Hello
I found this schematic here:
https://www.te1.com.br/2014/03/mini-strong-power-amplifier-100w-rms/amp/
The text is in Spanish, and I'm just wondering is there a way to adjust offset voltage with trim pot if the offset is out of safe range (+/-50 mV)?
I found this schematic here:
https://www.te1.com.br/2014/03/mini-strong-power-amplifier-100w-rms/amp/
The text is in Spanish, and I'm just wondering is there a way to adjust offset voltage with trim pot if the offset is out of safe range (+/-50 mV)?
This design's most glaring shortcoming is its poor thermal tracking of the bias current spreader, so it will tend to have excessive crossover distortion.
That said, it should be possible to address offset error. With respect to ground, would you report voltages at:
base of input differential pair (i.e. at the 56k resistor)
base of feedback differential pair (i.e. at the 56k 100k resistor)
amp output
Please use your meter's most sensitive range and report negative voltages as signed negative numbers.
Finally would you measure the voltage across the 2.7k resistor (i.e. between the 100k feedback and the 220uF).
Thanks!
That said, it should be possible to address offset error. With respect to ground, would you report voltages at:
base of input differential pair (i.e. at the 56k resistor)
base of feedback differential pair (i.e. at the 56k 100k resistor)
amp output
Please use your meter's most sensitive range and report negative voltages as signed negative numbers.
Finally would you measure the voltage across the 2.7k resistor (i.e. between the 100k feedback and the 220uF).
Thanks!
Error! The transistor in the current source of the long-tail pair is turned on incorrectly (swap the emitter and collector). Need BC556.
It is necessary to include 100 ohm resistors between the base and emitter of the output transistors.
It is necessary to include 100 ohm resistors between the base and emitter of the output transistors.
Yes, but notice the hack to fix that mistake, ie the 20K resistor, which becomes the current determining component. Obviously, this is an attempt to replicate a typical "Blameless" topology, but the author does not understand it. Does someone want to post a corrected schematic? If I get really bored, I suppose I could do it.
Thanks everyone for response.
Use the circuit above posted by Mooly as reference.
I measured 189mV between base and ground of Q1, 192mV between base and ground of Q2, and voltage across R7 (2k7) is zero volts.
Output voltage is around 100-120mV
I discovered strange behavior on Q8, he is quite warm enough to feel him but not get burned my fingers, and once is cooled down the output voltage drops to 20-25mV so I installed him an small heatsink secured with screw.
Without input signal there is no buzzing and hissing coming from speaker, for now.
tl;dr I have Logitech Z906 subwoofer with totally burned electronics, so I installed 2x38ACV 150VA classic transformer, good power supply with 2x10 000uF electrolytic capacitors and 4 seconds delay module with relay for the speaker and adjustable subwoofer filter from 35 to 200Hz.
Almost all parts used here are recycled from other boards, but every component are inspected and measured before I solder it. All transistors, IC's and diodes are brand new.
I agree that the schematic is not perfect but I choose this as starting point because is simple and cheap to build, and I wanted to learn a little bit more about solid state amplifiers.
Just to note that I'm not a totally noob, I have basic and advanced knowledge in electronics, and of course I'm not a pro for amplifiers.
(p.s Sorry for my English, not my primary language.)
A small transistor will run hot here. With the supply at -/+42 volts you have essentially 42 volts across the device with a current determined by R9 and R10 which also have 42 volts across them. So the current is 9.5 milliamps and power dissipated 0.4 watt which is a lot for a T092 package. 9.5 milliamps is a bit high, the 2k2s could be increased a little to say 3k3 which would bring the current down to nearer 6ma and dissipation to 0.25 watt. A big difference for heat generation.
Playing around with the simulation and one problem is that it is low on open loop gain. It is very difficult to get the offset down even with component value changes and circuit modifications.
Ok, here are 3 suggestions of different levels of improvement, although far from the ultimate designs. Others can help out explaining, but I will say:
1. Vas and driver transistors need to be larger transistors. TO-98 parts are not big enough for an amp running over 40V total, ie +/-20V.
2. You cannot leave the OP bases with no "discharge" path to prevent shoot though current.
3. You can tweak the offset in all cases, but a current mirror means it will never be far off. An offset pot can be in one of 3 places: the CCS, the LPT load, or the LTP degeneration.
I have added models to my library, but I think everything here is in Bob C's library?
1. Vas and driver transistors need to be larger transistors. TO-98 parts are not big enough for an amp running over 40V total, ie +/-20V.
2. You cannot leave the OP bases with no "discharge" path to prevent shoot though current.
3. You can tweak the offset in all cases, but a current mirror means it will never be far off. An offset pot can be in one of 3 places: the CCS, the LPT load, or the LTP degeneration.
I have added models to my library, but I think everything here is in Bob C's library?
