Here is what I am trying to accomplish.... I want to take 12 v dc and convert it to 300 volts dc. I only need it to be able to do 5 watts as I really only anticipate around 3 watts if I figured it out right. I need the power supply to be super quiet, as it will be running a 12ax7a tube as a class A amp. Maybe 6watts because I have two tubes in one unit.The problem is it is being used as a guitar preamp and I don't want the switching noise to be a problem at the output. So I suppose if it was above the range of amplifier bandwidth maybe it would be better. I have tried a couple designs but have not come to any conclusions. I have tried the 555 timer going to a 2n2222 where a mosfet gate was attached to the collector of the transistor. seems to run cooler that way. I achieved the 300 volts at one point, but with a 15k resistor as a load it dropped to 60 volts. I was using an old 50c5 speaker transformer as the power transformer. It might not handle the 6 watts but I think it will. The mosfet was a 75639s. I will heat sink them of coarse but for a quick test I did not as it is a pain and time consuming to do so as I experiment with different devices. All i am doing is seeing if they can reach the voltage and what the result is with a load. Now back to the design. Do you think if I rewound a transformer that had more dc resistance and more turns on the primary and still maintain the ratio of course on the secondary, would that help? I had been thinking maybe I should use two mosfets one for each half of the cycle. This is such a simple thing yet it is not performing how I wish. I am sure I will get one working, but trying to save time by calling on the minds to be. What is your thought on this? Go ahead I can take criticism. Thanks for advice. I see I blew already, guess that is what happens when you get sleep deprived. I see the mosfet is only good for 100 volt, hey it worked to 300 volt. I will use a rated part next time but I am looking for another design or one that works to my specs. I see I can use a basic switching type like the old 2n3055 transistors 60 hz diagram floating around I guess. and just swap to a different transistor and transformer to keep it small. Still I like ideas.
Last edited:
I have a nixie clock next to me which generates 200V at about 25mA or about 5 watts, from 9V. Using a MAX1771 chip and a 500V Infineon CoolMOS FET. Don't recall the part number. Operates as a straight boost converter - a flyback or tapped inductor boost might have a more sensible duty cycle or better efficiency, but hey, it works. Only issue is that the MAX1771 misbehaves a bit at input voltages above 12V.
There's plenty of designs for nixie clock power supplies out there, using 555 timers and such. Have a look around and copy one of those.
Add a second LC filter on the output to clean the supply up to meet your spec.
There's plenty of designs for nixie clock power supplies out there, using 555 timers and such. Have a look around and copy one of those.
Add a second LC filter on the output to clean the supply up to meet your spec.
yeah looking at those. but didn't know the current. Also looking at cfl lamp starters. I have another idea to try or two. Thinking of still using a 555 timer, heat sinking the mosfet this time and running the 555 timer into an opto coupler so I can control drive the drive transistor. Making sure my mosfet of choice turns on all the way and change the frequency and try again. I think I should be ok. I will show the diagram then. It might take me a day or two to get to it again, although I will retry tonight if I have the time. Come to think of it I have a diagram on the Internet of one I put together that works. Don't know why I didn't use that diagram? who knows, I guess just to ty something different. In my other diagram I used an op amp after the 555 to obtain full power needed to turn on mosfet. Might be the bottom line anyway, just isn't turned on all the way. I will reply back once I have it together.
Here you go:
http://cds.linear.com/docs/Application Note/an118fa.pdf
That appnote has schematics for 300V and 1000V converters that have 100 microvolts of output noise. As shown, they don't provide the output power you need. But they might give you some ideas.
But here's one that will give 330V @ 50 mA (16.5 Watts), with 8-15V input:
http://cds.linear.com/docs/Application Note/an39f.pdf
http://cds.linear.com/docs/Application Note/an118fa.pdf
That appnote has schematics for 300V and 1000V converters that have 100 microvolts of output noise. As shown, they don't provide the output power you need. But they might give you some ideas.
But here's one that will give 330V @ 50 mA (16.5 Watts), with 8-15V input:
http://cds.linear.com/docs/Application Note/an39f.pdf
Last edited:
No problem. Jim Williams, of linear.com, was one of THE greatest analog electronics gurus of all time. Incredibly, he and Bob Pease of national.com, one of the other greatest analog gurus of all time, died on the same day, earlier this year.
Linear.com is a _tremendous_ resource for information about DC-DC conversion. It's one of their flagship product lines. They have HUGE collections of fantastic application notes and design notes, and a series of "circuit collection" magazines, on line. They also produce the free LT-Spice software, and spice models for all of their stuff.
DC to DC converters
can probably be modified to increase output voltage by adjusting the secondary turns.
but I went the cheap and easy route. bought a chinese inverter and tore out the H bridge section. just used the DC-DC portion of the inverter.
can probably be modified to increase output voltage by adjusting the secondary turns.
but I went the cheap and easy route. bought a chinese inverter and tore out the H bridge section. just used the DC-DC portion of the inverter.
- Status
- Not open for further replies.