开云体育

Ran through a couple of mods and an article in the FT Newsletters that mention replacement of the OEM product detector quad with a set of Schottky diodes. Digging around those articles, one finds mention of HP 5082-2900 (a mesh type) as having the lowest noise figure.

Found this vendor as a source:

At least 32 and possibly more are on my to-buy list as upgrades for all of my '980s plus a bunch more period gear.

Hope this is of benefit to someone.

73 - Fred, N8YX

I've always found these diode-type articles interesting, and have wanted to actually take some before/after measurements. I found the QST article in the FT Newsletter that discussed this change as having a better low-frequency (AF) distortion (thanks for pointing it out!), although the FT newlsetter also mentioned lower audible noise. Both were anecdotal and didn't give any testing results or other evidence.

The 4082-2900 was HP's MPN for a 1N5711 Hot-Carrier Diode. You could buy them in "matched" lots. Do you know if there were any differences to a general 1N5711 part?


Back in the 60's and 70's HP's semiconductor division made some parts that no one else in the world did at the time for their test & measurement group, as well as general sale. But they spun it off with Agilent, which split it into Avago, which then merged with Broadcom. A lot of the old specialized parts are not available anymore (try to find a matched pair of detector diodes for a classic HP spectrum or power analyzer).

But like capacitors, technology has moved forward from that period. I suspect there are even better diodes available nowadays.? Certainly for mixer diodes (nowadays more likely to be passive MESFET mixers).

Please keep us updated: this one is really interesting to me ...

--
-Mat Breton, N8TW

Mat,

Interesting you mentioned the 1N5711. I have an entire parts drawer full of the things and have been using them in a number of mixer conversion projects - but all the data I can find on them states they're passivated (with a higher noise level than a mesh type).

Digging around on the 'Net this morning, I also came across mention of HP P/Ns 5082-2303, -2817 and -2818 as being mesh construction. May help with sourcing these. Plans are to get some -2900s and see what difference is made in my test FT-980 then convert the rest as they cycle through for maintenance and modifications.

73 - Fred, N8YX

Would it be possible for you to take before and after audio noise floor and signal measurements? I *think* it could be done with an HF signal generator and an audio analysis program on a PC (I would recommend?DL4YHF's Spectrum Lab because it already has noise floor and peak detection built in ...?).

I've never seen this done with any of the germanium->silicon diode replacement articles, and the guy who did the FT-980 mod took no measurements whatsoever.? This would be relevant to all sorts of older rigs, not just the FT-980 I think; if the passivation creates enough noise to show up in the 8mV (typical) signal level at Q3038, I think it has application in other rigs as well.

If done carefully, this might be the basis for a QST or QEX article ...

--
-Mat Breton, N8TW