Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Capacitors for DIY. Are there real audible differences? (Read 20420 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Capacitors for DIY. Are there real audible differences?

Reply #25
Are you building a tube amp?  I hope you have a good source for the output transformer.  That's probably the most critical component (assuming everything else in the design is done right).  I'd guess the output-transformer is also the single most expensive component.
I am building a pre-amp. I will use an impedance matching transformer on the output, I've been quite happy with Edcor.

(And a DIY tube duplication probably won't satisfy you either.)
I've built a few solid-state amps, using integrated circuits wherever possible to keep it as foolproof as possible, and to keep the cost down.  But, I wouldn't bother today because I can buy something for a LOT less money than it would cost to build it.  And, what I'm buying is probably better than what I could build.
Well, the reason I got into DIY in the first place was to make it sound better and more reliable than the vintage gear I was replacing.  At first I upgraded components or made additions to stock gear, but eventually started building my own.

Most recently, I was able to build something at much less cost which to me made sonic improvement than the vintage gear I replaced.  I financed the build, and had a couple hundred left over from the sale of the vintage piece I built my DIY around.
I can say definitively a DIY can be made to a far higher standard and better design (electrically) than a similar vintage design. Vintage commercial designs were made around cost efficiency, and using obsolete technology. Some of that obsolescence is desirable, (tube circuit) but some is not (drifting carbon resistors, aged, inferior components).  Sound quality is subjective, but improvements I've implemented sound markedly better to me, without changing the tubey-vintage sound.

Examples: the output of one tube preamp design I've been working with had undesirable impedance and drive to push a modern, solid-state power amp. Adding an additional tube/cathode follower, and an impedance matching transformer improved my design's output considerably. Using a regulated, solid-state rectified DC power source for the tube heaters resulted in lower noise and hum over an AC power supply, and I didn't need to use thick, twisted pair wiring.  Increasing the B+ voltage to the circuit to within the specs of what a tube can handle rather than 'starved plate' brought marked headroom improvement, which is more useful for bass guitar.

I've worked-on tube amps (years ago), but personally wouldn't try building one.  It's more difficult than going solid-state, the circuits are more "finicky", and the cost per-watt is a LOT higher.
No debate that solid-state is superior from an electrical design standpoint.

Capacitors for DIY. Are there real audible differences?

Reply #26
Examples: the output of one tube preamp design I've been working with had undesirable impedance and drive to push a modern, solid-state power amp. Adding an additional tube/cathode follower, and an impedance matching transformer improved my design's output considerably. Using a regulated, solid-state rectified DC power source for the tube heaters resulted in lower noise and hum over an AC power supply, and I didn't need to use thick, twisted pair wiring.  Increasing the B+ voltage to the circuit to within the specs of what a tube can handle rather than 'starved plate' brought marked headroom improvement, which is more useful for bass guitar.


Vintage tubed pramps that can effectively drive a SS power amp are there for the finding. 

Most but the early MacIntosh tubed preamps had cathode follower outputs and appropriately sized coupling capacitors:  (1956)

http://vintagevacuumaudio.com/schematics-m...-schematics.pdf



The Audio Research  SP3 did, as well.  (1972)



Conrad Johnson PV8 (1988)



Notice the modern (for the day) solid state power supplies!

Many other examples exist

Capacitors for DIY. Are there real audible differences?

Reply #27
The preamp's circuit I was building mine after was taken from a 'complete' Fender circuit. But it stopped before where the power amp section would start. It ended with a common cathode follower and went to the output from there, making the output Z very high. Adding a cat. follower was a simple addition, I don't know why they didn't.  Probably more expense and price point than they wanted or something.

Capacitors for DIY. Are there real audible differences?

Reply #28
Here's a schem of the Fender circuit I was talking about.  The preamp I was making was modeled after a vintage preamp using the same design. The output on the vintage one was placed after the second half of the first 7025 and the rest was discarded. My design's improvement was to add a White Cathode Follower and impedance matching transformer on the end to lower impedance and improve power transfer to a modern power amp. I can only guess why they didn't use that second 7025. Budget considerations?
I didn't think this up, I'm not that smart, but a friend who had implemented it turned me on to it.


Capacitors for DIY. Are there real audible differences?

Reply #29
Now I'm making the same design, but with a DC coupled cat. follower and some kind of adjustable compressor circuit in there somewhere...

ANYWAY - back on topic.  It seems I was momentarily buffaloed by all the fancy sounding techno-babble out there.  Some quality, modern polypro caps in the tone stack and I'm set.  Thank you Arnold B. Krueger, et. al.



Capacitors for DIY. Are there real audible differences?

Reply #32
Yeah, but it's SUITABLE IN HIGH MAGNETIC FLUX ENVIRONMENT!!  Did you watch the teaser?  Hilarious!

Capacitors for DIY. Are there real audible differences?

Reply #33
Vintage tubed pramps that can effectively drive a SS power amp are there for the finding. 

Most but the early MacIntosh tubed preamps had cathode follower outputs and appropriately sized coupling capacitors:  (1956)

http://vintagevacuumaudio.com/schematics-m...-schematics.pdf



The Audio Research  SP3 did, as well.  (1972)


I like how these two oldies feature sensible gain staging. Quite apparently the designers did care about noise levels, and since high supply voltages afforded a fair bit of headroom, they were able to include some gain in front of the volume control. This is not commonly seen in solid-state era devices, though e.g. classic Grundig hi-fi components made use of some pre-gain as well (and accordingly featured exemplary noise levels in the day).

