Class D audio chip MIX2052

[Liamjames192]

I love building small amps. I'm currently having a problem with this one. It is clipping at higher volume. I've tried different input resistors. But no change. The speaker I'm using is a 4ohm 3 watt speaker. The chip is a 2 ohm chip. I'm using 33k ohm input resistors with 0.1 uf caps.
Does anyone have any filter suggestions? There is only 1 schematic online for this chip. It's an 8pin chip. MIX2052 if chip.

 

[KlausST]

Hi,

Clipping at higher volume is quite expectable.

Clipping means that the supply rails limit the output signal.
Thus - besides the chip - the power supply is reponsible for clipping.
What exact power supply do you use? Electrical specifications please.
Since it seems to be a bridged output stage, your power supply should be able to provide V_supp / 2 Ohms of current ( I = V / R).

For a small speaker it is impossible to produce deep frequency bass. Thus it makes no sense to amplify those frequencies, because the membrane will just move without useful audible sound. This unnecessary movement will cause distortion in the speaker, but it additionally limits the amplifier's range for useful, audible frequencies.
--> focus on the audible frequency range.
Thus - depending on your speaker - an lower cutoff frequency of 200Hz..500Hz should be used.
This means an input capacitor of 22nF down to 10nF.

Klaus

 

[Liamjames192]

I'm using a power supply of 5.1v 1.5A. When I first built it. There was no clipping. I went to package it and afterwards there was clipping. I've tried replacing every cap and resistor. I've used several different sized speakers. All above 2ohms that the chip is rated at. The speakers vary in watts from 1 to 10 watts. I am still new to electronics and am self taught from internet and a few books. I really enjoy it and am appreciative with all your help. It may take me a few days to respond as I can't always get online. Thanks.

 

[KlausST]

Hi,

I'm using a power supply of 5.1v 1.5A

according Ohm´s law: 1.5A is not enough for the expectable speaker current.
Maybe you have sufficient big bulk capacitors to compensate for this.

When I first built it. There was no clipping. I went to package it and afterwards there was clipping.

We don´t know your first circuit
nor we do know your second circuit.
--> impossible to help in this regard

Show your (not the one from the datasheet) schematic, show photos, show scope pictures..(of both circuits)..otherwise we won´t be able to help.

Klaus

 

[Audioguru]

I answered this before maybe on another website? Over there, didn't you power it with a 3.7V battery?
The datasheet for the amplifier IC is written only in Chinese and says, "6 Whats, 2 ohms, 5.5V supply and 10% distortion".
It has a graph that shows an output into a 4 ohm speaker of only 0.8W when the supply is 3.6V when clipping begins. It produces about 0.5W into 8 ohms.
It shows an output of 2.2W into a 4 ohm speaker when the supply is 5.0V when clipping begins. It produces about 1.4W into 8 ohms.

 

[Liamjames192]

I appreciate all the help you've given. I am relaying this information from my brother who is in state prison. He has email but not internet. I'm able to see him when he video calls us and I can copy his schematic he drew. I am not sure how to post them here. Both the amp chip and speaker came from a Bluetooth speaker box.

5volt into pins 1 and 6 with a 10uf and 1.0uf caps. L and R inputs go into 0.1 caps each, then is combined going into a 33K ohm resistor then going to pin #3. Input Ground goes into a 0.1uf cap then a 33K ohm cap then to pin # 4. Pin #2 (Bypass) goes into a 1.0uf cap then to chip ground (pin #7). Pin #5 goes directly to (-) side of speaker. Pin #8 goes directly to (+) side of speaker. When it was rebuilt all the same caps and resistors were used. Unfortunately he doesn't have access to any equipment other then an a volt meter. Should I tell him there's nothing we can do without other test equipment? Is there any other filters he can try? Thanks again.

 

[BradtheRad]

schematic he drew. I am not sure how to post them here

See the icons in the Quick Reply window? Mouse over them and you should see captions. One of them says 'Insert Image. That uploads images from your computer.

 

[Audioguru]

The output into a 4 ohm speaker when the supply is 5.0V is 2.2W just below clipping so the 1.5A supply producing 7.5W is more than is needed and will be fine because the amplifier is an efficient class-D type.
On the other website the schematic is correct and is the same as is on the datasheet. I used Google Translate to see that the Shutdown pin is set for the amplifier to play.

 

[Liamjames192]

The shut down pin is supposed to go to the power pin, correct? Is there any filters I can try to tell him to use? Could a bad cap do this? I wish he had a way to test the caps, but he says all he has access to is a multimeter a digital one without many settings. Thanks again.

 

[BradtheRad]

Class D raw output is characterized by rapid transitions between one supply rail and the other.

Is it possible your source signal amplitude is excessive? Or gain is set too high? If so then the output waveform spends a long duty cycle at a supply rail.

Is there any filters I can try to tell him to use?

From the data sheet for MIX2052 (or similar class D IC), it runs at a high frequency such that the voice coil's impedance serves as a filter. In addition you might try attaching an LC second order low-pass filter. Select an inductor value so it reduces amplitude while smoothing the PWM waveforms.

 

[Audioguru]

The MIX2052 power amplifier has a low maximum output power.
Simply turn down the volume control like on any amplifier so that it is not clipping anymore.