Hardwiring, Layout Problems | Solutions To These Problems |
Power Supplies and Capacitors | Interstage Signal Feedback Through the Power Supply | Types of Distortions | Resistor comparisons |
Home | Reviews | Tube Preamplifiers | Tube Amplifiers | DIY Parts/Price list | Warranty | Specials | Contact, Directions | Audio Links | NEW: Forum
Below are reprints of two articles comparing different types of resistors, Bulk Foil, Melal Film, Wire Wound, Metal Glaze, Metal Oxide, and Carbon film. While not exhaustive, I think one can get a good general idea of how the different types compare, sonically. Although some metal film types are inferior sounding to carbon types, in general metal films are clearly superior. Since metal films are so abundant, the price has come down considerably. Carbons, on the other hand have gone up in price, since very few, if any, are manufactured in the USA. They are not extensively used because of the high fire risk. Now for the articles.
"A selection of resistors was requested, and those we received were subject to test at two values, 100 ohms and 33k ohms. The 100 ohms were used as the main load resistor for a moving coil cartridge, with a preamp input impedance of 1k ohm. By careful gain calibration, the load effect was assessed in absolute terms, while the resistors were also assessed via a single-presentation method of judging overall asound quality. The 33k ohm types were assessed while used as the main termination for a CD player of low output impedances (100 ohms), the associated preamplifier input impedance being 100k ohms. Again, comparisons were made without a test resistor to assess the degree and type of sonic aberration introduced by the resistors under examination."
"Some earlier listening work has been done; for example, a fabricated metal-film array for a 2w value was compared with an equivalently rated wire-wound (Pot Pourri, Sept '86). In another small test, included in the previously published capacitor comparisons, a few resistors were auditioned as moving-coil cartridge loads. Here the Holco type was compared with carbon composition, carbon-film and low cost metal-film types. Holco won that contest (HFN/RR, Oct '85). In the present test the rarely-used carbon composition types were not included; previously tested examples were unpromising. Wire-wound, several metal-films, metal-glaze, and carbon-film types are represented here, together with the costly bulk-foil designs. Not all types in the survey were available in the 33k ohm value."
"...Compared with other projects, the listening work here was extremely difficult, and as such I would judge it rather less reliable. It is said that resistor sound quality is easier to judge if complete amplifiers are built largely with one type; comparisons are then made more easily as the effects are multiplied. But the scoring for resistor sound is an arbitrary one, and has no connection with that for amplifiers, CD players or cables. It is merely relative."
Whilst the sample is fairly limited, it was interesting that the carbon-film types sounded a little better than some of the metal-films. But Roedenstein's good reputation for an inexpensive metal-film (No. 11) seems well justified. We were fortunate to have the Vishay bulk-foils (No. 10), since these really proved our case, namely that resistors do affect sound quality. The Vishays were significantly nearer the original, helping to show the magnitude of the test loading effects, which were comparatively minor, and also revealing the audible errors' introduced by most of the other resistor types tested.
In general, resistors add a 'glazed' and 'hashy' noise in the treble which seemed to dilute stereo focus. Other effects included as impression of mildly bumpy or lumpy bass, with some forwardness in the midrange and a noticeable loss of depth, ambience and dymanic power. In the case of the latter, some of the resistors under test gave a blander and more lifelss effect to the reproduction.
1) MFRS, 0.5W, 1% metal-film, by Welwyn.
33k: 33.09, 33.03, 100 ohms: 100.28, 99.81
Sound 70%: thinner lighter sound, depth compression, lowered treble focus.
2) MFR4, 0.25w, 1% metal-film, by Welwyn.
33k: 33.01, 33.01, 100 ohms: 99.61, 100.1
Sound 72.5%: slight improvement in depth and tonal balance.
3) MRS16, 1% metal-film, by Mullard/Philips.
33k: 32.9, 33.04: 100 ohms: 100.05, 100.43
Sound: 75%: mixed reactions at 100 ohm, better banance and improved persepective at 33k.
4) MRS25, 0.6 W, 2% metal-film, from Mullard/Philips.
33k: 33.06, 33.03, 100 ohms: 99.28, 98.89
Sound 77%: a lively sound with better bass but a touch of 'brashness' in the treble.
5) BTF4, .25W, 5% carbon-film, by Dublier.
33k: 33.43, 33.15. 100 ohms: 99.28, 98.89
Sound 72.5%: fair balance, some blandness with a mildly lumpy bass.
6) SFR25, 0.33W, 5% metal-film.
33k: 33.43, 33.15: 100 ohms: 99.02, 99.15.
Sound 80%: a welcome improvement, more dynamic, better balanced, improved focus.
7) CFR50, 0.6W, 5% carbon-film by Neohm.
33k: 33.07, 32.67. 100 ohms: 99.31, 98.95
Suond 77.5%: a good impression of depth, despite some 'glaze' in the treble and slight bass boom.
8) NK3, 0.125W, 2% metal-film, by Corning.
33k: 32.99, 32.74. 100 ohm: 99.90, 100.1
Sound 71%: mild 'glaze' and diffusion in the treble, better depth heard at 100 ohm value.
9) CFR25, 0.25W, 5% carbon-film, by Neohm.
33k: 32.77, 32.95, 100 ohms N/A.
Sound 77%: quite good all-arounder.
10) VSRJ, 0.25W, 0.01% bulk-foil, by Vishay
36k: 36.02, 36.02, 100 ohms: 100.05, 100.06
Sound 94%: an unmistakeably large step nearer to no resistor at all. Clear, tonally balanced, fine depth, negligible aberration.
