Paralleling the two stereo outputs of a TA2024 power amp to form one mono output

[Holograph]

I'm going to guess that a boost converter putting the input voltage to the TA2024 at 13V will make more of a difference than paralleling the inputs since it will actually increase the voltage swings. If not, at least it will make the volume more stable while the battery is discharged.

I am not "REALLY" unimpressed with the speaker by any means, I just thought I might as well get the most out of the chip. However, it seems adding a tweeter would probably be a better option if that's my goal.

The boost converter set at 13V will likely increase your available output power (possibly up to 17% if the internal circuitry in the chip doesn't limit that, but probably not quite that much in reality), but its main advantage will be that the voltage drop won't happen until the battery is depleted. Without this, your available power would decrease with voltage, which could cause a rise in distortion which may be audible.

[Mark Hennessy]

– I looked up porting; I'll have a closer look later but I don't know if I'll have enough room in the enclosure judging by the photos I saw (first glance, could be completely wrong).

Loudspeakers have a number of parameters that describe how the unit behaves around the resonant frequency, and these parameters can help a designer to choose the best enclosure for it. Some drivers work well in a sealed box of a particular frequency; others are more suited to a ported enclosure - which is a box that is sealed apart from an opening. The opening might be a slot, a simple round or square hole, or a hole with a tube behind it. Whatever it is, the air in the opening resonates at a chosen frequency (determined by the dimensions) to augment the output from the bass driver itself. Sealed boxes are pretty easy to get right - with ported enclosures, there are lots of ways to screw up the tuning!

To get a feel for this, have a play with WinISD, which will simulate the behaviour of a driver in a box. I have some examples of this on my website: http://www.markhennessy.co.uk/audax/background.htm

It's my experience that car loudspeakers aren't especially suited to either - because the manufacturer has no idea or control over how the units will be mounted (in doors, in the dash, on the rear parcel shelf), they design them so that they hopefully won't suffer mechanical damage when over-driven - in other words the suspension is very stiff, giving a high resonant frequency and a high "Q" for the size (compared to a driver that you'd use for hi-fi in a known enclosure). Essentially they are designed with open-baffle use in mind. As a consequence, the bass output is quite limited - but luckily car head-units have tone controls, and people aren't afraid to use them! Contrast that with the hi-fi crowd, where tone controls are a big no-no...

It's quite easy to measure the key parameters - called T-S after Theile and Small: http://en.wikipedia.org/wiki/Thiele/Small - to do so, you need an audio oscillator, an AC voltmeter, and the ability to make a box of known volume. An audio amp and a low value resistor is also needed. Even when the manufacturer supplies the measurements, you do need to double-check them anyway.

Regarding making an active system, I hadn't seen your loudspeaker when I wrote that, so didn't know it was a two-way coaxial job. Probably not worth doing now...

The mention of a high-pass (or subsonic) filter is about trying to conserve energy - no point shoving in loads at 50Hz if the drive unit is 20dB down at that point, all that will do is heat up the voice coil needlessly and perhaps cause distortion and other problems - especially if the box is not sealed. So a 12dB/oct filter - just a single op-amp - will be useful. I did that in a small project ages ago: http://www.markhennessy.co.uk/activespeaker/index.htm

[microbug]

I was thinking along the same lines (although I didn't realise that a lower voltage could cause distortion). I will definitely need to add some sort of warning system for a low battery though; a red light would probably suffice since the capacity of my battery is 2.2Ah – I'll notice long before any damage is done.

I'll have a look at porting tomorrow, but my gut instinct is that once I've cut a hole I can't go back. I'm guessing by "audio oscillator" you mean an ability to generate waveforms of a given frequency; I have an AD9850 module (pending eventually putting with a precision op-amp amplifier) which can do that. I'll look the examples on your website tomorrow as well (it's 9:47PM here for all you Yankees!).

