• This topic has 1 reply, 2 voices, and was last updated 1 week, 1 day ago by Guenni75.
Viewing 1 reply thread
  • Author
    Posts
    • #8275
      LectronFan
      Moderator

      Hi all,

      The more I experiment with Lectron, the more I spot differences between the different transistors and their use.

      They each have their advantages

      Let’s look at the Legacy Germanium transistors.

      Their advantage is the lower Base Emitter voltage (about 0,3V), ideal for quick switching circuits. They also have a build in resistor (680K, 330K and 100K), which saves mounting space.
      But this is also a disadvantage ! Due to the resistors, some circuits simply wont work or need  adaptation.

      Also, Germanium transistors tend to drift. This means that their amplification factor Hfe is dependant of the ambient temperature. The same transistors differ also. You can try this by building circuit 51a and 51b in the legacy manual and watch the differences in meter reading, even if they have the same resistor value build in !

      So, as for the Silicon transistors there’s much more stability.

      Since they don’t have the resistor build in, its much easier to adapt their Hfe to the circuit. They don’t drift by temperature which makes the circuit perform better. They need more space on the board though.

      Since they don’t leak current (Germanium transistors do), they can be better used for DC coupled stages.

      In our Lectron environment, the Ubungs System 1102 and OpAmp labor are 2 examples of a fine marriage between Germanium and Silicon transistors.

      And this is what I tend to strive for too ! Many future constructions will show off the advantages of both types.

      What are your opinions ?

      Many greetings

       

    • #8278
      Guenni75
      Moderator

      Hi all,

      I have been working with Germanium transistor since about 1960 – as one of my several hobbies. The history with their different techniques was very diverse since 1948: point contact / grown junction / alloy junction / diffusion alloy / drift / surface barrier / mesa / etc., even npn symmetric*.

      Every production process had a special goal, so every “type class” had advantages – except temperature, as Frank mentioned.  I love the “old” outlines, no plastic, no uniform look. So I am collecting even the very first pocket radios since 1954 with the TO-22 outline from Texas Instruments, Hitachi and Sony. Furthermore the glass versions like OC44 … OC75 or OC602 … OC614. And these transistors are still working very good since that time without problems.

      Germanium npn transistors in glass case were the OC139…OC141, described as symmetric* and could be replaced in some applications by two diodes (…?). Philips designed some pocket radios with the OC140 and OC74 as complementary push-pull output stage with OC75 as driver and a subminiature hearing aid transistor OC59 as pre-amplifier. Looking into such radios I must always laugh… that’s the diversity I like – heaven knows why. The look into radios since 1970 is most boring to me.

      This very special amplifier is since many years the last active Lectron block in my superhets with pnp types:

       

      The 1-transistor-reflex-superhet I introduced a while ago was another good example for the very special usage of (selected) germanium AM BC pnp transistors – only some Japanese types (drift types ?) worked without changing the bias resistor (or other parts of the circuit). European types only very few with some changes, Russian types were very “special”: extreme amplification / sensivity, oscillation etc.

      The germanium transistors in the Lectron blocks (AF: AC173, AC151 and RF: AF126) with the built-in bias resistor may be bypassed by using the (universal) transistor blocks with a socket. Then npn or pnp, Ge or Si, RF or AF, low power or medium power types may be used depending of the actual problem / circuit.

      In this photo from ~ 1970 Max Guerth used in this experiment a couple of this socket blocks:

       

      Silicon types are always first choice, but in circuits / models / applications before ~ 1970 I like to go “old”  – with all pro’s and con’s.

      Best,

      G.

      • This reply was modified 1 week, 1 day ago by Guenni75.
      • This reply was modified 1 week, 1 day ago by Guenni75.
Viewing 1 reply thread
  • You must be logged in to reply to this topic.