TDA2030 based amp

[Geleia]

Hello guys. I'm new to the forum, and I hope to get along with you all.

After playing with electronics for some time, building my own small amp (2W) for my guitar and all, I think it's now time to move on to something bigger.
While I won't be building a 100W amp or something like this, as I live in a small apartment and I only play for fun, I thought about building a TDA2030 based amplifier with preamp and tone control.

The schematic looks like this for now: http://geleia.lzzbr.com/m8010/8010.png

Now, here:
http://geleia.lzzbr.com/m8010/8010_1.png

The one at top is my schematics, while the one at the bottom is the original. As I don't want the headphones output, is my schematic correct in this adaption?
Any other things to point at my schematics?

And I have this small transformer:
http://geleia.lzzbr.com/m8010/trans.png

As the image shows, it measures are 60mm x 25mm x 50mm. It works 110V or 220V giving me 20V+20V. What I want is to run it with 110V at it's 220 input. This should give me 10+10V. But I have no idea of it's current output. Can you guys estimate it roughly? I don't mind if I end burning it, but I would like to use it with this amp.

Thanks a lot.

EDIT: I modified a little the power amp part to use less components and replaced TL072 with 4558, although I think they are pin equivalents.
http://geleia.lzzbr.com/m8010/amp_modified.png

[PRR]

The schematic appears correct.

The PT is likely 40V at 1A. If fed half rated primary, it will be 20V at 1A. And twice the cost and weight necessary, but you already got it and aren't carying it anywhere, so that's fine. 20V AC will rectify-up to 28V DC (or +/-14V DC) which is safely under the '2030's 36V rating. At this voltage a 1A PT rating will be fine on 8 ohm load, a bit strained on FULL-power 4-ohm load, probably never strain in apartment use (you can't run FULL 20 Watts for more than a few minutes or someone will complain; loud drunk bar use with a drummer could be different).

[Rob Strand]

The transformer could be as low as 20VA even when driven at normal voltage.  Core size gives a large error in the estimate.  Can you measure the primary resistance across the 230V terminals?

Your second ckt will sound different, makes it s bit duller. because the stuff you removed affects the frequency response.

[Geleia]

At this voltage a 1A PT rating will be fine on 8 ohm load, a bit strained on FULL-power 4-ohm load, probably never strain in apartment use (you can't run FULL 20 Watts for more than a few minutes or someone will complain; loud drunk bar use with a drummer could be different).

Yeah, I intend to use a 4 ohm speaker, but I won't use it at full power for sure. Not even because I live in an apartment, I can't stand too loud sounds for much time.

The transformer could be as low as 20VA even when driven at normal voltage.  Core size gives a large error in the estimate.  Can you measure the primary resistance across the 230V terminals?

Your second ckt will sound different, makes it s bit duller. because the stuff you removed affects the frequency response.

It's resistance is around 70ohms for 110V and 180ohms for 220V.

Maybe I'll stick to the first circuit, but the extra parts will give me some trouble designing the PCB. The preamp with tone control is proving to be a pain alone.

[Rob Strand]

My guess is the transformer spec is something like 18V + 18V  and 18VA (possibly 15VA) .  The 180ohms means isn't that great (unless the secondary resistance is disproportionally lower to compensate).   Your measured 20V is the unloaded voltage which equates to a 18V loaded measurement.    (Just as an example here's a 20VA transformer spec which a core about the same size.  As you can see the primary resistance is some what lower, click on the spec PDF http://uk.rs-online.com/web/p/chassis-mounting-transformers/0504583/ )

If you run the 240V winding at 110V that's going to lower VA rating to a bit more than half that, maybe 10VA (as PRR pointed out).

I think by the time you allow for the poor transformer regulation, rectifier drops, and the TDA2030 output swing limits.  You might only end up with about 2W into 4 ohms.  I doubt the transformer rating is an issue because the regulation of the transformer will prevent enough power being delivered to the speaker.

You can check the secondary resistance as follows:
- set meter to ohms
- short meter to get residual resistance (typ 0.2 to 0.4ohms)
- measure the resistance of the outer most winding of he secondary.
- subtract the residual resistance

I suspect you will have about 4 ohms.

[fixed a few typo's, bolded)

[Geleia]

My multimeter is really sh*tty, so it's shorted resistance was 4ohms alone.

When I measured the secondary, I got 12ohms, so, it's around 8ohms.

EDIT: http://geleia.lzzbr.com/m8010/8010_pcb.png

That's how the PCB is looking. I have to check all the connections to see if the circuit is correct and then, I'll continue strengthening the trails.

[Rob Strand]

Yeah, maybe not a trust worthy measurement.  Sometimes you high shorted resistances when the leads are stuffed, or, the meter is on the conductivity (beeping) range.

I think my last post sums it up.    I suppose the bottom line is how much power you really want out of it.

[Geleia]

The transformer is enough at least to test the circuit, right? If I get 2W out of it, it's enough. Then I can consider buying a real 24V 1.5A transformer, as there's no point having the transformer and the circuit, for some reason, not working.

[Rob Strand]

transformer is enough at least to test the circuit, right?

Sure, no problem.

[Perrow]

This sounds interesting, I've got a 2030 in a drawer waiting for a project

[Geleia]

This is how I left the PCB. There was two errors that I already fixed. I was a little lazy strengthening the trails, that's why it's ugly like this.
http://geleia.lzzbr.com/m8010/8010_pcb2.png

If I find the time, I'll try etching the PCB today. I will now buy the parts I don't have.

