Quite the contrary. It might help to read the literature. In my own testing (and that of others with me as a test rat), I was able to distinguish these things ears-only. And I don't claim any kind of golden ear status. I was just curious and I do trust my ears. See, for example, my article in Linear Audio Volume 2.
Entirely possible from a poor end connection- or for inappropriate sizing for a specific circuit position. That part of what he's saying is eminently plausible. True in general of carbon film? I doubt that, but I don't think Ed claimed it, either.
You've taken all the fun out of it. The issue was with a carbon film vs another type. I actually think he misread the specs as CF usually are nowhere near 25 or 50 ppm.
Now if resistors measure the same low numbers on both AC distortion and DC excess noise, then I would be skeptical of reported differences. It is when the reports match the measurements that even without formal testing, it gets interesting.
Are your DBTs different ? Isn´t DBT the "gold standard" and infallible ?
Quote: Dr. Floyd Toole: "It turns out that, within very generous tolerances, humans are insensitive to phase shifts. Under carefully contrived circumstances, special signals auditioned in anechoic conditions, or through headphones, people have heard slight differences. However, even these limited results have failed to provide clear evidence of a 'preference' for a lack of phase shift. When auditioned in real rooms, these differences disappear.."
In order, yes and no. An out of context quote from FT is not persuasive.
You can see the phase shift test (and enjoy the spectacle of Curl frantically warning people not to try it) somewhere in the Blowtorch thread. Both Pano and I were able to successfully identify the phase-shifted tracks. The availability of the ABX plugin for foobar allows you to easily determine your own sensitivity to phase shifts (how much, what frequencies...). It was a lot of fun to do.
I've never seen a carbon film specced tighter than 1% or 200pmm Tc, or any figure given at all for voltage coefficient--but I suspect rather worse than any metal film. Same goes for carbon composition and metal oxide. That alone eliminates those resistor technologies for almost all audio applications where I get to choose parts.
Metal film, bulk foil and wirewounds can best those objective measurements by orders of magnitude, and I so use them at their best specifications when I can afford the best, or something approaching that metric. I do try to avoid magnetic parts when I can, just on some general principle but I don't always have that choice.
(Caddock apparently makes some very good resistors, and tantalum resistors might be really good too, but the latter have no objective data to back them up and the former are just hard to obtain and may be rather bulky, and both are rather expensive. So they don't make my leaderboard.)
In practice, I'd like to use .1% 25ppm at a minimum, especially in feedback, input stages and where the resistor is directly in the audio signal path. I rather like Vishay/Dale and PRP because I think they just make better quality parts. I don't really like color-banded resistors because often I can't make out the precise color bands!
Having done that, I have to hope my choices are valid, and wish I could afford really, really good speakers that might allow me to actually hear a difference with my aging, tinnitus-plagued ears. Resistors ideally should be legible, but never actually heard. Some people really seem to like carbon resistors' coloration; I want absolute neutrality.
I don't like mixing science, religion and audio, but it usually does come down to a matter of faith...
Metal film, bulk foil and wirewounds can best those objective measurements by orders of magnitude, and I so use them at their best specifications when I can afford the best, or something approaching that metric. I do try to avoid magnetic parts when I can, just on some general principle but I don't always have that choice.
(Caddock apparently makes some very good resistors, and tantalum resistors might be really good too, but the latter have no objective data to back them up and the former are just hard to obtain and may be rather bulky, and both are rather expensive. So they don't make my leaderboard.)
In practice, I'd like to use .1% 25ppm at a minimum, especially in feedback, input stages and where the resistor is directly in the audio signal path. I rather like Vishay/Dale and PRP because I think they just make better quality parts. I don't really like color-banded resistors because often I can't make out the precise color bands!
Having done that, I have to hope my choices are valid, and wish I could afford really, really good speakers that might allow me to actually hear a difference with my aging, tinnitus-plagued ears. Resistors ideally should be legible, but never actually heard. Some people really seem to like carbon resistors' coloration; I want absolute neutrality.
