UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 7th Jul 2017, 3:32 pm   #21
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: TDA2030 + Baxandall audio amp circuit

And don't forget that you need a 1 ohm resistor in series with the 220n at the output.

Together, they form a "Zobel network" that keeps the IC stable. But with no resistance, it'll likely oscillate at high frequencies, potentially damaging the IC.

And while it'll probably be OK for testing, don't forget to include the 1N4001 diodes (shown on the datasheet) on the final project. These protect the IC when it's being driven hard.

In short, just build what is shown on page 1 of the datasheet
mhennessy is offline  
Old 7th Jul 2017, 3:39 pm   #22
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by ms660 View Post
Table 6 also shows what can happen if the values given for other components are changed.
Looks like I'm OK with going for higher than quoted values
indigo.girl is offline  
Old 7th Jul 2017, 3:46 pm   #23
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: TDA2030 + Baxandall audio amp circuit

Just remember that in the world of RF gremlins in general, ESR and inductance can be a gremlins friend, I'm thinking of the supply decoupling capacitor here, the non electrolytic one.

Lawrence.

Last edited by ms660; 7th Jul 2017 at 3:51 pm. Reason: clarification
ms660 is offline  
Old 7th Jul 2017, 3:56 pm   #24
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by indigo.girl View Post
(The supply decoupling cap is already there (circled in yellow)
Hi Nicola,

What TS and Lawrence mean is that this component should be right across the relevant pins, on the chip, not just between Vcc and ground. This way it provides a low-impedance pathway for high-frequency currents.

You may, of course, already have put it in the right place, but just to be clear...
__________________
Al
Al (astral highway) is offline  
Old 7th Jul 2017, 5:02 pm   #25
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Thanks for clarifying -I did not add one as I presumed that this was effectively the same as the 100nF circled in yellow. Why does one also need to be close to the chip too since pin 5 and pin 3 is effectively Vs and ground?
indigo.girl is offline  
Old 7th Jul 2017, 5:05 pm   #26
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: TDA2030 + Baxandall audio amp circuit

Apologies for any ensuing confusion/uncertainty.... I'd been looking at the post #1 circuit and thinking, I learned the hard way that these beasties need HF decoupling between supply and earth very close to the device itself, even an unassuming circuit with a few inches of wire to a half-decent PSU reservoir capacitor could misbehave without very close-connected additional decoupling. I hadn't figured that you'd connected an HF decoupler anyway. As markhennessy highlights, by time all the necessary taming/safeguard components have been added in, what looked like a simple building-block of a circuit can get quite busy looking! A similar thing can happen with simple-looking 3-pin regulator ICs that end up with closely-connected input and output decouplers to stop them oscillating, reverse-catch diodes to stop them being killed by charged loads, minimum-load resistors to keep them regulating properly....
turretslug is offline  
Old 7th Jul 2017, 5:26 pm   #27
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Nothing ever ends up simple in electronics! I'm learning that fast

I have added at 470nF cap across the pins 5 and 3 from the underneath - I will redesign the board when finished to tidy things up as its looking a bit messy now.

Anyway - I have now built a passive baxandall circuit with a volume pot at the front end and connected the output to my TDA2030 amplifer circuit. I'm getting a real loud buzz over the top but I can hear the audio playing. The audio (but not the buzz) is volume sensitive - the buzz stops me listening out for treble and bass changes so can't really test the effectiveness of the baxandall tone control yet.

I have lots of long wires connecting the variable pots to the board (as these will be located away from the board). Plus I have temporarily connected the baxandal to the amplifier circuit with input and ground wires. In the final design I will have them all on the same board (but waiting for diodes and the 1ohm resistor before I rebuild the amp circuit on the baxandal board.

Is the noise just interference picked up in the wires because I'm using unshielded? I don't have enough shielded but will get some on order. Should this sort it out or is there anything else I should consider? A unity gain buffer between volume and baxandall was mentioned earlier - what improvement would that make?
Attached Thumbnails
Click image for larger version

Name:	tda2030 test5.jpg
Views:	124
Size:	78.6 KB
ID:	145794   Click image for larger version

Name:	tda2030 test6.jpg
Views:	312
Size:	32.1 KB
ID:	145795   Click image for larger version

Name:	tda2030 test7.jpg
Views:	213
Size:	50.7 KB
ID:	145796  

Last edited by indigo.girl; 7th Jul 2017 at 5:35 pm.
indigo.girl is offline  
Old 7th Jul 2017, 5:39 pm   #28
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: TDA2030 + Baxandall audio amp circuit

Nicola, re post#23, this video is about real world capacitors and will show what I was implying regarding capacitors used for decoupling, in a nutshell there's capacitance and then there's capacitors, it's well worth watching from start to finish, very much so, the guy I believe is a Tektronix bloke and knows his stuff and makes very good utube videos (W2aew):

https://www.youtube.com/watch?v=vi24SpKYYoQ

EDIT: Plus another one:

https://www.youtube.com/watch?v=9EaTdc2mr34

Lawrence.

