Output capacitor voltage regulator problems

Thread Starter

leonhart88

Joined Feb 23, 2007
118
Hi All,

I'm encountering a strange problem where my voltage regulator works without a certain output cap, but with the cap, it doesn't output the correct voltage.

I've attached a picture of the schematic. I have a really large output tantalum cap (10uF). I've discovered (through a lot of scratching of my head), that the voltage regulator does not work when I have this large cap in. It outputs something like 4.7V instead of 12V like it's supposed to. I also notice that as time goes on, the voltage slowly drops. If anything, I would expect the voltage to slowly rise as the cap gets charged up.

Is the cap too large? The datasheet recommends 0.1uF, but I didn't think a very large cap would cause problems. Or is it because I'm using a tantalum capacitor and has something to do with the ESR?

Thanks.
 

Attachments

Papabravo

Joined Feb 24, 2006
21,157
If there is no other load in the system then I would suspect the capacitor. With No DC load on the output it should output the 12V plus or minus a little bit. It should also not be drawing any current and remain at room temperature.

A dropping output of 4.7 volts sounds like the output has a heavy load, is drawing way too much current, and is probably is finger burning HOT. In short -- find the short -- so to speak.
 

Thread Starter

leonhart88

Joined Feb 23, 2007
118
Hi Papabravo,

There is currently no load, as I'm putting together parts of my PCB and I'm doing the power circuits first. Actually, the only load is a 4.75K resistor with an LED.

I also encountered another large cap which was causing problems for another switching power supply on my board. I removed a 100uF cap and it works as expected now. The 100uF on that circuit and the 10uF on the 78L12 are much larger than what the datasheets suggest. Can the problems I'm encountering be caused by having too large of an output capacitance?

Would having no load on the supplies also be a culprit? I did not see anything in the datasheets which talk about minimum loads. I have attached the other circuit as well.
 

Attachments

Papabravo

Joined Feb 24, 2006
21,157
I'm not saying there is a problem with the schematic. Linear regulators of the 78Lxx type are not likely to care about the size of the capacitor on the output. There are some regulators that require a minimum amount of capacitance for stability. More should not be a problem. As always RTFDS.

Look for non-obvious faults on the board or in the surrounding environment. An ohmmeter can tell you if a cap is shorted. If the cap is not shorted have a resistor (470-680 ohms) and an alligator clip handy to discharge it.

Your problem is unlikely to be the size of the capacitor. Let us know what you find.
 

bountyhunter

Joined Sep 7, 2009
2,512
Last edited:

Thread Starter

leonhart88

Joined Feb 23, 2007
118
Hi All,

Thanks for your suggestions and help. It turns out I just had the polarity of the capacitors wrong (doh!). My silkscreen was confusing so I was putting the caps the wrong way...I guess next time this will be the first thing I double check.

I've also opted to get rid of the 100uF tantalum on the LTM8021. It was the only place I was using that part and it is kind of unnecessary and would reduce my parts count too.

Thanks again!
 

bountyhunter

Joined Sep 7, 2009
2,512
I've also opted to get rid of the 100uF tantalum on the LTM8021. It was the only place I was using that part and it is kind of unnecessary and would reduce my parts count too.

Thanks again!
As I said above: the LTM8021 is a 1.1MHz switcher. A tantalum capacitor is beyond it's usable frequency at 1.1MHz, it actually looks inductive not capacitive.

I read the data sheet for the LTM8021 and it has extensive information on the ceramic caps you are supposed to use with it.

From the data sheet:

Capacitor Selection Considerations

The CIN and COUT capacitor values in Table 1 are the
minimum recommended values for the associated operating
conditions. Applying capacitor values below those
indicated in Table 1 is not recommended, and may result
in undesirable operation. Using larger values is generally
acceptable, and can yield improved dynamic response or
fault recovery, if it is necessary. Again, it is incumbent
upon the user to verify proper operation over the intended
system’s line, load and environmental conditions.

Ceramic capacitors are small, robust and have very low
ESR. However, not all ceramic capacitors are suitable.
X5R and X7R types are stable over temperature and applied
voltage and give dependable service. Other types,
including Y5V and Z5U have very large temperature and
voltage coeffi cients of capacitance. In an application circuit
they may have only a small fraction of their nominal
capacitance resulting in much higher output voltage ripple
than expected.
I believe you could use Tants only if there was enough ceramic capacitance in parallel to swamp out the resonant effect.
 
Last edited:

ramancini8

Joined Jul 18, 2012
473
When you put large caps on regulator outputs you run the risk of blowing the regulator during shutdown if the input voltage goes to ground. the solution is to bypass the regulator with a reverse diode.
 
Top