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

Re: The joy of learning

Expand Messages
  • Joseph Charles
    Hi Henrik, Here is a variation on the idea I mentioned. This does work, but takes a VERY long time to fire. Referring to the attached gif: the solar cell
    Message 1 of 16 , Apr 8, 2006
    • 0 Attachment
      Hi Henrik,

      Here is a variation on the idea I mentioned. This does
      work, but takes a VERY long time to fire.

      Referring to the attached gif: the solar cell charges
      the 100uF cap and the two-transistor oscillator
      oscillates away. Every time the NPN transistor turns
      off, the 100uH inductor resists the change in current
      through it and produces a large voltage spike at its
      junction with D2. A small amount of charge is feed via
      D2 to C3, the main storage capacitor for the solar
      engine. The voltage on C3 slowly rises until the Solar
      Engine eventually (and I stress 'eventually' !) fires.

      I have one oscillating away in low light, but the
      voltage on the storage cap rises EXCRUCIATINGLY
      slowly. Under my ceiling light, 100W incandescant,
      with the circuit on a table, I think I may be here all
      night. I got a 'pop' in about half an hour this
      afternoon, with the circuit near a window. The day is
      overcast and quite dull - it's heading to winter here,
      and we've been given a sneak preview. So daylight is
      better than artificial light. It very much depends on
      the current output of your solar cell. Calculator
      cells generally produce very little current.

      I state on the attachment that my solar cell's current
      was around 90uA. This is enough current to run the
      oscillator (indeed, anything over about 10uA will do
      it seems), but with this current you may well be
      waiting forever for a 'pop'. The daylight 'pop' I got
      was with a current just under 1mA. Sorry for any
      confusion. It would be a pain to change the GIF - read
      on.

      I don't think this is much of a success, but does make
      a very nice low light pummer. Just get rid of C3 and
      the SE, and make D2 an ultrabright LED (anode as
      shown, cathode to GND).

      Sorry for the poor quality of the schematic. I've had
      to do it on one computer and scan it into the one I'm
      using to post. Technology problems!

      I cannot upload it to the files section as there is
      not enough space. If you do not receive the BEAM post
      by email, and therefore have not received this
      attachment, I can email it to you if you are
      interested.

      The lack of current from the solar cells in low light
      is very much the issue here.

      Cheers

      Jo C


      --- In beam@yahoogroups.com, "Henrik Pettersson"
      <henrikp@...> wrote:
      >
      > That was way to advanced for me in such abstract
      form. If you try it out
      > later please let me know.






      ____________________________________________________
      On Yahoo!7
      Messenger - Make free PC-to-PC calls to your friends overseas.
      http://au.messenger.yahoo.com
    • Joseph Charles
      Hooray! A pop! (After...err...a couple of hours, maybe.) OK, I realised I had a 1381N in my Miller SE and was going to have to wait until the voltage on the SE
      Message 2 of 16 , Apr 8, 2006
      • 0 Attachment
        Hooray! A pop! (After...err...a couple of hours, maybe.)

        OK, I realised I had a 1381N in my Miller SE and was going to have to wait
        until the voltage on the SE cap hit around 4V. The voltage was around 3.6V at
        this stage, and I was sick of waiting, and it's nearly bedtime, so I subbed a
        1381J in its place, thus reducing the needed voltage by around 800mV. And
        voila!

        So, it does work. Just don't sit there watching and waiting....! Find something
        useful to do!

        (Don't worry, I was teaching myself how to program in Python while I waited. A
        very nice language, if you're interested.)

        Cheers

        Jo

        --- In beam@yahoogroups.com, Joseph Charles <jodicalhon@...> wrote:

        > Under my ceiling light, 100W incandescant,
        > with the circuit on a table, I think I may be here all
        > night.
      • David Treadwell
        Jo, YEAH for Python! I absolutely love the language! I started picking it up about six months ago, and I ve abandoned almost everything else. It is the most
        Message 3 of 16 , Apr 8, 2006
        • 0 Attachment
          Jo,

          YEAH for Python! I absolutely love the language! I started picking it
          up about six months ago, and I've abandoned almost everything else.
          It is the most enjoyable language I've ever had the pleasure to
          program in.

          If you remember (or not) a few months back, I was working on a
          program to solve kinematic linkages. So that more BEAMers could use
          it, I originally wrote it in Visual Basic for Excel. I hated every
          second of having to write in that language. I'm now rewriting the
          code in Python, and it's much more powerful and easy to understand.

          And congratulations on getting you SE to pop!

          --David T.

          On Apr 8, 2006, at 9:23 AM, Joseph Charles wrote:

          > Hooray! A pop! (After...err...a couple of hours, maybe.)