[Holograph]

(although I didn't realise that a lower voltage could cause distortion)

What I meant by that is if you've got your gain set such that you're getting output voltage peaks of 11v, but then your battery voltage decreases to 10vand you don't change your gain, you're now clipping the amp to 10v (or less), raising distortion significantly. If you put the boost converter in to make sure your battery voltage only drops when the battery runs out of charge, then you won't have this issue.

For the port, you can of course cover a hole - it just might not look as good when you're done They make ports that are variable length so you can play around with it and then you just glue the 2 pieces together when you decide on a length.

[Mark Hennessy]

I'm going to guess that a boost converter putting the input voltage to the TA2024 at 13V will make more of a difference than paralleling the inputs since it will actually increase the voltage swings. If not, at least it will make the volume more stable while the battery is discharged.

The boost converter set at 13V will likely increase your available output power (possibly up to 17% if the internal circuitry in the chip doesn't limit that, but probably not quite that much in reality), but its main advantage will be that the voltage drop won't happen until the battery is depleted.

Say you get an extra 10% power. That's 0.5dB. Think you'd hear that? I certainly wouldn't!

By all means try it - after all, it's easy enough to power the amp from a variable bench power supply - and certainly regard it as a future retro-fit option, but in my opinion it's simply not worth the extra complexity and hassle. KISS.

I first got this lesson aged 16 or 17. I wanted a louder car stereo, and back then, cheap high power amplifiers simply didn't exist. So I designed a bridged amplifier, and invented a topology that gave almost rail-to-rail operation - normally, you're lucky if you can get within 2-3 volts of the rail, especially into low impedances. Anyway, this thing did work quite well, and was very nearly completely debugged, but I'd run into a problem - perhaps I would have solved it if I'd stuck at it, but at that point I decided to so the dB calculations. At this point, I realised that for the sake of gaining an extra 1.5 or 2dB, I'd be better off abandoning my crazy design, and instead I used a pair of TDA2005s in bridged mode. In fact, I now had extra room in the box as my design was largely discrete, so was able to fit 4 channels of amplification instead of just 2. Result! It was a great amp, and I still have it somewhere in the attic...

[microbug]

OK – I'm not expecting any noticeable difference but I imagine it would be better just in terms of stability and noise, as previously discussed. I won't bother testing the amp on 13V to check for a sound difference.

[dannyf]

so I am now looking to put the outputs in parallel.

The output stage in that chip is a full bridge itself. So paralleling it really isn't a great idea.

If your goal is to parallel it (to what end?), I guess you have no choice.

If your goal is to increase power output, there are other ways to do it, without paralleling it.

If your goal is to decrease distortion, there are other ways to do it as well, without paralleling it.

You just need to know what you want to do.

[Anks]

Do you think 50 TA2024 will power these or am I better fixing the crown K2 next to them?



On a more serious note these chips are pretty crappy and as previously mentioned are already a bridge arrangement. You could try paralleling them and giving them a bit more voltage to throw around or lowering the impedance of the load but not worth the effort really and most likely wont last long.

[microbug]

If the TA2024 is "crappy", what would be a better amplifier? I would like it to be efficient (class D) as this is battery powered.

[microbug]

OK, I found a tweeter:
http://www.electromarket.co.uk/speakers-audio-equipment/parts-components/tweeters/cca0387

Thoughts?

[Anks]

OK, I found a tweeter:
http://www.electromarket.co.uk/speakers-audio-equipment/parts-components/tweeters/cca0387

Thoughts?

I would sign up to Speakerplans forum and look what people do there. There are lots of people there that make battery powered systems but if your just going for it works albeit a bit crap then the TA2024 build it yourself way will probably teach you something. Depends what your going for.

[Holograph]

If the TA2024 is "crappy", what would be a better amplifier? I would like it to be efficient (class D) as this is battery powered.

If it's doing the job well enough now, then just keep it. No, it's not the best chip, but you already have it and you might as well get your money's worth. If you end up killing it at some point, oh well. If you end up upgrading later, fine.

