Hey guys,
Whenever I see a tube amp, (diy or otherwise) there are always large resistors (not sure what type or wattage rating) used throughout, not the typical small ones you see in solid state (or other) equipment. Why is this? Is it not fine to use normal 0.6W metal film resistors in tube amps? Note that I am only talking about resistors that aren't specifically mentioned to be rated for 1 or more watts.
Like these:
Whenever I see a tube amp, (diy or otherwise) there are always large resistors (not sure what type or wattage rating) used throughout, not the typical small ones you see in solid state (or other) equipment. Why is this? Is it not fine to use normal 0.6W metal film resistors in tube amps? Note that I am only talking about resistors that aren't specifically mentioned to be rated for 1 or more watts.
Like these:
Hi,
Wattage ratings etc are all the same but in tube amps you're working with high voltages so you need to watch the voltage rating as well. regular metal film types will be fine as long as their voltage rating is adequate.
Wattage ratings etc are all the same but in tube amps you're working with high voltages so you need to watch the voltage rating as well. regular metal film types will be fine as long as their voltage rating is adequate.
Hi Welcome
If you are not familiar with the sort of voltages that typical resistors are suitable for please have a look at their datasheets.
You can find the datasheets in many places but one easy place to navigate through is http://www.rs-online.com.
Just search for (eg) 100K 0.6W resistor which will bring up a selection of those resistors, you can then pick and choose and look at the datasheets for the different types.
This one for example --- http://uk.rs-online.com/web/p/through-hole-fixed-resistors/0148972/ --- would be suitable up to 350V --- http://docs-europe.electrocomponents.com/webdocs/0062/0900766b80062c77.pdf
NOTE:- The datasheet is only applicable for that type of resistor. If you are sailing close to the wind with any component it is essential that you know exactly what you are buying.
If you are not familiar with the sort of voltages that typical resistors are suitable for please have a look at their datasheets.
You can find the datasheets in many places but one easy place to navigate through is http://www.rs-online.com.
Just search for (eg) 100K 0.6W resistor which will bring up a selection of those resistors, you can then pick and choose and look at the datasheets for the different types.
This one for example --- http://uk.rs-online.com/web/p/through-hole-fixed-resistors/0148972/ --- would be suitable up to 350V --- http://docs-europe.electrocomponents.com/webdocs/0062/0900766b80062c77.pdf
NOTE:- The datasheet is only applicable for that type of resistor. If you are sailing close to the wind with any component it is essential that you know exactly what you are buying.
Last edited:
There is nothing wrong with buying resistors from E-Bay - at least no-one is faking the usual fare. If there are any fakes out there they will be the more exotic resistors.
The reputable vendors will give you the relevant part of the datasheet.
0.6W Metal Film Resistor 100K (Pack of 30) | eBay
The reputable vendors will give you the relevant part of the datasheet.
0.6W Metal Film Resistor 100K (Pack of 30) | eBay
As above,
Resistors have a voltage rating, the wattage is dissipation and is Voltage drop X Current.
If the voltage rating is not above the voltage used it could flash over.
The type is preference based.
Also Higher wattage tend to give lower noise.
However a high wattage resistor does not always have a high working voltage.
So don't get confused that high wattage always means higher voltage rating.
Regards
M. Gregg
Resistors have a voltage rating, the wattage is dissipation and is Voltage drop X Current.
If the voltage rating is not above the voltage used it could flash over.
The type is preference based.
Also Higher wattage tend to give lower noise.
However a high wattage resistor does not always have a high working voltage.
So don't get confused that high wattage always means higher voltage rating.
Regards
M. Gregg
I like Farnell. They have a large selection and you can search with filters. Sometimes they have some very cheap end of line offers. 🙂
Through Hole Resistors | Newark
Through Hole Resistors | Newark
Last edited:
All the above, plus if somebody has really done their homework, they know the operating temperatures (tube amps can run hot) and have de-rated the resistors accordingly. Add all those factors together and what seemed to be a simple application of a 1/4 watt resistor turns into several times that.
Do You will agree, but there is, IMO, a fact, that resistors "sound" different - what do You think, what kind do You prefer? I think, carbon film res-s "sound" better metal....
I Prefer carbon composite exept for phono input stage where they are too noisy, so in this position i use metal film. The carbon composite are non inductive, and less prone to parasit oscilations.
Absolutely.
I wasn't convinced, until I did some comparative listening (with very good headphones!), of different resistors and LEDs used for cathode biasing a single tube... Well, the common metal films sounded quite bad when compared to the others. The most musical/pleasant were the Kiwame carbon films and the most detailed were some low current 3mm red LEDs. The green LEDs that I tried were all either noisy or bad sounding, to my surprise...
Note: you need a good amp, some VERY good speakers and a good pair of ears to hear this kind of differences, but they are real.
If your resistors have a "sound" that is not imaginary, you have chosen them poorly or there is a major flaw in the circuit design. Any nonlinearity of a resistor in a circuit (assuming you've chosen the type according to basic engineering practice) is something like 100,000 times lower than the nonlinearity of the best tubes. I'm sorry, unless someone gives you solid evidence, audibility of differences between different brands of a correctly-specified resistor is a remarkably unlikely claim.