11) 0.25W, 2%, metal-film, by Roederstein
33k: 32.93, 32.81, 100 ohms: 99.55, 99.84
Sound 82%: strong dynamic sound with fairly good clarity, focus , and depth, but a touch of boom and a trace of 'glazed' treble brightness.
12) W21, 2.5W, 5% wire-wound vitreous, by Welwyn
33k, N/A, 100 ohms: 100.04, 100.07
Sound 86%: some dulling of air and a mild 'wirey' emphasis on string tone; good bass, slightly forward mid, but good clarity, dynamics and depth.
13) RS, 2.5W, 5% wirewound silicone cement coated, CGS
33k: N/A, 100 ohms: 98.7, 99.2
Sound 86%: similar to the above but slightly sweeter, with good instrumental separation. Mildly band limited, faint rolloff at the frequency extremes.
14) Econistor, 0.1%, precision wirewound, non-inductive, by Rhopoint.
33k: N/A: 100 ohms: 100.01, 100.005
Sound 86%: slightly soft bass, good focus, very slightly 'wirey' on strings; mild mid presence, good depth.
15) H8, 0.125W, 0.5% precision metal-film (Holco).
33k: N/A. 100 ohms: 100.31, 100.07
Sound 88%: a very fair copy- neutral, well balanced, clear and clean strings, good depth, slightly soft bass, very mildly defocused, accurate.
16) LR1, 0.25W, 1% metal-film, by Neohm
33k: N/A. 100 ohms: 100.04, 100.42
Sound 73%: some treble brittleness and grain, loss of depth and ambience: bass 'thumpy', loss of focus.
17) RGP0207, 0.5W, 1% metal- glaze, by Neohm.
33k: N/A. 100 ohms: 100.00, 100.12
Sound 76%: a touch forward and bright, but not too brittle or grainy. Fair bass and focus, some loss of depth and dynamics.
18) 'Commercial', 0.5W, 5% carbon-film, from Japan.
33k: N/A. 100 ohms: 98.49, 102.59
Sound 80%: some 'nasality' on violin, but clearer and less grainy than average. Quite good depth, fair bass and reasonable tonal balance.
19) N/A, 0.5W, 5% metal-oxide (obsolete), by Welwyn.
33k: N/A. 100 ohms: 100.10, 102.50
Sound 80%: quite neutral, fairly good depth, with some softening of detail; a touch bland and undynamic.
"For interest's sake, some subjective tests were performed on resistors during the capacitor auditioning. Only a few types were tried, more generic than specific, and only in one application, namely as the 100 ohm shunt loading function for an m-c cartridge, as reported on by Christopher Breunig a few months back (HFN/RR June, 1985). The test cartridge was an Empire van den Hul MC1000 and the preamp an SP-8."
"The judgments are arbitrary to some extent, based on the listeners' experience of sound quality 'improvement'. Even if the reader disagrees as to the true meaning of the observation - is it a change or an improvement? - the results do suggest that the resistors differ in their effect on the system's sound quality. Given the minute power and voltage levels involved in the test application, it is highly unlikely that thermal or voltage stress effects- which would result in the nominal resistance being modulated by the signal - are involved."
"We emerged from this brief series of tests feeling that while resistor differences were difficult to identify, significant differences were nevertheless present. The effects may well be diffferent in alternative circuit positions and at different power and voltage levels, but in this application, only the Holco proved capable of loading the cartridge without apparently degrading the sound quality of the SP-8 preamplifier to some extent. The other four resistors were felt to reduce the sound quality sufficiently to make the question of optimum resistive loading for the cartridge irrevelent. They were best omitted altogether."
R1 Make: Holco Score 91%
Type: 100 ohm metal film
Comments: Very small effect; tight precise stereo and good overall control.
R2 Make: RS Components Score 86%
Type: 100 ohm metal film
Comments: Trace of 'muddle'; slight loss of space; a touch of treble grain, and a softer bass.
R3 Make: Roederestein Score 84%
Type: 100 ohm metal film
Comments: Good bass sound, with pleasing depth but a slight grain and 'zzz' in the treble. A slightly 'muzzy' effect in the midrange.
R4 Make: Anonymous Score 70%
Type: 100 ohm carbon film, 1/4 watt
Comments: 'Zitty' defocused sound; 'louder' and mildly fatiguing; grainy treble.
R5 Make: RS Components Score 64%
Type: 100 ohm carbon film, 1/4watt
Comments: Defocused and grainy sound, with a loss of bass definition, stereo depth and focus.
Left out was the Caddock series of resistors, although, from my understanding, they have high standing in the audio world.
This paper is presented in the interests of the public.
SAS Audio wishes to thank Svetlana and New Sensor (Sovtek) for allowing us to use their vacuum tube graphics on our site.
* tm, sas audio labs, SAS Audio Labs, SAS AUDIO LABS, SAS AUDIO LABS banner, sasaudio, sas audio, SAS Audio are trademarks of SAS Audio."
copyright©: 12-28-1996 Updated 07-19-2013. All contents of this page article (except Sovtek and Svetlana Tubes) are copyrighted. All layouts of all our components, and term "lead to lead wiring" are copyrighted. All rights reserved. No portion of this article may be reproduced without written permission from Steve Sammet at SAS Audio, Martin Colloms titled "Piece De Resistance"? and the second by "Hi-Fi News and Record Review". SAS Audio Labs is registered with the state of Illilnois.