But speaking of noise, IMO this is the bigger issue when combining devices from different eras. Hollow state power amps tended to have a fair bit less gain than their solid state successors, with matching preamps making up for it. I'd say it's about 10 dB that got shifted from pre to power with the transition to solid state, maybe even 14 dB. Levels in between were reduced accordingly, hence the requirements for preamp output noise levels would have become more strict. 30 dB pre + 16 dB power or 16 dB pre + 30 dB power = OK, but 30 dB pre + 30 dB power = hissy mess, in all likelihood.

Capacitors for DIY. Are there real audible differences?

Reply #34
I like how these two oldies feature sensible gain staging.


The gain staging is pretty well defined by the environment they work in, and does not vary that much from normal practice with SS.

For example, the midband gain of a RIAA preamp is strongly influenced towards approximately 40 dB by the RIAA curve itself.  A SS RIAA preamp from the 80s would be typically within 3 dB of that - an insignificant difference. 

While tubes operate at higher voltages than SS their saturation voltages are also vastly higher, with the result that the range of linear operation is not that different.  Plate curves of a 12AX7 and dissipation concerns limited most high quality tubed preamps to a range of about 70 volt p-p. Comparable two-transistor SS designs might operate over a linear range of 42 v p-p which is again not that different, especially considering their slightly lower gain.

Quote
Quite apparently the designers did care about noise levels, and since high supply voltages afforded a fair bit of headroom, they were able to include some gain in front of the volume control.


Tubes are generally inherently far noisier because of their higher operational temperatures, and as I already pointed out often included about the same amount of gain in front of the volume control.

There were no significant shifts in phono cartridge or microphone output voltages across the transition from a mostly tube world to a mostly SS world.

Quote
This is not commonly seen in solid-state era devices,


Already debunked.

Quote
But speaking of noise, IMO this is the bigger issue when combining devices from different eras. Hollow state power amps tended to have a fair bit less gain than their solid state successors, with matching preamps making up for it.


Among common US power amps there were no such differences, and tubed and SS amps and preamps were and are commonly mixed and matched. Stand alone power amps have required input voltages on the order of 1.5-2 volts for full output. Given the far greater power output of common modern SS power amps, they probably have more gain than the classic tubed amps rather than less.

My specific comments are in the context of US products. I know that in Europe lower line level voltages and therefore higher amplifier gains were common, but again there were no problems achieving them with SS gear and it was common practice to do so.  In general, SS gear is less noisy and has lower distortion.



Capacitors for DIY. Are there real audible differences?

Reply #35
Electronic components have tolerances that vary from one manufacturer to other manufacturer. Components also differ in tolerances depending on their use; say resistors, for common use there’s the 20% tolerance, and there’s the 10% and 5% for more precise use. Precision components will definitely cost more. Operating temperature and frequency response also affects the rated value of the electronic component. These can be eliminated with careful designing.

Capacitors for DIY. Are there real audible differences?

Reply #36
I'm involved in making DIY preamps for musical instruments. I see LOTS of different capacitors in different sizes, price, materials, etc. with the same value. (i.e. .01uF).

What I'd like to know from the non-audiophile community is whether component choices can make audible differences in the sound?  - Specifically, capacitors.  I'd like to go with some small, inexpensive, modern film capacitors from someplace like Mouser or Digikey.  In designing a vintage style circuit, can "improvements" be had by the selection of more expensive "audio-grade" or "vintage-style" capacitors? or is it a waste of money and common sense? 

I'm of the opinion that beyond circuit design, the actual components chosen (as long as they fit the specs) are of little/no significance to what the end result sounds like.  Am I wrong?
I'm looking for objective answers. (TOS #8) Subjective is meaningless to me.  I figured this would be the only forum where that's likely.


Hmm, probably best to avoid capacitors in the signal path altogether.

Capacitors for DIY. Are there real audible differences?

Reply #37
Electronic components have tolerances that vary from one manufacturer to other manufacturer. Components also differ in tolerances depending on their use; say resistors, for common use there’s the 20% tolerance, and there’s the 10% and 5% for more precise use. Precision components will definitely cost more. Operating temperature and frequency response also affects the rated value of the electronic component. These can be eliminated with careful designing.

Sorry, but a designer of electronics who needs to be reminded of this can be considered substandard.

The fact that components come in different tolerance ratings was clear to me when I was a kid. Look into a catalogue and it is obvious.

What is less obvious is the effect such tolerances have on the operation of some electronic circuit. That's for the design engineer to figure out. A good design tries to make sure that only a few parts in the circuit are critical in their tolerances.

When it comes to capacitors in analog audio equipment, the tolerance only matters when the capacitor's role in the circuit involves defining a frequency, i.e. in a filter. Decoupling caps, or coupling caps, usually don't require any tightening of tolerances. Whether they're off by 20% shouldn't have a noticeable effect. Yet it is those places where those audiophile "specialists" want to see special significance.

 

Capacitors for DIY. Are there real audible differences?

Reply #38
Electronic components have tolerances that vary from one manufacturer to other manufacturer.


The idea that tolerances are primarily dependent on who manufactures the part is false.

To be taken seriously as a first or second tier mainstream supplier, a parts manufacturer has to provide parts in tolerances that conform to industry standards. 

The industry standards for parts tolerances are generally tighter than was given in the post. 5% and 1% are now the common tolerances for resistors used in mainstream applications.

There are specialty vendors who sell  special parts with tolerances that are much lower. They are typically used in specialized test equipment, and so the volumes for these parts are relatively small.

One of the impacts of the ongoing digitization of all electronic gear is a vast reduction in the need for high tolerance parts.