So here's my story. I have naked resistors for cathode bias. Lowest noise, incredibly clear, but can tend to the lean side. My favorites for tone are Roederstein. But those alone don't have the transparency I'm after. So for a while I paralleled them calculating the resistance of the Roederstein so it wouldn't change the overall resistance.
I just finished the project and decided to listen with just the naked resistors. Clearer, cleaner, more transparent. Moved to the leaner side.
So my question is: am I just injecting noise by paralleling these trying for the best of both worlds? I've tried just about every resistor in this position. Since I have an incredibly simple circuit the tube bias is the only place I have R/C in the signal path so the R or the C make a huge shift in the sound.
I just finished the project and decided to listen with just the naked resistors. Clearer, cleaner, more transparent. Moved to the leaner side.
So my question is: am I just injecting noise by paralleling these trying for the best of both worlds? I've tried just about every resistor in this position. Since I have an incredibly simple circuit the tube bias is the only place I have R/C in the signal path so the R or the C make a huge shift in the sound.
By blending multiple types of resistors, you will get a blend of all their best attributes. You will also get a blend of all their worst attributes. That means you will end up with some of the leanness and lack of tone you dislike from the Nakeds, and some of the noise and loss of transparency you've heard in the Roederstein. As you say, pluses and minuses. The choice is up to you.
Having 221R in parallel with 91K means that the 91K only contributes 0.24% of whatever happens i.e. it can only have virtually no effect. It could be very noisy or very non-linear and you probably would not notice. If it was a perfect resistor then it would have almost no effect in 'correcting' the 221R. Maybe I have missed something, but such a combination seems totally pointless.
DF96 -- This is the part of theory that meets audio reality. Many things make no sense yet have sonic effects. The 91K definitely adds some body and loses some transparency. The explanations is what I was after. The fact its only contributing .24% makes it not so parasitic in my understanding.
RTF -- thanks also, the preamp does sound amazing to me and I've had many classic and expensive ones (which price doesn't result in quality) come through here by way of others and I've owned some good ones.
Much appreciated to everyone.
RTF -- thanks also, the preamp does sound amazing to me and I've had many classic and expensive ones (which price doesn't result in quality) come through here by way of others and I've owned some good ones.
Much appreciated to everyone.
The 91K could approximately double the stray capacitance, but this is unlikely to have an audible effect. Are the resistors bypassed?
If the 91K is genuinely changing the signal in a noticeable way, then it must be generating significant distortion which is then attenuated to 0.24% of its 'bare' value and then either adding pleasant distortion or partially cancelling existing unpleasant distortion. Some appropriate measurements would help. Note that this explanation would mean that a 91K resistor of this type used on its own in an audio circuit would generate severe very noticeable distortion - I find this very hard to believe unless you have a faulty 91K resistor. Note that audio circuits operate using physics, not magic. Whatever the 91K is doing, it will be attenuated to 0.24%. So 10% distortion would become 0.024%.
If the 91K is genuinely changing the signal in a noticeable way, then it must be generating significant distortion which is then attenuated to 0.24% of its 'bare' value and then either adding pleasant distortion or partially cancelling existing unpleasant distortion. Some appropriate measurements would help. Note that this explanation would mean that a 91K resistor of this type used on its own in an audio circuit would generate severe very noticeable distortion - I find this very hard to believe unless you have a faulty 91K resistor. Note that audio circuits operate using physics, not magic. Whatever the 91K is doing, it will be attenuated to 0.24%. So 10% distortion would become 0.024%.
The resistors are just in parallel then in parallel with the bias cap. I don't have measuring equipment, but again, I appreciate your explanation.
On its own a Roederstein has a very pleasing tone to me. What it lacks in transparency it makes up for in tone. When compared with the naked one can hear that the Roederstein is noisier. Which was my basic original question which I feel you answered here.
The naked resistors, like any resistor type, require a judicial use of them in specific areas. Each time I've used them in the most noticeable positions I hear the same thing which require shaping the tone so the outcome is not too sterile.
Being the only resistor in the signal circuit exposes their qualities even more.
On its own a Roederstein has a very pleasing tone to me. What it lacks in transparency it makes up for in tone. When compared with the naked one can hear that the Roederstein is noisier. Which was my basic original question which I feel you answered here.
The naked resistors, like any resistor type, require a judicial use of them in specific areas. Each time I've used them in the most noticeable positions I hear the same thing which require shaping the tone so the outcome is not too sterile.
Being the only resistor in the signal circuit exposes their qualities even more.
That sounds like adding a little pleasant distortion in order to avoid the unpleasantness of signal accuracy.
If these resistors are bypassed, then it is virtually impossible for the extra resistor to make any difference. However, adding it could make a difference if the heat of soldering affected the existing resistor. This could improve or degrade it a little. I think something like this has to be my favoured physics-based explanation.
I don't want to get into some of this because its well worn territory. One point I should make is I'm using digital which may very well account for compensating in the pre-amp. Perhaps if it was vinyl these issues would go away. I have a pretty good digital setup but...
Voicing and optimizing a circuit, imho, is the magic of the better sounding audio gear. Its all a delicate balance of parts used to final sonic effect. I know how some very expensive gear that I think sounds incredible yet they chose to use very inexpensive parts to shape the signal. Still has resolution and incredible grounding schemes but its not a collection of the 'best audio parts'.
But I really don't want this to degrade into a debate about it.
I really like this board apart from other ones because people are helpful and knowledgeable. I greatly appreciate that from you all.
Voicing and optimizing a circuit, imho, is the magic of the better sounding audio gear. Its all a delicate balance of parts used to final sonic effect. I know how some very expensive gear that I think sounds incredible yet they chose to use very inexpensive parts to shape the signal. Still has resolution and incredible grounding schemes but its not a collection of the 'best audio parts'.
But I really don't want this to degrade into a debate about it.
I really like this board apart from other ones because people are helpful and knowledgeable. I greatly appreciate that from you all.
True , much too true .
But i have never heard of multiple resistor brands .
@ Entire topic
Are you using 10W resistors or just plain old 2W resistors ?
Sorry, I misunderstood. However, my views on other issues tend to be similar!
I accept that resistors can sometimes sound different from each other, as they can have different non-linearities, but these effects will be small. In the case you have cited, any difference will be swamped by the large ratio in values and the bypass cap. Hence I find it difficult to come up with an explanation.
I accept that resistors can sometimes sound different from each other, as they can have different non-linearities, but these effects will be small. In the case you have cited, any difference will be swamped by the large ratio in values and the bypass cap. Hence I find it difficult to come up with an explanation.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.