Sadly, The TDA7268 (16 pin DIP stereo) and TDA7267 (8 pin DIP mono) is now marked obsolete when I view the datasheets. Just happened over the last few months.
The TDA7267A is the mono 16 pin DIP and is available. It is really the same as the 8 pin version, but the bigger thermal path allows it to be operated at 14.5 volts with 8 ohms for more power output. The 8 pin version is limited to 12 volts at 8 ohms. Of course, you can use 4 ohm loads up to around 11 volts.
I love these chips. I tested them and scoped them under stress. They are stable, quiet, very easy to use and sound good. With floating inputs, they can be paired and used as a bridge amp with 9 volt supply 8 Ohm load for more power.
Yes, you do need to reduce the input voltage a bit for line level devices. I find some of my mp3 collection to be recorded not so high, so the extra headroom is welcome. A couple cheap resistors is no big deal.
Fortunately over the last 15 years every linear IC manufacturer around has pooped out a profusion of class D audio amplifiers designed specifically to drive little speakers crazy with hardly any power loss or support components. Just go to Digikey and type in "Class D Amplifier" and go to town. If you find your favorite old chips getting axed it's probably because incremental advances in low voltage power FETs and smaller packages with higher dissipation capacity are slowly winning the day. The only real bummer is you have to constantly scan a bunch of datasheets to make sure you're using the optimum part.
Last edited:
Well, I don't think I'll be using either of those chips. It's looking like an LM386 for me. Once I can see which diagram I ought to use, I'll need to pick a speaker. Human hearing range is 20-20k hertz, but there is no way I'll be able to find a small speaker that can do anything near that. Unfortunately, I have little (no) experience in choosing speakers. I've just been browsing across digikey, I found this:
Digi-Key - GF0401M-ND (Manufacturer - GF0401M)
Which is looking OK, but I just really have NO CLUE whatsoever if this fits in with the kind of circuit I'm trying to make. Any help here? If this is NOWHERE near what I need, is there anything that might work, hopefully 50 MM diameter, and maybe 1 cm tall, or smaller?
Digi-Key - GF0401M-ND (Manufacturer - GF0401M)
Which is looking OK, but I just really have NO CLUE whatsoever if this fits in with the kind of circuit I'm trying to make. Any help here? If this is NOWHERE near what I need, is there anything that might work, hopefully 50 MM diameter, and maybe 1 cm tall, or smaller?
Here is a circuit to consider...
Input: Stereo output of your Ipod or audio source should never be shorted together to make mono or damage may result. They are mixed together using the two 1K resistors. The 4.7k shunt resistor is needed due to the high input impedance of the IC. Without it, it may pickup electrical noise or even have HF oscillation from it's own output. The input cap is included because some portable devices may have a bit of DC on the output and must not get into the IC. I don't recall the input impedance of this IC, but .1uf film cap should work. No volume control is necessary since the headphone players have that already.
Output: If you use those mini speakers, get two and put them in series. Pay attention to proper phasing (+/- connection on speakers). Using only one may burn out at high volume given their low power rating. Okay to use .1uf in output snubber circuit. If Rat Shack sells them by the two pack, one can be used there and other on input.
Power and grounding: You may want to add a LED in series with a 4.7K resistor across the supply (after the switch, of course) so you won't forget to turn it off like I would
The capacitor across pin 7 to ground can be omitted if you never plan to power the amp with a wall wart. Keep the signal grounds away from power and output grounds. Connect pin 4 between them and you will be fine.
Input: Stereo output of your Ipod or audio source should never be shorted together to make mono or damage may result. They are mixed together using the two 1K resistors. The 4.7k shunt resistor is needed due to the high input impedance of the IC. Without it, it may pickup electrical noise or even have HF oscillation from it's own output. The input cap is included because some portable devices may have a bit of DC on the output and must not get into the IC. I don't recall the input impedance of this IC, but .1uf film cap should work. No volume control is necessary since the headphone players have that already.
Output: If you use those mini speakers, get two and put them in series. Pay attention to proper phasing (+/- connection on speakers). Using only one may burn out at high volume given their low power rating. Okay to use .1uf in output snubber circuit. If Rat Shack sells them by the two pack, one can be used there and other on input.
Power and grounding: You may want to add a LED in series with a 4.7K resistor across the supply (after the switch, of course) so you won't forget to turn it off like I would
The capacitor across pin 7 to ground can be omitted if you never plan to power the amp with a wall wart. Keep the signal grounds away from power and output grounds. Connect pin 4 between them and you will be fine.An externally hosted image should be here but it was not working when we last tested it.
Sweet! Thanks a ton! If I used a different speaker, say, this one:
Digi-Key - 668-1112-ND (Manufacturer - AS02708CO-WR-R)
Which is rated for one watt, but I noticed the LM386 is only rated for .7, so... will it work at all? Or should I just look for a .5 ISH watt speaker. One of the main things I was concerned about was the range of the speaker - human hearing goes down to 20hz, and this speaker only goes down to 250. I don't know that much, but I see a difference. Is it a big deal?