          <snip>

          > (Don't worry, I was teaching myself how to program in Python while
          > I waited. A
          > very nice language, if you're interested.)
          >
          > Cheers
          >
          > Jo
        • Joseph Charles
          Just one more comment, and then I ll go to bed. The voltage on the SE cap is rising at around 1mV every 2 or 3 seconds with 100uA solar cell output If this
          Message 4 of 16 , Apr 8, 2006
          • 0 Attachment
            Just one more comment, and then I'll go to bed.

            The voltage on the SE cap is rising at around 1mV every 2 or 3 seconds with
            100uA solar cell output

            If this rate continued, it should reach ~3.6V in about 2 - 3 hours. However, a
            capacitor's charging rate slows as it approaches capacity. My capacitor is
            rated 6.3V, so it's likely still on a mostly linear part of its charging curve
            (...err...I think!). I haven't accurately measured the voltage on the cap after the
            'pop', but it seemed to be down to around 1V.

            G'night all.

            Jo



            --- In beam@yahoogroups.com, "Joseph Charles" <jodicalhon@...> wrote:
            >
            > Hooray! A pop! (After...err...a couple of hours, maybe.)
          • Joseph Charles
            Hi David, Yes, I do remember that. It came out of the thread on whatsisname s (I ve forgotten that!) walkers, wandering down the beach, etc. I think you are
            Message 5 of 16 , Apr 8, 2006
            • 0 Attachment
              Hi David,

              Yes, I do remember that. It came out of the thread on whatsisname's (I've
              forgotten that!) walkers, wandering down the beach, etc.

              I think you are possibly just a 'little' more advanced in Python than I. I've
              previously only used a little BASIC, and then not much. I'm quite enjoying
              myself, though. My brain doesn't hurt TOO much!

              Thanks for the congrats!

              Now...where's that bed...this time for sure...

              Cheers

              Jo

              --- In beam@yahoogroups.com, David Treadwell <cp2fe@...> wrote:

              > If you remember (or not) a few months back, I was working on a
              > program to solve kinematic linkages. So that more BEAMers could use
              > it, I originally wrote it in Visual Basic for Excel. I hated every
              > second of having to write in that language. I'm now rewriting the
              > code in Python, and it's much more powerful and easy to understand.
              >
              > And congratulations on getting you SE to pop!
              >
              > --David T.
              >
            • hempa_pett
              Very cool that you tried it! It seems rather random what messages reaches my computer. Didn t see this until I logged onto the beam group online. And here I
              Message 6 of 16 , Apr 8, 2006
              • 0 Attachment
                Very cool that you tried it! It seems rather random what messages
                reaches my computer. Didn't see this until I logged onto the beam
                group online. And here I can't get to the attached gif. :-( Can I
                trouble you to send it to me directly, please? With the schematics
                and a couple of spare hours I might understand it even.

                An update: My first beam bot has seen the light of day! It worked
                well enough with four small calculator cells which I mostly on
                gutfeeling and random testing decided to put in two parallell serial
                pairs. In the light from my ceiling lamp it took about an hour to get
                a pop I think. The bot is very ugly since I wanted results fast.
                Furthermore my own eye design is less than satisfactory. It is just a
                transistor on each motor with a photoresistor on the base. It might
                be obvious to you guys why it is not working (or not working well
                enough to be able to tell any difference even with one eye covered),
                but I thought it was very clever of me. More careful breadboarding
                next time. Anyhow it has a marching speed of about 6 mm/min in cloudy
                daylight.

