Low power mic to 2w speaker mono audio circuit design

[Newtopprojects]

We need to build a mono microphone to 2w speaker circuit with 2v to 5v and less than 200mA power supplies. We had tried the simple LM386 circuit with 5v below but the quality is much less than desired. Can our advanced members please advise some (other) workable approaches? Thank you,

 

[rjenkinsgb]

Sorry, but you are asking the impossible.
200mA at 5V is only 1W input power, so a maximum output to a speaker of about half a watt, with a class AB amp.

You also need adequate voltage drive to the speaker to get a particular power level; with 5V a non-bridged amp, you cannot get 1W in to a 4 Ohm speaker.

At 2V & 200mA you are down to fractions of a watt.

5V and a bridged type amp may get you somewhere near, if you have an adequate current available.

A TDA7052 would work down to 4.5V and give around 1W max output.

https://www.nxp.com/docs/en/data-sheet/TDA7052A_AT.pdf?

Or you could try two LM386s in bridge configuration, which is rather less efficient.
Concept circuit below.

Note with a built up "bridge" amplifier such as this, the inputs are swapped, so the outputs drive opposite voltages, effectively doubling the voltage available across the speaker. It should still work, at lower power, down to 5V or less.

 

[schmitt trigger]

There are several class-D amplifier ICs specifically designed for low supply voltages.
Go to the TI, OnSemi, NXP, STMicro webpages and using the parametric tools select a device that suits your needs and budget.

But if you have the 5v @ 200 mA supply limitations, even with a class-D your output power will be below 1 watt. Simple physics laws.

 

[Newtopprojects]

@rjenkinsgb

Thank you for your advice. I am new to this. We have to scale the output down to 1W due to the power restriction. Checked with Arrow and someone there recommended TS4890. Do you happen to work with it before? Does it deliver better sound quality than LM386? I plan to follow the sample circuit provided in their datasheet. Recommendations on better configurations will be much appreciated?

Best regards,

 

[audioguru]

The LM386 works fine when it is powered from a new 9V battery and you keep the output into an 8 ohms speaker below 0.45W. Using a higher supply voltage and/or a lower speaker impedance simply makes the LM386 heat more with almost the same output power.

The TS4890 produces only 0.2W into an 8 ohm speaker when its supply is 2.2V. You might need to hold the speaker close to you to hear it.
Why are you using the very complicated Demo circuit (the output capacitors are not needed to feed the speaker)?

Texas Instruments have many modern amplifier ICs.

If the mic can hear the speaker then you will have acoustical howling feedback.

 

[Newtopprojects]

There are several class-D amplifier ICs specifically designed for low supply voltages.
Go to the TI, OnSemi, NXP, STMicro webpages and using the parametric tools select a device that suits your needs and budget.

But if you have the 5v @ 200 mA supply limitations, even with a class-D your output power will be below 1 watt. Simple physics laws.

Raising the power supply to 5v @ 300 mA in my next try. Thank you for your advice.

 

[Newtopprojects]

The LM386 works fine when it is powered from a new 9V battery and you keep the output into an 8 ohms speaker below 0.45W. Using a higher supply voltage and/or a lower speaker impedance simply makes the LM386 heat more with almost the same output power.

The TS4890 produces only 0.2W into an 8 ohm speaker when its supply is 2.2V. You might need to hold the speaker close to you to hear it.
Why are you using the very complicated Demo circuit (the output capacitors are not needed to feed the speaker)?

Texas Instruments have many modern amplifier ICs.

If the mic can hear the speaker then you will have acoustical howling feedback.

Hi Audioguru,
5v is the highest voltage provided. we will increase the current to 300~400 mA. The mic and the speaker will be separated and insolated (hopefully good enough).

Can you please recommend a widely available good TI modern 5V/1~2W amplifier ICs?

Thank you for your advice.

 

[audioguru]

5V at 400mA is only 2W. Texas Instruments has 37 amplifier ICs that can do it.

The first one is a TPA6211A1DGN and it is class-AB so the power of its heating is almost as much as its output power.
It produces 1.3W at 0.1% distortion into 8 ohms when its heating is 0.55W. So the 5V current will be 370mA playing at this maximum level. Speech and music usually will have average current levels of 37mA when playing that loud. It can be a little louder than a cheap clock radio.