In very low level circuits, noise can be an issue if the circuit is not thought through properly.
The design and type choice aspects for passive component use in tube circuits, with data to back it up, is covered quite thoroughly in Morgan Jones's excellent book.
In very low level circuits, noise can be an issue if the circuit is not thought through properly.
The design and type choice aspects for passive component use in tube circuits, with data to back it up, is covered quite thoroughly in Morgan Jones's excellent book.
I have SE 2A3 RCA, SE 300BJJ and OTL.OCL 6AS7 GElectric with hornes with Fostex 206E - 96 dB and Headphone Loftin White SE 6AS7 RCA with Sen 580 and I can make,IMO, difference of sounding of resistors. My SE 2A3 is with carbon film....I changed all resistors to Kiwame and sound was bad.... I think that every kind of resistors have own sound....
Jan, I'll give this some plausibility- let's say you have a high plate resistance tube loaded with a CCS; worst-case is when the tube is high mu and the cathode is undegenerated. Now, the current noise in a standard bipolar is dominated by the transconductance of the bipolar times the voltage noise of the LED. The resulting current noise is then converted to a voltage noise across the tube's plate resistance.
So... you can try to reduce the noise at the CCS. That is not easy, but it WILL vary among LEDs and LED types. Lesson learned (and I learned it myself the hard way recently): a competent design will not use a CCS load on a high rp (or effective rp) tube where noise can be a factor!
So... you can try to reduce the noise at the CCS. That is not easy, but it WILL vary among LEDs and LED types. Lesson learned (and I learned it myself the hard way recently): a competent design will not use a CCS load on a high rp (or effective rp) tube where noise can be a factor!
You can hear,
The difference when you use a resistor to feed the LED from the supply to get enough current.
I have a preference of resistor types and positions, because I think I hear a difference.
Also this is linked to voltage rating, wattage etc.
If it makes you happy why not, you have to use a resistor so why not pick a type (noise etc)
If they all sound the same then nothing lost or gained.
I also think Kiwame change sound after a period, but as we know that's not true.
I seem to imagine a harsh edge that seems to disappear and the sound becomes warmer. (2nd harmonic)
As I said this is just imagination 🙂 it takes about a 4 weeks or so.
Another is using 5W mills in cathode positions low signal level tubes.(as with all things you can over do it)
But that again is subjective. (imagination)
Regards
M. Gregg
The difference when you use a resistor to feed the LED from the supply to get enough current.
I have a preference of resistor types and positions, because I think I hear a difference.
Also this is linked to voltage rating, wattage etc.
If it makes you happy why not, you have to use a resistor so why not pick a type (noise etc)
If they all sound the same then nothing lost or gained.
I also think Kiwame change sound after a period, but as we know that's not true.
I seem to imagine a harsh edge that seems to disappear and the sound becomes warmer. (2nd harmonic)
As I said this is just imagination 🙂 it takes about a 4 weeks or so.
Another is using 5W mills in cathode positions low signal level tubes.(as with all things you can over do it)
But that again is subjective. (imagination)
Regards
M. Gregg
Last edited:
I didn't wait some period of time, that Kiwame maybe will start sound how I like....but IMO, there is a fact of "time work" to change the sounding, as like about tubes....
Last edited:
Just my opinion,
Kiwame are good in anode and G2, and they have a horrible harsh sound worse than a standard resistor this changes over about 4-6 weeks and becomes smoother and warmer sounding.
Carbon film are also quite good in the anode.
In the cathode metal film/tantalum or (mills 5 Watt<<high detail)
On the grid Takman metal film and/or in the cathode.
Tantalum take a long time to "Bed in" and are a bit like the Kiwame sounding not very good when first used. However I think tantalum need careful placement.
Again this is dependant on bypass cap type and value.
Just for fun each to his own I guess.
Regards
M. Gregg
Last edited:
I would have thought that a current source needs a noise-free voltage reference would be lesson number 1. CCS noise current will then be due to the shot and flicker noise current of the transistor, plus the amplified noise from whatever resistor it uses (or other active device, which will be even noisier).
This leads to the inescapable fact that the output noise current of a CCS is always similar to or even greater than the noise current of a typical valve. CCS loading always worsens the SNR, compared with a simple resistor load.
You missed defective resistors, this has happened to me. Long ago I bought a lot of Mepco-Electra metal film resistors that I most definitely could hear, it turned out there was a good reason as I measured a number of them in comparison to another brand, and it turned out that the M-E resistors were generating harmonic distortion as a function of the amplitude of the audio voltage across them in a simple voltage divider, and the other brand was not. Even with just a volt rms across the resistor samples varied from from a few tenths of a % to nearly one % thd. (I assume this had something to do with the end terminations, but had no way to determine this. They measured and worked fine in every other way.)
I use Caddock, Vishay and the older copper terminated Holco resistors in my designs. They are consistent, have low excess noise, low tc, low voltage coefficients, and tight tolerances, and none of them make measurable levels of thd. (I generally will not pay for <1% tolerance, but in diff amps 0.1% matched or 0.01% tolerances are de rigeur if you need anything approaching good cmrr.)
- Status
- Not open for further replies.