Very complicated amp

[HM666]

What do you think abou this amp(schematic is aded as attachment)? Sory for image quality.

[Nysan]

Hmm .. its just constructed with an NPN inputstage.
Not really common to do that but there should be no reason not to.
There are some diodes though that I really dont understand the purpose with using.
The ones at the output and the ones at the input ..
Anyone?
And I dont think the diode over the cap on the Feedback
is nessecary even if you use an electrolytic.


/Nysan "Hater of diodes"

[richie00boy]

Diodes at the input
These are zener diodes and are used to create reference voltage either side of the trimmer so that the DC offset can be adjusted.

Diodes at the output
These are used as catch diodes to return any large back-EMF induced by the loudspeaker back into the supply rails rather than reverse-bias the output transistor and possibly damage it.

[traderbam]

Curiously, the 47uF dc blocker in the feedback path is around the wrong way, assuming the dc offset is trimmed to zero at the output.

[runebivrin]

How do you figure that? With zero volts on the output, the voltage across the DC blocker would be zero volts too. How is there a right and a wrong way to zero volts?

Edit: However, on closer inspection I think the decoupling caps on the voltage rails to the output stage are the wrong way around.


Rune

[Nick Walker]

As a standard practice the DC blocker looks the right way around to me. Normally the base current at the input transistor is enough to cause a small positive voltage at steady state. So this will get amplified x11, correctly polarising the nfb capacitor, then cancelled at the -ve differential input.

I'm not sure of the need for the Zener biasing to cancel the DC given that there is a DC blocker in the nfb loop. However given the zeners I think the nfb cap and the diode could be removed if the trimming is correct, provided you're not concerned about an OPS failure that could impress DC on the NFB that is.

Some times zeners on the input are used to provide input level protection, but the configuration is not the same as shown.

[ChocoHolic]

I think the 47uF DC-blocker should be connected with the opposite polarity. If we assume that the output is at zero and the DC bias for the basis is delivered through the feedback resistor, then this current will have to flow from output (zero) to the basis and result in a DC voltage at the basis just little below zero..... But in fact this voltage is small, so the cap will probably work without major issues not matter in which direction you connect it.

My general proposal is to use film caps, or use two electrolytic in anti seriell connection (double value each) and give some bias volts (5V....15V) with a high value resistor (1Mohm ....10Mohm) to the center tap of the series connection.

Bye
Markus

[destroyer X]

I will simulate this amplifiers and i will give you some results.

Maybe you already constructed it... tell me some.

When complicated this way, normally explode (Yes!.. thats i am seeing often)

But it is very interesting unit.... working, will be very good.

Carlos

[ChrisW]

Is it just me? Or do those 2200 uF caps (smoothing caps) look the wrong way round to you? Now that would lead to a very dynamic explosion on a high current supply!

[sss]

Originally posted by ChrisW
Is it just me? Or do those 2200 uF caps (smoothing caps) look the wrong way round to you?

well its me too