Maker Pro
Maker Pro

Not getting correct voltage from voltage regulator

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
you asked for datasheet info, I provided it ... there are many other references on the net
if you cant accept that, it's not my problem ;)
 

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
I've always used caps on the input and output of 78xx series regulators too, but never stopped to analyze the reason. I just blindly do what the datasheet says. (Call me sheep.)
 

cjdelphi

Oct 26, 2011
1,166
Joined
Oct 26, 2011
Messages
1,166
The output cap should really be there to smooth out any ripple (which is important) but....

Any reg with a feedback should get a clean voltage reference (hence the cap) but say 78xx (not adj) without fb should be ok with a cap on the input and not much on it's output as the reference is to gnd...

What i'll try is varying loads without the cap (can't see it oscillating personally with a linear reg)
 

Farticus

Jun 21, 2015
36
Joined
Jun 21, 2015
Messages
36
I have fitted the Volt reg in my board and it is working fine.
I am not using for anything except to try to understand the basics of how things work.
Most of what came after my last post was way over my head but thanks anyway.
 

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
The output cap should really be there to smooth out any ripple (which is important) but....

Any reg with a feedback should get a clean voltage reference (hence the cap) but say 78xx (not adj) without fb should be ok with a cap on the input and not much on it's output as the reference is to gnd...

What i'll try is varying loads without the cap (can't see it oscillating personally with a linear reg)
Like I said yesterday, try an LP2950 linear LDO regulator.
Again, from the datasheet excerpt that I posted yesterday:-
Without this capacitor the part will oscillate.

And believe me, from my own experience, it does oscillate.
 

cjdelphi

Oct 26, 2011
1,166
Joined
Oct 26, 2011
Messages
1,166
Lp2950 (ok ok, i've ordered one lol)

Well i ordered 10 (3.3v ldo) ...
 

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
Lp2950 (ok ok, i've ordered one lol)

Well i ordered 10 (3.3v ldo) ...
They're pretty handy, besides as a regulator they make a good reference since they're fairly precise. That's what I was using the LP2950-3.3 for when I got the oscillations the other week - a reference on the input of a comparator.
On my DMM, it measured as a higher-than acceptable voltage. I didn't check the oscillations with my oscilloscope - it's buried under a ton of stuff and I haven't got around to digging it back out yet.

Edit: I should add that I had no load on it apart from the comparator input. This was a battery monitoring circuit that had to have very minimal loading on the battery, another reason for choosing the LP2950 - it uses very little current internally. And after adding the cap, even with that minimal load, it settled down nicely to 3.28V as mentioned.

I think the LM2936 suffers from the same problem too, from memory.
 
Last edited:

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
I have fitted the Volt reg in my board and it is working fine.
I am not using for anything except to try to understand the basics of how things work.
Most of what came after my last post was way over my head but thanks anyway.
Good one, glad it's sorted. Now you can get down to the real fun. Ignore the stuff that's over your head - just follow datasheet guidelines for components and you can't really go wrong. :D
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
it certainly is possible to get instability without an input capacitor, but what you normally see is output instability due to incorrect loading.

As you point out, LDO regulators often suffer a particular form of instability, but there are other forms.

you will note that almost all three terminal regulators talk about additional input capacitance where the regulator is some distance from the filter capacitors.

The real trap for young players are the regulators that require a minimum ESR. I suspect this was always true, but that capacitors have developed to such an extent that it's now a lot easier to get caps with inappropriate (for this purpose) specs.
 

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
Is it possible that you mistook the input and output pins ?
He already explained that he had the regulator connected using flying leads and had a bad connection. Going by the output voltage he was getting, it would have been a bad ground connection.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Steve, don't you mean 'appropriate'?

No, I actually mean inappropriate.

There are cautions against using high value ceramic caps as the output caps for certain LDO regulators. These high value ceramics (say, 10uF 6V) are attractive for use on a board populated with SMT components. They're small, they're cheap, and we use a 0.1uF ceramic, so a 10uF must be great, right?

Well in some cases "WRONG". The ceramic caps have a very low ESR, and I've seen regulators (I'd have to go searching to find them now) which specified a range of ESR for the output capacitor to achieve both stability and their published specs.

Actually, google comes to my assistance with this, and this.

Here is another pdf which goes into some detail about the input capacitor.
 

Old Steve

Jul 23, 2015
734
Joined
Jul 23, 2015
Messages
734
No, I actually mean inappropriate.

There are cautions against using high value ceramic caps as the output caps for certain LDO regulators. These high value ceramics (say, 10uF 6V) are attractive for use on a board populated with SMT components. They're small, they're cheap, and we use a 0.1uF ceramic, so a 10uF must be great, right?

Well in some cases "WRONG". The ceramic caps have a very low ESR, and I've seen regulators (I'd have to go searching to find them now) which specified a range of ESR for the output capacitor to achieve both stability and their published specs.

Actually, google comes to my assistance with this, and this.

Here is another pdf which goes into some detail about the input capacitor.
Right, I get your meaning now.
 
Top