3.3 voltage regulator problem

[elsey.jack]

Hello everyone,

I have done several projects involving 7805 voltage regulators. Noob that I am, however, I have never worked on a project where logic high is +3.3 V.

I have been trying for the past hour or so to get a LD33V voltage regulator working, but I can only conclude that there is something wrong with either me or the chip (although I will cede that is is most likely the former). Here is a **broken link removed** of the voltage regulator I am using.

The pinout from the datasheet (PDF) is as follows, from left to right: ground, Vout, and Vin. However, when I connect the regulator following this pinout to a power supply (which works fine with my 7805 voltage regulators) I only get a measly 700 mV between Vout and ground!

I have tried varying the capacitors, removing the capacitors altogether, and even wiring it like a 7805, all without the desired result. Other possibly relevant information is as follows:

• The voltage being supplied by the power supply drops from about 9 volts to about 6 volts when I have it connected to the regulator. This voltage only drops to about 8 volts when connected to my beloved 7805's.
• The regulator warms, but does not get hot.
• There is a direct connection between the heat sink and the central pin, which (according to the datasheet) is Vout.
• The regulator draws about 100 mA when connected.

Am I missing something here? Any guidance would very much be appreciated.

 

[Menticol]

I think the regulator is defective, and connecting it like a 7805 didn't helped very much. Do you have any other LD33V in stock? So you can track the problem

EDIT: Yours is the regulable or fixed version?
EDIT: Try the exact configuration shown on the Typical Application Section, figure 11 page 17 onyou pdf

 

[Speakerguy]

1117 regulators are very dependent on output capacitor type. Too low of an ESR and they will oscillate. Don't use any ceramic caps at the regulator, and make sure you have a big cheap electrolytic (high esr, high capacitance value) at the regulator. Drawing more current from it will help with stability as well.

I once blew up an 1117 regulator with no load just by hooking up a ceramic output cap. Damn thing burned my finger.

 

[elsey.jack]

I have another LD33V voltage regulator, which also only gives me 700 mV.

 

[elsey.jack]

It is a fixed voltage regulator (or at least the Sparkfun product page did not say it was).

I tried connecting it in the way depicted in Figure 4 on page 3 of the datasheet without the desired result. I did not have a 100 nF capacitor on hand, so I substituted a 470 nF capacitor. Both capacitors were electrolytic.

 

[Menticol]

Is Yours one of this? (use the Find function on your web browser)

LD1117S30TR LD1117D30TR (*)
LD1117S33TR LD1117D33TR LD1117DT33 LD1117DT33TR LD1117V33 3.3 V
LD1117S33CTR LD1117D33CTR LD1117DT33C LD1117DT33CTR LD1117V33C 3.3 V

Remember that maximum input voltage is 15V. The Allmighty 7805 is stronger
"Absolute Maximum Ratings are those values beyond which damage to the device may occur. Functional operation under these condition is
not implied. Over the above suggested Max Power Dissipation a Short Circuit could definitively damage the device."

 

[elsey.jack]

Is Yours one of this? (use the Find function on your web browser)

LD1117S30TR LD1117D30TR (*)
LD1117S33TR LD1117D33TR LD1117DT33 LD1117DT33TR LD1117V33 3.3 V
LD1117S33CTR LD1117D33CTR LD1117DT33C LD1117DT33CTR LD1117V33C 3.3 V

Remember that maximum input voltage is 15V. The Allmighty 7805 is stronger

It is most likely the LD1117V33, possibly the LD1117V33.

The power supply (typical wall wort AC-DC voltage converter) only inputs 9V.

 

[Menticol]

Well, I'm not an expert by any means, but I've read in this forum that those wall-wart adapters are the evil. Under no load the voltage may go to the sky, and regulation is not necessarily good. Can you get an old computer power supply? Or the computer that you are using right now, if it's a desktop PC, remove the cover off and connect the thing to the PSU 12V rail (yellow and black wires)

 

[elsey.jack]

Well, I'm not an expert by any means, but I've read in this forum that those wall-wart adapters are the evil. Under no load the voltage may go to the sky, and regulation is not necessarily good. Can you get an old computer power supply? Or the computer that you are using right now, if it's a desktop PC, remove the cover off and connect the thing to the PSU 12V rail (yellow and black wires)

Ha ha, even if it wasn't a laptop I'd be wary of opening it up in fear that I'd never be able to put it together again.

It is weird that the typical application circuit (Figure 4 on page 3) has a higher capacitance on Vout than on Vin. Wouldn't you want the bigger capacitor on the pin that will experience the most unstable voltage, i.e. the pin that is connected to the wall wart?

