Ok, so the capacitor argument is a case of diminishing returns. Let's say it is much like the power vs SPL relationship. So if the original design has a 100uf cap in the power supply, by doubling that value you could lose 3db worth of noise. By overcapping by a factor of 10, you could lose 10 db of noise. After that, the increased value of capacitor makes less and less difference the bigger the oversize factor. So a 2000uf substitute could lose you 13db of noise.
Now in the signal path, it's a whole different animal. Since the audio signal is passing through or being shunted or bypassed by the capacitor, it has a much more dramatic effect on the sound. This is where most of the cork sniffing and tinkering comes into play. No component is perfectly ideally pure, so the intrinsic properties of different materials or construction add "flavors" to the audio signal as a side effect. Changing the capacity changes crossover or resonant frequencies of the node the component serves. The whole character of an amplifier can be modified by altering a very few key components. For many, good enough is found quickly; for few, the quest for perfection is slow and methodical.
Now in the signal path, it's a whole different animal. Since the audio signal is passing through or being shunted or bypassed by the capacitor, it has a much more dramatic effect on the sound. This is where most of the cork sniffing and tinkering comes into play. No component is perfectly ideally pure, so the intrinsic properties of different materials or construction add "flavors" to the audio signal as a side effect. Changing the capacity changes crossover or resonant frequencies of the node the component serves. The whole character of an amplifier can be modified by altering a very few key components. For many, good enough is found quickly; for few, the quest for perfection is slow and methodical.
Last edited:
T'aint so. That much over capping will probably increase noise as your transformer and rectifiers struggle to meet the peak current demands; radiating RF through out your amplifier.
And that much capacitor probably implies high ESR or ringing from parasitic inductance. (Been there, done that).
There's a sweet spot for the simple (agricultural) capacitor input filter. Beyond that you need to start looking at more complicated filtering. Whether that's cRC or cLC or shunt regulation or series regulation is up to you.
Last edited:
Technically, they do... air pushes down on the spoiler pushing down on the rear wheels to avoid slippage... which helps the car go faster.
... though I know you're probably talking about Honda Civics with 50lbs of fiberglass on the trunk. The weight alone more than offsets any gain achieved from having a spoiler.
But, real cars with real spoilers, it does make the car faster.
I guess we could say that the difference between a Civic and an F1 car, is like the difference between Chinese and Japanese caps?
EDIT... Wow... discussion really kinda got away on me over the past hour.
Last edited:
'Zackly! Not to mention, the downforce from the "spoiler" is a negligible fraction of the car's weight at any speed those vehicles can achieve.
Real race cars (F1, CART etc) are something else entirely, there is so much downforce that they could drive upside down on a ceiling, if you had a big enough ceiling!
Along the same lines as the fibreglass spoilers: putting much larger wheels and tyres on a car with a typical consumer-sized engine almost always slows it down (reduces acceleration, increases zero-to-sixty time). This is because bigger wheels have much more rotational inertia - so, from a standing start, a lot of power from the engine goes into spooling up the wheel rpm, rather than accelerating the car itself!
A buddy of mine bought a box of (made in China) pencils from a dollar store. When he got home he found out the "pencils" were just wood - no graphite in the centre.I guess we could say that the difference between a Civic and an F1 car, is like the difference between Chinese and Japanese caps?
At the other end of the scale, Apple's much-lusted-after plastic fondleslabs / iThings are all made in China - and a majority of consumers seem to think they are the pinnacle of desirable quality and technology.
Apparently Chinese factories will produce whatever level of quality you are willing to buy, from useless pencils to overpriced phones!
-Gnobuddy
What he said^^^
A Honda Civic is a real car, it just isn't a race car. Context is everything. The fin on mom's Honda won't make her car faster, even if they would on a race car.
If you could gain a few db less noise with a larger filter cap, fine. Now the next guy sees that larger cap, and says double THAT. OK, and on and on, at some point your cap becomes mom's spoiler.
And remember, you were running on a battery so ther is zero ripple to filter. How much better than zero can you make it?
A Honda Civic is a real car, it just isn't a race car. Context is everything. The fin on mom's Honda won't make her car faster, even if they would on a race car.
If you could gain a few db less noise with a larger filter cap, fine. Now the next guy sees that larger cap, and says double THAT. OK, and on and on, at some point your cap becomes mom's spoiler.
And remember, you were running on a battery so ther is zero ripple to filter. How much better than zero can you make it?
In your case, there was a definite problem to be solved (you had audible hum).
