jcarr said:
It gets even worse. I have a Wireless World article somewhere from the 50s describing a low-distortion audio oscillator design (possibly by Bailey) that specifically warns against using carbon resistors because of the distortion problem. And this is in a valve oscillator, mark you.
There are lots of ways in which real components can upset practical circuits, but the best way to improve something is to design it correctly in the first place. Circuit design is far more important.
There are a few things that vary their characteristics with applied voltage. Just think of all those ceramic caps, transistors and tubes (valves).
Knowing something about your parts makes you a better designer, like an aircraft mechanic needs to know about his bolts and rivits.
I don't know why carbon resistors do this either. But they used to make microphones from carbon packed a little loose.
-Chris (old thread 'til Nixie found it!
)
Knowing something about your parts makes you a better designer, like an aircraft mechanic needs to know about his bolts and rivits.
I don't know why carbon resistors do this either. But they used to make microphones from carbon packed a little loose.
-Chris (old thread 'til Nixie found it!
)
http://www.barthelectronics.com/pdffiles/Pulse Catalog.pdf#search='Carbon%20composition%20resistor%20voltage%20coefficient'
(pages 4 - 5)
These guys do pulse testing with resistors and no interest in audio... an are therefore credible.
(pages 4 - 5)
These guys do pulse testing with resistors and no interest in audio... an are therefore credible.
Hi Diy-ers
Very interesting topic for sure. But nobody spoke about wirewound ones . These gave , in my opinion, better results in tube amps . I would ask what Vishay do you speak about ? SFERNICE ones (SFERNICE is now part of Vishay) ? imo the best for mf production . I found a 10000 parts stock for cheap and tried all the values i could use . In low signal (phono/mic preamps) these are killer and AB cc junk . Cheers
Very interesting topic for sure. But nobody spoke about wirewound ones . These gave , in my opinion, better results in tube amps . I would ask what Vishay do you speak about ? SFERNICE ones (SFERNICE is now part of Vishay) ? imo the best for mf production . I found a 10000 parts stock for cheap and tried all the values i could use . In low signal (phono/mic preamps) these are killer and AB cc junk . Cheers
I am wondering if any of you, 'experts' know about 'excess noise'? Do you? What about carbon composition resistors, any ideas about 'excess noise'? Now, for pulse response with a bandwidth of 350 MHz or so, exess noise might not be too important, but the risetime of the resistor (1nS) would be VERY important. So much for the pulse measurers. Audio requires low distortion and low 'excess noise'. Many quality metal film, bulk metal, and wirewould resistors fit this model. Vishay bulk metal 'may' be overkill. Try and see.
Actually, this was well-known and extensively discussed in the tube world for decades. See, for example, the article I referenced earlier. The comprehensive book "Valve Amplifiers" discusses excess noise and modulation noise.
John, you gotta venture out of the Dark Side now and then and revisit the Cult of the Glowing.
John, you gotta venture out of the Dark Side now and then and revisit the Cult of the Glowing.
SY said:
1955-57! I was just a tadpole. Given the date is it safe to assume it deals exclusively with carbon comps? I'm curious if this behaviour is to varying degrees common to any resistor which doesn't use a metallic element. Tantalum and carbon film for example.
(Offtopic: While spending a few days visiting my folks I clipped all the worthwhile articles from my 15 years of 70's-80's vintage Audio mags in the garage. I'ld forgotten what a wonderful mag it was when Heyser was on staff.)
Tantalum is a metal and makes a great resistor. I measured its 'excess noise' with a Quantech noise analyser. It is as good as a typical metal film. I use tantalum film pots at the input of my ultra low noise phono inputs with great success. The equivalent noise of my input is 0.4nV/rt Hz or 10 ohms.
Rdf,
thanks for that reference you found and quoted. I think it gives a good idea of the problem with carbon comp. and unless one prefers to go deeper into the physics, I think it serves as an answer to my earlier question.
All,
While there has been many discussions about resistors and also about volume pots (and similar), I can't remember seeing any discussion about trim pots. The types usually available seems to be cermet and carbon (comp or film, I don't know). AFAIK cermet is more reliable and less noisy, but maybe there are other factors to consider too? Well, I can, of course, come up with one such factor immediately. The price.
thanks for that reference you found and quoted. I think it gives a good idea of the problem with carbon comp. and unless one prefers to go deeper into the physics, I think it serves as an answer to my earlier question.
All,
While there has been many discussions about resistors and also about volume pots (and similar), I can't remember seeing any discussion about trim pots. The types usually available seems to be cermet and carbon (comp or film, I don't know). AFAIK cermet is more reliable and less noisy, but maybe there are other factors to consider too? Well, I can, of course, come up with one such factor immediately. The price.
- Status
- Not open for further replies.