[Perrow]

This is how I left the PCB. There was two errors that I already fixed. I was a little lazy strengthening the trails, that's why it's ugly like this.
http://geleia.lzzbr.com/m8010/8010_pcb2.png

If I find the time, I'll try etching the PCB today. I will now buy the parts I don't have.

Could you post the layout (with components that is) as well?

[Geleia]

Could you post the layout (with components that is) as well?

Sure. As I couldn't find the TDA2030 and the potentiometers in my city store, I'll have to order it online. It will take some days to arrive, probably around Friday or Saturday.

Until then, I'll strength the trails of the PCB with care and make a part list. As I haven't tested the PCB as it is now, I can't say for sure it's working alright.

Also, I noticed that in the original schematics R10 was 0R1. So I think I'll just skip the 150R and the 0R1 resistors and jumper it to ground. The 150R was probably to reduce the current through the headphones.
http://geleia.lzzbr.com/m8010/8010_2.png

EDIT: About the potentiometers, I can't find the exact values for them.
I need a B220K, two A1M and two B22k. B stands for linear and A for logarithmic, right?

I can only find 200K and 20k linear and 500K logarithmic. Would it be alright to use these values?

[Perrow]

Pots usually have 10 to 20 percent tolerance so 20 and 200 will most likely be just fine. Who are you shopping from that don't have 1M log pots?

[Geleia]

I don't have many options for electronics here at my country, so it's a little hard to find some things. Sometimes I have to buy thing from eBay and wait 1 month or more till it arrives.
I'll ask my dad, who lives in another city, to check if they have the 1M log pot, as they are coming to visit me this weekend they can bring the parts. If they don't have, I think I'll just leave a 1M resistance in the bass potentiometer place until I find one.

And I found something interesting. It's the same project as mine, with PCB already done. Mine was a little smaller, using 15x5cm, but this one is 20x5 and makes it possible to solder the pots and connectors directly in the board. As this one fits in the box I was going to use, I think I'll leave mine at that and use his. If someone still wants to use my PCB, just tell me and I'll post the layout with components and a strengthened trails PCB (still have to do this, but this is no problem).

http://geleia.lzzbr.com/m8010/8010_pdf.pdf

[petemoore]

  If the pots are offboard it's easy to mess with them, 1 meg basspot...it's easier for you to jumper a 1meg resistor to see which way is preferrable [bass 'on' or 'off'], using the 500k pot + 470k fixed/jumperable resistor would allow the bass to have a limited amount of the original range [about 1/2 of 1meg = 1/2 the sweep]...just to see what it does and where the bass range would be placed, you might even like the 'fine-tune' effect and have sufficient control.

[Rob Strand]

You need to mod the tracks on the power supply rails so they are like this:

- rectified diode cathode (+) -> track -> + terminal of + rail cap ->  take +rail from other side of +V of + rail cap
- rectified diode cathode (-) -> track -> - terminal of  - rail cap ->  take -rail from other side of -V of - rail cap
- - terminal  of + rail cap and  + terminal  of -  rail cap join,
   the CT/0V from transformer this join on one side, the other side of the join becomes the circuit 0V

You don't make connections at points between the cap and diodes, or between the transformer 0V and the caps.
You can get hum, and sometimes other issues.

See the flow here for examples:
http://www.eleccircuit.com/wp-content/uploads/2009/04/pcb-simple-power-supply-regulator-12v-15v-30v-by-zener-diode.jpg
http://www.electronica.mk/all_articles/Audio_Accessories/20W_LM1875_Amplifier/lm1875%20power%20supply%20PCB.PNG

[Geleia]

You don't make connections at points between the cap and diodes, or between the transformer 0V and the caps.
You can get hum, and sometimes other issues.

Sorry, I don't think I understood, so let me ask.

First you connect the diodes, then the caps and ONLY AFTER these you connect the other components? Like, the diodes and caps must be as near as possible to the the AC inputs and the other components must only be placed after the caps.

If the pots are offboard it's easy to mess with them, 1 meg basspot...it's easier for you to jumper a 1meg resistor to see which way is preferrable [bass 'on' or 'off'], using the 500k pot + 470k fixed/jumperable resistor would allow the bass to have a limited amount of the original range [about 1/2 of 1meg = 1/2 the sweep]...just to see what it does and where the bass range would be placed, you might even like the 'fine-tune' effect and have sufficient control.

I see, I might do this if I don't find the 1M pot.

[Rob Strand]

First you connect the diodes, then the caps and ONLY AFTER these you connect the other components?

Yes.  The other way to say that is to simply connect the circuit side of the + and - rails to the corresponding caps terminals

The most important thing of all is that the circuit ground only connects to junction of the two power supply cap only at one point.  You definitely do not want to connect anything along the track between the transformer 0V connection and the junction of the two caps.  Keep the track between the two caps short a thick.

Like, the diodes and caps must be as near as possible to the the AC inputs

Not important 

and the other components must only be placed after the caps.

Yes, it's as safe plan.

[Geleia]

I see. I'm etching that design from the Scorpion guy, but I might feed the circuit from a different rectifying circuit.

Thanks for the tips about this. I'll remember this when I design my next PCB.