Simple Symetrical Amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
From listening experience , amp with multiple caps (16x 5000uf ) don't have as good a bass as (4x20000) , sounds weaker or put another way , everytime I have heard an amplifier that has exceptional bass , slam and control , it is never with large quantity of small caps , in the PSU...
I think if you took the time to properly compare four identical amplifiers, one pair with a big caps and the other pair with lots of little caps you will reach a different conclusion.
 
I think a.wayne was correct in his observation. There are many things affecting the performance. One of them is the trace inductance as Esperado has mentioned. So there must be a balance considering all possible parameters affecting the performance.

If there is a parameter to suspect, I think it is the maximum ESR. For cheap caps, big capacitance is a big no-no. But for good caps, big capacitance is fine.

Also, they mentioned about higher voltage rating having lower distortion. I have no observation with power supply caps, but I have observed and practiced this with output caps.

About one big versus many smalls, it is almost similar to tone control phenomena. When you get good treble, you get bad bass. When you get good bass, you get bad treble.
 
I have done the multiple little cappy thing , I did not like the bass , worst the amplifier had no energy, best was as discussed previously , big ballsy bwoys in the main psu , smaller caps closer to outputs on board ...

Copper bars instead of wire , everything as short as possible ....

Just saying ...
 
Thank goodness they [audiophile caps] are not available here [in India]
What the hell are doing the marketing guys ? They canvass Emirs and ignore Maharajas?
About one big versus many smalls, it is almost similar to tone control phenomena. When you get good treble, you get bad bass. When you get good bass, you get bad treble.
So true !
Your speaker protection idea is the most complex one I have seen on this planet,
My goodness! And I thought it was as simple as the SSA? Just a single comparator with some appetizers around.
but it fails to eliminate a (cheap?) relay at the amp output.
Different approach (such as used by Pioneer if I'm not mistaken) is to ground the input and output of the amp when DC is detected at output. I don't know if this approach is effective enough but why not?
Why do you treat my relay cheap ? They thought they where expensive and they are so sad, now.

This protection isolate the loudspeaker during power on/off for no click, noise etc.. and provide a soft current charge in the PSU caps. I would not like to short circuit the Amp output during this time where offset can Rock'nroll in a funky way ;-)

I'm unable to hear any difference between the relays and a big wire soldered across their contact. So, I'm not emotional about it.
Yes, my loudspeakers are impedance compensated both motional and zobel (i highly recommend) , so they present a quasi flat impedance curve, and, so, slight difference in damping factor (internal impedance of the amp) by the (very) little resistance of the relay's contacts is not so sensible.

Last, you can parallel an active device to reduce further this resistance and contacts problems. I use Big relays with two paralleled contacts for each channel.

Of course, you can forget about protection, (a strait line wire is less expensive and sound the best ) if Murphy ignore where you are living. I write this under his supervision.
 
Last edited:
What the hell are doing the marketing guys ?

Their best.

They canvass Emirs and ignore Maharajas?

No dear Esperado. They are doing their country a favor by not importing electronic 'Maharajas'. And we, the Indians are thankful for that. They learn their lesson by watching those who suffered the maharaja's wrath. ;-D
Also, we don't export maharajas and 'tin-ear' is an unpopular/unknown phrase here.
;-)
 
Last edited:
Best no protection , no fuse straight to speaker , of course very dependent on system resolution.
Well, high resolution speakers (are mine part of them ?) are very expensive, right ?
Who want to get the risk to destroy them ?
About fuse in the DC lines, i can hear the sound degradation they bring on dynamic, a lot more than relays in the HP lines, so i use fuse only during the first power on after every change on the Amp.

About work like comparing sound differences between paralleled and big caps, different condenser technologies etc... i first lie on measurements and physic laws when i can (response curves, distortion) Then i listen to correlate.
If the difference is obvious, no problem.
If my ears preferences go against measurement,, time to ask-myself questions.

I'm very suspicious about hearing judgments on subtle differences, can lie to myself, auto convince, imagine, prefer my habits etc...

I do not believe on blind test neither.
(Remember a blind test between mini-disk copies (ATRAC compression) and original where most of the sound engineers participating to this test where preferring the copy to the original ;-)
You can loose a lot of time in that kind of comparisons and be absolutely on the wrong way.

