Connect with us

Stability of voltage regulators

Discussion in 'General Electronics Discussion' started by swagguy8, Mar 23, 2016.

Scroll to continue with content
  1. swagguy8

    swagguy8

    73
    6
    Dec 10, 2014
    Hey guys
    is it safe to put a 16v rated capacitor behind (to the output) of a 7815/7915 dual supply regulator? I have some lelons laying around that i want to use for the a hifi project. will the capacitors blow, or is the regulator stable enough to be at a constant 15v?

    thanks
     
  2. cjdelphi

    cjdelphi

    1,096
    104
    Oct 26, 2011
    If it's a linear, it will only output 15v max so it's safe, just keep an eye on the cap make sure it's not getting too warm

    I have a 16v on a 19v feed (but with a constant load most of the time lol) works fine
     
  3. davenn

    davenn Moderator

    13,782
    1,934
    Sep 5, 2009
    it's far from advisable
    you should be using a cap rated at around double the working voltage
    so a 25 - 35V rated cap


    seriously !! , that is not good advice


    Dave
     
  4. cjdelphi

    cjdelphi

    1,096
    104
    Oct 26, 2011
    Why not? 16v cap is fine for a 15v supply... it's done all the time, will it last as long? Prob not but that was not the question
     
  5. davenn

    davenn Moderator

    13,782
    1,934
    Sep 5, 2009
    not by any professional and qualified techs that are worth their salt
     
  6. swagguy8

    swagguy8

    73
    6
    Dec 10, 2014
    i see where you're coming from, but i see big tech companies like hp and dell use 25v capacitors for their 20v power supplies all the time (i've opened up their power supplys). They also have a regulated circuit. Is 2 times the original really necassary, after it is linearly regulated. I know that it is not very advisable to do so, as it will suffer from longevity issues, but the high end capacitors like lelons supposedly are very stable and accurate.
     
  7. AnalogKid

    AnalogKid

    2,462
    711
    Jun 10, 2015
    The vast majority of electronic components way outlast their ratings, life cycles, and other datasheet data. That's why a TV lasts 10 years when everything inside is rated for 1 year or less. And an amateur is free to run at the edge of the envelope with no serious consequences. But for a professional, serious means having to fly to Australia to repair a multimillion dollar system that is down because someone used a $1 part instead of a $3 part. It's a local call for Daven, but Ohio is 12 time zones and one hemisphere away.

    Yes, according to the manufacturer, 16 V cap will be just fine on a 15 V regulator output - for 42 days. After that, all bets are off, there is no warranty coverage, and you are on the hook for any damage caused by the component's failure. 1000 hours, 6 weeks, good night. Sure it's done all the time, by people in situations where the total cost of a failure can be absorbed.

    ak
     
    Last edited: Mar 24, 2016
    hevans1944, Arouse1973 and davenn like this.
  8. cjdelphi

    cjdelphi

    1,096
    104
    Oct 26, 2011
    Look, if this is powering a $$$ system then sure, a 19 - 24v is better than 16v but i kinda presumed it's not...

    As for professionals, absolutely they do, the amount of caps from samsung, acer who use 22v caps on 20v so many it's not funny and then i agree with you completely ...

    But is this a $$$ tv/amp/etc or just a linear regulator to power a simple circuit? If so then 16v is absolutely fine, just don't expect it to last 10 years...

    I still stand by my answer in this instance...
     
  9. swagguy8

    swagguy8

    73
    6
    Dec 10, 2014
    Don't worry, only a pair of NE5532s. and i have other capacitors that are less expensive ones before it, and they are rated @ 25v. i just want a cleaner signal, by getting rid of the fine residue with a lelon. I might considering not putting the lelon there, but i might just put it there and give it a shot, after all, the 7815/7915 is relatively stable.
     
  10. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    For what it's worth, I personally choose a support capacitor with a working voltage at least 50% higher than required. It does depend also on the circuit and load. You might need to go even higher than 50%. You see it's not just the regulator that can cause issues with cap ratings.

    Example: 12V Wall wart supplying 2 Amps in pulses to a regulator supplied with a 1 m cable. So we choose a 16 Volt capacitor because the input is only 12 Volts right?......Wrong! The inductance of the cable can cause considerable fly-back voltages which could approach 50 Volts. So over time the capacitor fails and no one knows why. The same can happen on the output of the regulator.

    I think it's even more important if you are going to be driving different loads with varying frequency.

    The cost of the capacitor won't always be significant. Actually my boss would be pleased for me to increase the voltage on a capacitor. As the 35 Volt one I found was cheaper than the 25 Volt one specified for the same value :)

    Just my opinion.

    Thanks
    Adam
     
    Last edited: Mar 24, 2016
    hevans1944 and davenn like this.
  11. cjdelphi

    cjdelphi

    1,096
    104
    Oct 26, 2011
    However, we're talking about a linear reg, if that starts outputing > 15v something failed in which case the capacitor is least of your worries...
     
  12. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    Yes agreed CJ. I was merely sharing another view point of voltage rating of capacitors that can be overlooked.
    Adam
     
    davenn likes this.
  13. davenn

    davenn Moderator

    13,782
    1,934
    Sep 5, 2009
    that's not even related to the discussion ... wouldn't matter what type of PSU, linear or switching
     
  14. swagguy8

    swagguy8

    73
    6
    Dec 10, 2014
    thank you for your explanation, can you elaborate a little more on the part that you said about the fly back voltages, or lead me to some reference?
     
  15. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    No I don't have a reference but here is an example. 24 Volts with a rise time of 1 us and wire inductance of 200 nH. The output shows nearly 40 Volts, I know it only for a short amount of time but worse cases can damage capacitors over time. The capacitor can be its own demise.

    Ring1.PNG

    Ring2.PNG
     
    davenn likes this.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-