If it's not a big deal, I found these:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1563-ND
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1553-ND
Which of the speakers that I've posted would work best standalone (because I can't fit 2 in the tin)?
Digi-Key - 668-1112-ND (Manufacturer - AS02708CO-WR-R)
Which is rated for one watt, but I noticed the LM386 is only rated for .7, so... will it work at all? Or should I just look for a .5 ISH watt speaker. One of the main things I was concerned about was the range of the speaker - human hearing goes down to 20hz, and this speaker only goes down to 250. I don't know that much, but I see a difference. Is it a big deal?
If it's not a big deal, I found these:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1563-ND
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=102-1553-ND
Which of the speakers that I've posted would work best standalone (because I can't fit 2 in the tin)?
Last edited:
You can't expect response to go very low with these tiny speakers. The 1563 had a reasonably flat response through most of the listening range.
You don't have to match the speaker directly with amplifier wattage, but putting a .1 watt speaker on a amp that makes seven times that is not a good idea. Any speaker with a .5 watt rating or greater will be fine. 8-16 Ohm range is fine as well.
You don't have to match the speaker directly with amplifier wattage, but putting a .1 watt speaker on a amp that makes seven times that is not a good idea. Any speaker with a .5 watt rating or greater will be fine. 8-16 Ohm range is fine as well.
You should really do some research on your own. But since i feel like being nice today, i'll help you.
In home sound applications it is actually recommended that the amp can deliver more output to the speakers (even 2x), because short-term peaks of up to twice the rated power are easily swallowed by the speakers without even distorting the sound, while amplifier clipping (overdrive) is much more noticeable, and can break speakers if sustained for too long.
On the other hand putting a speaker with a higher power rating on a lower powered amp, simply means that the speaker will not be driven to its maximum potential. But here comes the tendency of turning the volume knob higher, overdriving the amp and damaging the speaker in the long term.
As far as human hearing goes - even a 15" subwoofer can't reproduce 20Hz too loud. You'd need multiple 18"s (or a 15" in a car, since the car is a smaller space it gets more bass). As a matter of fact i bet your home stereo doesn't go much lower than 60Hz. And your ipod headphones (at least the stock ones) can't do anything under 70Hz except farting.
For a speaker that small you'll want to roll it off from 150Hz. For a 8 ohm speaker you can achieve that with a 100uF capacitor in parallel with a 22uF capacitor, on the output of the amp before the speaker. See here: High Pass Filter - Java Experiment Positive lead of capacitor goes to LM386 output, negative lead goes to positive lead of speaker, and negative lead of speaker goes to ground. I'd still go with the TDA2822 instead of the LM386 but the choice is yours.
If you hook the speaker directly to your ipod it'll be about as loud as trying to listen to your earphones from 2 meters away. In the past you could do that - i could hook my first walkman directly to my home speakers and get loud sound. Not "party loud", but enough for casual listening. However it ate those AA batteries really quickly.
As portable players get smaller and smaller and boast longer battery life at the same time, there's something that has to be compromised, and that is output power. Earphones have a high impedance (32 ohms, but there also are HPs in the 100-300 ohm range), they have little moving mass, and thin voice coil wire, thus they are easy to drive. Thus you just have to provide them with ~1 volt RMS of swing, and they use very little current. Most 'phones are rated for 50mW to 300mW, and nowadays the output power of MP3 players is 1mW to 10mW. Yes you read it right, 10 milli watts, a hundredth of a watt. My old walkman could provide 500mW. You don't wanna know how many phones i broke with that.
Now i think you understand that a speaker with low impedance (requires more current) and heavy cone (again requires more current) will not be driven to any useful range by an ipod.
In home sound applications it is actually recommended that the amp can deliver more output to the speakers (even 2x), because short-term peaks of up to twice the rated power are easily swallowed by the speakers without even distorting the sound, while amplifier clipping (overdrive) is much more noticeable, and can break speakers if sustained for too long.
On the other hand putting a speaker with a higher power rating on a lower powered amp, simply means that the speaker will not be driven to its maximum potential. But here comes the tendency of turning the volume knob higher, overdriving the amp and damaging the speaker in the long term.
As far as human hearing goes - even a 15" subwoofer can't reproduce 20Hz too loud. You'd need multiple 18"s (or a 15" in a car, since the car is a smaller space it gets more bass). As a matter of fact i bet your home stereo doesn't go much lower than 60Hz. And your ipod headphones (at least the stock ones) can't do anything under 70Hz except farting.
For a speaker that small you'll want to roll it off from 150Hz. For a 8 ohm speaker you can achieve that with a 100uF capacitor in parallel with a 22uF capacitor, on the output of the amp before the speaker. See here: High Pass Filter - Java Experiment Positive lead of capacitor goes to LM386 output, negative lead goes to positive lead of speaker, and negative lead of speaker goes to ground. I'd still go with the TDA2822 instead of the LM386 but the choice is yours.
If you hook the speaker directly to your ipod it'll be about as loud as trying to listen to your earphones from 2 meters away. In the past you could do that - i could hook my first walkman directly to my home speakers and get loud sound. Not "party loud", but enough for casual listening. However it ate those AA batteries really quickly.
