I was playing with the values of C1 and C2 to see what effect it would have on the simulation too. C2 seems to add a bit of feedback to the system in addition to the L3 feedback. I remember studying poles and zeros in circuit and my gut tells me that that might be able to help explain it. But I learned that stuff 50 years ago, and it is long gone. I tried refreshing my memory from my textbooks, but it is beyond going from the textbook to a practical application for me anyway.

Ford would not spend a penny if it wasn't necessary, so I am sure it is there for a reason. I know that copper resistance is strongly influenced by temperature which can vary a lot in a car. Perhaps C1 and/or C2 is there to help with that?

 

My knowledge of EL panels = ~0, but I'd be surprised if the operating frequency were critical?

 

My knowledge of EL panels = ~0, but I'd be surprised if the operating frequency were critical?

EL panel perceived brightness is proportional to frequency. It's basically acts like a capacitor that emits light while charging. You can charge it faster with a higher voltage or you can charge it more often per second. Either result will be brighter. Unfortunately the life of the panel is degraded by either a higher voltage or higher frequency. Picking both voltage and frequency while balancing brightness and lifetime is definitely a design challenge.

 

I have been scouring the internet for a source for an off-the-shelf transformer that I could either use as a drop-in replacement or in a similar circuit for a new design without any luck. The closest ones I can find are designed for CCFL applications and have no where near the inductance needed to get it down to the 500Hz area. If anyone knows of an off-the-shelf transformer that might work I would sure apprecaite pointing me in that direction.

As wayneh pointed out higher frequencies are not good for the EL panels longevity and these panels are made out of unobtanium. I would hate to be responsible for burning out the illumination someone's rare Cobra cluster. And the cluster is tied into the car's PATs system so the car won't start without the original cluster. What a cluster...

If no COTS transformers are available does anyone know of a custom transformer supplier who might be able to make about 100 of these transformers for me without costing an arm and a leg?

 

I used a wall wart run 'backwards'. That is, I applied my signal to what was the low voltage secondary and used the old primary to get ~10X voltage boost. Not ideal by any stretch, but it works. I thought the frequency might be too high, and I don't know what the limit is, but 250 Hz was no problem.

 

I was looking through my E-waste pile for something that might work, but had nothing close.

I plan on helping people to keep their cluster illumination working for as long as possible, so I will need something that will get me to about 300V P-P at about 500Hz. I need a turns ratio somewhere in the 25-30 range. It also need to fit inside of the instrument cluster so size is also an issue. If I am to use a blocking oscillator design, I also need a third winding for feedback with a high enough turns ratio to drive the transistor.

Also, I will need 100-200 units if I am going to be able to offer a repair service or a replacement module for Terminator/Lightning owners of that vintage, so I will need more than just one or two units. Thanks for the thought though.