+12v power supply - LM7812

[pomprocker]

Can someone help me to figure out how to get a stable 12volts dc?

I have an old laptop power supply that outputs 15vdc at 3amps.

I bought a LM7812 12volt regulator....Now I need to figure out what caps/resistors i need..


thanks

[MaltiK]

U1   LM7812 +12 VDC Voltage Regulator
BR1   4 amp bridge rectifier
T1   18 volt, 2 amp ac transformer
F1   2 amp slow-blow fuse
S1   SPST toggle switch
R1   330 ohm resistor
C1   3,000 uF electrolytic capacitor, 35 volt min.
C2   100 uF electrolytic capacitor, 35 volt min.
LED1   Light Emitting Diode
MISC.   fuse holder, heat sink for U1, binding posts, ac cord with plug, chassis

[pomprocker]

Yeah thats the first one I saw, but because the the voltage regulator has a +/- 5% tolerance, I was getting 11 something volts out of the regulator, so I couldnt even get 12 volts.



This is for a high voltage programming of AVR circuit btw..

[pomprocker]

What is the difference between a 2N3903 and a 2N3904?

[MaltiK]

that should give you 12v, it must be your battery? Did you check to see if it was 12v before hooking it up to the circuit? Did you make sure the caps are 10% tolderance, and resistors 5-10%, with 1/4W?

[ArcMan]

As Matlik said, you should get 12V (+/- 0.5V).  One possible cause for not getting ~12V is the dropout voltage of the 7812, which is 2V.  That means you must have at least 14V at the input to get 12V at the output.  What is your input voltage, really?

[pomprocker]

I have an old laptop power supply that outputs 15vdc at 3amps.

My input voltage is 15vdc

[pomprocker]

With the circuit above I was able to get 11.86vdc...still not quite 12vdc, any way to bump it up a little more?

[ArcMan]

The 7812 is spec'ed at 12V +/- 0.5V, so it's operating within spec.  If you need your voltage right at 12V, I can suggest the following:
Try another 7812.  It might be closer.
Switch to an LM317.  You can then tweak your voltage to exactly where you want it.

[leye0]

If someone else have this problem, maybe he can try a 7812 in the CK version instead of others like CT. I think there's a smallest dropout voltage.

[Soeren]

Hi,

Dropout voltage of the LM7812 is 2.5V minimum!
(minimum = "when the sun is shining and you are going downhill").

You should never go that low, keep at least 3V overhead or suffer on line and load regulation as well as output noise and ripple.
Whenever you feed a regulator from a mains transformer, you have to account for mains variations of +/-10%. a transformer keeps the input/output factor, so if the mains at the primary side is down 10%, so will the secondary voltage be - a 15V transformer may thus be 13.5V to 16.5V within specs.
A switch mode supply of the universal type used in most laptop supplies is much better in this regard, since it keeps its output (still with a tolerance of its own though), usually from 90V to 240V (read the label for the exact specs).

You should measure the supply voltage when loading the output of the regulator with 1A (remember a heat sink on the LM7812), to see if it's possible to get it a little higher. If you've got a high enough voltage under load, you could raise the output voltage like this:

Unfortunately, I used an adjustable regulator for the drawing, but the pin labelled ADJ should be GND!

D1 will raise the voltage around 0.6V to 0.7V (using a schottky diode you'll get around 0.3V to 0.4V).
C1 should be around 2,000µF per 1A output current (larger is OK and improves the ripple rejection)
C2 hinders the regulators tendency to oscillate, when C1 is more than an inch away from the regulator.
C3 dampens noise and lowers the impedance on the ground pin
C4 output cap should be at least 20..25µF to keep the regulator in control (up to 100µF is OK)
R1/LED1 is just if you want a monitor telling you that the circuit is powered.

Allways put a fuse in front of a power supply circuit. A fuse of 1.5A to 2A (Slow Blow) will be suitable here.

[tank]

Soeren,

Will your suggested circuit above work for a 12 vdc lead acid battery? I want to regulate its voltage since it spikes up to 17 to 18 volts. I want to have a regulated 12 vdc 500ma - 1 A source.

Regards

[Soeren]

Hi,

Please ask here exclusively (and keep to a single PM if at all, it slows down answering speed).

Will your suggested circuit above work for a 12 vdc lead acid battery? I want to regulate its voltage since it spikes up to 17 to 18 volts. I want to have a regulated 12 vdc 500ma - 1 A source.

No, as reading the entire post would have told you.

Dropout voltage of the LM7812 is 2.5V minimum!
(minimum = "when the sun is shining and you are going downhill").

You should never go that low, keep at least 3V overhead or suffer on line and load regulation as well as output noise and ripple.

To spell it out... For 12.0V out, you need a minimum of 15V in.

The best way to get 12V from a "12V" lead-acid battery is with a good quality LDO regulator (assuming your battery never gets under ~12.5V) or a switcher (of either buck-boost, SEPIC or Ćuk topology).

[Afroman]

You are overthinking this guys. Check out what his application is.

This is for a high voltage programming of AVR circuit btw..

He doesn't need a tightly regulated 12.0V line for that. A volt or two out of range will be fine. My programmer's high voltage line is 13.something volts. Double check the datasheet for the acceptable range of Vpp.

[TrickyNekro]

What is the difference between a 2N3903 and a 2N3904?

From a very quick glance that I gave they seem that the have different Hfe... Or b factor...
In fact the 2N3904 Hfe is double compared to the Hfe of the 2N3903

The b factor is the amplification factor that transistors have... Transistors are current driven and act as current sources...

So say you put 1mA current to the base... then... you are gonna get b * 1mA current at the collector and (b + 1)*1mA current at the emitter, at a common emitter config.

Usually, you want to design something with little dependence on the b/Hfe value... so you don't have thoughts as should I use the 2N3903 or the 2N3904....

That's design considerations... If you don't want to mess with them... stick to the diagram, unless the transistor is merely used as a switch...

Using the transistor as a switching device means that you're driving it between cut-off and saturation... where b factor is not a concern as you imply a certain b factor to the transistor rather than using it's natural one.

Typically in saturation you should consider a minimal voltage drop between the collector and the emitter of 0.1V to 0.3V... Usually it's 0.2V, but in either cases it's negligible...

All that for as a friendly designing advice... ;-)

Best Regards,
Lefteris, Greece

[Soeren]

Hi,

He doesn't need a tightly regulated 12.0V line for that. A volt or two out of range will be fine. My programmer's high voltage line is 13.something volts. Double check the datasheet for the acceptable range of Vpp.

I didn't check the specs for Atmel devices, but for Microchips devices the tolerance is +/-0.5V, or should I say 0.500V(?) as going over +0.5V is a definite no-no and going under will fail on most (all?) devices. Luckily, it's dead easy to keep within that fairly wide range.

However, a zener regulator should be just fine and there's no need for an external supply, as it's pretty simple to switch up from the 5V line (we're talking <1µA).

With the circuit above I was able to get 11.86vdc...still not quite 12vdc, any way to bump it up a little more?

That's OK.
It's: (12-11.86)/(12/100)% = ~1.2% within specs.

Don't chase a reading of "12.00" or you'll drive yourself insane trying.
Your meter has a tolerance as well and even the best digital instrument will have a possible error of close to one count on the least significant digit - Unless your meter cost >$500 and was recalibrated less than a month ago and was warmed up for at least 20 minutes (to a temperature of 25°C) and you sacrificed a virgin, expect less precision.

For some weird psychological reason, most people tend to accept an overrange easier than an underrange - Ask yourself... For a 12V line, would you prefer 11.89V or 12.12V?