So, since years, before to validate a change, i just listen to music with no attention to the system, and let impressions to impose themselves after some days or week on a lot of different sources.

It is better to use a system to listen music than to listen endless to the system itself , it is a terrible disease to be an audiophile maniac ;-)
 
About fuse in the DC lines, i can hear the sound degradation they bring on dynamic, a lot more than relays in the HP lines, so i use fuse only during the first power on after every change on the Amp.

Do you notice degradation even if you have a small cap in parallel with the fuse and have the fuses upstream of the main reservoir caps ?

I find a good solution is to use constantan resistance wire soldered between two posts as an o/p fuse - I like 0.1 - 0.2R on the o/p - and it to my ears it does not degrade the sound at all.

any thoughts on that ?

mike
 
Well, high resolution speakers (are mine part of them ?) are very expensive, right ?
Who want to get the risk to destroy them ?
About fuse in the DC lines, i can hear the sound degradation they bring on dynamic, a lot more than relays in the HP lines, so i use fuse only during the first power on after every change on the Amp.

About work like comparing sound differences between paralleled and big caps, different condenser technologies etc... i first lie on measurements and physic laws when i can (response curves, distortion) Then i listen to correlate.
If the difference is obvious, no problem.
If my ears preferences go against measurement,, time to ask-myself questions.

I'm very suspicious about hearing judgments on subtle differences, can lie to myself, auto convince, imagine, prefer my habits etc...

I do not believe on blind test neither.
(Remember a blind test between mini-disk copies (ATRAC compression) and original where most of the sound engineers participating to this test where preferring the copy to the original ;-)
You can loose a lot of time in that kind of comparisons and be absolutely on the wrong way.

So, since years, before to validate a change, i just listen to music with no attention to the system, and let impressions to impose themselves after some days or week on a lot of different sources.

It is better to use a system to listen music than to listen endless to the system itself , it is a terrible disease to be an audiophile maniac ;-)

So make it good to hear the music , but not better to hear more realism of said music ...:confused:

Err, Ok, Noted ... :)

Anyway , I may or may not be upset if the amp lets go and take out my speaker, I definitely will be upset with less sound and since this is DIY, I prefer no current limiters, no fuses, no relays,( of course until amp is secure) no input capacitors, nothing between outputs and Speaker wire. I will fuse PSU line, which will be provided with the biggest wire i can fit, plugged directly to the wall, into the best receptacle with in reason...

Small, high impedance and high efficiency speakers will suffer differently, require a different approach and have different sound, So there will always be a different approach to satisfaction ....
 
Last edited:
Do you notice degradation even if you have a small cap in parallel with the fuse and have the fuses upstream of the main reservoir caps ?

I find a good solution is to use constantan resistance wire soldered between two posts as an o/p fuse - I like 0.1 - 0.2R on the o/p - and it to my ears it does not degrade the sound at all.

any thoughts on that ?

mike
If the fuse is before the caps, it will not protect the amp, as the caps are free to discharge across power units even after it has fired..

The problem with the fuses are they present an impedance that greatly vary with temp/amps.
I never tried your tip. Is-it as fast as a fast fuse ? No corrosion problem if your wire is in free air ? Did they present a constant impedance ?
 
Last edited:
The whole idea of constantan resistance wire is that it has a very low temp coefficient.

I have used on amp o/p and I have not noticed any corrosion or degradation in sound.

The values I use has successfully blown at various appropriate times but I have not done all scenarios testing.

It works for me but true, this is a non standard measure and so does not come with guarentees.

mike

p.s. this wire sounds as good as any resistor I have found and much better than most that I have tried in this position.
 
Last edited:
The whole idea of constantan resistance wire is that it has a very low temp coefficient.
Constantan have a 1 300 °C melting point. I presume it work at very high temp, on normal usage (not so far from the melting point ?). Any idea how is the impedance curve at those temps ? (supposed to be near flat between 20 °C et 100 °C ).
How to calibrate the max amp with enough precision ?
No manufacturer offer calibrated "audio" fuses using constantan wires ?
Your tip sound good, anyway.
 
Last edited:
How to calibrate the max amp with enough precision ?.

Would need to establish a standard construction method and then do some experimentation to establish blow times at different currents - can't beat empirical method :D

This is a very old fashioned approach - when fuse blows have to get out the fusewire & soldering iron !

But for DIYers this is not a problem.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.