I did know prior that smt resistors have less distortion yet 3h dominant profile compared to carbon film resistors. I did not know the difference would be so dramatic tho once i used them in my preamp.
Below are the pics of a same circuit used in my modular preamp. Based on je918. The layout and values are all the same, except that the green board has most of the components in smt format. I even opted for melf resistors because i read that its 'best sounding'. The caps on both versions are c0g so its really down to the resistors. Well, i much prefer the carbon film version because the melfs sound 'thin' and shrill. They do have more 'air' and holography so the fact that theyre lower distorting holds true.
But all in all subjectively i find the carbon films the superior sounding.
My question is how to tune the melf version to sound 'better'? Would sushumu(sp?) Thin film resistors be better drop in? Increase the voltage rating on them melfs?
I realize this is rather sidestepped from purely objective discussion but some of us have concluded a while back that audio design is very subjective.
Below are the pics of a same circuit used in my modular preamp. Based on je918. The layout and values are all the same, except that the green board has most of the components in smt format. I even opted for melf resistors because i read that its 'best sounding'. The caps on both versions are c0g so its really down to the resistors. Well, i much prefer the carbon film version because the melfs sound 'thin' and shrill. They do have more 'air' and holography so the fact that theyre lower distorting holds true.
But all in all subjectively i find the carbon films the superior sounding.
My question is how to tune the melf version to sound 'better'? Would sushumu(sp?) Thin film resistors be better drop in? Increase the voltage rating on them melfs?
I realize this is rather sidestepped from purely objective discussion but some of us have concluded a while back that audio design is very subjective.
Attachments
Your TH resistors aren't carbon anyway: they are metal-film
Whelp that throws my theory out the window. Youre right, theyre metal film!....
What do you suppose is the reason for what im hearing?
Imagination and confirmation bias.
You already wanted the (supposed/assumed) carbon resistors to be best beforehand. Truth is that they are metal film. Now the thought will be that TH metal film sounds better than SMD resistors 🙂
You already wanted the (supposed/assumed) carbon resistors to be best beforehand. Truth is that they are metal film. Now the thought will be that TH metal film sounds better than SMD resistors 🙂
Last edited:
Resistors just sound different. A good designer uses this to his advantage.
I can only imagine, for the low frequencies encountered in audio, what counts most for resistors is whether they are truly linear. Parasitics like self inductance and capacitance should not affect anything as the operational frequencies are too low.
Theres still a chance that what im hearing isnt placebo. Maybe i also mistrusted my assumption that this forum is a place friendly/open minded enough to not shut me down so fast
I designed and built several audio projects that used THT and SMD components. I can say that I did not hear/measure any difference between the two. In a recent test, I built a PHONO pre-amp with SMD - think film and I measured it. I replaced the resistors with thin-film and did the measurements again. I could not measure any difference between the two. The general consensus is that think film is almost the worst type of resistor that you can use. This is due to the fact that thick film resistors are a lot more noisier than metal resistors (due to the material used). Thin film resistors are made out of a metal film and hence will be a lot more stable and should have lower noise. I was not able to see any difference using my QA401 analyzer. Considering that PHONO pre-amp have high gain, I expected to see some clear differences. So I think that there must be something else that impacts your sound and not the resistor type.
My current theory is that the oem melf i used thru jlcpcb were low quality with lots of impurities. Also theyre rated for 1/4w yet the cylinder size is that of a 1/8 th resistor? Hmm...
I dont have measuring instruments so the only way to tackle this situation for now would be to design a TH version that still uses the same layout and misc components including the smt caps and zeners. And... im off to do that 😆
A forum is not necessary a place where all ones hopes and beliefs are instantly confirmed and cheered upon. Deal with it.
//
//
To buy parts produced by reputable brands also helps. You have OEM (?!) 0.25W SMD resistors the size of 0.125W and they were bought at JLPCB 🙂
This is too much assumption that those dreaded SMD resistors are to blame as their form factor does not seem to be the cause. It could have been OEM TH resistors bought at JLPCB performing just as mediocre. The metal film resistors of today are often very cheap as well with steel lead wires. I cherish my complete E96 Beyschlag series for that reason. In your case both TH and SMD resistors have no brand, no type, no datasheet, nothing. So what are you comparing?! Bad Bing Hai recycled SMD resistors with mediocre Yang Huan recycled TH resistors maybe... calling them OEM is an excellent way of using cheap stuff that works.
Sorry but it is assumptions like this that can start a following based purely on hearsay. If you look carefully you can already see a thread in the future stating SMD resistors to sound worse than TH. It just works like this, certainly when you realize most people go back to their youth "when everything was better". Some even look quite sour when you are making an SMD parts based device just because they are not willing or able to solder these parts.
In all honesty I think this phenomenon is interesting as I see it with guys I know as well. Many stop accepting newer technology, new music as a certain fear for the unknown starts to develop and then things they already know are a kind of certainty. So Class D can't be any good, Class A is always OK, tubes are warm (they are, I tested this), SMD is garbage, Sigma Delta DACs can not sound right etc. The real challenge is of course to accept recent technology and pick what performs OK.
