See if you can help me with this. I'm building a cute little output transformerless headphone amp. The design calls for a 150V power supply.
Here's the deal. I have a 140V transformer going into a bridge rectifier, and a 220μF filter, this gives about 180VDC. I need to get it down to ~150. I've got zeners, a few power MOSFETs, and an assortment of regulator ICs. I'd like to have a regulated supply, specifically one I can adjust with a trimmer to dial in the voltage, but I'm not sure of the best way to go about it. I'm really terrible at HV power supply design, so I need pointed in the right direction before I blow up any more resistors.
Here's the deal. I have a 140V transformer going into a bridge rectifier, and a 220μF filter, this gives about 180VDC. I need to get it down to ~150. I've got zeners, a few power MOSFETs, and an assortment of regulator ICs. I'd like to have a regulated supply, specifically one I can adjust with a trimmer to dial in the voltage, but I'm not sure of the best way to go about it. I'm really terrible at HV power supply design, so I need pointed in the right direction before I blow up any more resistors.
You can use one of the circuits I gave in this message and the subsequent ones:
http://www.diyaudio.com/forums/powe...gh-voltage-regulator-circuit.html#post2417349
You simply have to adjust one resistor value for your voltage, and you will end up with an excellent regulator, safer than any integrated one without compromising on performance.
http://www.diyaudio.com/forums/powe...gh-voltage-regulator-circuit.html#post2417349
You simply have to adjust one resistor value for your voltage, and you will end up with an excellent regulator, safer than any integrated one without compromising on performance.
While I'm here, is there any way to make this circuit:
http://hem.passagen.se/ebcpecz/Tube/6AS7 Headphone/images/pwr_schematics.gif
work with a transformer without a center tap? I threw one together on a tagboard with a diode bridge in place of D2/D5/C6/C5/R9/R11 and literally vaporized R14. I'm certain that I did something positively hilarious, but as I said I suck at power supplies. (I do promise to spend the next few months reading up on them)
http://hem.passagen.se/ebcpecz/Tube/6AS7 Headphone/images/pwr_schematics.gif
work with a transformer without a center tap? I threw one together on a tagboard with a diode bridge in place of D2/D5/C6/C5/R9/R11 and literally vaporized R14. I'm certain that I did something positively hilarious, but as I said I suck at power supplies. (I do promise to spend the next few months reading up on them)
The link seems to be missing an _ to work properly.
Maybe this helps to answer your question: Elliott Sound Products - Linear Power Supply Design
Maybe this helps to answer your question: Elliott Sound Products - Linear Power Supply Design
No, you didn't. But the author of the schematic did: R14 is most probably 560K.
Here's what I've gotten so far. Starting from the design in the LM317 datasheet I worked this out from parts I had laying around. I'm not entirely sure about it, so I thought I'd run it by you guys first. I had to substitute a lot of parts, I have a bunch of Darlington transistors but couldn't find them so I used an N-channel MOSFET (IRF840), which I'm not sure I did right. I'm also unsure about the value of R6, the voltage adjustment resistor. 4.7K produces 150V but putting such a low resistance across the output makes me nervous, it seems like it would pass too much current.
I don't stock many power resistors, so in the testing stage I used a 0.25W metal film just to see what would happen (smoke, is what happened). Needless to say that did little to allay my fears.
So what do you guys think? Any suggestion is welcome.
I don't stock many power resistors, so in the testing stage I used a 0.25W metal film just to see what would happen (smoke, is what happened). Needless to say that did little to allay my fears.
So what do you guys think? Any suggestion is welcome.
The amp itself draws about 50mA, using a 6DJ8 and 6AS7GA, which (feel free to correct me) should be about 8W. The AC supply here is extremely noisy (air conditioners, refrigerators, computers and tons of fluorescent lamps) but it doesn't vary much, staying within a volt or so of 125V RMS. Even short transients are pretty rare. The transformer is an EI core type (better noise rejection than toroids) and can put out 140V (or 70V to the center tap) with a current rating of 250mA/30VA, which gives a rectified and smoothed DC voltage of around 188V. I'm not terribly worried about the filtering, I'm using some big chunky Cornell-Dubilier electrolytics with a higher than necessary ripple current. U1 and IC1 are both TO-220 and have aluminum heatsinks.
My primary concern is the amount of current consumed by the resistor network in the voltage regulator circuit, and whether I left out something crucial in that circuit that would either shorten the life of the components or cause unreliable performance. The real point I worry about is R6. There's 143V across that resistor, so 143/4700 = .0304A * 143 = 4.35 watts, which seems like a lot of waste heat. Is there a better way to adjust this thing or should I go ahead and keep the 4.7K power resistor?
This isn't a critical project, so it doesn't have to be perfect, just reliable and safe. I'm just building this from some spare tubes I had laying around to kill some time.
My primary concern is the amount of current consumed by the resistor network in the voltage regulator circuit, and whether I left out something crucial in that circuit that would either shorten the life of the components or cause unreliable performance. The real point I worry about is R6. There's 143V across that resistor, so 143/4700 = .0304A * 143 = 4.35 watts, which seems like a lot of waste heat. Is there a better way to adjust this thing or should I go ahead and keep the 4.7K power resistor?
This isn't a critical project, so it doesn't have to be perfect, just reliable and safe. I'm just building this from some spare tubes I had laying around to kill some time.
hmmmn. I don't so much like your doubler design, but that could just be me. I would have bridge rectified the entire winding and floated the centre tap
THe article that helped me understand the Maida was this one from SY.
THe article that helped me understand the Maida was this one from SY.
New revision:
This one gives 153 steady volts. Nothing exploded, so I guess I'm on the right track.
An externally hosted image should be here but it was not working when we last tested it.
This one gives 153 steady volts. Nothing exploded, so I guess I'm on the right track.
Oh, you're too kind. I don't see any reason to resort to insults. I was the first to admit that I know absolutely nothing about power supplies, I've been building electronics for years but this is the very first time I've tried to design something from scratch. My degree is in computer science, thinking linear is not something I'm used to.
Last edited:
The comment was not aimed at you, but rather at the original inspiration of the circuit.
The guy at Nat Semi is obviously trying to sell as many LM317's as possible.
It works in theory, but it is a high wire exercise for the LM317.
There are lethal voltages everywhere, and as soon as something goes slightly wrong, it ends up in smoke.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.