voltage regulator +/-

Status
Not open for further replies.
still don't work... odd.
 
Last edited by a moderator:
you want this to be continuously variable from -15v to +15v right? getting that will be no problem

the big problem is how much power do you want from your supply?
to be a general purpose supply you should shoot for at least 200 watts.

if this is a classroom project then the circuit i submitted will suffice
 
i went out last night and bought a printer/scanner because i found out you can attached stuff to emails.
i didn't know you could do that.

a brand new hp psc 1410 for 90 bucks. was that too much?
 
As far as I know that's a good price and HP makes good products. I think that you will find that your scanner's resolution is really good. I bought a cheap Lexmark before that worked decently but I think the resolution was about 300 DPI. These days you should be able to get 2400 DPI without even trying.
 
i went out last night and bought a printer/scanner because i found out you can attached stuff to emails.
i didn't know you could do that.
Click to expand...
no offense, but I thought everybody knew that =]
 
here is the schematic for those who can't get the above link to work.
http://www.killerkittykats.net/quik/cato/out4.gif

edit: man, it was hard to post that, I was bidding on a laptop and if you have ever had a bidding war, you know how nerve racking that can be. anyway, I won the laptop:
* CPU: PIV - 2.4GHz
* RAM: 512MB
* Screen: 15" LCD
* DVD-ROM/CDRW combo drive
* 3.5" Floppy disk drive
$262!

the only problems are lack of hard drive, and the screen has a line through it =]
 
Last edited:
i have noticed on both mine and metakrons circuits the components aren't labeled.
as in R1, R2, Q1, IC1, etc.
it would be a good idea to do that
 
I did say that it wasn't finished. I gave some workable values for the resistors and capacitors. The transistors are generic and so are the op amps. I like the LM324 op amps because they go down to the rail on output voltage and don't require double-sided supplies. They are also very common and cheap.

Select the transistors for the power output and heatsink them correctly. You have a driver transistor and a power output transistor in each pair, and they are arranged as Darlingtons. They are all arranged as emitter followers. Each resistor is for current limiting and is selected to allow enough current through to drive each transistor. You are going to need 1 to 2 milliamps to the base of the driver transistor at the very least to switch the output transistor all the way on if it is passing 10 amps.

By heatsinking them correctly, I do mean that you figure out how much wattage each transistor dissipates and provide for that.

I am going to have to leave it for the next day or two to draw up the circuitry that I would recommend for generating the right voltages. This should be fed by regulated voltages or you will (may) get feedback and oscillations.
 
Last edited:
I'd have to look it up, but I'm pretty sure that the open loop gain for the 324 is at least 1,000,000, and that's voltage gain. With an op-amp you look at the gain-bandwidth product. If it has a gain-bandwidth product of 1,000,000 you get a DC gain of that much but only 1,000 at 1,000 Hz. A project like this is happy with any op amp that can swing close to the supply rails. The current gain of most op amps is "very high."

The gain issues with this project have to do with the amount of current the resistors let through the transistors. With a 12 volt supply, not more than about 2 milliamps can get through to the bases of the driver transistors using 4.7k resistors. The current gain of those transistors will be about 100. The power output transistors will have a gain of about 50 to 100, and I would measure that before using them. It will take 100 milliamps to the bases of the output transistors to get five amps through them, or 200 mA to get 10 amps through them. It's not quite that simple, but that's a close estimate. Also, at 10 amps the .1 ohm resistors I specified will drop a solid volt each. These specifications are marginal at 10 amps and you won't get more than 9.8 volts each direction. The best thing to do is to reduce the 4.7k resistors to about 2.2k to 3.3k, the 47 ohm resistors to 22 to 33 ohms, and the last current limiting resistors to .05 ohms or maybe do without them. Most op amps can handle 25 to 30 milliamps indefinitely, so the approximately 5 milliamps required will not strain them.

At 1 to 5 amps the values specified should be solid. You can still lower them the way I just said. Doing so will give you headroom. I just picked the values I did in sort of a hurry. Those values are not very critical, but when you are running 200 mA into the bases of the output transistors, you are going to need the current limiting resistors at the collectors of the driver transistors to be able to take at least 2 watts. Those are the ones that are marked "47". An alternative is to use Darlington power transistors for your output transistors.
 
for future use:
http://tinycad.sourceforge.net/
its a program to draw circuits. just save a screenshot and you can post it ( or send it to me as like a jpeg or something)

that way you don't have to hand draw stuff (leo) =]
 
i downloaded the tinycad but the problem is that i never used a program like that before.
the last schematic that i submitted would make a good starting point for you.
all you need now is the regulator part and power output stage.
neither one should be too hard to design
the major problem of varing the voltage from minus to plus has been solved.

gee cato aren't you going to school for this kind of stuff?
we gotta let you do SOME of the work.
 
Status
Not open for further replies.
Back
Top