TDA7052 Troubleshooting Advice - Choppy Output
bradharper
Posts: 64
I'm having troubles with my use of this chip and wanted to see if anyone might have suggestions or insights.
I've built three different boards with the identical audio circuit below and each gives me different results.
Audio Circuit Schematic:
On the first, and the one the almost works as needed, the audio starts to get choppy only at higher (~75% of maximum) volumes. By choppy I mean as if one were toggling a mute switch as fast as you could with your finger, and with slight pops on each transition. Other than this issue, the audio is clear, loud and works exactly as my breadboard prototype did throughout months of initial testing.
On the second, the choppiness starts at about ~50% volume.
On the third, the chop starts @ ~20% volume, plus the chip gets *very* hot even when no audio signal is being sent to it. The symptom that may strike some of you as a more obvious hint is that while I'm loading code onto the propeller of #3, the same choppy pattern starts as a single slightly-muted pop, then proceeds to continually increase in frequency until the upload process is complete, then the chip is muted once the propeller starts to boot. Even during the code upload, the chip is very hot, much hotter than either of the others get even at high volumes.
Each is running the same code, and using the same power source.
I've replaced all components in the circuit except for the TDA7052, but since I'm seeing this on all three boards, I'm guessing it's not the chip - unless I've semi-fried them from a mistake.
There are other subtle changes (added a few switches with pull-ups) on each of the boards, with each being a separate revision, but I'm not clear how any would effect this circuit.
I'm not very keen on testing yet, but the only obvious detection I could see was that at 0 volume, the voltage at the +/- pins for the speaker was .84V on the "hot" #3 chip, and only ~.01V for the other two.
Any of this sound familiar? Don't be gentle - I'm here to learn.
I've built three different boards with the identical audio circuit below and each gives me different results.
Audio Circuit Schematic:
||
.---||--------.
| || |
| 1uF || |
|--------||---| 7.4V
GND || | |
| 100uF |---||-
| | |
o------. |
| | |
.----------o------o .--6------1---.
| | | | 7 NC
P7o-' 10nF --- '---3 |
| --- | TDA7052 4 NC
| | .--2 |
| | | '--5------8---'
| | | | | __ /|
| ___ | || | | '------| | |
'---|___|--o----||-' '-------------|__| |
4.7k || \|
1uF 1W 8ohm
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
On the first, and the one the almost works as needed, the audio starts to get choppy only at higher (~75% of maximum) volumes. By choppy I mean as if one were toggling a mute switch as fast as you could with your finger, and with slight pops on each transition. Other than this issue, the audio is clear, loud and works exactly as my breadboard prototype did throughout months of initial testing.
On the second, the choppiness starts at about ~50% volume.
On the third, the chop starts @ ~20% volume, plus the chip gets *very* hot even when no audio signal is being sent to it. The symptom that may strike some of you as a more obvious hint is that while I'm loading code onto the propeller of #3, the same choppy pattern starts as a single slightly-muted pop, then proceeds to continually increase in frequency until the upload process is complete, then the chip is muted once the propeller starts to boot. Even during the code upload, the chip is very hot, much hotter than either of the others get even at high volumes.
Each is running the same code, and using the same power source.
I've replaced all components in the circuit except for the TDA7052, but since I'm seeing this on all three boards, I'm guessing it's not the chip - unless I've semi-fried them from a mistake.
There are other subtle changes (added a few switches with pull-ups) on each of the boards, with each being a separate revision, but I'm not clear how any would effect this circuit.
I'm not very keen on testing yet, but the only obvious detection I could see was that at 0 volume, the voltage at the +/- pins for the speaker was .84V on the "hot" #3 chip, and only ~.01V for the other two.
Any of this sound familiar? Don't be gentle - I'm here to learn.
Comments
1) Problems that start at high output levels suggest something wrong with the power source, usually that it's inadequate.
2) The suggested power supply filter is 220uF. You might try something higher than what you're using (100uF)
3) The TDA7052, like other ICs, switches pretty fast. You want to keep leads short.
4) The chip shouldn't get hot with no input. Maybe the chip is oscillating. Again, you want to keep leads short.
5) If you're using a breadboard, maybe there's something wrong with it ... a short or an open.
6) Maybe your circuits are wired incorrectly. The picture looks correct, but that's not what's broken.
Do you mean insufficient voltage?
Just tried a 220uF with no discernible change.