I have my own standards in audio, but I try not to spend other people's money too much. I myself am really into audio and use mostly higher-end stuff. But If you're reasonably happy with the setup and can't afford anything significantly better anyway, then don't listen to people saying the amp chip is crappy. This is your project, not ours.

There are lots of sources you could check out such as DIYAudio, DIYMobileAudio (more about car audio but that community is generally more knowledgeable than any other car audio forum I know of), HTGuide, the DIY section on AVSForum, etc.  You can learn a lot about speaker development. There are also sites like sound.westhost.com and bcae1.com to learn some generics. All just depends on how much time and money you want to spend. If you're happy enough now then leave it be until you're no longer happy with it.

BTW, check out www.parts-express.com. I order a lot of my speaker components from Madisound, but Parts Express is also great and has a good selection of cheap drivers as well, should you want to add a tweeter or upgrade or whatever. I don't personally think it's worth adding a separate tweeter with that coaxial, though.

[Mark Hennessy]

OK, I found a tweeter:
http://www.electromarket.co.uk/speakers-audio-equipment/parts-components/tweeters/cca0387

Thoughts?

As I said earlier, I suggested a tweeter before I knew you had a loudspeaker with a tweeter built in. Actually, having had a closer look, I see that it's not a separate tweeter, but a "whizzer" cone. Even so, I would suggest you listen to it before deciding if a separate tweeter is required.

I agree entirely with the sentiment from Anks - go with what you have as it will probably teach you something. Don't expect perfection on the first outing, just be prepared to experiment until you're happy. That's how it works - it's very unusual for projects to go according to the book, despite how they might appear when presented afterwards. Even in the commercial world a lot of the time. Learn to enjoy this uncertainty, because that's where the real joy and discoveries are found.

[microbug]

I read over all the posts again. Whether or not I add a tweeter depends on how good the loudspeaker I have at the moment are at high frequencies. As I said, I have no experience in this domain but, by my ear, it certainly sounds a lot better when I enable the "treble booster" EQ on my iPod. Unfortunately I can't do that with every device because not all have an EQ setting (and as far as I can tell, it just limits the bass rather than making the treble any louder). Taking this into account I will try to find a tweeter that will fit the max 35mm diameter, but if that isn't possible / tweeters that small aren't worth having, I'll just go without (and not parallel the loudspeaker I already have). The only other change I'll make is adding a boost converter set to 13V so the amp isn't affected by the battery voltage drop.

I don't expect perfection, I just want to do the best I can with my constraints!

[Holograph]

but if that isn't possible / tweeters that small aren't worth having

The size of a driver (be it a tweeter, mid, or woofer - though not so much subwoofers) should be matched to the system. Smaller drivers weigh less and can typically use a relatively larger motor because the smaller cone has less flex. This means they can play higher in frequency. Also, with typical dynamic speakers (especially cones) there is a property known as "beaming" where essentially larger cones have less angular dispersion than smaller cones. If you had speakers in your living room with a relatively large 2" tweeter and you sat directly in front of your speakers, you would be okay. But if you had friends over and they sit way off to the side, they will hear a lot less treble than you. But if you were using a nice 3/4" dome or ring radiator tweeter (like the excellent Vifa XT19 19mm ring radiator my DIY front speakers use) your friend will get sound that is much closer to what you hear (though still not perfect).

The upside to larger drivers is that they can typically handle more power, have more excursion capability (Xmax / Xmech), and can play lower in frequency range. Edit: Oh, and as previously explained, they are often more sensitive, though not always. Depends on the motor design.

There is also a downside to larger angular dispersion: Your room is likely to cause more reflections which can cause phase issues and the "combing" effect. There are ways to mitigate this but not that easily, so for a budget system it's something you'll just have to deal with. It shouldn't be a super major issue, though.

