79L05 negative regulator outputs -2V and overheats with large capacitor

[Monisa]

Hi, today I used 79L05 as the negative electrical level. The output filter capacitance is very large at first, just like the figure below:

It is -12V. Under this condition, the chip is very hot,and the output voltage is just about -2V.

I changed the 220uf to 47uf, then the chip worked well and the output voltage was at -5V without heating. I wonder why?
Sorry for my poor English, but can anyone answer me? Thanks very much!

 

[berkeman]

Hi, today I used 79L05 as the negative electrical level. The output filter capacitance is very large at first, just like the figure below:

It is -12V. Under this condition, the chip is very hot,and the output voltage is just about -2V.

I changed the 220uf to 47uf, then the chip worked well and the output voltage was at -5V without heating. I wonder why?
Sorry for my poor English, but can anyone answer me? Thanks very much!

Welcome to the PF.

Is there any chance that you got the first capacitor backwards in the circuit? That would have drawn a pretty big current. Or maybe the capacitor had been previously damaged?

 

[davenn]

Is there any chance that you got the first capacitor backwards in the circuit?

yes, that would have been my suspicion as wellDave

 

[jim hardy]

IC voltage regulators have some esoteric fine points.

Note that the negative regulator uses a common emitter output stage which has higher output impedance than positive regulators with common collector.
Look at this schematic, it has the series pass element Q21's collector tied to the output pin.
http://www.st.com/content/ccc/resou...df/jcr:content/translations/en.CD00000450.pdf

That's same as in LDO regulators. That topology requires a minimum ESR to be stable, unlike the positive regulators which are basically common collector output stage
see http://www.ti.com/lit/an/snva167a/snva167a.pdf
and http://www.ti.com/lit/an/slyt194/slyt194.pdf
and http://www.ti.com/lit/an/snva020b/snva020b.pdf

I don't know whether your particular regulator is internally stabilized for low ESR loads. Is there a precaution in its datasheet ?
Those could be compensating capacitors around Q16 and Q20 in that ST schematic just above, i don't know..

Do you suppose that your 220uf capacitor was just "Too good" for the lightly loaded regulator ? Try tacking it back in with maybe an ohm in series and see if things get any better.

Anyhow, that's just something to be aware of when using LDO and negative regulators.

old jim

 

[rbelli1]

capacitor was just "Too good" for the lightly loaded regulator

Some positive ones have this problem as well. As well as some switching ones. I have seen them where they recommend low ESR types with a known series resistor rather than rely on an unknown value from a sloppier cap.

BoB

 

[Averagesupernova]

It is interesting that when you look at the internal schematic for a positive or negative they are very similar. The three terminals are not the same but the layout is very similar. I hadn't paid attention but quite often the negative regulators are low dropout by default. I found this recently while looking for a way to drive LED backlit LCD display modules.

 

[jim hardy]

It is interesting that when you look at the internal schematic for a positive or negative they are very similar. The three terminals are not the same but the layout is very similar. I hadn't paid attention but quite often the negative regulators are low dropout by default.

That's it !

Pass transistor emitter to output pin in this 7800 positive regulator,, as opposed to collector in negative 7900 above. Emitter follower has lower Zout.
https://www.sparkfun.com/datasheets/Components/LM7805.pdf