Loading ...
Sorry, an error occurred while loading the content.

Expand Messages
  • jong kung
    Dave / Larry, ... I looked there and on one of the pages they have typical specs (input and output). The footnote says these were taken using 0.22uF on the
    Message 1 of 10 , Jan 24, 2013
    • 0 Attachment
      Dave / Larry,

      > Go to the 78XX datasheet. There they recommend a cap
      > input and output. Try

      I looked there and on one of the pages they have typical specs (input and output). The footnote says these were taken using 0.22uF on the input and 0.1uF on the output. Another 78xx datasheet says to use 0.33uf on the input. These are for 19v input, 12v output. I don't really know how going with 24v input would affect the size needs of input capacitor (I'm gonna guess the larger the input ripple and bigger the input capacitor). But also the exact value might NOT be SUPER critical.

      You have not mentioned what is your current requirement.


      Jong
    • Larry Beaty
      I went from a 7800 series regulator to a low-drop out regulator. I used the same i/o caps I used in the 7800 circuit. In production, the darned regulator
      Message 2 of 10 , Jan 24, 2013
      • 0 Attachment
        I went from a 7800 series regulator to a low-drop out regulator. I used the
        same i/o caps I used in the 7800 circuit. In production, the darned
        regulator oscillated. Spec sheet specified a cap value I was not on that
        precise value. So now I read manufacturer's specs very closely.



        Larry



        From: Electronics_101@yahoogroups.com
        [mailto:Electronics_101@yahoogroups.com] On Behalf Of jong kung
        Sent: Thursday, January 24, 2013 11:12 AM
        To: Electronics_101@yahoogroups.com
        Subject: [Electronics_101] (unknown)





        Dave / Larry,

        > Go to the 78XX datasheet. There they recommend a cap
        > input and output. Try

        I looked there and on one of the pages they have typical specs (input and
        output). The footnote says these were taken using 0.22uF on the input and
        0.1uF on the output. Another 78xx datasheet says to use 0.33uf on the input.
        These are for 19v input, 12v output. I don't really know how going with 24v
        input would affect the size needs of input capacitor (I'm gonna guess the
        larger the input ripple and bigger the input capacitor). But also the exact
        value might NOT be SUPER critical.

        You have not mentioned what is your current requirement.

        Jong





        [Non-text portions of this message have been removed]
      • John Popelish
        ... The specs for low drop-out regulators often include an internal resistance spec for the output capacitor. This can be very problematic, especially over a
        Message 3 of 10 , Jan 24, 2013
        • 0 Attachment
          On 01/24/2013 12:52 PM, Larry Beaty wrote:
          > I went from a 7800 series regulator to a low-drop out regulator. I used the
          > same i/o caps I used in the 7800 circuit. In production, the darned
          > regulator oscillated. Spec sheet specified a cap value I was not on that
          > precise value. So now I read manufacturer's specs very closely.

          The specs for low drop-out regulators often include an
          internal resistance spec for the output capacitor. This can
          be very problematic, especially over a wide temperature
          range and different batches of capacitors. I get around
          this by using a very low ESR capacitor (ceramic, X7R or X5R)
          and add a fixed resistor in series with it (usually around 1
          ohm), as the output bypass, that is right at the regulator.

          You can look at the frequency response for those regulators,
          on the data sheet, and see a peak, just below the frequency
          where the regulation rolls off. Put the RC time constant,
          actually f=1/(2*pi*R*C), on that peak, and you put a loss
          right at the frequency where the regulator is least stable.
          That loss sucks up the peak.

          This has worked, without fail, for me.

          This peak damping approach works well with voltage
          reference, shunt and series regulators also (i.e. 4040 and 431).

          --
          Regards,

          John Popelish
        • Larry Beaty
          Ouu, I like that John. My problem was the regulator input cap. Right on the hairy edge of spec sheet recommendation. But I like your approach to roll off
          Message 4 of 10 , Jan 24, 2013
          • 0 Attachment
            Ouu, I like that John. My problem was the regulator input cap. Right on
            the hairy edge of spec sheet recommendation.



