battery amplifier, searching an IC

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
B

barbizio

Member
Joined 2010
Hello,

I've played with electronics for a while now, building tube amplifiers and stompboxes.

Now I'm in need to build a couple of battery guitar amps for me and a friend to go out in the street!

I've looked around for quite a while and I've seen some really different (sometimes opposite) comments on which chip to use. That's why I write you here.

All the project is on a tight budget!

The speaker will probably be a 10" 35w 8ohm 95dbm

I'm searching for an IC with the following characteristics:
- not SMD (I won't be able to solder it)
- quite efficent (first idea was obviously class D/T but seems that all the IC's are SMD :-( )
- not too current ungry (to save battery life)
- can deliver around 10W (I think it will be plenty of power for busking, we need to be heard, we don't want to impose)

Until now I'm looking into the TDA2005 and TDA7396.
Anyone know if they will fit my needs or have a better idea?

aaand time for noob question:
How can I have a approximate idea of how much current the chip will require for a "normal" operation? of course we got quiescent current, but the rest?

Thank you all and have a good day
 
B

barbizio

Member
Joined 2010
Thank you for your answers.

So I looked over the two chips and I'm trying to lay down the building blocks. The first thing that I'm missing is the input sensitivity (altough on the TDA2030a there's a little reference) that I'll need as a parameter to design the preamp.
After some thinking I came up with the following:

I'll follow the signal backwards and calculate the values.
Knowing the output power and the load I can calculate the output voltage.
Knowing the output voltage and gain I can calculate the input voltage.

so If I want 8W over 8ohm there will be 8V (P=V^2 /R).
having a gain of 20db (10fold voltage) the input is 800mV.

(that in an ideal world)

Does that make sense or not at all?

p.s. should I open a new thread for further questions?

Have a nice day!

Patrizio
 
B

barbizio

Member
Joined 2010
Gingertube, thank you for the link, I will keep it as a failsafe!

Erikb, thank you for your reply and link. I know that I'll need a preamp and I was planning on adding some eq (i found somewhere a really useful program for that).

I'm not trying to reinvent the wheel, but I'll like to know how to estimate voltages and gain needed.
Looking at tje Tpa3125 datasheet a little more carefully I realised my previous post is wrong!
But how much input voltage wants to develop 8w and before overloading?
 
2

29285

Disabled Account
Joined 2010
In reality the audible difference between 5 and 10W is pretty small, so it does not make much sense being precise with amp wattage.
If you power a full brigde amp (like TPA3125 or myriads of others) by a 12V SLA you can expect something between 5~10W of undistorted power at 8ohm speaker.
 
T

tedkok

Member
Joined 2015
You can try CS6A4983, that can deliver 8WRMS with two 9V batteries and a 4ohm speaker.
Linear current assisted Class A operation with very low current consumption that it can run at 8WRMS with the two 9V batteries for more than 6 hours depends on if you have a normal or alkaline battery.
 
B

barbizio

Member
Joined 2010
Thank you Voltwide. But that doesn't answer my question.... Or Yes?

When I say input voltage is the guitar signal coming from the preamp (that I have yet to "design").

Maybe I'm asking something really obvious and I came up with an idea (I more or less understand the process of class D amplification but not really what every component in the chip do) :

Is it possible that the chip (like a Tpa3125)"doesn't care" too much about the signal voltage as long as it stay between the maximum allowable (for the Tpa3125 is - 0,3 - 7v)? And that you just set the supply voltage and gain?

Sorry I don't want to sound pedantic.
Thank you!
Have a nice evening
 
2

29285

Disabled Account
Joined 2010
barbizio said:
Thank you Voltwide. But that doesn't answer my question.... Or Yes?

When I say input voltage is the guitar signal coming from the preamp (that I have yet to "design").

Maybe I'm asking something really obvious and I came up with an idea (I more or less understand the process of class D amplification but not really what every component in the chip do) :

Is it possible that the chip (like a Tpa3125)"doesn't care" too much about the signal voltage as long as it stay between the maximum allowable (for the Tpa3125 is - 0,3 - 7v)? And that you just set the supply voltage and gain?

Sorry I don't want to sound pedantic.
Thank you!
Have a nice evening
Click to expand...

You do not have to care for the details of a specific class-d-amp
the wattage is a matter of battery voltage, impedance and half/full bridge topoly.

Assuming 12V supply and a gain of 20dB (10x), full bridge topology:
Output swing is <12Vpp * 2
Input level for clipping thus is 24Vpp/10 = 2,4Vpp i.e. approx 1Vrms.

So if you pre-amp delivers more output voltage, the power amp is overdriven.
Depending on your speaker this may sound more or less awful.
But no prob as long as kept insde the allowed range.

To make things simple I state any pre-amp supplied with 9~12V will do the job.
 
Last edited:
2

29285

Disabled Account
Joined 2010
tedkok said:
two 9V batteries means +/-9V supply.

The amplifier can achieve a maximum output of +/-8V, i.e. 16Vpp

the RMS output at 4ohm load is 16*16/(4*8) = 8WRMS.

You may want to check your own math
Click to expand...

Your math is based on the false assumption that 2 9V-blocks can deliver 2x9V into 8W load for a significant time.

1) 9V batteries are some of the most inefficient sources to power amps of significant wattage.
2) The proposed linear amp is not first choice for battery operations, simply because of low efficiency at low to medium output levels. For longest possible battery life class-d amp are superior.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom