Powersupply questions

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Mendelt

Well-known member
Joined
Nov 16, 2004
Messages
221
Location
the Netherlands
I need to build a powersupply for some PM1000 line-amps (yeah i'm still messing around with pieces of my sawed up pm1000). They need 44V.
I was wondering if it was a good idea to use 7824's for this with a 20V zener to ground to regulate the voltage to 44V instead of 24V.
What is the max input/output differential voltage these regulators will take? I have some 2x25V toroids so if I use those I will have 70 volts max on the inputs of the regulators. This seems a bit high to me. But the nearest lower voltage toroids i can get are 2x18V and this is probably a bit low.

Another thing i can't find is what is the maximum power disipation of a TO220 component without a heatsink? When do i need to add one?

Thanks
 
Data sheets are all over the web.
Of the top of my head, me thinks 40 volts max input, so if zener on pin 2, 20 plus 40 equals 60 volts max input, and you should allow a little safety in there.

I would find another way around this.

Picking the right pwr iron is the most important part of building a supply. Once you blow it, there is no real cool way to keep things cool, short of opening up the x-former and taking off turns.
 
2x18AC is ok for 44vdc out . You will have after diodes 36x1.41=50.76VDC. Is just enough and safe for use LM783c to regulate it to 44vdc

Cheers
 
As far as heatsinking a TO-220, I wouldn't have more than a watt tops dissipated without some help. If you can burn your finger it is hotter than I would like for reliability. If a drop of water promptly sizzles and evaporates then it is in the vicinity of 100 C and the chip temp inside is significantly higher. Look at the spec for thermal resistance junction/chip-to-case and junction/chip-to-ambient. Calculate the dissipation as the current in times the voltage drop across---this will include the quiescent current required for the regulator to work along with the load current.
 
Thanks for all the help!

[quote author="CJ"]How much current do you need?[/quote]

For this application i need about 100mA. But i think i need more later because i was planning on using the same design for 4 racked up pm1000 channels. I heard they take 200mA each so that's almost one Amp.

[quote author="CJ"]Data sheets are all over the web.[/quote]
I know. But the datasheet i found for the 78xx only had minimum input voltages specified. I'll try to find some datasheets from different manufacturers.

I'll try the 2x18v version in the simulator. Dunno if the voltage to the regulators will stay high enough if i pull 0.8A from it.

[quote author="CJ"]I would find another way around this. [/quote]
That's probably a good idea. I'm a bit stubborn though so i'll just keep trying to get this to work for a while :grin: The worst that can happen is that i fry some stuff and maybe learn something.
 
[quote author="FotisGR"]2x18AC is ok for 44vdc out . You will have after diodes 36x1.41=50.76VDC. Is just enough and safe for use LM783c to regulate it to 44vdc
[/quote]

I'll try that. I'm just a bit worried that when i take too much current from this the input voltage to the regulators will drop too much.
 
I'm just a bit worried that when i take too much current from this the input voltage to the regulators will drop too much
Use plenty reservoir capacitance to avoid input ripple , otherwise use higher ac voltage input. If you need most curent from 783c or similar regulator, you can easily obtain it by driving one or two power transistors going to 200-300W power. If you need help about how to do this, i can help you.


Cheers
 
[quote author="FotisGR"]2x18AC is ok for 44vdc out . You will have after diodes 36x1.41=50.76VDC. Is just enough and safe for use LM783c to regulate it to 44vdc[/quote]

I think that's cutting it a little close. First off, you'll lose a few volts due to diode drops; second, there's no margin for the AC from the wall running low. Even ignoring the diode drops, line voltage that's 10% low will give you about 45V unregulated, not enough to keep the regulator regulating.

2x20VAC might work better, but those transformers seem to be less available than in previous years. 2x24VAC would be fine; you can drop much of the excess voltage by making the filter two-stage, with a hefty dropping resistor. Better that should dissipate the heat than the regulator.

Peace,
Paul
 
Ok.. i give up.

I found the regulator/zener idea appealing but it seems like it won't work for 44V.
It seems like i can get this to work when i use two filter stages and a 2x25V toroid. but when i disconnect the load the filter caps charge to 70V and the regulator will probably start to smoke. I can try to find a 2x22V toroid. but it's probably easier to use a different regulator.

The other two options i know of are a LM317 with the same zener trick to get it to work above 37V or a TL783 with a power transistor as Fotis suggested. I found some LM317's that can regulate up to 1,5A so i guess i'll try that one first.

Thanks again for all your help.
 
317 doesn't need a zener trick. You might want to put a big zener across its input to output in the event of an momentary output short, or if your unreg supply comes up much faster than the output loads allow the output to come up, but the beauty of the part is that it floats its reference and error amp with the output voltage divider. I've done ~300V regulators with them, although I eventually found the noise uncomfortably high and went to a discrete design.

For the amount of current you are contemplating though, I would split the chores up among at least a couple of well-heatsunk parts.
 
[quote author="bcarso"]For the amount of current you are contemplating though, I would split the chores up among at least a couple of well-heatsunk parts.[/quote]

That might be a good idea. I read most of the other threads on powersupplies here and it seems lots of people here prefer the TL783 over the LM317. If i just put two powersupplies in my 4-preamp boxes I can use them without any problems.
 
[quote author="FotisGR"]If you need most curent from 783c or similar regulator, you can easily obtain it by driving one or two power transistors going to 200-300W power. If you need help about how to do this, i can help you.
Cheers[/quote]

I found some schematics on how to do this in the 783 datasheet. Thanks for pointing it out.
 
What about powering a PM1000 module with Phantom Power? Would the suggested 200ma current requirement be too much? I just have one input module to power.

I don't know if this would be useful to people with more traditional setups, but this would definitely work for me. Much easier than building a whole seperate power supply, though, perhaps not as flexible.
 
You mean powering it from an external phantom power supply?

I don't think this would work. If I remember correctly most phantom psu's have 6k8 resistors connected in series to limit the current you can get from them so there's no way you can pull 0.2A from it to power a preamp.
 
I used to think the same, but then I saw an example of someone powering a preamp from phantom. I don't have the schemo for the specific example.

I was going to build the same power supply that I used for my phantom power box, and that steered me down this avenue. Why build a whole new identical supply, if I could power this with phantom?

By the way, the supply ciruit is here:
http://tangible-technology.com/power/Phantom_frying.html
 
Back
Top