V. to improve that this latest version uses a PIC to control the boost converter, this allowed a easily variable frequency/duty cycle. I also changed the switch and inductor.
Files
SilentELWireInterver_Datasheet.pdf
Simple Datasheet with Wiring Diagram
Adobe Portable Document Format -
959.62 kB -
09/30/2019 at 03:40
These two ghosts are powered by this circuit. They have about 28-33nF of cable in each. They ran about 290-300Vpp, instead of the full 360Vpp. I lowered frequency to 1kHz to help voltage.
I built 3 PCBs using a new PCB layout that incorporates the bypass cap on the uProc. All three PCBs work with no issues. I put these 3 units up for sale on tindie for $25.
I built 5 of these PCBs, using a 100uH inductor. 3/5 PCBs flickered. I found that the uProcessor was being reset intermittently due to the fast transients in the boost converter. The cross talk made it to the uProc's power supply line and caused it to reset. I fixed it by adding a bypass capacitor on the backside of the programmer pin head array. I also turned down the H-Bridge frequency to reduce the power loss. It now works with no flickering and no hot devices. I have had on a moon in my sons room for a month now with no issues.
I have received a revision to the latest design, the FET was getting a bit hot when trying to drive a meter of EL wire. So I added some metal to the layout to try to reduce the temp on it. The temperature of the FET was too hot to keep my finger on it for more than 30 seconds. I have built one it does work, and I am trying to optimize the Boost converter signal period on duty cycle to get the most power output while keeping the FETs temperature under control.
hi im trying to message you privately but im having a hard time... is there any way i can get your email for some questions i have on this project? what type of mosfet did you use and how did you turn down the frequency on the h bridge? your help would greatly be appreciated
The could be a good use for it. I would assume your thinking battery powered, I have not tested it with lower voltages, say 3V. The max wire length would decrease with lower input voltages.
Love the project mate. Can you explain the code abit? Thanks bro.