TDA2030 + Baxandall audio amp circuit

[indigo.girl]

Hi - following on from a conversation on another thread I have attempted to make my own audio amplifier circuit from scratch.

I want tone control included to have based it on the Baxandall circuit following the description on this website. http://www.learnabout-electronics.or...plifiers42.php

I have used a TDA2030 as the amplifier chip and have added a log pot at the front end for volume control and a 1000uF capacitor at the output to feed into an old 4ohm speaker (note in the photo is says 10uF but it is now 1000uF). I am using a 12V power supply.

Its not working!! I have checked all my solder joint etc and can see nothing wrong so am suspecting my circuit design is wrong. Can anyone spot where I've gone wrong. Here is my circuit diagram:



Many thanks,
Nicola

[Biggles]

First thing is to check the output pin of the chip for half supply volts. If this is at full rail or 0 volts then you have a problem. Check the voltage between 0 volts (negative) and the pin of the chip which goes to the output capacitor connected between it and the loudspeaker. Instability i.e. supersonic oscillation can also cause funny things to happen. If the output pin is sitting at about half supply volts, then maybe you have a signal connection problem, maybe open circuit speaker or audio input connectors? Edit; Just noticed that the output capacitor is a bit on the small side. Something like 470-1000 microFarad would be better.
Alan.

[ms660]

I would first get the amplifier working without the tone network if you can as per the schematic shown at Fig17 in the 1st link below, you can use your existing 50k volume pot in place of the 22k volume pot shown on that schematic, if you can get the amp working as per that schematic and all's well then I would connect up the passive Baxandall network (4.2.2 in the 2nd link below) The Baxendall's input connected to the wiper of the volume pot and the Baxendall's output connected to C1 -ve (C1 in Fig17 of 1st link) The top of the volume pot being the signal input, Baxendall network ground connects to amplifiers ground, then see how it responds:

http://www.st.com/content/ccc/resour...CD00000128.pdf

http://www.learnabout-electronics.or...plifiers42.php

Lawrence.

[indigo.girl]

Originally Posted by Biggles

First thing is to check the output pin of the chip for half supply volts.

Hi Alan - The output pin (pin 4) to ground is at 0.43V. Is this too low?

[indigo.girl]

Originally Posted by ms660

I would first get the amplifier working without the tone network

Good idea - I will build a new one without the Baxandall.

In my circuit I used the active Baxandall design where its ground connects to the chips output rather than actual ground. Have I misinterpreted the active Baxandall. I've just connected a loudspeaker to its output. Should I instead connect an amplifer circuit as in the TDA2030 specs? That might be why the active Baxandall only has a 10 uF cap at its output because it is not designed for a loudspeaker connection. That first chip is simply to reduce the loses due to the Baxandall.

[mhennessy]

Sorry, but you can't strap a Baxandall network around a TDA2030.

The TDA2030 must be used in a circuit with with a minimum gain of x16. At values below that, it won't be stable.

To achieve that, you could use a feedback resistor of 22k, and 1k2 to ground (via a capacitor of around 22uF).

So yes, the tone control circuit needs to be implemented with a separate op-amp. Make sure it's one that is unity-gain stable.

I'd also add a unity gain buffer between the input and the tone control circuit. Dual op-amps are very handy... The NE5532 is the classic audio op-amp, but is overkill for a TDA2030-based amp. I recommend the LM833, but there are countless options out there

You can put the volume control between that and the TDA2030. But if you did add the buffer in front of the tone control section, then the volume pot could be placed at the input if you like.

[julie_m]

Remember ideal op-amp behaviour: (1) no current flows into or out of either input, (2) If the output can assume some voltage that would cause the difference between the + and - inputs to be zero, it will do that; otherwise, it will hit the positive supply rail if the + input is higher, or the negative supply rail if the - input is higher.

The only place any current flowing into the input of your circuit can go, is into the amplifier output. If you have something like

Circuit input ----- resistor R1 ----- amplifier - input ----- resistor R2 ----- amplifier output

then the amplifier - input will be at 6V, and the current through R2 will be equal to the current through R1. You can work the current out from the input voltage divided by R1 (pretend the coupling capacitor never charges or discharges, just always has a steady 6V across it, and try not to think too hard about how it can have a voltage across it if we've just said it never takes a charge), and that multiplied by R2 will tell you the difference between what voltage at the amplifier output must be and 6V.

Now you can see that if we let the current through R1 and R2 be I, then I = Vin / R1, so Vout = 6 - I * R2 = 6 - Vin * R2 / R1. So the gain is set by R2 / R1, and it relies on there being a DC path from the output to the inverting input.

That tone control circuit in the input looks like it could be creating some strange values for R1 and R2. Try and get the TDA2030 to amplify (and not oscillate!) without it, first .....

[mhennessy]

Following on from that, if it helps, there's an introduction to op-amps here: http://www.markhennessy.co.uk/articles/op-amps.htm

[Biggles]

Re post#4. Yes, that is too low for the output pin to sit at IG. The general rule of thumb (but not always, depending on the type of circuit) is that the output pin should be at half supply volts. This is to ensure that the output can swing between near 0 volts and maximum supply volts, hence giving maximum power out. Also beware that some amplifier chips are bridge types, which means that the output terminals both sit at half supply volts, with the speaker connected between them, and you will damage the chip if either of the speaker wires touches 0 volts. This is particularly common with car radio types. The chip you are using is not a bridge type.
Hope that helps.
Alan.