Last edited by ms660; 7th Jul 2017 at 6:08 pm.
ms660 is offline  
Old 7th Jul 2017, 7:58 pm   #29
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Fantastic video's Lawrence - thanks. I have watched them through and have a much better understanding of some of the potential issues with caps (god its complex!).

The location dependency of decoupling caps for the HF noise is a new one for me - So in the TDA2030 spec diagram there are two caps for cleaning up the noise in the power supply. The big electrolytic one for the low frequency noise and a small (100nF) one for the high frequency noise. The latter needs to be close to the chip for it to work best. So I should actually just move the 100nF power supply cap closer to the chip rather than add another across the pins.

indigo.girl is offline  
Old 7th Jul 2017, 8:01 pm   #30
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by mhennessy View Post
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.
Hi Mark - what is the purpose of the unity gain buffer between volume and tone control?
indigo.girl is offline  
Old 7th Jul 2017, 8:37 pm   #31
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by indigo.girl View Post
The location dependency of decoupling caps for the HF noise is a new one for me - So in the TDA2030 spec diagram there are two caps for cleaning up the noise in the power supply. The big electrolytic one for the low frequency noise and a small (100nF) one for the high frequency noise. The latter needs to be close to the chip for it to work best. So I should actually just move the 100nF power supply cap closer to the chip rather than add another across the pins.


Some circuit diagrams are excellent, containing implicit practical layout advice, others are strictly "theoretical" in nature and following them absolutely in practical layout may end in tears.... It's a shame that the 100nF decoupler in the generic TDA2030 circuit isn't drawn actually between the +ve and -ve supply pins, rather than "somewhere over in the corner" with a generic earth. Theory says that they amount to the same thing, but the real world of high gain, high bandwidth circuits says that the behaviour could be very different! Being shown across the the top and bottom IC pins contains a message- "This capacitor needs to be physically close to be of use".

As well as recognisable national trends in circuit diagram layout (some curse at US, German, Russian etc. tendencies, some find that they "click" quite well with them), I suspect that some circuit draftsmen understand electronics well and incorporate as many wise layout hints as they can in the limitations of 2D line drawings, others are simply neat and ordered at laying things out in a space without specific circuit behaviour insight. It's something that comes with practice and a very broad guide is that low impedance, high current semiconductor circuitry can be quite unforgiving of connection arrangement and layout whereas high impedance, low current valve circuitry may be more tolerant. Exceptions to every rule, of course!- and this is discussing AF circuitry, RF brings more lessons....
turretslug is offline  
Old 7th Jul 2017, 8:40 pm   #32
indigo.girl
Retired Dormant Member
 
Join Date: Apr 2016
Location: Loughborough, Leicestershire, UK.
Posts: 754
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by indigo.girl View Post
Hi Mark - what is the purpose of the unity gain buffer between volume and tone control?
I've googled a bit but still have questions!

http://www.learningaboutelectronics....-buffering.php

I know that a unity gain buffer is a very high impedance component that draws virtually no current. So applied to this case it will draw virtually none of signal current that's running through the volume pot. So the same voltage at the variable pot wiper is seen at the input of the Baxandall circuit but without the current. Why do we only want the voltage and not the current from the audio signal? Where is the current coming from that flows through the Baxandall or is it just movement of electrons back and forth within the Baxandall so there isn't an need for a current source?
indigo.girl is offline  
Old 8th Jul 2017, 8:43 am   #33
GeoffK
Heptode
 
Join Date: Mar 2016
Location: Essex, UK.
Posts: 602
Default Re: TDA2030 + Baxandall audio amp circuit

The buffer is an emitter follower, (or cathode follower in a valve circuit) the input impedance is high so the voltage doesn't sag or drop as there is only a small current at the base input of the transistor, it is maintained at the input level. The output, in phase with the input signal, has a low impendence allowing large current to flow from the power source. It's a sort of power assisting circuit.
__________________
Geoff
GeoffK is offline  
Old 8th Jul 2017, 11:20 am   #34
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: TDA2030 + Baxandall audio amp circuit

Quote:
Originally Posted by indigo.girl View Post
Hi Mark - what is the purpose of the unity gain buffer between volume and tone control?
Sorry for the delay - it needs to be a long post. Sorry for the long post, but it's slightly complicated

The short answer - TL; DR - is this: the active Baxandall presents a "difficult" load to the previous circuitry.