The remainder of the amplifier ICs are class-D with very low heating but are in tiny surface-mount packages that are almost impossible to solder by hand.

 

[Newtopprojects]

Hi Audioguru,

Thank you for bringing up an important factor. Heating is a big concern since I am putting this circuit in a plastic enclosure. Are there any of those TI class-D ICs come with SOP/DIP packaging? If they only come with SMD type packaging, I have a good friend who might be able to help me. I have looked up some ICs as shown below. Can you please advise if any of them is suitable for my application? And if I need an amplifier for the capacitor microphone I will be using?




Thank you again for your help.

 

[rjenkinsgb]

This is one that may be suitable; very cheap and available in a leaded-style surface mount package that could be hand soldered:

DiodesZetex,1.8W, 8-Pin MSOP PAM8303CBSC | RS Components

uk.rs-online.com

https://docs.rs-online.com/c10e/0900766b8154d63e.pdf

 

[audioguru]

I knew about the PAM8403 stereo amplifier IC but I did not know that the PAM8303 mono amplifier is still available.
With a 5V supply, the PAM8303 produces 1.25W into 8 ohms with low distortion. It seems to go crazy driving a 4 ohm speaker.

 

[Newtopprojects]

A friend got the parts and we are trying a capacitor mic + TS472 + MD8002A + 8 ohm 1W speaker solution with 5 v supply. There seemed to be a spike of current, from under 100 mA to over 200mA, at random time (sometimes 20 seconds into talking, sometimes 40 seconds). Is this kind of configuration workable? Do we need more capacitor to reduce noise?

 

[Nigel Goodwin]

Pretty well any device, and particularly a battery powered one, needs a large electrolytic across the supply rail - none of your circuits show that.

 

[audioguru]

The Sunrom MD8002A amplifier IC is from a company in India. Its datasheet shows a problem of excessive heating when its power supply is 5V, its speaker is 8 ohms and its output power is about 0.25W, maybe oscillation?

 

[Newtopprojects]

Thank you all for pointing out the issues of my configurations.

Can someone please kindly recommend an IC that can do the needed functions?
1. Taking input from a capacitor mic
2. Consuming less than 350 mA @ 5V
3. Output to 8 ohms speaker with about 1W
4. Does not generate too much heat (class-AB excluded?)

Will any of the following works? Do I need an additional mic IC?

 

[rjenkinsgb]

All the ones in your list seem to be surface mount types that cannot practically be hand soldered.

The PAM8303 I mentioned earlier should be suitable.

Any of them may need a preamp to give adequate output from a mic capsule.

 

[audioguru]

What is a "capacitor mic"?
Years ago there were "condenser mics" that needed 48V on them and a very high input impedance preamp.
For at least 15 years we have "electret mics" where 48v is stored in the electret material and a Jfet is used inside the mic as an impedance converter.
An electret mic is inexpensive, common today and has good performance. The Jfet needs to be powered from a few +volts at about 400uA.

I think you should use the PAM8303 amplifier that is available at many electronic parts distributors.

 

[Newtopprojects]

Hi rjenkinsgb,

Arrow's website (one of the largest distributors) says PAM8303 is not in stock. MOQ is 2.5k and the lead time is 42 weeks :-(

I've found TPA2005D1 is widely available with a workable packaging. Can you please advise a better way to connect the only mic in my application? Assuming it will need something like TS472 to preprocess the mic output before import them to the TPA2005D1?

Can you also please recommend the sizes and the locations of capacitors to reduce the interference?

Thank you,

***UPDATE: Found PAM8303D from Mouser. Will try it.

Are additional capacitors needed for the PAM8303D to reduce the interference ?

Thank you

 

[audioguru]

You will have feedback howling interference if the mic can hear the speaker.
You will have low frequency hum interference if the mic is not connected with shielded audio cable.
You will also have hum interference if the 5V has poor filtering.

The datasheet for the PAM8303 shows both VDD and PVDD pins connected together with a 10uF capacitor to ground.

The TS472 preamp circuit shows a complicated balanced input and balanced output that are not needed.

 

[Visitor]

Octopart.com is a website you should know about. Type in a part number and you get real-time stock information at all major vendors.