This is too much assumption that those dreaded SMD resistors are to blame as their form factor does not seem to be the cause. It could have been OEM TH resistors bought at JLPCB performing just as mediocre. The metal film resistors of today are often very cheap as well with steel lead wires. I cherish my complete E96 Beyschlag series for that reason. In your case both TH and SMD resistors have no brand, no type, no datasheet, nothing. So what are you comparing?! Bad Bing Hai recycled SMD resistors with mediocre Yang Huan recycled TH resistors maybe... calling them OEM is an excellent way of using cheap stuff that works.
Sorry but it is assumptions like this that can start a following based purely on hearsay. If you look carefully you can already see a thread in the future stating SMD resistors to sound worse than TH. It just works like this, certainly when you realize most people go back to their youth "when everything was better". Some even look quite sour when you are making an SMD parts based device just because they are not willing or able to solder these parts.
In all honesty I think this phenomenon is interesting as I see it with guys I know as well. Many stop accepting newer technology, new music as a certain fear for the unknown starts to develop and then things they already know are a kind of certainty. So Class D can't be any good, Class A is always OK, tubes are warm (they are, I tested this), SMD is garbage, Sigma Delta DACs can not sound right etc. The real challenge is of course to accept recent technology and pick what performs OK.
Last edited:
The Forum is friendly and open minded, that does not mean we must agree 2+2=5 just to be polite.
And the very thread title you chose is controversial, WHAT is "SMT sound" anyway?
Please define it first.
Mathematically, any non-sinusoidal waveform/signal can be expressed as a sum of cosines and sines. This transformation depends only on the shape of the repeating cycle. The sum of cosines and sines does not change if the original waveform is magnefied (amplified) by a fixed number (constant). This means, linear devices which develop voltages across them in linear proportion to the currents flowing through them, cannot alter the sum of cosines and sines and their coefficients (these are known as the harmonics). Having sines and cosines accounts for harmonic phases.
If readers are in doubt, the best way is to ask experts who have studied and have certification. Do not ask for opinions of the opinionated or of those who do not understand Physics and electronics: doing so, is like doing astronomy without using investigative tools like high power optical telescopes and radio telescopes.
Electronics and electricity are two well understood disciplines: ask the experts even if you have to pay for it.
If readers are in doubt, the best way is to ask experts who have studied and have certification. Do not ask for opinions of the opinionated or of those who do not understand Physics and electronics: doing so, is like doing astronomy without using investigative tools like high power optical telescopes and radio telescopes.
Electronics and electricity are two well understood disciplines: ask the experts even if you have to pay for it.
"ask us self-proclaimed experts as the plebs don't know so don't ask them". 1984 🙂
It may help to start comparing stuff that has real data and that is of known origin. This comparison was between a glass bottle with a brown fluid and a tin can with a brown fluid. Someone recognizes the bottle and assumes Coca Cola which he likes. Even after learning both fluids to be the same cold coffee he still thinks the fluid in the bottle tastes better. He is a certified expert in aerodynamics also having a PhD in physics and mathematics.
It may help to start comparing stuff that has real data and that is of known origin. This comparison was between a glass bottle with a brown fluid and a tin can with a brown fluid. Someone recognizes the bottle and assumes Coca Cola which he likes. Even after learning both fluids to be the same cold coffee he still thinks the fluid in the bottle tastes better. He is a certified expert in aerodynamics also having a PhD in physics and mathematics.
Last edited:
The thermal noise is exactly the same. If you ever find a resistor (not a circuit that behaves resistively, but a thing with no power source) with more thermal noise or less thermal noise than another resistor with the same value and temperature, publish it in a physics journal, because you then have proven that the second law of thermodynamics is wrong and a perpetuum mobile is possible.
1/f noise (excess noise, flicker noise) is different, but that noise manifests itself as random resistance variations. If there is little or no DC bias voltage across the resistors, that won't affect the noise floor, but only cause noise sidebands around the signal.
None of the above appears in any datasheet, manufacturers literature or is derived from any branch of electrical/electronic engineering.
Why do hobbyists seem to think electronics associated with audio is the only segment exempt from any sort of expertise, the laws of physics, science and engineering ?
Fixed it.
You keep assuming my subjective experience has no validity. That is a dangerous assumption itself. I can see how my observation has no objective merit. But to smear the observator for it is a form of close mindedness
No it is keeping an open eye to assumed matters with a wrong conclusion and a thread title that is simply a silly, unfounded and closed minded provocation: "That dreaded smt sound". One can know for sure people will react to that. You effectively state that SMD parts sound dreaded which is complete nonsense as it is based on at least 1 wrong assumption and only 2 devices with parts of unknown origin. There, I said it.
It is no personal attack (oversensitive?!) but it is my right to call this comparison flawed and of no value. This regardless of my respect for fellow humans. If you would repeat the test with for instance SMD and TH parts to be both by Vishay then we can discuss further. If you can't accept this you can always start a debate about the ever important tone.
It is no personal attack (oversensitive?!) but it is my right to call this comparison flawed and of no value. This regardless of my respect for fellow humans. If you would repeat the test with for instance SMD and TH parts to be both by Vishay then we can discuss further. If you can't accept this you can always start a debate about the ever important tone.
Last edited:
- Status
- Not open for further replies.