            But I like your approach to roll off frequency response.



            Larry



            From: Electronics_101@yahoogroups.com
            [mailto:Electronics_101@yahoogroups.com] On Behalf Of John Popelish
            Sent: Thursday, January 24, 2013 1:22 PM
            To: Electronics_101@yahoogroups.com
            Subject: Re: [Electronics_101] (unknown)





            On 01/24/2013 12:52 PM, Larry Beaty wrote:
            > I went from a 7800 series regulator to a low-drop out regulator. I used
            the
            > same i/o caps I used in the 7800 circuit. In production, the darned
            > regulator oscillated. Spec sheet specified a cap value I was not on that
            > precise value. So now I read manufacturer's specs very closely.

            The specs for low drop-out regulators often include an
            internal resistance spec for the output capacitor. This can
            be very problematic, especially over a wide temperature
            range and different batches of capacitors. I get around
            this by using a very low ESR capacitor (ceramic, X7R or X5R)
            and add a fixed resistor in series with it (usually around 1
            ohm), as the output bypass, that is right at the regulator.

            You can look at the frequency response for those regulators,
            on the data sheet, and see a peak, just below the frequency
            where the regulation rolls off. Put the RC time constant,
            actually f=1/(2*pi*R*C), on that peak, and you put a loss
            right at the frequency where the regulator is least stable.
            That loss sucks up the peak.

            This has worked, without fail, for me.

            This peak damping approach works well with voltage
            reference, shunt and series regulators also (i.e. 4040 and 431).

            --
            Regards,

            John Popelish





            [Non-text portions of this message have been removed]
          • John Popelish
            ... If I used the RC on the output, the regulators worked well as long as I put plenty (at least a couple times the recommended) input capacitance. I saw no
            Message 5 of 10 , Jan 24, 2013
            • 0 Attachment
              7On 01/24/2013 02:32 PM, Larry Beaty wrote:
              > Ouu, I like that John. My problem was the regulator input cap. Right on
              > the hairy edge of spec sheet recommendation.

              If I used the RC on the output, the regulators worked well
              as long as I put plenty (at least a couple times the
              recommended) input capacitance.

              I saw no benefit from using an RC on the input. The
              unstable feedback loop is all based on the output voltage.
              But you certainly don't want to be tight on the input
              capacitance. And it needs to be physically, very close, to
              minimize the inductance between regulator and capacitor.
              Don't want to introduce an LC resonance to add to the
              regulation resonance.

              --
              Regards,

              John Popelish
            • jong kung
              John, ... Is this a well known method? Maybe you should forward this technique to the regulator manufacturer and get a footnote (if not money) assigned to you.
              Message 6 of 10 , Jan 24, 2013
              • 0 Attachment
                John,


                > > Ouu, I like that John. My problem was the
                > regulator input cap. Right on
                > > the hairy edge of spec sheet recommendation.
                >
                > If I used the RC on the output, the regulators worked well
                > as long as I put plenty (at least a couple times the
                > recommended) input capacitance.

                Is this a well known method?

                Maybe you should forward this technique to the regulator manufacturer and get a footnote (if not money) assigned to you.

                :-)


                Jong



                --- On Thu, 1/24/13, John Popelish <jpopelish@...> wrote:

