Why don't you use Sockets for the chip? Much easier.!
Why don't you use Sockets for the chip? Much easier then cutting legs off.. Barbaric..Sorry but that's my feelings about such ways of fixing something.
Cutting legs of is only for devices which are Toats or DEAD, and sure not for something which is unsure.. But I know that's the point where the electronic Ghost's part ways..
Regards Chris
Why don't you use Sockets for the chip? Much easier then cutting legs off.. Barbaric..Sorry but that's my feelings about such ways of fixing something.
Cutting legs of is only for devices which are Toats or DEAD, and sure not for something which is unsure.. But I know that's the point where the electronic Ghost's part ways..
Regards Chris
I read of others that use the circuit as is and prefered it.
anyway, i would like to learn a few things from the exercise so I am starting from the beginning and will make improvements as needed on my own.
do you believe that will solve my issues or just a suggested improvement?
That I did not do. what do you mean by power supply? you mean the transformer ground (center tap)? I connected it to the incoming ground at the amplifier power pins (as star ground), which comes across from the center tap.
I had assumed as much. As said I want to start from the reference and work my way up. And this will balance the input bias? I already thought about doing it that way but did not know that was ideal.
Because I don't have one. In fact I have never seen a socket for this pin layout. And besides sockets always seem to me crude or show a lack of faith/adventure/commitment. The real hesitation is that I only have one extra chip at the moment 😱
Funny thing is, it actually worked better before I fixed all the errors! which is of course ironic...
The circuit 'as is' will be very dependent on the exact way it is built and also the load it is used with.
Not having an output inductor means that a 'real' load consisting of speaker cable and speaker rather than an 8 ohm resistor test load could cause stability issues.
Returning the speaker to the power supply removes any possible interaction of the speaker current with the circuitry around the chip.
It is very easy to cause stability problems unless you fully understand all the aspects of what can happen so this is an easy way to remove that possibility.
The circuit you have posted above still isn't correct around the input stage. Yours will alter the DC offset as the pot is turned. You must AC couple the wiper. Also the little cap should be on the other end of the 1k. The exact values aren't so critical, the way it all connects up is.
Copy the one I posted a little earlier 🙂
You should start with a tried and tested design and layout and work backwards from that if you wish, not the other way around.
Without a shock or two and exploding components where is the fun ?
I did my electronics City and Guilds in 1980 on a Government funded course.
There were 12 of us on the course.
We all had to build up a power amplifier.
The fun part came at first turn on.
Exploding electrolytics the wrong way around and burning smells.
Great fun.
Every time a cap exploded there was a big cheer from the class.
To test the amps we were supplied with one record deck and one vinyl record which was Queens "Bohemian Rhapsody"
Never in my life have I so quickly come to hate a song !
I did my electronics City and Guilds in 1980 on a Government funded course.
There were 12 of us on the course.
We all had to build up a power amplifier.
The fun part came at first turn on.
Exploding electrolytics the wrong way around and burning smells.
Great fun.
Every time a cap exploded there was a big cheer from the class.
To test the amps we were supplied with one record deck and one vinyl record which was Queens "Bohemian Rhapsody"
Never in my life have I so quickly come to hate a song !
I have decided the best way forward and simplest way is to just build a second board with my remaining chip. That way, hopefully, I learn something in the process. It will be slightly different, particularly in the input section, taking advantage of what I have been suggested here.
I will update in a few days when I get to the result....
I will update in a few days when I get to the result....
Not only the input, the output Zobel is FUNDAMENTAL.
Same with grounding,
Quote:
Proper way is to build a power supply on its own, connect transformer center tap straight to filter capacitor joint, which is the real ground node, and return chipamp/speaker return/etc grounds there using separate wires, that is a true star ground.
Same with grounding,
Quote:
That´s the worst way to do it, since "everything", including speaker return, connects to the real ground through the same wire.
Proper way is to build a power supply on its own, connect transformer center tap straight to filter capacitor joint, which is the real ground node, and return chipamp/speaker return/etc grounds there using separate wires, that is a true star ground.
The power supply is on a separate board. The center tap goes to the amp board to a star ground with separate wires for each ground path.
Which includes the center of the smaller filter caps as close to the power pins (as suggested by the datasheet and every other application I have read.
If that is not the established DIY methodology then I will apply the suggestion to move the caps away from the amp to the first ground point star.
I accept the tip, but fail to see how that could cause the problem I experience at mid volume.
also a side question: Is it possible/recommended to made this chip amp DC stable? I would be interested in this possibility. but for the current purpose of a test and measurement amp it is unwanted.