Hello,
[SOLVED: it's some kind of compressor - thanks all]
I'm building a small guitar bass amplifier, just for fun.
I would like to use TDA2030 for the power amplifier. I've already used this chip for a guitar amplifier and it sounds pretty good.
My amplifier will have a dual rail supply -18V 0 +18V.
As usually, I've checked some famous brand schematics that used this chip on their product.
I've notice something "unusual" on the schematics I've found.
As example find here the Ampeg BA-110A (thanks to Ampeg the deploy schematics to anyone) https://ampeg.com/support/files/Schematics/B%20Series/BA110/262sch_1.pdf
You can also find something similar in the Peavy Microbass circuit, and in the Fender Rumble 15 and others.
It seems to me this is some kind of protection circuit that pulls down the signal if a certain voltage raise on the output.
1) Am i right? Is it a protection circuit?
2) Why does they I need it, usually guitar amplifier does not have this circuit (see as example the Marshal MG10: ElectroSmash - Marshall MG10 Analysis.)
thanks!
[SOLVED: it's some kind of compressor - thanks all]
I'm building a small guitar bass amplifier, just for fun.
I would like to use TDA2030 for the power amplifier. I've already used this chip for a guitar amplifier and it sounds pretty good.
My amplifier will have a dual rail supply -18V 0 +18V.
As usually, I've checked some famous brand schematics that used this chip on their product.
I've notice something "unusual" on the schematics I've found.
As example find here the Ampeg BA-110A (thanks to Ampeg the deploy schematics to anyone) https://ampeg.com/support/files/Schematics/B%20Series/BA110/262sch_1.pdf
You can also find something similar in the Peavy Microbass circuit, and in the Fender Rumble 15 and others.
It seems to me this is some kind of protection circuit that pulls down the signal if a certain voltage raise on the output.
1) Am i right? Is it a protection circuit?
2) Why does they I need it, usually guitar amplifier does not have this circuit (see as example the Marshal MG10: ElectroSmash - Marshall MG10 Analysis.)
thanks!
Last edited:
"1) Am i right? Is it a protection circuit?"
It is a compressor. The more ac signal across the opto coupler, the lower the value of the shunting resistor.
It is a compressor. The more ac signal across the opto coupler, the lower the value of the shunting resistor.
I was wrong 
perfect 🙄 I thought the voltage divider R31_47K + OC1_max1M (completely lower down the signal), indeed the output resistance of the OC1 depends on the current input.
Thank you!
What I've notice is that Peavy Microbass and Fender Rumble use a slightly different approach, this one of Ampeg seems much more simpler to me 🙂
I'll try to build this one, but If you got any suggestion, you're welcome.
One ore question is about the R32 + C23, this is a low pass filter If I'm not wrong but I don't get it why they need it, it's a 230Khz low pass, that is way to high, but maybe I'm wrong like before
perfect 🙄 I thought the voltage divider R31_47K + OC1_max1M (completely lower down the signal), indeed the output resistance of the OC1 depends on the current input.Thank you!
What I've notice is that Peavy Microbass and Fender Rumble use a slightly different approach, this one of Ampeg seems much more simpler to me 🙂
I'll try to build this one, but If you got any suggestion, you're welcome.
One ore question is about the R32 + C23, this is a low pass filter If I'm not wrong but I don't get it why they need it, it's a 230Khz low pass, that is way to high, but maybe I'm wrong like before
Hi
C23 and R32 does not form a LP filter.
Whereas R31 in parallel with OC1 and R36 plus R38 forms a LP filter.
In the situation where the compressor is not active (OC1 >> 47K) the LP cutover frequency will be 1/(2*PI*(47k||47k + 1k)*200pf) ~ 30KHz.
This is probably still too high, but will reduce the risc of oscillation.
Regards
Martin
C23 and R32 does not form a LP filter.
Whereas R31 in parallel with OC1 and R36 plus R38 forms a LP filter.