                > From: John Popelish <jpopelish@...>
                > Subject: Re: [Electronics_101] (unknown)
                > To: Electronics_101@yahoogroups.com
                > Date: Thursday, January 24, 2013, 9:42 AM
                > 7On 01/24/2013 02:32 PM, Larry Beaty
                > wrote:
                > > Ouu, I like that John.  My problem was the
                > regulator input cap.  Right on
                > > the hairy edge of spec sheet recommendation.
                >
                > If I used the RC on the output, the regulators worked well
                > as long as I put plenty (at least a couple times the
                > recommended) input capacitance.
                >
                > I saw no benefit from using an RC on the input.  The
                > unstable feedback loop is all based on the output voltage.
                > But you certainly don't want to be tight on the input
                > capacitance.  And it needs to be physically, very
                > close, to
                > minimize the inductance between regulator and capacitor.
                > Don't want to introduce an LC resonance to add to the
                > regulation resonance.
                >
                > --
                > Regards,
                >
                > John Popelish
                >
                >
                > ------------------------------------
                >
                > Please trim excess when replyingYahoo! Groups Links
                >
                >
                >     Electronics_101-fullfeatured@yahoogroups.com
                >
                >
              • John Popelish
                ... As far as I can tell, the manufacturers are only interested in showing the absolute lowest cost recommendation for the use of their products, not the best.
                Message 7 of 10 , Jan 24, 2013
                • 0 Attachment
                  On 01/24/2013 06:17 PM, jong kung wrote:

                  > Is this a well known method?
                  >
                  > Maybe you should forward this technique to the regulator
                  > manufacturer and get a footnote (if not money) assigned
                  > to you.

                  As far as I can tell, the manufacturers are only interested
                  in showing the absolute lowest cost recommendation for the
                  use of their products, not the best.

                  By the way, the Chinese have made an art of cost reducing
                  power supplies to just below what actually works for more
                  than an month. I correct the designs, on electronics I buy,
                  all the time. I just fixed two flat screen monitor supplies
                  and two large, flat screen TVs supplies. Computor
                  motherboards are not the only products that have over worked
                  capacitors and unstable regulators.

                  --
                  Regards,

                  John Popelish
                • jong kung
                  John, ... It s worse than that. Recently I bought 3 thing ULTRA cheap online: (1) LONG USB printer cable (2) replacement battery for my laptop (3) replacement
                  Message 8 of 10 , Jan 24, 2013
                  • 0 Attachment
                    John,


                    > By the way, the Chinese have made an art of cost reducing
                    > power supplies to just below what actually works for more
                    > than an month.

                    It's worse than that. Recently I bought 3 thing ULTRA cheap online:

                    (1) LONG USB printer cable
                    (2) replacement battery for my laptop
                    (3) replacement power supply for my laptop

                    They ALL failed on arrival. Pushing the boundary and falling of past the boundary are two different things.


                    Jong
                  • Kerim F
                    As a manufacturer (not trader) I paid well for the components, the boards produced according to my designs and power supplies that I got directly from China.
                    Message 9 of 10 , Jan 24, 2013
                    • 0 Attachment
                      As a manufacturer (not trader) I paid well for the components, the boards produced according to my designs and power supplies that I got directly from China. Their global rate of failure, even after 3 years, doesn't exceed 3%. But I noticed when I buy a Chinese product from a local trader; it could be just garbage ;) In general, most traders look gaining fast money instead of a long-term good reputation as in the case of most manufacturers. I ended up believing that most traders look for the lowest quality but with an acceptable appearance and China comes to their rescue. So anytime a user complain, the trader says: I told you it is Chinese :) Some `passing by' traders don't mind even cheating. Perhaps some of you have already noticed that in some calculator, the solar cell is just a printed sheet of plastic :) Also some flashlights that have a moving bobbin to charge by hand an internal battery have just a normal cell battery (not rechargeable) and the moving coil has no connections at all ;) I mean a trader always knows what he imports (on his request) and is ready to accuse the product's origin when questioned ;)
                      Please note, I am not defending China or blame anyone, I just present facts I knew personally.