The longer version:

Rather than think in terms of voltages and current, it's best to think about impedance.

Your source equipment - be it a CD player or an MP3 player or whatever - will have output impedance. We don't know what that will be, but it should be reasonably low - perhaps between 50 ohms and 1k? Who knows? The fact this is unknown doesn't normally cause a problem, but I'll come back to this in a moment.

Next, your amplifier has input impedance. This is something that you can decide at the design stage. It's a good idea to make it 10k or more.

Consider what happens when you join a source to an amplifier. The combination of the output and input impedances form a potential divider (see attachment). This means that there is some signal loss. Normally this signal loss is insignificant - for example, if the output impedance is 50 ohms and the input impedance is 10k, then that equates to a loss of 0.04dB, which is inaudible, no matter how golden your ears are!

Click image for larger version

Name:	equivalent.gif
Views:	95
Size:	4.7 KB
ID:	145834

However, sometimes it is significant.

Right - all of that was pre-amble. Understanding loading effects is really important when doing design work. Above I've described the situation when joining equipment - and it's something that I have to explain all the time in my day job - but exactly the same principles apply here; you get the same loading effects between stages within an amplifier.

But here's where it gets complicated:

This is something that is rarely discussed - I'm not sure why. I discovered it (in other words, it caused me a problem!) many years ago when I was just a kid. But while I'm sure designers were aware of this, I've only actually seen it written down in one place - Douglas Self's book (Small Signal Audio Design). This is a book (and author) that I cannot recommend highly enough once you've got a bit of experience under your belt.

The active Baxandall circuit has a peculiar input impedance. It varies with frequency, depending on how you've set the controls. Even with the controls in their central positions, it varies over a 3:1 range approximately.

So if the preceding stage has significant output impedance, the frequency response of your amplifier will be all over the place because the loading effect varies with frequency. By "significant", even 1k or perhaps half that could be enough to make a readily audible difference, depending on the values chosen for the tone control network. Trust me - that's what I found when I was building amps as a kid. I didn't have anything more sophisticated than an analogue multimeter at the time, but still the effect was obvious and problematic.

So let's think about your topology. Here's what I thought it was when I made my first reply to this thread:

Input -> Active Baxandall -> Volume control -> TDA2030

This might well be OK if the source has a low output impedance. However, there's a risk the tone control circuit could overload because it's ahead of the volume control (depends on the supply rails and how "loud" the source is - remember that CD players are nominally 2V RMS (5.7V pk-pk), so just 3 or 4dB of bass and/or treble boost might be enough to cause clipping if your supply rail(s) are on the low side).

So I suggested this:

Input -> Volume control -> Buffer -> Active Baxandall -> TDA2030

Which essentially has infinite headroom - it doesn't matter how "hot" your source is, because the power amp will clip before the tone controls. This is "gain structure" - one of the first things to consider when building an amp...

For more details about that, see this thread, where I tried to explain how it works when the volume control is the first thing you meet. As you'll see, people with lots of experience still manage to confuse this stuff, so please don't worry if it's not immediately clear

Anyway, while a buffer would be desirable in the first (where the source impedance is unknown, but probably low), a buffer becomes essential in the second, because whatever the source is, we know the source impedance of the volume control is going to be high enough to cause problems because of the input impedance of the active Baxandall circuit. If you pick a 10k volume pot, when set to the electrical centre, it presents a source impedance of 2.5k to the next stage.

The buffer could well be an emitter follower - that's what I used all those years ago - but it does require a few resistors to correctly bias it. Today, I'd use an op-amp as you've already got one for the tone control, and duals cost much the same as singles. But either would work, and the emitter follower will have a distortion performance that is good enough for the context.

Finally, regarding the noise mentioned earlier, what power supply are you using?

Again, sorry for the long post. But while what you're trying to achieve might seem simple, it takes many years to fully understand what's going on, and to anticipate the pitfalls. Hopefully this helps to short-circuit some of that pain

Mark
mhennessy is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:20 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.