Taken from
http://focus.ti.com/lit/an/szza036b/szza036b.pdf
Page 44:
Oh ok. So inputs are voltage controlled not current for the XOR.
http://focus.ti.com/lit/an/szza036b/szza036b.pdf
Page 44:
Oh ok. So inputs are voltage controlled not current for the XOR.
Last edited:
Attachments
The input cap is used to decouple the input DC. And remember this "POP" effect when the amp is powered on: this is caused by charging of this 3,3uF input cap. You can use 10uF here too, but you will get more power "POP". Also, it is possible to use 1uF cap here (and before pot too), but this will cause noticiable bass reduction (real high pass filter effect). So you can explain 3.3uF so, that you was needed to reduce the POP effect but not lost the bass frequencies too.
Generally it is correct, but it is needed to explain, where you get Ipk=2.5-3.0A from...
Irms = Vrms/R, so you do not need to use sqrt(2) here... Also, the simplified formula is P=U*U/R or P=I*I*R, it can be get using Ohm's Law.So P = Vrms*Vrms/R = 14,14*14,14 / 8 = 24,99 or P=14,14*14,14 / 4 = 49,99 or P=14,14*14,14 / 2 = 99,96.
Last edited:
24.99999?
Grrrr! 1/sqrt(2)*1/sqrt(2)=2 exactly, and not 1.999999.
And how could you calculate with these precision, where none of the source datas are precise more then 0.5 %? A sensible person wouldn't write more then 3 digit.
And duty cycle is strongly limited, so 25W is not possible, 20 W is more probable.
I would take account dead time, Rdson, and the resistance of the filter choke too.
Grrrr! 1/sqrt(2)*1/sqrt(2)=2 exactly, and not 1.999999.
And how could you calculate with these precision, where none of the source datas are precise more then 0.5 %? A sensible person wouldn't write more then 3 digit.
And duty cycle is strongly limited, so 25W is not possible, 20 W is more probable.
I would take account dead time, Rdson, and the resistance of the filter choke too.
No. The volume pot sets NO any bias... The bias is set by your duty cycle pot (~2V) on the positive input of TL494. The same DC voltage should be presented at negative input too. This is achieved by the fb resistor (22k). And it is needed to decouple this DC voltage, which is done via 3.3uF cap.
Generally, there is no need to use the cap before the volume pot, but if some music source has an DC voltage at the output (very rare), then this cap is needed...
Last edited:
ok, thanks for the clarification.
I was reading Page 100 from G. Randy Slone “High-Power Audio Amplifier Construction Manual”
He had a schematic which had the voltage divider 10k potentiometer with a capacitor at the output of the pot. But the input capacitor before the 10K pot was missing, so I got concerned. Because I'm going to reference to this if he asks how did you manage to volume control your amplifier.
Also I just found a textbook that mentions about possible DC components in the musical signal, just like you said.
So i just found a reason to blurt and quote with evidence why I used the 10uF cap and also the 3.3uF cap.
I was reading Page 100 from G. Randy Slone “High-Power Audio Amplifier Construction Manual”
He had a schematic which had the voltage divider 10k potentiometer with a capacitor at the output of the pot. But the input capacitor before the 10K pot was missing, so I got concerned. Because I'm going to reference to this if he asks how did you manage to volume control your amplifier.
Also I just found a textbook that mentions about possible DC components in the musical signal, just like you said.
So i just found a reason to blurt and quote with evidence why I used the 10uF cap and also the 3.3uF cap.
The duty cycle is fixed at 50% with no modulation.
When modulation occurs, the DC voltage produced at the output of the error amplifier (F/B pin). The 3.3uF capacitor blocks the DC component of that, so that only AC component goes into the error amplifier.
So what I see in the scope when modulation occurs is the AC variation (As the modulation swings from either side of the 50% Duty Cycle).
So when the error amplifier compares the error signal and the 2V reference, to compensate for any voltage difference between the two inputs. The 3.3uF cap blocks the DC component so only AC component passes through.
so for example at some given point in time, the music signal after the volume pot is 0.4V, then the output from the F/B pin is 4V, which results in a change in duty cycle from the 50% mark to something above 50%.
And let's say now it goes to 0.1V after volume pot from 0.4V. so F/B pin output is 1V, so duty cycle swing is from 50% to something below it.
These are all AC signal variations?
{But ultimately the zero reference is 50%.
There is no uncecessary biasing at the input of the Tl494 which can move the duty cycle away from the 50% zero reference mark. }
But the Output voltage from the error amplifier which goes to the PWM comparator in the internal circuitry. That is still DC voltage.
I have never seen the output from the Error Amplifier change (I will double check this to make sure), since the output from the Error Amplifier is a DC voltage (Think my scope is DC coupled). Although it would make sense if it did change value, which would explain the modulation swing from the 50% as music comes through.
Sounds like there's an AC component coming out of the error amplifier which I am not seeing at the F/B pin (again scope is DC coupled I think) and this is what's causing the modulation of the PWM at the TL494 output transistors, because the output from the error amplifier is a fixed voltage value which is confusing me as to how is the PWM modulating then.
When modulation occurs, the DC voltage produced at the output of the error amplifier (F/B pin). The 3.3uF capacitor blocks the DC component of that, so that only AC component goes into the error amplifier.
So what I see in the scope when modulation occurs is the AC variation (As the modulation swings from either side of the 50% Duty Cycle).
So when the error amplifier compares the error signal and the 2V reference, to compensate for any voltage difference between the two inputs. The 3.3uF cap blocks the DC component so only AC component passes through.
so for example at some given point in time, the music signal after the volume pot is 0.4V, then the output from the F/B pin is 4V, which results in a change in duty cycle from the 50% mark to something above 50%.
And let's say now it goes to 0.1V after volume pot from 0.4V. so F/B pin output is 1V, so duty cycle swing is from 50% to something below it.
These are all AC signal variations?
{But ultimately the zero reference is 50%.
There is no uncecessary biasing at the input of the Tl494 which can move the duty cycle away from the 50% zero reference mark. }
But the Output voltage from the error amplifier which goes to the PWM comparator in the internal circuitry. That is still DC voltage.
I have never seen the output from the Error Amplifier change (I will double check this to make sure), since the output from the Error Amplifier is a DC voltage (Think my scope is DC coupled). Although it would make sense if it did change value, which would explain the modulation swing from the 50% as music comes through.
Sounds like there's an AC component coming out of the error amplifier which I am not seeing at the F/B pin (again scope is DC coupled I think) and this is what's causing the modulation of the PWM at the TL494 output transistors, because the output from the error amplifier is a fixed voltage value which is confusing me as to how is the PWM modulating then.
Last edited:
NO. The error amp has DC 2V at both positive and negative inputs, and has additional AC component via 3.3uF+2.2K at negative input. This AC component should be compensated via FB resistor (22K). So at the error amp's output we see DC 2V AND amplified AC component. So this 2V+AC voltage will be compared with sawtooth (0 to 4V?) inside of TL494, which produces the PWM then.
What this 3.3uF cap is needed for, is to produce DC decoupling between volume pot and TL494 input...
ok kewl. The error signal from the output of that summing junction is a AC signal (going to Pin 15) and this AC signal is what's required to be compensated by the error amplifier.
So pwm modulation I see is because of this varying AC voltage. Since it looks like the DC voltage of about 2volts roughly never changes ( will double check this).
Ok I had a look at the F/B Pin today.
The output of the error amplifier is the combination of the 2V DC bias and the AC variation occurs here (Just Like you said). With no modulation (No AC amplification) it is fixed at 2V.
While when modulation occurs, the output of the error amplifier is the amplified AC waveform of the music signal, which keeps varying over time.
The output of the error amplifier is the combination of the 2V DC bias and the AC variation occurs here (Just Like you said). With no modulation (No AC amplification) it is fixed at 2V.
While when modulation occurs, the output of the error amplifier is the amplified AC waveform of the music signal, which keeps varying over time.
Besides attenuating Harmonics in the AC signal. What other uses do the LC filters have?
Because it makes sense what my teacher is saying, as to why I would I need a LC filter in the first place. If we can't hear above 20KHz and I'm switching at 200Khz.
I'll get a bigger boom with +20V instead of the +5V going into the Loudspeaker now.
Because it makes sense what my teacher is saying, as to why I would I need a LC filter in the first place. If we can't hear above 20KHz and I'm switching at 200Khz.
I'll get a bigger boom with +20V instead of the +5V going into the Loudspeaker now.
Last edited:
20V square wave has a lot of harmonics, which are not heared by humans, but are very good radiated and 'heared' by other electronic devices (TV, radio, etc).
So it is much better to have 5V 200Khz sine wave after LC filter, than 20V square wave...20V square wave has a lot of harmonics, which are not heared by humans, but are very good radiated and 'heared' by other electronic devices (TV, radio, etc).
Alright but those appliances picking up those frequencies. What's the result of it?
Tv giving funny effect when you play something on it?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.