                      --- In Electronics_101@yahoogroups.com, John Popelish wrote:
                      >
                      > On 01/24/2013 06:17 PM, jong kung wrote:
                      >
                      > > Is this a well known method?
                      > >
                      > > Maybe you should forward this technique to the regulator
                      > > manufacturer and get a footnote (if not money) assigned
                      > > to you.
                      >
                      > As far as I can tell, the manufacturers are only interested
                      > in showing the absolute lowest cost recommendation for the
                      > use of their products, not the best.
                      >
                      > By the way, the Chinese have made an art of cost reducing
                      > power supplies to just below what actually works for more
                      > than an month. I correct the designs, on electronics I buy,
                      > all the time. I just fixed two flat screen monitor supplies
                      > and two large, flat screen TVs supplies. Computor
                      > motherboards are not the only products that have over worked
                      > capacitors and unstable regulators.
                      >
                      > --
                      > Regards,
                      >
                      > John Popelish
                      >
                    • Donald H Locker
                      Yes. It s not just the manufacturer. The person who specifies what is to be manufactured and what parts to use (sometimes by specifying a cost) set the
                      Message 10 of 10 , Jan 24, 2013
                      • 0 Attachment
                        Yes. It's not just the manufacturer. The person who specifies what is to be manufactured and what parts to use (sometimes by specifying a cost) set the quality. The manufacturer can do worse, but often does better.

                        Donald.
                        --
                        *Plain Text* email -- it's an accessibility issue
                        () no proprietary attachments; no html mail
                        /\ ascii ribbon campaign - <www.asciiribbon.org>

                        ----- Original Message -----
                        > From: "Kerim F" <ahumanbeing2000@...>
                        > To: "Electronics 101" <Electronics_101@yahoogroups.com>
                        > Sent: Thursday, January 24, 2013 10:46:27 PM
                        > Subject: [Electronics_101] Re: (unknown)
                        >
                        > As a manufacturer (not trader) I paid well for the components, the
                        > boards produced according to my designs and power supplies that I got
                        > directly from China. Their global rate of failure, even after 3 years,
                        > doesn't exceed 3%. But I noticed when I buy a Chinese product from a
                        > local trader; it could be just garbage ;) In general, most traders
                        > look gaining fast money instead of a long-term good reputation as in
                        > the case of most manufacturers. I ended up believing that most traders
                        > look for the lowest quality but with an acceptable appearance and
                        > China comes to their rescue. So anytime a user complain, the trader
                        > says: I told you it is Chinese :) Some `passing by' traders don't mind
                        > even cheating. Perhaps some of you have already noticed that in some
                        > calculator, the solar cell is just a printed sheet of plastic :) Also
                        > some flashlights that have a moving bobbin to charge by hand an
                        > internal battery have just a normal cell battery (not rechargeable)
                        > and the moving coil has no connections at all ;) I mean a trader
                        > always knows what he imports (on his request) and is ready to accuse
                        > the product's origin when questioned ;)
                        > Please note, I am not defending China or blame anyone, I just present
                        > facts I knew personally.
                        >
                        > --- In Electronics_101@yahoogroups.com, John Popelish wrote:
                        > >
                        > > On 01/24/2013 06:17 PM, jong kung wrote:
                        > >
                        > > > Is this a well known method?
                        > > >
                        > > > Maybe you should forward this technique to the regulator
                        > > > manufacturer and get a footnote (if not money) assigned
                        > > > to you.
                        > >
                        > > As far as I can tell, the manufacturers are only interested
                        > > in showing the absolute lowest cost recommendation for the
                        > > use of their products, not the best.
                        > >
                        > > By the way, the Chinese have made an art of cost reducing
                        > > power supplies to just below what actually works for more
                        > > than an month. I correct the designs, on electronics I buy,
                        > > all the time. I just fixed two flat screen monitor supplies
                        > > and two large, flat screen TVs supplies. Computor
                        > > motherboards are not the only products that have over worked
                        > > capacitors and unstable regulators.
                        > >
                        > > --
                        > > Regards,
                        > >
                        > > John Popelish
                        > >
                        >
                        >
                        >
                        >
                        > ------------------------------------
                        >
                        > Please trim excess when replyingYahoo! Groups Links
                        >
                        >
                        >
                      Your message has been successfully submitted and would be delivered to recipients shortly.