                Cheers!/Henrik

                --- In beam@yahoogroups.com, Joseph Charles <jodicalhon@...> wrote:
                >
                > Hi Henrik,
                >
                > Here is a variation on the idea I mentioned. This does
                > work, but takes a VERY long time to fire.
                >
                > Referring to the attached gif: the solar cell charges
                > the 100uF cap and the two-transistor oscillator
                > oscillates away. Every time the NPN transistor turns
                > off, the 100uH inductor resists the change in current
                > through it and produces a large voltage spike at its
                > junction with D2. A small amount of charge is feed via
                > D2 to C3, the main storage capacitor for the solar
                > engine. The voltage on C3 slowly rises until the Solar
                > Engine eventually (and I stress 'eventually' !) fires.
                >
                > I have one oscillating away in low light, but the
                > voltage on the storage cap rises EXCRUCIATINGLY
                > slowly. Under my ceiling light, 100W incandescant,
                > with the circuit on a table, I think I may be here all
                > night. I got a 'pop' in about half an hour this
                > afternoon, with the circuit near a window. The day is
                > overcast and quite dull - it's heading to winter here,
                > and we've been given a sneak preview. So daylight is
                > better than artificial light. It very much depends on
                > the current output of your solar cell. Calculator
                > cells generally produce very little current.
                >
                > I state on the attachment that my solar cell's current
                > was around 90uA. This is enough current to run the
                > oscillator (indeed, anything over about 10uA will do
                > it seems), but with this current you may well be
                > waiting forever for a 'pop'. The daylight 'pop' I got
                > was with a current just under 1mA. Sorry for any
                > confusion. It would be a pain to change the GIF - read
                > on.
                >
                > I don't think this is much of a success, but does make
                > a very nice low light pummer. Just get rid of C3 and
                > the SE, and make D2 an ultrabright LED (anode as
                > shown, cathode to GND).
                >
                > Sorry for the poor quality of the schematic. I've had
                > to do it on one computer and scan it into the one I'm
                > using to post. Technology problems!
                >
                > I cannot upload it to the files section as there is
                > not enough space. If you do not receive the BEAM post
                > by email, and therefore have not received this
                > attachment, I can email it to you if you are
                > interested.
                >
                > The lack of current from the solar cells in low light
                > is very much the issue here.
                >
                > Cheers
                >
                > Jo C
                >
                >
                > --- In beam@yahoogroups.com, "Henrik Pettersson"
                > <henrikp@> wrote:
                > >
                > > That was way to advanced for me in such abstract
                > form. If you try it out
                > > later please let me know.
                >
                >
                >
                >
                >
                >
                > ____________________________________________________
                > On Yahoo!7
                > Messenger - Make free PC-to-PC calls to your friends overseas.
                > http://au.messenger.yahoo.com
                >
              • Joseph Charles
                What configuration have you used for the eyes, Hendrik? (A) (B) ... (LDR) c
                Message 7 of 16 , Apr 11, 2006
                • 0 Attachment
                  What configuration have you used for the eyes, Hendrik?

                  (A) (B)
                  _______________ +ve __________________
                  | | | |
                  (LDR) c | (motor)
                  |___b (NPN) | |
                  e (LDR) c
                  | |____b (NPN)
                  (motor) e
                  | |
                  ----------------*---------- GND ------------------*------------

                  Configuration A will give greater variance in motor speed vs light levels for
                  the lowest parts count.

                  If you have used configuration B, you will need a resistor of between 1k
                  and 10k, depending on your LDR, between the transistor's base and GND
                  to improve the performance.

                  Jo

                  (I hope the ASCII art comes through OK!)

                  --- In beam@yahoogroups.com, "hempa_pett" <henrikp@...> wrote:

                  > Furthermore my own eye design is less than satisfactory. It is just a
                  > transistor on each motor with a photoresistor on the base. It might
                  > be obvious to you guys why it is not working (or not working well
                  > enough to be able to tell any difference even with one eye covered),
                  > but I thought it was very clever of me.
                • Joseph Charles
                  Blimey! What a mess! Looks fine in the reply box though. Try clicking reply to my last post. See if that works. Jo
                  Message 8 of 16 , Apr 11, 2006
                  • 0 Attachment
                    Blimey! What a mess!
                    Looks fine in the reply box though. Try clicking reply to my last post.
                    See if that works.

                    Jo


                    --- In beam@yahoogroups.com, "Joseph Charles" <jodicalhon@...> wrote:
                    >
                    > What configuration have you used for the eyes, Hendrik?
                    >
                    > (A) (B)
                    > _______________ +ve __________________
                    > | | | |
                    > (LDR) c | (motor)
                    > |___b (NPN) | |
                    > e (LDR) c
                    > | |____b (NPN)
                    > (motor) e
                    > | |
                    > ----------------*---------- GND ------------------*------------
                    >
                    > Configuration A will give greater variance in motor speed vs light levels for
                    > the lowest parts count.
                    >
                    > If you have used configuration B, you will need a resistor of between 1k
                    > and 10k, depending on your LDR, between the transistor's base and GND
                    > to improve the performance.
                    >
                    > Jo
                    >
                    > (I hope the ASCII art comes through OK!)
                    >
                    > --- In beam@yahoogroups.com, "hempa_pett" <henrikp@> wrote:
                    >
                    > > Furthermore my own eye design is less than satisfactory. It is just a
                    > > transistor on each motor with a photoresistor on the base. It might
                    > > be obvious to you guys why it is not working (or not working well
                    > > enough to be able to tell any difference even with one eye covered),
                    > > but I thought it was very clever of me.
                    >
                  • Joseph Charles
                    Works when viewing with IE, which I did the art in. Not so good in Safari, though still understandable - to me anyway! Jo
                    Message 9 of 16 , Apr 11, 2006
                    • 0 Attachment
                      Works when viewing with IE, which I did the art in.

                      Not so good in Safari, though still understandable - to me anyway!

                      Jo

                      --- In beam@yahoogroups.com, "Joseph Charles" <jodicalhon@...> wrote:
                      >
                      > Blimey! What a mess!
                      > Looks fine in the reply box though. Try clicking reply to my last post.
                      > See if that works.
                    Your message has been successfully submitted and would be delivered to recipients shortly.