Buffer before amp TDA7231 produces no sound

Started by Ksander, September 25, 2022, 02:37:01 PM

Previous topic - Next topic

Ksander

Dear all,

Its not a stompbox, but I guess people here will be familiar: I'm experimenting with variants of the ruby amp, and have found some TDA7231 amplifiers instead of the 386. I've tried the schematic from the datasheet with a guitar as input, which works fine, and I've also tried putting two in bridged mode, which also works. Now I want to put a buffer in front, to play around with the gain and implement some filtering. However, when i put a basic jfet (j113) or op amp (tl072) buffer (as described on the muzique website) in front of the amp, I get no sound. I know the buffer works, because I've tried it in isolation.

Why might this be?

ElectricDruid

Show us a schematic and we might be able to tell you. Otherwise, we'd just be guessing.

Thanks!

Ksander

I hacked this one together in paint, so it's ugly, but this is it:



antonis

#3
Cut the wiring between 4 & 3 and and 4 & 1 and put a 10k resistor from pin 4 to GND..

"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

FiveseveN

Quote from: antonis on September 26, 2022, 08:49:12 AM
Cut the wiring between 4 & 3 and and 4 & 1

I'm pretty sure the rectangle around the TDA7231 symbol is supposed to represent the component and not all its pins shorted to each other. Unfortunate use of a single line weight/color/style.
Other than that yes, ground reference yer input (with that 10K resistor)!
Quote from: R.G. on July 31, 2018, 10:34:30 PMDoes the circuit sound better when oriented to magnetic north under a pyramid?

Ksander

#5
Indeed, the rectangle was just unfortunate drawing; it didn't represent connections. With the 10K resistor placed as suggested, I indeed get sound. However, it sounds horrible; grainy. It's much worse than without the buffer. Any ideas why that would be? And why is it necessary to geound-reference the input? Why the value 10K?

Anyway, thanks! it is nice to make progress ;D

Edit: maybe I blew up the speaker  :icon_sad:

antonis

Could you plz post a realistic schematic of your particular amp..??
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Ksander

Quote from: antonis on September 26, 2022, 02:47:44 PM
Could you plz post a realistic schematic of your particular amp..??

Certainly. This is the schematic from the datasheet, which I breadboarded. It works, also without the 10k input resistor, but it gives off a bit of a grainy sound. I guess this could be the speaker being blown, or maybe it is due to some limitations imposed by the breadboard?

When I put the buffer in front of the amp (as in the previous image), at first I got no sound. However, putting the 10k resistor at the amp input (as is actually also in the datasheet), it does work. Unfortunately it sounds worse with the buffer. I don't understand why, and I also don't see how putting the 10k resistor at the amp input solved the issue of getting no sound.


antonis

Quote from: Ksander on September 26, 2022, 04:48:14 PM
When I put the buffer in front of the amp (as in the previous image), at first I got no sound. However, putting the 10k resistor at the amp input (as is actually also in the datasheet), it does work. Unfortunately it sounds worse with the buffer. I don't understand why, and I also don't see how putting the 10k resistor at the amp input solved the issue of getting no sound.

Non-inverting input DOES need some DC bias (zero volt hereto)..
Try to make buffer output cap 470nF - 1μF to see what happens..
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Rob Strand

#9
QuoteNon-inverting input DOES need some DC bias (zero volt hereto)..
It's going to be that.

The datasheet doesn't spell it out but reading between the lines a resistor is required from TDA7231 pin 4 to ground.
In other words put the 10k back in after the cap.

Some chips have input resistors to ground inside the chip but maybe the TDA7231 does not.


For example, you can see the TDA2822 doesn't have resistors,

https://pdf1.alldatasheet.com/datasheet-pdf/view/25057/STMICROELECTRONICS/TDA2822M.html

But the LM380 does,

https://pdf1.alldatasheet.com/datasheet-pdf/view/8880/NSC/LM380.html

No internal schematic for the TDA7231 so we have to infer no internal resistors from the example circuit.


Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.