In the situation where the compressor is not active (OC1 >> 47K) the LP cutover frequency will be 1/(2*PI*(47k||47k + 1k)*200pf) ~ 30KHz.
This is probably still too high, but will reduce the risc of oscillation.
Regards
Martin
Thanks @diyMartin,
I mean R38 not R32 sorry and C32, also I must be drunken I mean ~700Khz, not 230Khz, 1/(2*PI*1k*200pf) ~ 720KHz
I don't take the two 47k(+OC1) || 47k in account. Now I get it.
You have to excuse me but my background is in the Software world.
Should i use Thevenin equivalent circuits to solve this circuit, right?
One more clarification, If i put a 2.2uF before the R31, it will work like a High pass filter 2.2uF + R31 in series with OC1, forming a ~1.5Hz high pass filter, when OC1 is on the filter pass more band. Is it right?
Thank you!
I mean R38 not R32 sorry and C32, also I must be drunken I mean ~700Khz, not 230Khz, 1/(2*PI*1k*200pf) ~ 720KHz
I don't take the two 47k(+OC1) || 47k in account. Now I get it.
You have to excuse me but my background is in the Software world.
Should i use Thevenin equivalent circuits to solve this circuit, right?
One more clarification, If i put a 2.2uF before the R31, it will work like a High pass filter 2.2uF + R31 in series with OC1, forming a ~1.5Hz high pass filter, when OC1 is on the filter pass more band. Is it right?
Thank you!
My background is actually electronics - but it dates back to around 1980, with the unfortunate result that I use some (probably several) methods for circuit analysis without even thinking about the name of the method.
So you just reminded me that what I did to calculate the effective resistance C23 is connected to is actually called the Thevenin equivalent.
Thanks!
So: yes you are right - to determine the resistance C23 is "looking into" you have to find the thevenin equivalent.
This (the thevenin equivalent) also applies for a capacitor in series with R31.
The resistance here will be R31 + OC1 || R36 (and I dont take the input impedance of the TDA2050 into consideration).
So: depending on OC1 the resistance for this HP filter will be between 47K and 94K.
The cut-over frequency will therefore be between 0.77 and 1.5 Hz.
I can only recommend that you do this (adding a capacitor in series with R31) - otherwise a DC error on the CD input can cause damaging DC on the speaker output.
BTW: If you intend to copy the schematics - including the flat cable to connect the power amplifier to the rest of the circuitry, then be aware that the TDA2050 gets its DC bias via the cable (pins 2 on the cable).
I would likely myself have made sure the bias was always present by having either R38 and R36 on "the other side" of the flat cable or by adding maybe a 100K resistor directly from pin 1 (TDA2050) and ground
Cheers,
Martin
So you just reminded me that what I did to calculate the effective resistance C23 is connected to is actually called the Thevenin equivalent.
Thanks!
So: yes you are right - to determine the resistance C23 is "looking into" you have to find the thevenin equivalent.
This (the thevenin equivalent) also applies for a capacitor in series with R31.
The resistance here will be R31 + OC1 || R36 (and I dont take the input impedance of the TDA2050 into consideration).
So: depending on OC1 the resistance for this HP filter will be between 47K and 94K.
The cut-over frequency will therefore be between 0.77 and 1.5 Hz.
I can only recommend that you do this (adding a capacitor in series with R31) - otherwise a DC error on the CD input can cause damaging DC on the speaker output.
BTW: If you intend to copy the schematics - including the flat cable to connect the power amplifier to the rest of the circuitry, then be aware that the TDA2050 gets its DC bias via the cable (pins 2 on the cable).
I would likely myself have made sure the bias was always present by having either R38 and R36 on "the other side" of the flat cable or by adding maybe a 100K resistor directly from pin 1 (TDA2050) and ground
Cheers,
Martin
A protection circuit would disconnect amp from speaker if a DC fault.
Simply reducing volume wouldn't stop that.
Simply reducing volume wouldn't stop that.