| AudioGeek |
I need to build a voltage regulator to go from 40v to 30v at 40mA or so...
should i just use a GOOD 3 terminal reg? (which ones)
or
build one using discrete parts?
this will be supplying a mosfet p-channel source follower.
Thanks! |
|
|
| pat allen |
Most of the 3 lead regulator that i know cant take more than 35 volts at theirs inputs, so you might build your own with zener/transistor to lower it down to 34v and then instal a lm350?
even that is beyond the limits of the lm. |
|
|
| AudioGeek |
looking at the LT317 might work.
says that the max voltage DIFFERENTIAL is 30v... mine is only 10v...
question is... should I place this right after my rectifiers or, add say a 1000uf cap on the input of the recitifer?
http://www.linear.com/pc/downloadDo...,C3,P2039,D3784
is the link to the datasheet. |
|
|
| pat allen |
I see what you mean, i have a disastrous experience with lm's and even if they said that it is the differential voltage that should not exceed 35 volts, my experience tells me that is it not, i have build a prototype that used a plain lm 7812, the input voltage was 38 volts, exceeding the 35 volts limit, but not the differential (38-12=26v).
The regulator simply refused to turn on, giving something like only 1,25 volts randomly. SInce there was very few current draw, i have inserted a 8.2 volts, 5w zener diode in line at its input, so the input was getting 29.8 volts. It worked like a champ.
You MUST instal it after the rectifier, and dont forget the oh so habitual 0,1 uf caps in/out to avoid any high frequencies oscillation of the regulator, if you find a proper one.
Since your current needs is very low, you may just put a 1k resistor and a zener diode to the ground...the resistor will have to dissipe 10 volts @1k=0,1w, so a 1/4 w resistor should be fine, giving exactly a limit of 40ma, while a 30 volts zener, does that exist?
1n4751 looks fine to me (1w) or 1n5363(5w)
hth |
|
|
| peranders |
A plain LM317 will work as long as the voltage across the regulator is less then 35 Volts (in your case it's 10 Volts) and other specifications are alright. One think to remember though is that this regulator is rather thirsty due the feedback resistors.
If you are lazy you can use my Excel spreadsheet. |
|
|
| pat allen |
Hi peranders, about the lm317, isnt that playing with fire, since the time it takes to turn on, there will be a time when the output will be 0v while the input full power (40v):confused:
The problem stated above about the lm7812 was exactly this, after measurements with a scope on my prototype, there was a time upon turn on when the voltage exceeded the 35 volt in/out limit. As soon as the output cap were loaded, it ran fine if reseted because the output was already at its potential.
:confused: :confused: |
|
|
| quasi |
This is a regulator that I developed as part of my amp project. I have already built a few of these and they work a treat.
In theory this cct can provide any voltage within reason ( under say 200 volts) at currents of up to a few hundred milliamps.
The circuit attached shows a plus and minus design. If only a plus design is required all of the negative rail components can be omitted.
The only change you need to make is to select the zener value. I.e. if you want 30 volts output then your zener would 30 - regulator voltage. So if you used a 7815 then your zener (or combination of zeners) would be 15 volts.
Cheers |
|
|
| Nicke |
I have used a couple of LM317/337 as regulated supply in my poweramp .The input voltage are about 56V and output is set to 48V.
They have worked that way without trouble the latest 8 years... |
|
|
| peranders |
| quote: | Originally posted by pat allen
Hi peranders, about the lm317, isnt that playing with fire, since the time it takes to turn on, there will be a time when the output will be 0v while the input full power (40v):confused:
The problem stated above about the lm7812 was exactly this, after measurements with a scope on my prototype, there was a time upon turn on when the voltage exceeded the 35 volt in/out limit. As soon as the output cap were loaded, it ran fine if reseted because the output was already at its potential.
|
I'll guess certain combinations of load, feedback and other passive parts may cause less successful performance but first of all. The "in" voltage have a rise time and while it rises the output must be able to charge any capacitance.
So you probably not will have zero volts out and 40 volts in.
Have you checked LM431 with a booster transistor? LM431 can handle 37 volts if I'm not misstaken. |
|
|
|
