Importance of component choices and placement for a class A amplifier

Status
Not open for further replies.
Are components created equal? Do they really make a difference to how your amplifier will sound?

Which are better in terms of resistors, metal film, the infamous Z foils or wirewounds which are non inductive?

CoEE94Z.jpg


I keep seeing charts saying wirewounds and metal foils have the lowest noise

Capacitors such as Elna silmics, PIO, Nichicon muse, Polyprops and polystyrene! With so much choice and information overload its hard to decide what to choose.

Should resistors have specific locations in terms of there type? I.e power stages of amplifier wirewounds, audio paths sections use z foils&metal films etc etc.


I've been playing with the idea of upgrading my amplifier boards which are from china, now not knowing what brand the resistors are and the capacitors made me think i actually buy better parts and upgrade it.

Thing is i am still learning how it all works and thought i would ask you guys for some help.


These are jims audio Jean Hiraga 30w amp boards.


xDoRlCx.jpg



This is the circuit diagram

CkjfJ6C.png




This are some of my notes with choices of upgrades.

ZBtHZbT.jpg



My thoughts were to by vishay dale rn65 series resistors having read about how good they are (low noise etc)

For the kubota stages i was going to replace the capacitors with elna silmics ii 50v 100uf simple and better then what was given in the kit, i would also replace all the resistors with Vishay dales rn series resistors (all low noise types).


Now the next bit i cant figure out, what are my audio paths?

It made sense to me that my audio input resistors i need really good resistors such as the charcroft z foils, those being

R1 - 680r
R2 - 33k

The input capacitor is 150pf my thoughts were to either use silver mica or polystyrene (read and heard polystyrene is better but not with high temps)

would R3 and R4 also be audio paths? Could i use z foils here to?

I wasnt sure what my audio output paths would be.

I thought R11,R17 AND R18 must be important so some good quality resistors here would be important, also C4 is 2200pf could i just use a PIO capacitor instead here?

My thoughts also were that the power resistors supplying voltage to the transistors should be wirewounds non inductive types such as vishay mills R13, R14,R15, R16.

Maybe im overthinking/overkilling this?

Can anyone tell me what resistors do what i.e audio paths, power paths etc that would be really helpful!

Thanks
 
HI. I would consider the practice of build it in stages
First the PSU stage then the amp.
Or FIRST the amp but with an unregulated supply (if available)

The key for success in class A is perfect match of area of transistor to the heatsink - but that's true for every tipe of power device - and the eventual isolator pad with heat conductive paste, and the circulation of air around the heatsink ( which usually it is big, so area/mass of air is fairly good.
Sooo... that's the real ( thermic) resistance!
 
Thanks Pico, the PSU and chassis, heatsinks are all ready and well above spec. Here i am trying to upgrade parts to better ones.

Just want to know what people experience are with specific types of components for sections of the amplifier.

Could someone specify which resistors are in the audio path and which are important to be low noise for the trannies?

For me the audio path is the most important, which is what i am trying to find out what the most important resistors are for this circuit design as then i would go Z foil resistors if it suitable as they are only 0.4w.

What would people change, use or add to make this better. Wirewounds for power to and from the transistors, especially high wattage ones?
 
The concept of an "audio path" is flawed. All components are in the audio path in one way or another, to varying degrees but ultimately they all influence the signal quality.

No, for example, the power supply is not in the audio path, and is isolated by the PSRR, (80db). Many of the " better" components will make no difference in sound, ( if the resistor noise is already unnoticeable how will a resistor with less noise make a difference? ). And sometimes make it worse. If a circuit needs a certain amount of ESR in a cap to be stable, replacing it with a "better" lower ESR cap can mke it unstable.

I wuold love to hear and measure the difference between the parts they gave you, and the "better" parts. There may be very little.
 
Last edited:
Totally agree with you Scott, to me after mucking around for years with my set up, I've come to the same conclusion & to all who do not think so pls try & focus on main PS & find out for yourself it's effect on SQ. Reality is one should start with choice of trafo first the type of rectifier diodes followed by PS caps. Ideally we should start with the mains but that's out of our control so we focus on what's next on line. For me I'm able to get custom trafo wound without costing much.
 
R17 at this very critical feedback position faces a large voltage swing.
When not ultra linear, will introduce distortion .
And with 300Ohm, dissipation will be around 1Watt at full power.
A 1 to 2 Watt metal film is to be advised at this position instead of the 0.25Watt currently on the PCB.


I do not understand the function of C4, 220pf.
In a CFA a cap parallel to Rf is a no no.
C9 and C10 should take care of reducing bandwidth / stability.


Hans
 
The concept of an "audio path" is flawed. All components are in the audio path in one way or another, to varying degrees but ultimately they all influence the signal quality.
Some components are on the path and carry signal, others support those components by powering, biasing etc. The concept of a signal path is straight-forward and useful. Yes its a bit nebulous(*) with feedback and differential signals, but clearly there is a difference between say a constant current source and a device amplifying the actual audio! There is a difference between influencing and carrying a signal.


(*) class B output stage is an interesting case of a signal path swapping at every zero-crossing.
 
Last edited:
The concept of a signal path is helpful to those who understand its limitations. It is deeply harmful to those who do not understand, yet these are the ones most likely to use it. An example is the person who believes quite sincerely that series components are in the signal path but shunt components are not, hence prefers to use a volume control which places a variable shunt after a fixed resistor; here a false understanding of signal path leads him to do the very opposite of what is best for signal integrity.

A CCS is every bit as much in the signal path as any other component.
 
Avoid mutual or shared ground returns for signal path and electrolytic and bypass capacitor ground returns. This is more important than a component choice and can make a 20-30 dB difference in power line frequency residuals and its multiples in the output spectrum, reflecting directly in S/N.

That is a broad statement and partially not true. Each stage should be tied back to it's appropriate decoupling capacitor. Those in turn tied back to the bulk PS decoupling capacitor. Input ground to the 1st stage star and output to the last stage star. PS should also go to last stage star as that is the largest potential therefore least susceptible to noise.

Valve Wizard has a great write-up here.
 
Status
Not open for further replies.