voltage keep dropping why?

pot

- . ( ). :

: .

Is it possible that you had one of those capacitors mounted backwards so that they were failing and shorting out?

It is also possible that your adjustable resistor is getting warm and changing value. What is its maximum resistance (the full resistance of the element, not counting the wiper contact)?

It is not easy to drop the voltage all the way to zero with this regulator. Doing this requires that you provide a negative reference voltage to connect to the adjustable resistor that equals the voltage the LM317 produces when you adjust the resistance all the way to zero (around -1.2 volts). The LM317 always tries to produce an output voltage

1.2 volts more positive than whatever you connect to its reference (adj) pin. So you have to set the reference pin to -1.2 to produce a zero volt output.

Do you mean a Proto Board such as the PB10?

If so you might have an intermittent high resistance in one or more connections.

it

You haven't said whether you were measuring the voltage across a load. If it was loaded and the regulated adaptor is working correctly and within its capability, then the output voltage will depend upon the load value and whether or not the 317 is heatsinked.

Very basic 317 regulator. There is absolutely no capacitance for proper smoothing under load or to supply peak response current. There is no protection for the 317 either.

at

the

pot

PB10?

I am measuring with Multifunction Volt Meter, and I have put three 5 volt bulbs, strangly voltage remains same at that time.

The reason why I need this is b/c I am testing some series-parallel circuits, rheostats, and superposition theorem, and few other stuff. Mostly college work. I figure if I have a voltage regulator at home I can do my work twice, and verify the result -- lab is only one hour durning summer, and no extra time....

It has a head sink, very good one I might add.

I think I need to build a bit more advance regulator, something that can handle 30v load, and is very precise. This circuit is very much like unregulated adaptor -- expect this takes the voltage down. I am very unhappy with this.

I am trying to do my next week lab, and 1/2 of times I am worried of the voltage going down.. No good.

You said you were using 3 x 5V bulbs as a load but you didn't give the wattage ratings of the bulbs or whether you had them in series or parallel or a combination of these modes. You also have not provided any details of the output capability of the regulated plugpack supply.

You say you do have a substantial heatsink on the 317 but I would question the wisdom of using a TO220 device in a proto-board while it has a substantial (read heavy) heatsink attached. The weight of the h/sk can cause the device to lean over thus creating possible bad connnections.

I, and many others, have constructed regulated and adjustable power supplies based on the 317, and I can vouch for the ability of these devices perform as intended. There is little chance that you can construct a decent power supply in time for your project completion date by going about it in the manner you appear to be. It takes a lot longer than that to build something from scratch unless you know what you are doing and you must have a good grasp of the technicalities of designing and building such a power supply. I really would suggest that you borrow a ready built unit for your immediate requirements and then design your own as you gain more experience.

There are many suppliers of kits such as this unit

complete with a pcb which can provide what you are looking for. This one is available from KitsRus in HK and it won't cost much.

If you want a unit with higher long term current capability (bigger heatsink) then use K124

Okay bit of an update. I just soldered it in a pcb board, and it seems the thing is staying at 10.05 -- i set it at that speed at 6 o'clock in the morning -- same circuit as above, I have change the pot to 50k, and no capacitors. It's very stable now. Althought it goes from 10.05

- 10.06.

Also, beside bulb test, I am testing series, and parallel circuits (which I have just finished it), and now I am starting on superposition theorem circuits. I thing I'll be okay now. Ever since I soldered them on a pcb, it's been very good to me. I should have done it months ago...

unit

with a pcb can provide what you are

won't cost much.

K124

I will do some research on this, cause I really need to build one, and soon. Cause next semester I will be taking AC circuits...

PS: How come there are only 2 boys in my Electronic I thought this used to be men domain?

(snip)

This (the stability of soldered connections over plug connections) is a lesson almost as useful as the superposition theorem.

the

provided

supply.

it

50K is too high imo - you don't need anything higher than 10K.

What do you mean by "and no capacitors"?

You must have capacitors connected as specified in order to maintain loop stability and dc filtering. I strongly suggest you include capacitors as used in the KitsRus designs because this circuit is typical of those used by knowledgable designers.

Well you may have learned something from this exercise. Never trust temporary connections (eg. non-soldered) in power circuits. Even a tiny amount of resistance in a joint can drop a relatively large voltage when appreciable current flows through it.

:-) I think so, this circuits has been very good to me, my professor just cancel the lab class! And we girls had to two more labs. Good things I brough my stuff in the car, and we are doing it in the library :-) So far so good.

I am also building 2nd voltage regulator.

here is the diagram

I don't have 1.2v LM313 Reference Diode, but I do have 3v or 3.3v lm133 (how do I check it?), and I am wodering if I can do something.

snipped-for-privacy@gmail.com wrote: (snip)

A single 1.5 volt chemical cell might be good enough for the negative reference, though it will get discharged when you turn the AC off, so you will have to disconnect it.

A simple, low precision 1.25 volt shunt regulator can be made by connecting two diodes in series, or by using an IR LED (like those that produce the light signal for remote controls) or by making a Vbe multiplier with any small transistor.

The last is made by connecting two approximately equal resistors in series, one collector to base and one base to emitter, so that when the voltage across the transistor rises to two diode drops, the transistor turns on and the collector current keeps the voltage from rising much higher. You need one additional resistor to the negative bias supply to feed it. Adjusting the ratio of the two resistors trims the shunt regulated voltage. None of these is as temperature stable as the integrated shunt reference regulator, but at room temperature, are probably good enough for your use.

I see that the above schematic does not show where the low current negative 10 volt bias supply comes from. Do you need help deriving that from your transformer and rectifier?

Do yourself a favor and and change the 50K pot to a 5K linear taper pot. You'll find it much easier to adjust unless your 50k has a log taper. Also use the capacitors for sure and maybe some protection diodes. See Fig 1 of this datasheet for diode connections.

Mike

"The scientist is possessed by the sense of universal causation...His religious feeling takes the form of rapturous amazement at the harmony of natural law, which reveals the intelligence of such superiority that, compared with it, systematic thinking and acting of human beings is an utterly insignificant reflection." Albert Einstein (theoretical physicist)