Since your woofer/midrange is fairly small, you should be able to get away with a small tweeter like 3/4". If you had a 6.5" or especially 8" woofer (without a midrange i.e. staying with a 2-way system instead of a 3-way) then you would be more likely to need a larger tweeter to match, because a larger woofer will roll off (and start beaming) sooner than yours, so ideally the tweeter would be able to cross over at a lower frequency.

As a side-note, I prefer 3-ways for high-end home speakers, but there's always a tradeoff - the crossover. They become more complex and expensive and you really want to avoid having to using L-pads to match the volumes in a passive crossover. An active crossover isn't so much of an issue if you have enough amp channels (in your case you don't but this is more general speaker knowledge anyway).

If you do use a separate tweeter with that woofer, I'd clip the tweeter lead for the coax. It's probably just using a 1st-order filter consisting of a capacitor. Don't try to remove the cone, though, or you will probably destroy the speaker (and if you didn't, you'd need to buy a dustcap for it to cover the hole).

[Mark Hennessy]

If you do use a separate tweeter with that woofer, I'd clip the tweeter lead for the coax. It's probably just using a 1st-order filter consisting of a capacitor.

As I said a couple of posts back, it's not a coax unit. I thought that at first from looking at the OP's pictures, but having looked at the manufacturer's info, it's just a "whizzer".

[Mark Hennessy]

I read over all the posts again. Whether or not I add a tweeter depends on how good the loudspeaker I have at the moment are at high frequencies. As I said, I have no experience in this domain but, by my ear, it certainly sounds a lot better when I enable the "treble booster" EQ on my iPod.

Rather than relying on source devices to do this, I'd include tone controls. Dead simple - just a "Baxendall" job would probably do. And with a loudspeaker project like this, it would be customary to build in some equalisation into the amp itself, often separate from the tone controls. Have a look at that powered speaker project I linked to earlier for an example. There was a time when all half-decent transistor radios had EQ built in to give the little 'speaker some help in the cabinet, and that's partly why old radios sound so much nice than many of the modern ones you can buy today.

As I mentioned earlier, tone controls are not considered socially acceptable in the hi-fi world, where complaints of poor tonal balance are addressed by equipment "upgrades", or magic wire or other voodoo. Strange that tone controls aren't allowed in an amplifier when most crossovers have tone controls built in to account for "baffle step" and drive unit anomalies, but these are invisible to audiophiles who think their equipment is designed with pixie dust.

However, in the real world, there is no such thing as the perfect recording, the perfect loudspeaker, the perfect listening room, and tone controls are extremely useful. For something that is small and portable, it'll need all the help it can get.

[GamerAndds]

I know tha I'm "Preaching to the choir here" so to say but all these chips should be externally balanced, that way throwing on a simple pair of TO220's or something similar can be done easily. Personally love the STK4050 for this reason! But I also have a thing about paralleling amps... Each amp would have to be "100%" the same because of there not your going to be dragging one another down. And for that point it dosent really make any sence. Your not going to be amazed by the improved response IF you even hear it.

[dannyf]

that way throwing on a simple pair of TO220's or something similar can be done easily.

There is nothing preventing you from connecting a gate driver + two mosfets on each output pin and hook it up to a high voltage supply for more power.

Topology wise, this particular chip has no feedback (post or pre filter) so the performance may not be great.

[diyaudio]

If the TA2024 is "crappy", what would be a better amplifier? I would like it to be efficient (class D) as this is battery powered.

@microbug

I can recommend, however, I'm lazy today all your answers you want are here, fragmented.. its a rather long thread 100+ pages I'm sure you will have fun.   

http://www.diyaudio.com/forums/class-d/104402-boominator-another-stab-ultimate-party-machine.html

[microbug]

See you in a few days then!

EDIT: That thread is huge – I'm serious about a couple of days. If anyone is less lazy, feel free to help!

EDIT 2: I have found a more recent (Jan 2006) version of the TA2024 datasheet. I don't think it will be much more help but I have attached it anyway.