Transistor quality over the years: Motorola vs. Philips vs. Fairchild, etc...

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Joined
Aug 7, 2009
Messages
35
Location
Oklahoma City, Oklahoma
Hey all,

About to lifetime buy old-stock transistors (BC184/214, BC550/560, BD137/138, BD139/140) and have the choice of grades, brands and various years going back to around the early 80's and would humbly appreciate opinions based around these parameters relative to your brand usage, volume, likes, dislikes, issues, non-issue issues, good/bad batches, etc over time... And of course all else equal comparative sound impressions that you may have noted.

Pretty satisfied with these specific part numbers and all brand experiences are certainly welcome!

Thank you!

Chris

 
Hate to sound like a broken record, but make sure you trust your source.

I have not experienced major issues with properly sourced common brands of any of the above. If given the choice, any from Fairchild, Siliconix, TI, Motorola, ST Microelectronics, National, OnSmi, etc would be equally fine with me. I’d be less inclined to buy generic equivalents from ECG, NTE etc simply because they tended to provide one part that covered the middle of the specs of many different part numbers.
 
Thanks rackmonkey,

A trusted source and no generics to be sure.

Over the years surely every manufacturer understandably has had hiccups and every designer has a preferred part/maker given the choice just as every volume user may and I'd love to hear about it all.

Not all identically-named part numbers are created equal.

Regards,

Chris



 
MediaWorksOKC said:
every designer has a preferred part/maker given the choice just as every volume user may and I'd love to hear about it all.

Not all identically-named part numbers are created equal.
A well-designed circuit should not depend on the transistor's specific performance, and be ready to deal with tolerances.
 
JohnRoberts said:
old parts are old.
Luckily.... good old quality....

ruffrecords said:
And they quite probably change properties over time. The 1980s parts you buy today probably sound different to the way they sounded originally.
But not more different than new ones with the same component name but only with a portion of the silicon which was used in the 70's or 80's and maybe even another topology which fulfills the specs too.  Best example is the 2N3055. 
 
ruffrecords said:
And they quite probably change properties over time. The 1980s parts you buy today probably sound different to the way they sounded originally.

Cheers

Ian
I am not aware of any mechanism that degrades semiconductors from just sitting unpowered, ASSuming hermetic packaging. Exposed to difficult environments they might suffer some surface oxidation, but I have never seen or heard of that in any significant numbers.

If anything semiconductor processes have dramatically improved over the decades with improved precision and feature resolution (photographic resolution). Modern silicon wafers may have different sheet resistivity than old media, and much higher purity, so the same old mask designs run across modern semiconductor production lines could be measurably improved (different).

The myth that older things are automatically better can be easily disproved, just ask any old person. :-( The device performance of some iconic old parts (like 2n3055) were embarrassingly slow. No self respecting device maker would sell a modern device with that "old" level performance, but they are not above branding new parts with old part numbers to win sales. 

JR
 
JohnRoberts said:
I am not aware of any mechanism that degrades semiconductors from just sitting unpowered, ASSuming hermetic packaging.
JR

Lots of work has been and is being done in this area. Most of it applies to modern devices (mostly FETs) but there is no reason to think that similar mechanisms do not occur in transistors made last century.

Here is one example paper: https://parts.jpl.nasa.gov/mmic/4.PDF
which came from doing this search:

https://duckduckgo.com/?t=canonical&q=semiconductor+performance+degradation+with+time&ia=web

Cheers

Ian
 

Latest posts

Back
Top