So of course it's then entirely logical to look for the cause, and for possible fixes to the problem. The hum could have been caused by any of several issues - poor shielding, poor grounding, excessive power supply ripple, et cetera. You found out that using more filtering reduced the hum; that tells you that the wall-wart you were using had inadequate filtering built into it. It was probably a cheaply designed model that was intended to power a ripple-tolerant device (eg. a battery charger - any amount of ripple is okay there.)
So your fix (bigger cap, less hum) makes perfect sense. But the logic falls apart if we try to extrapolate your experience to every power supply: if a wall-wart / power supply already has adequate filtering built into it, adding more capacitance is unlikely to cause an improvement, and may cause additional problems.
So the takeaway is "If there's a problem, find the cause, and fix it". That's very different from "More capacitance is always better", isn't it?
-Gnobuddy
This is another example of an audio myth that is also hogwash. Here's a thought puzzle for you: how does an electron know it is in the signal path and how does it decide to stay there?
Attachments
So.... in summary
Use a good brand of caps for the guitar amp, but don't need to be the audiophile level ones because won't make enough of a dif to care about
Mini-amp... can recap if wanting to, but for what it is, don't bother, just make sure the speakers are good. Can recap the power input cap, in which event, higher uf and higher volt ratings are always better to a point where it can't get better.
That about sum it up?
Use a good brand of caps for the guitar amp, but don't need to be the audiophile level ones because won't make enough of a dif to care about
Mini-amp... can recap if wanting to, but for what it is, don't bother, just make sure the speakers are good. Can recap the power input cap, in which event, higher uf and higher volt ratings are always better to a point where it can't get better.
That about sum it up?
I think I already said it, but... make the thing with ordinary parts. Any "improvements" that could be made with fancier parts won't stop it from working if you don't. Work any bugs out of it. Find out if it does something useful or not as a unit. Then when motivated, swap out a few caps with better or different ones and see what change they do or do not make. Then you decide if the brands matter or not. There are only a couple caps, so this experiment would cost little.
Assuming you wind up maintaining interest in this, as you build more complex circuits, and learn more about them, you will come to grips with components. I myself am not aware of any strong opinions as to brands for solid state amps - meaning usually low voltage circuits. Over in the tube amp world there is generally a discussion of orange drops versus mallory caps, but that is a different story.
Assuming you wind up maintaining interest in this, as you build more complex circuits, and learn more about them, you will come to grips with components. I myself am not aware of any strong opinions as to brands for solid state amps - meaning usually low voltage circuits. Over in the tube amp world there is generally a discussion of orange drops versus mallory caps, but that is a different story.
I'd love to do something with tubes... my Dad had a giant bin of them when he passed away, now they're just sitting there. I suspect they go back as far as the '40s-50s. He knew this stuff, I never really picked up on it until his memory started to go so, a bit of a shame there.
My first guitar amp was a tube amp he built himself in the '60s... when my folks moved I think it was either lost/buried in the basement, or (and I hope not the case) they tossed it out.
My first guitar amp was a tube amp he built himself in the '60s... when my folks moved I think it was either lost/buried in the basement, or (and I hope not the case) they tossed it out.
If that remains an interest, you will get there. Just take as much time as you need to learn enough to be safe. High voltages (which tubes need) can be very dangerous.
Some tubes now go for stupid money (hundreds of dollars). Others are essentially worthless (one US store sells them for $1 each). My suggestion is - just don't chuck anything from your dad's bin of tubes out, or use it up, until you know what it's worth. Depending on what's in there, it may be worth next to nothing, or there may be enough to pay for a nice vacation.
Well, that's a nice memory to cherish!
If you get involved in audio electronics, keep in mind that it's an old, mature, engineering field. In 1940, it was cutting edge, and very smart and educated researchers and engineers filled the field. As the field matured, and answers were found to all the burning questions, most of the qualified people left. And then the kooks and the ignorant and the superstitious moved in, slowly replacing science and engineering with superstition and nonsense.
What that means to you is that probably the majority of the audio electronics information you find on the Internet is actually, to varying degrees, nonsense. You can usually put a bit more trust old sources (1950s and 60s), or actual engineering textbooks (Horowitz and Hill has been the gold standard for semiconductor electronics for decades).
So Caveat Emptor. Buyer beware!
-Gnobuddy
Your battery project is a good place to start on circuits in general, and is quite safe to poke in. I think for many people tube circuits are more intuitive, at least superficially easy to understand. But there are some dangers, and having some skill and technique before going into them I think is good. There are many tube projects to choose from, some pretty basic yet pleasing.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.