Announcement

Collapse
No announcement yet.

Possibly THE WORST PI EVER designed?

Collapse
X
Collapse
 
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
Previous template Next
  • #1

    Possibly THE WORST PI EVER designed?

    I present to you the Hobby Electronics "Diana" from 1981 (which I built as a student).

    I tried MANY times to get the setup to work properly but it simply drags the "9V" - bottom ZD1 - rail down and will not set properly. It worked, of sorts, but never had anything like decent performance.

    Would anyone care to take a look at it an propose /hypothesise why it sucks?

    As far as I can see, the intergrator is beyond weird, the switch arrangement doesnt seem to make any sense(itivity).

    Maybe I'm missing something and this is a very clever design, but somehow I think NOT!

    If fact it's SUCH a bad design, it made me spell the word WORST wrong!

    HE_Diana.pdf

    HE_Diana_2.pdf
    Tags: None
  • #2
    This project was in the September 1981 issue. I have found the November issue, and this errata noting a wrong resistor value was published:



    I have the October issue somewhere, If I can find it, I'll look for any errata in it.

    Attached jpg
    Attached Files

    Comment

    • #3
      good project for 81 year. thank you Sean. but yes, as for first look those 4016 clues seem sure weird connected.
      Diana is obstinate girl, like they all are.

      Comment

      • #4
        Hi Brother,

        I read the errata (I bought the kit from a company called Magenta Electronics) so it had the corrected parts list. It STILL has problems. I've modded just about EVERYTHING on it and the performance is still around 10cm on a 1 Euro coin, It seems the whole design was just flawed.

        Thanks for adding the November print. I have EVERY hobby magazine for around 15 years in my loft. I have been scanning them in one by one when I have the time (around 1500 magazines) with the intention of putting them on a website but somehow I just dont seem to find the time to finish the job.

        As this was the September issue, I was bought the kit for my birthday. I built it and always thought it could be better but I had fun with it and I would like to make it work properly if someone can figure out what the design issue is and if it can be easily corrected.

        Comment

        • #5
          Here is the Sept 81 issue of HE, from the American Radio History site ( a great resource, many UK mags, like Elektor, HE, Everyday electronics, PE, Ambit/Radio&Electronics World ):
          https://worldradiohistory.com/UK/Hob...cs-1981-09.pdf

          Comment

          • #6
            Originally posted by Sean_Goddard View Post
            Hi Brother,

            I read the errata (I bought the kit from a company called Magenta Electronics) so it had the corrected parts list. It STILL has problems. I've modded just about EVERYTHING on it and the performance is still around 10cm on a 1 Euro coin, It seems the whole design was just flawed.

            Thanks for adding the November print. I have EVERY hobby magazine for around 15 years in my loft. I have been scanning them in one by one when I have the time (around 1500 magazines) with the intention of putting them on a website but somehow I just dont seem to find the time to finish the job.

            As this was the September issue, I was bought the kit for my birthday. I built it and always thought it could be better but I had fun with it and I would like to make it work properly if someone can figure out what the design issue is and if it can be easily corrected.
            They only claim that it is capable of detecting small coins at a fair depth, whatever that means. 4" may be all it is capable of achieving.
            What is the pulse rate and TX pulse width? Also, have you noticed that the damping resistor is 100R? Seems a bit low to me.

            Comment

            • #7
              I know very little about PI's , but a few general comments:
              The homebuilt coil has to be a weak point, that could cripple even a good circuit design. I notice that ready-built coils were available from one retailer, your project may have worked better with such coil. And true, that damping resistor does seem low in resistance.
              There's a fair bit of CMOS in it, the schmitt inverters, the 3130 opamps, easy for a teenage novice to damage with static. And it appears the unused 4016 gate is left unconnected, it would be wise to join it up to something, like 0 volts.
              And I agree, that arrangement with the three switches is baffling. Is that supposed to do some kind of SPCO switch action?
              Does the PCB foil match the schematic ? It's not unusual for the PCB to be right, but the published diagram to have errors.

              No wonder the lady on the magazine's front cover is concentrating so hard on operating the machine.

              Comment

              • #8
                maybe layout is good vs the circuit. let to see the pcb both sides.

                Comment

                • #9
                  KT315 it is a single sided SRPB PCB. I will photograph it later and post on here.

                  Skippy...NICE ONE BRO many thanks and you are right about the CMOS. I destroyed the first build then took the correct precautions. Those CA3130 amps have ALWAYS been flaky, even when new. Never one of my favourites.

                  FTx = 249Hz

                  TxPW = 152us

                  Sample Delay = 53us (HOLY HADES)!!

                  Sample PW = 38.3us

                  I think these aremiles of anything that would make a sensitive unit. Will post more later.

                  And I'll change the dampoig resistor as a starter

                  Comment

                  • #10
                    Originally posted by Sean_Goddard View Post
                    FTx = 249Hz

                    TxPW = 152us

                    Sample Delay = 53us (HOLY HADES)!!

                    Sample PW = 38.3us

                    I think these aremiles of anything that would make a sensitive unit. Will post more later.

                    And I'll change the dampoig resistor as a starter
                    The damping resistor and the sampling delay are most likely the main culprits. You could also try increasing the sample pulse width.

                    Comment

                    • #11
                      I noticed that 74hc14 is driving 4016 which is higher impedence than 4066, perhaps it makes no difference.
                      Driving current is small too, ~5mA. Maybe a recipe for misfire at times with 4066.

                      Comment

                      • #12
                        Well I found one major problem, the 3V9 Zener is cooked and is dropping 9V so that is one thing. I have "upped the Tx frequency to 1024Hz and the Tx pulse to 60us. Fropped the sample delay to 15us and the pw to 40us and perfromance is MUCH better thgat it has ever been.

                        Still just playing with this design as the SMPI blows it out of the water. I built the Silverdog kit and got 340mm on a 1 euro coin with a simply coil. With the super fast coil I built more like 420mm which ius better than the Whites machine!!

                        I have a PD3200, a Barracuda and a Dave Emery "Worm Burner" to play with. The WB is stupidly deep but uses some proprietory IP of Daves.

                        Comment

                        Previous template Next
                        Working...
                        X