[indigo.girl]

Originally Posted by mhennessy

Following on from that, if it helps, there's an introduction to op-amps here: http://www.markhennessy.co.uk/articles/op-amps.htm

Thanks Mark - that op amp summary is really useful. I am still a little confused though when I try to apply the principles to the TDA2030 circuit on the spec sheet. The non-inverting input is held at 6V (due to the 100K:100K potential divider across the 12V power supply. So this means the op amp will try to make the inverting input the same at 6V. Based on the 4.7k:150k resistor pair (from output to ground) this means that 196V will be at the output pin. That's not right is it? Is the output not limited to a maximum of the Vs (12V?)

[peter_sol]

Using a single supply you need a cap between the 4.7K resister earthy end and earth (ground)

[Al (astral highway)]

Hi Nicola, if you do a little thought-experiment you can see that 196V is impossible, - fortunately, from a safety viewpoint!

It's a 14W max device from a 12V rail. Let's for simplicity assume it's handing 10W for the time being.

A Watt is a Volt x Amp. So you'd expect 1V out with 10A current =10W by one combination of volts and amps. But this doesn't seem remotely likely for various reasons- that would require very beefy circuit board tracks to provide a low-impedance pathway.

So how about 10V out with 1A current =10W? This seems much more likely! ...or a cocktail of each in between.

How did you get those current and voltage calculations?

You do seem to be confusing the DC operating conditions (as set by your resistive voltage divider) with the AC circuit

[Herald1360]

Originally Posted by peter_sol

Using a single supply you need a cap between the 4.7K resister earthy end and earth (ground)

Which causes the TDA2030 to have unity gain at dc so its output doesn't slew hard over and an ac gain in the audio region of 1+(150/4.7).

[mhennessy]

Originally Posted by indigo.girl

Thanks Mark - that op amp summary is really useful. I am still a little confused though when I try to apply the principles to the TDA2030 circuit on the spec sheet. The non-inverting input is held at 6V (due to the 100K:100K potential divider across the 12V power supply. So this means the op amp will try to make the inverting input the same at 6V. Based on the 4.7k:150k resistor pair (from output to ground) this means that 196V will be at the output pin. That's not right is it? Is the output not limited to a maximum of the Vs (12V?)

Yes, as mentioned, you need to add a capacitor in series with the 4k7 resistor. With that done, it'll work perfectly (providing you add the input and output caps, as I'm confident you already know). As Herald1360 says, it'll make the gain at DC equal to 1, so you'll get 6 volts at both inputs and the output.

In many ways, split supplies are much easier for someone who is still new to all this. But even so, the cap in series with the 4k7 is still very much recommended.

In terms of value, it's not critical. 4u7 would be adequate, anything more is a bonus.

Nice little chips, these. Shame they went obsolete - I've no idea who's making them today. I'm guessing that Cricklewood have them as NOS? There's quite an industry in making fake chips - even ones as cheap as the 555: https://zeptobars.com/en/read/Ti-555...-china-chinese

[turretslug]

Also, it's probably wise to have a supply decoupling capacitor with low HF impedance connected as close as possible across the IC's supply pins- particularly if the supply leads are long. 220-470nF plastic film should do. Many of these power op-amp type ICs don't need too much excuse to misbehave!

[indigo.girl]

Originally Posted by peter_sol

Using a single supply you need a cap between the 4.7K resister earthy end and earth (ground)

I see I have mislead with that previous diagram - I only drew in the resisters to demonstrate my calculations....

This is the actual circuit I am building from the spec sheet:



However, I forgot the realise that the cap in series with the 4.7K will block DC and so I have applied the gain of the TDA2030 to the DC component hence the 196V calculation. I can see now that for the DC there is a unity gain:

Vin+ = 6V = Vin- = Vout

But for the AC audio component there is a 32x gain due to the 4.7K:150K resistor pairing.

All the above comments helped me get there so thanks to everyone

[indigo.girl]

Originally Posted by turretslug

Also, it's probably wise to have a supply decoupling capacitor with low HF impedance connected as close as possible across the IC's supply pins- particularly if the supply leads are long. 220-470nF plastic film should do. Many of these power op-amp type ICs don't need too much excuse to misbehave!

I have updated my diagram with the components used including Turretslugs suggestion of a 220nF cap across the two chip inputs and a change in some of the cap values with what I have to hand (1000uF output cap and 10uF input cap and cap in series with 4.7K resister).

[indigo.girl]

Have I got this right Turret slug? Or did you mean a cap from 12V supply to pin1 and another from 12V supply to pin2?
It doesn't make sense to me to have the two input pins connected by a cap so I think I have misunderstood your suggestion.

[ms660]

Connect the decoupling capacitor between pin 5 (12v +ve) and Pin 3 (12v -ve which is ground)

EDIT: All as per Fig 17 schematic in the data sheets (page 8) The reason for it being there is given in Table 6 (page 11)

Table 6 also shows what can happen if the values given for other components are changed.

Lawrence.

[indigo.girl]

Originally Posted by indigo.girl

Have I got this right Turret slug? Or did you mean a cap from 12V supply to pin1 and another from 12V supply to pin2?
It doesn't make sense to me to have the two input pins connected by a cap so I think I have misunderstood your suggestion.

I've removed the cap so it doesn't figure in my circuit. I have switched on and I'm getting output from my speaker so the basic circuit is working now. I'll now incorporate at variable pot at the input for a volume control....



(The supply decoupling cap is already there (circled in yellow)