Okay, I have built the amp circuit, having everything working and for some reason I haven't been able to figure out is why the output of the amplifier is having a 30V DC offset.
I am power the chip with a +/-35.5V supply, and more or less have the exact circuit built as the one in the datasheet, save for a different resistor value at the signal input (if i recall, 5K linear pot). any ideas?
http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXqwvzw.pdf
I am power the chip with a +/-35.5V supply, and more or less have the exact circuit built as the one in the datasheet, save for a different resistor value at the signal input (if i recall, 5K linear pot). any ideas?
http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXqwvzw.pdf
Hi,
Sorry I,ve mistaken your post with another one about LM3886,
I meant TDA7294.You're still impatient
Calm down as it does not help matters.You need to be rational of course experience helps.
If you say you saw oscillations in your scope? are you sure? Anyway if it is then only the chip amp is doing it so it may be damaged.Assuming the connections are correct then replacing the ic is the way to go.I don't
think the cap is faulty.Although it's possible but I doubt it,of course
you are free to substitute if you have spares to try.
regards.
Ps. Since you mention it was not present on your breadboard all the more to recheck wiring and component values and position.
As I'm in a different timezone.I can only reply tomorrow after this.
Soooo.... stay calm and learn to meditate.
Sorry I,ve mistaken your post with another one about LM3886,
I meant TDA7294.You're still impatient
Calm down as it does not help matters.You need to be rational of course experience helps.If you say you saw oscillations in your scope? are you sure? Anyway if it is then only the chip amp is doing it so it may be damaged.Assuming the connections are correct then replacing the ic is the way to go.I don't
think the cap is faulty.Although it's possible but I doubt it,of course
you are free to substitute if you have spares to try.
regards.
Ps. Since you mention it was not present on your breadboard all the more to recheck wiring and component values and position.
As I'm in a different timezone.I can only reply tomorrow after this.
Soooo.... stay calm and learn to meditate.
Last edited:
If you say you are getting some oscillation on the scope, this could be the reason for the large "DC" offset (just very high frequency AC which a meter can´t keep up with). Oscillation is usually due to inadequate bypassing of the power rails. As well as the main capacitors (large in size, around 100uF to 470uF for onboard caps and then over 4700uF for the main PSU caps), you should also have a 100nF polyester capacitor on each power leg of the chip, between the rail and 0V/GND. On a TDA7294 the power pins are 7 and 13 for +V along with 8 and 14 for -V.
These capacitors help to remove any high frequency noise that may be on the power rails and cause the output to oscillate.
Also, does the design you are using have a zobel network on the output? This helps to flatten the impedance variations from the speaker seen by the chip. Not having one can also make the output oscillate (although only usually with a load connected).
These capacitors help to remove any high frequency noise that may be on the power rails and cause the output to oscillate.
Also, does the design you are using have a zobel network on the output? This helps to flatten the impedance variations from the speaker seen by the chip. Not having one can also make the output oscillate (although only usually with a load connected).
The oscillations really is not that big of an issue at this point. I need to figure out why the amp's output is going to +rail voltage and there is non from then negative rail.
Thanks though, when I figure out the problem. I'll try it out
Oscillation MIGHT be the 'big deal' so don't discount it. 30 Volts isn't an 'offset', it's either a device or design failure. As I understand it, the standby mode puts it into a high impedance disable state. Can you throw it into standby mode and if so, does the output go to 0? If standby goes to 0 the device is possibly not toast and your circuit is making the amp go to 30 Volts either by design or construction errors.
G²
The mute and standby modes are working. And the design is straight from the datasheet except I'm using 20K resistors instead of 22K so I'll just check out the board again
It could well be the issue, it may be very high frequency AC oscillation that happens to appear as DC as whatever you are using to measure it cannot measure each individual peak and trough so it all just merges into one voltage that appears to be DC.
Also, you say you used a 20k resistor in place of a 22k? Which 22k resistor did you replace? The one in the feedback network? The input impedance setting resistor. If you replaced the one in the feedback network, it may well be that you have lowered the gain below the recommended minimum of 24dB and the amp has become unstable (this is what happens when the gain is too low).
20k:680r is a gain of ~+30dB, that 6dB overhead on the minimum of +24dB should be comfortably away from oscillation due to gain margin.
Is the output unloaded?
Is the input shorted?
Insert a 10r0 250mW film resistor into each supply rail.
Measure the voltage drop across each of these feed resistors.
Is the output unloaded?
Is the input shorted?
Insert a 10r0 250mW film resistor into each supply rail.
Measure the voltage drop across each of these feed resistors.
Right now im testing the circuit on a test bench qith a 70Vac input:
MUTE AND STANDBY MODE, NO LOAD NO SIGNAL:
1: 0v
2: 0-.2v
3: 0V
4: 0V
5: 0V
6: 1-1.4V
7: 25.4V
8: -25.6V
9: 0v
10: 0
11: 25.4v (NOT CONNECTED)
12: 25.4v (nc)
13: 25.2v
14:-23.6v (but under load, 4ohm speaker, it drops to -.8 to -1V)
15: 24.4V
MUTE AND STANDBY MODE, NO LOAD NO SIGNAL:
1: 0v
2: 0-.2v
3: 0V
4: 0V
5: 0V
6: 1-1.4V
7: 25.4V
8: -25.6V
9: 0v
10: 0
11: 25.4v (NOT CONNECTED)
12: 25.4v (nc)
13: 25.2v
14:-23.6v
(but under load, 4ohm speaker, it drops to -.8 to -1V)15: 24.4V
What have you done?
70Vac cannot be right.
throw away that speaker
test into open circuit, then try a 1k0 dummy load then a 100r dummy load.
tell us the results.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.