I don't like mixing science, religion and audio, but it usually does come down to a matter of faith...
This thread has been a tremendous help for me. I read it front to back and took notes. Then I went over my notes and came to the conclusion that for an I/V stage, I should consider Riken ($5), Caddock TF020 ($6), and perhaps the Kiwame ($1) which appears to be a sorted/tested KOA Speer available at Mouser for much less.
Well I started with Riken 1W since stocks are depleting and it could be my last chance to try them. Wow!
These Rikens are warm, detailed, and with a very sweet and lush sounding upper mid / lower treble (think female vocal). These are fantastic.
In fact they sound so good that I don't think I'll ever get around to trying the Caddocks or the Kiwames.
I guess I should also report on the loosers (those which I removed). In an I/V application, a 1/2 watt PRP set sounded thin, overly bright, and had an electronic sounding upper mid / lower treble, almost metallic sounding. Most of this electronic edge was gone after >100 hours of break-in with pink noise, but there was still a slight signature of it. Although the PRPs were quite open and had a very large and clearly delineated soundstage, the electronic-sounding highs ruined the illusion for me. I just could not be transported to the imaginary venue, or the venue could not be transported to my living room with this constant electronic signature reminding me that I was actually listening to an audio playback system.
* * * In this I/V application, the resistors see an average of about 130 mW, with peaks to about 250mW. (if that helps)
I was actually very happy with the sound before I installed the PRPs. But I changed out these resistors because I didn't know what I had in there, and I prefer to know. I just knew they were not Vishay CMF55's as I had originally spec'd. It turns out that these were 250mW Xicon metal films. So Xicons can actually be quite good. (But the Rikens are much better, and this is with only one night of listening. Chances are good the Rikens will morph and sound even better as time goes on.)
Thanks to everyone who contributed to this thread. Very very helpful and reliable. I have found that looking for and reading other's actual listening impressions on DIYaudio is reliable, particularly when folks are describing their listening impressions and when you are looking for the characteristics they specifically describe.
Well I started with Riken 1W since stocks are depleting and it could be my last chance to try them. Wow!
These Rikens are warm, detailed, and with a very sweet and lush sounding upper mid / lower treble (think female vocal). These are fantastic.In fact they sound so good that I don't think I'll ever get around to trying the Caddocks or the Kiwames.
I guess I should also report on the loosers (those which I removed). In an I/V application, a 1/2 watt PRP set sounded thin, overly bright, and had an electronic sounding upper mid / lower treble, almost metallic sounding. Most of this electronic edge was gone after >100 hours of break-in with pink noise, but there was still a slight signature of it. Although the PRPs were quite open and had a very large and clearly delineated soundstage, the electronic-sounding highs ruined the illusion for me. I just could not be transported to the imaginary venue, or the venue could not be transported to my living room with this constant electronic signature reminding me that I was actually listening to an audio playback system.
* * * In this I/V application, the resistors see an average of about 130 mW, with peaks to about 250mW. (if that helps)
I was actually very happy with the sound before I installed the PRPs. But I changed out these resistors because I didn't know what I had in there, and I prefer to know. I just knew they were not Vishay CMF55's as I had originally spec'd. It turns out that these were 250mW Xicon metal films. So Xicons can actually be quite good. (But the Rikens are much better, and this is with only one night of listening. Chances are good the Rikens will morph and sound even better as time goes on.)
Thanks to everyone who contributed to this thread. Very very helpful and reliable. I have found that looking for and reading other's actual listening impressions on DIYaudio is reliable, particularly when folks are describing their listening impressions and when you are looking for the characteristics they specifically describe.
Last edited:
Don't bother with them (particularly Kiwames), a possible improvement can be reached with naked Vishays bulk metal foil (insanely priced).
You can try RC55Y from Mouser, though, they're really good sounding non magnetic resistors.
This thread has been a tremendous help for me. I read it front to back and took notes. Then I went over my notes and came to the conclusion that for an I/V stage, I should consider Riken ($5), Caddock TF020 ($6), and perhaps the Kiwame ($1) which appears to be a sorted/tested KOA Speer available at Mouser for much less.