Alright. What should I consider short? The longest lead for the TDA on my board is just under .5"
Based on feedback from others, and what I've been able to gather about the issue, I think this is most likely the case. Also, I can touch the chip and the chop goes away for ~3-10 seconds. I'm on a small pcb at present. I think it's very likely I have at least something wired incorrectly elsewhere!
-Phil
2) You could try an even higher value. If you have a couple of 220uF on hand, put two or three in parallel and see what happens. It probably won't solve the situation, but it gives you more information.
3) 0.5" is short enough for most purposes.
4) Touching the chip does two things. It cools the chip by trying to heat your finger and it increases the capacitance around the chip. The latter suggests that maybe there's a bad connection and the added capacitance helps stabilize things. The 3-10 second delay suggests that maybe the chip heats up again and starts to misbehave.
5) A PCB avoids the problems of a breadboard.
6) You may have problems elsewhere in your device, but that wouldn't explain the problems you've presented here.
My data sheet specifies 4.5 to 18V and the schematic up there reads -7.4Volts.
Perhaps turning the Voltage supply around?
Pins 3 and 6 should be tied to gnd as well and your drawing does not reflect that.
I updated the schematic for clarity.
The symbol is just reversed. Vin is +7.4V. Thanks.
Have you addressed the floating volume control input (pin 4) issue yet? 'Just wondering if that's part of the problem.
-Phil
Just tested with a 1uF from pin 4 to ground and, despite a bit of new noise in the signal, the "motor-boating" is completely gone. I also just discovered that the reason I got away with an unconnected pin 4 throughout my testing is that the 8-DIP version I used on the breadboard doesn't have the DC volume control on pin 4.
I guess eventually I'll learn to read a data sheet thoroughly and not make assumptions on any specs that aren't immediately clear to me.
I *really* appreciate everyone's time.
-bh
What I hear now in the audio is a slight muting of the sounds at the time when their inherent sample volume would be the loudest. It seems like that DC volume control is giving in a bit at the peak of the signal and the result is a temporary decrease in gain - then, when the peak passes, the gain boosts back up to where it was originally.
I currently only have the volume pin coupled with the 1uF - could this gain fluctuation be because I'm not using the parallel resistor?
The volume pin is a high-impedance, low current source with a control scale of mute @ V < .3 and max @ V > 2. This seems like in order to facilitate the max V of 2, then as little resistance as possible would be necessary. That conceptually correct? If so, how does introducing a big resistor somehow "tie" the V up to a value necessary for max gain?
What aspect am I missing?
-Phil
(which you probably have mentioned above, but I've yet to grasp).
Thanks for sticking with me...
You could control it by connecting a Prop pin. That way, it would operate at either full volume or muted. Replace the cap with 0.1uF, and add a 4.7K pulldown resistor. Drive the pin from the Prop via a 3.3K series resistor.
-Phil
Like this?
.---- P5 | .-. | |3.3k .------------------. | | | | '-' .1uF | TDA7052A | | || | 4---o-----||-------. | | | || | '------------------' | | | ___ | '----|___|-----| 4.7k === GND-Phil
-Phil
The DC volume control, pin 4, is now driven by a prop pin and that feature works well, but after about 20 seconds of audio the jitter returns.
-Phil
Also, the gain seems to be much higher now than ever before - good bit noisier too. Sounds to me like the "floating gain" I was getting away with earlier was lower. Like before, if I keep the volume levels low, now below about 20% the audio is fine, but as soon as I increase above a certain threshold, the volume will suddenly fade slightly and start motor-boating - slowly at first, then steadily increasing in frequency.
I definitely have to get the chop resolved, but I must say, being able to mute the chip with pin 4 was worth a board revision. No pops, clicks or hiss at all now when I need silence.
-Phil
Yes.
Just to recap the difference from successful experimentation before - when the issue was resolved earlier, it was using a polarized 1uF from pin 4 to GND and no parallel resistor, currently using a ceramic 1uF and the 4.7k.
Could the amplitude of my input signal be a problem? Maybe it's been too hot all along, but the floating pin 4 allowed me to get away with it a bit. (grasping at straws with much less expertise)
-Phil
Looks like the only one rated @ 8ohms is the 8-DIP version. So, basically, I'm sizzling the chip?
Curious to me that NXP offer two packages, but @ different specs...
-Phil
-Phil