An old AM radio that passed down to me from my grandparents finally ceased to function after I accidentally dropped it and old parts inside got damaged. I ripped all damaged parts out and replaced them with modern equivalents. There is one thing I am having trouble replacing: the detector diode.

The original diode is 2AP9 (Chinese part, was THE detector diode in Chinese radio designs, similar to 1N34A,) a germanium detector diode that got shattered. I don't have any germanium parts any more (the germanium PNP transistors was replaced with silicon ones, but the amplifier is adjusted accordingly to make them work correctly after a few resistors get replaced)

However I do have a few Schottky rectifiers 1N5819. Are those good enough as detector diodes in AM radio? Do I need to adjust some caps and resistors to compensate the 110pF junction capacitance? (2AP9 have 1pF)

  • \$\begingroup\$ I guess a BAT85 would be a better try. Maybe you have a G transistor left from which you could use the BE junction? \$\endgroup\$ – Wouter van Ooijen Jan 16 '15 at 16:42
  • \$\begingroup\$ I dumped and replaced all old transistors actually. The 3AG1's have a planar BE junction and I am a bit worried about its junction capacitance. Those old parts didn't have such a specification! \$\endgroup\$ – Maxthon Chan Jan 16 '15 at 17:01
  • \$\begingroup\$ @WoutervanOoijen That radio is older than me! I wouldn't trust anything in it at all (maybe except the PCB board itself) but I just didn't bother to replace all the caps and resistors. \$\endgroup\$ – Maxthon Chan Jan 16 '15 at 17:05
  • 1
    \$\begingroup\$ Why not just try it? \$\endgroup\$ – Andy aka Jan 16 '15 at 17:27

If 2AP9 is similar to 1N34A, you can surely still find those, although I'm not sure about availability in China (your location).

As for using Schottkys as substitute, 1N5711 (a ~0.2V [@ 25C] UHF Schottky) does work as substitute in some designs; 1N5711 has only 2pF capacitance. YMMV if it will actually work in your radio, but it seems to have a better chance than 1N5819.

There are some research papers on using (CMOS-fabricated) Schottkys as ultrawide-band AM detectors: "The input and output matching of the detector is better than -10 dB from 0-10.3 GHz". I'm guessing that a worry might be getting too much band and thus possibly HF noise when using a Schottky instead of a Ge diode for an AM radio in the classic AM radio band.

  • \$\begingroup\$ +1 Lots of 1N34A diodes offered from China these days. Presumably they're not all rebadged 1N4148s. Maybe a BAT62 (0.35pF typical at 0V). \$\endgroup\$ – Spehro Pefhany Jan 17 '15 at 2:59
  • \$\begingroup\$ Yes BAT62 has even better-looking stats capacitance-wise as you say (and about the same 0.2V drop at 25C). But the OP might not like that it is only a surface-mount part. Trying to adapt it to an old THT board might end up adding significant parasitic capacitance. \$\endgroup\$ – Fizz Jan 17 '15 at 3:15
  • \$\begingroup\$ Those selling "1N34A" offers rebadged 1N4148 at inflated prices and the 0.7V drop will probably eat away the signal completely, since the receiver resonator is pretty weak. The board is probably 25-30 years old so it was all single sided THT. I will give 1N5711 a go. \$\endgroup\$ – Maxthon Chan Jan 17 '15 at 13:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.