Well I started with Riken 1W since stocks are depleting and it could be my last chance to try them. Wow!
In fact they sound so good that I don't think I'll ever get around to trying the Caddocks or the Kiwames.
I guess I should also report on the loosers (those which I removed). In an I/V application, a 1/2 watt PRP set sounded thin, overly bright, and had an electronic sounding upper mid / lower treble, almost metallic sounding. Most of this electronic edge was gone after >100 hours of break-in with pink noise, but there was still a slight signature of it. Although the PRPs were quite open and had a very large and clearly delineated soundstage, the electronic-sounding highs ruined the illusion for me. I just could not be transported to the imaginary venue, or the venue could not be transported to my living room with this constant electronic signature reminding me that I was actually listening to an audio playback system.
* * * In this I/V application, the resistors see an average of about 130 mW, with peaks to about 250mW. (if that helps)
I was actually very happy with the sound before I installed the PRPs. But I changed out these resistors because I didn't know what I had in there, and I prefer to know. I just knew they were not Vishay CMF55's as I had originally spec'd. It turns out that these were 250mW Xicon metal films. So Xicons can actually be quite good. (But the Rikens are much better, and this is with only one night of listening. Chances are good the Rikens will morph and sound even better as time goes on.)
Thanks to everyone who contributed to this thread. Very very helpful and reliable. I have found that looking for and reading other's actual listening impressions on DIYaudio is reliable, particularly when folks are describing their listening impressions and when you are looking for the characteristics they specifically describe.
riken have nice sound, I love it but I noticed recently this riken sounds bit dirtier (or bit noisy) compare to wirewound resistor (I use mills) and tants (shinkoh)
and don't ever bother of kiwame...not close enough to riken, believe me, I still have many kiwame (around 950 pcs) and not even want to use it anymore
if you use the resistor on small signal you can try TX2575 from Texas Components ....your jaw will drop
if you use the resistor on small signal you can try TX2575 from Texas Components ....your jaw will drop
Would you still recommend Texas Components TX2575 for I/V?
If you scour this thread, it seems that the TX2575 does not come out on top for I/V.
(Thanks ClaveFremen)
riken have nice sound, I love it but I noticed recently this riken sounds bit dirtier (or bit noisy) compare to wirewound resistor (I use mills) and tants (shinkoh)
and don't ever bother of kiwame...not close enough to riken, believe me, I still have many kiwame (around 950 pcs) and not even want to use it anymore
if you use the resistor on small signal you can try TX2575 from Texas Components ....your jaw will drop
Save them for signal path. Made noticeable difference in phono stage and that was over already low noise SMD
Interesting. Would have thought a very low Tc would be important in a I/V converter. Best I can think of would be a Z-1 foil element in a hermetically sealed oil-filled package:
http://www.vishaypg.com/docs/63146/vhp203.pdf
I don't even want to guess what these could cost!
http://www.vishaypg.com/docs/63146/vhp203.pdf
I don't even want to guess what these could cost!
The Rhopoint squaristor and econistor seem to be very good, also as I/V.
They're on my next project BOM list
They're on my next project BOM list
unless you've got exotic transducers you're listening to 3000 ppm/C Cu voice coils heated with tens of Watts that do swing 10s of C with the music
I really don't think the extra thermal mass of the oil filled can matters with 0.2 ppm/C TC and small fraction of 1 C rise for the ~1 mw you often see in I/V R
the reason they're sold "naked" for audio is that people distrust the dielectric of the packaging
I really don't think the extra thermal mass of the oil filled can matters with 0.2 ppm/C TC and small fraction of 1 C rise for the ~1 mw you often see in I/V R
the reason they're sold "naked" for audio is that people distrust the dielectric of the packaging
I suspect the Vishay and TI bulk metals to be technically identical. Reason: in their spec sheets, they both show the exact same pictures explaining how the trimming works.
- Status
- Not open for further replies.