I'm pretty sure that the power supply that I am using is okay because it works for other circuits I have around when I plug it in.

 

[Menticol]

Well, that question may be easily answered by the "bosses" of this forum, but I think they are sleeping now. You need to take out suspect elements from the equation, and that wall wart is one of them. Believe me, a PSU scrapped from any old desktop PC is awesome for every electronic project: you have +12, +5, +3.3 0 -5 -12 to choose from. And lots of amps, if it's an ATX model.

 

[elsey.jack]

1117 regulators are very dependent on output capacitor type. Too low of an ESR and they will oscillate. Don't use any ceramic caps at the regulator, and make sure you have a big cheap electrolytic (high esr, high capacitance value) at the regulator. Drawing more current from it will help with stability as well.

I once blew up an 1117 regulator with no load just by hooking up a ceramic output cap. Damn thing burned my finger.

Both my capacitors are electrolytic and I still cant get a voltage over 800 mV even when I place a 10K resistor across Vout and ground.

I had some interesting experiences this summer with a LiPo battery accidentally shorting out. Stuff happens.

 

[elsey.jack]

Well, that question may be easily answered by the "bosses" of this forum, but I think they are sleeping now. You need to take out suspect elements from the equation, and that wall wart is one of them. Believe me, a PSU scrapped from any old desktop PC is awesome for every electronic project: you have +12, +5, +3.3 0 -5 -12 to choose from. And lots of amps, if it's an ATX model.

Yes, thank you very much for the help though.

 

[creakndale]

If you're trying to use the regulator as it's shown in the picture, all 3 pins are shorted together by the breadboard. You placed it that way in the breadboard just to take the picture, right?

Maybe there's defective contact(s) inside the breadboard so that it's not making proper contact to all 3 leads?

creakndale

 

[DirtyLude]

I don't really have an answer, but I use these in the SOT-223 package and they have been pretty reliable for me. I think you just need to step through all of the components here to figure out what the issue is. Based on the markings, it's definitely the fixed 3.3v part. Move everything to another part of the breadboard, check voltages on all to the lines without the TO-220 plugged into the board to make sure nothing is straying, use the capacitors as close to the component as possible... you'll figure it out.

 

[audioguru]

It is weird that the typical application circuit (Figure 4 on page 3) has a higher capacitance on Vout than on Vin. Wouldn't you want the bigger capacitor on the pin that will experience the most unstable voltage, i.e. the pin that is connected to the wall wart?

It is normal for a low dropout regulator to have a certain faily large capacitor value on the output to prevent it from oscillating. The output capacitor has a minimum allowed value. The output capacitor must be connected directly to the pins, not to a breadboard.

It is weird that the LD33V is not an LD1117 so maybe it is a cheap copy made by a Chinese company.

 

[elsey.jack]

It is normal for a low dropout regulator to have a certain faily large capacitor value on the output to prevent it from oscillating. The output capacitor has a minimum allowed value. The output capacitor must be connected directly to the pins, not to a breadboard.

It is weird that the LD33V is not an LD1117 so maybe it is a cheap copy made by a Chinese company.

I had the capacitors in the breadboard sockets directly adjacent to the regulator. Is that what you mean by "directly connected to the pins"?

I will go today and purchase two more regulators from a different shop to see if I can make them work. Thanks for the help everyone!

 

[audioguru]

Breadboard sockets that are directly adjacent to the regulator's pins are fine.
I have seen many breadboard circuits with too long wires all over the place and they do not work.

 

[elsey.jack]

One last question. . .

OK, got a pair or NTE1904's from the shop today. These are putting out a rock-steady 3.28 volts.

The minimum "output bypass capacitance", which I assume is the capacitor on the output pin, is 2µF, with a typical value of 10µF.

There is no specified "input bypass capacitance" because that is dependent on how non-steady (Is there a better word?) the input voltage is, right? I have a 100µF capacitor on this pin because I am using a power supply, but would I need one if I was using a battery as a voltage source?

 

[Vizier87]

Microchip's got a good documentation:
https://www.electro-tech-online.com/custompdfs/2010/01/3_3vto5vAnalogTipsnTricksBrchr.pdf

but I think this is a late reply, seeing that you've bought the NTE1904. Just an afterthought.

 

[mjtessmer]

Looking at the picture provided, all the pins of the regulator would be shorted by the solder-less breadboard.
You need to rotate the regulator 90 degrees, then make your connections.