As portable players get smaller and smaller and boast longer battery life at the same time, there's something that has to be compromised, and that is output power. Earphones have a high impedance (32 ohms, but there also are HPs in the 100-300 ohm range), they have little moving mass, and thin voice coil wire, thus they are easy to drive. Thus you just have to provide them with ~1 volt RMS of swing, and they use very little current. Most 'phones are rated for 50mW to 300mW, and nowadays the output power of MP3 players is 1mW to 10mW. Yes you read it right, 10 milli watts, a hundredth of a watt. My old walkman could provide 500mW. You don't wanna know how many phones i broke with that.
Now i think you understand that a speaker with low impedance (requires more current) and heavy cone (again requires more current) will not be driven to any useful range by an ipod.
Last edited:
Alright, that clears a lot of things up. If I got an extremely high impedance speaker, do you think the ipod would be able to better power it? I'm just looking down several roads here, but I kinda figure the amp is the way I'll have to go. But if it's avoidable.... anyway.
I found this speaker:
Digi-Key - P9613-ND (Manufacturer - EAS-4D02C0)
Says it's full range, I looked in the datasheet and couldn't find a graph for it to see close it comes to flatline, but I was wondering what you guys thought?
As for the "high impedance" road, I found this:
AR027150MR-R PUI Audio Speakers & Transducers
Which looks OK, assuming the datasheet is correct (says 20-10000hz) (download here: http://www.projectsunlimited.com/audioproducts/movieclips/products/drawings/AR027150MR-R.pdf)
So it runs off minimal power, and has a high impedance. Sounds pretty good to me... But then again, it only runs up to 10khz. Any advice?
I found this speaker:
Digi-Key - P9613-ND (Manufacturer - EAS-4D02C0)
Says it's full range, I looked in the datasheet and couldn't find a graph for it to see close it comes to flatline, but I was wondering what you guys thought?
As for the "high impedance" road, I found this:
AR027150MR-R PUI Audio Speakers & Transducers
Which looks OK, assuming the datasheet is correct (says 20-10000hz) (download here: http://www.projectsunlimited.com/audioproducts/movieclips/products/drawings/AR027150MR-R.pdf)
So it runs off minimal power, and has a high impedance. Sounds pretty good to me... But then again, it only runs up to 10khz. Any advice?
Hey Snipeye, I don't know if you received my PM but I've been working on this solid since about 5:00 PM yesterday. It turns out this is a project I've been interested in for a long time.
You need to be aware of a few important details before you can really begin to make any decisions with this:
These tiny speakers are often used as headphone drivers and the efficiency is measured at 1cm. Other times they measure at 10 or 50 cm, which still gives optimistic number compared to the Standard distance of 1m for usual audio applications. Efficiency is something you are VERY interested in since it will directly affect how much sound you get for how much battery.
Notes on sound attentuation/distance: Damping of sound level with distance - decibel dB damping calculation calculator distance versus sound reduction free field - decrease in sound over distance microphone distance versus dB sound at different distances - sengpielaudio Sengpiel Berlin
Tang Band makes some interesting small woofer drivers. This one is the most output/low-end extension for size that I have found:
http://www.tb-speaker.com/detail/1230_04/w2-1625sa.htm
This is a cool part from CUI. Although it has much higher efficiency than the Tang band woofer, it has a resonant frequency of 160Hz, which tends to define your lower cutoff point:
CUI Inc | Speakers | CMS0461KLX : 46 x 46mm, 2W, 22.8mm depth, 8Ω
Both of these drivers need to be paired with a tweeter. If you don't want to get very technical/fancy with this a truly practical person would just buy something from ihome and be done with it. I am approaching this from an absolute maximum performance perspective. The advice you received in an earlier post about there being no signal output available below 70Hz or something is pretty much bull. LF response is something that even the cheapest compressed digital audio systems can still do well simply because the signal speed is so low, so in all likelihood you have electrical response down to 20. The problem is you have no business using it! Trying to drive a speaker below resonance in this application an excercise in absolute waste and if you want to optimize overall performance you need to filter off below resonance, wherever you choose that to be. Higher resonances will typically give you increased efficiency for a given size. So already you have two crossover points required. I haven't had time to dive deep into the power amplifiers but I'm going to start that right now.
The battery is another issue. Really the Sours tin is probably not the smartest way to go if you're going to spend the time making something like this. You'd be better off to investigate the possibility of happening upon a collapsible/telescoping gig where you can increase the encosure volume to run and pack it down to carry.
Take a look at all this crap and let me know what you think.
You need to be aware of a few important details before you can really begin to make any decisions with this:
These tiny speakers are often used as headphone drivers and the efficiency is measured at 1cm. Other times they measure at 10 or 50 cm, which still gives optimistic number compared to the Standard distance of 1m for usual audio applications. Efficiency is something you are VERY interested in since it will directly affect how much sound you get for how much battery.
Notes on sound attentuation/distance: Damping of sound level with distance - decibel dB damping calculation calculator distance versus sound reduction free field - decrease in sound over distance microphone distance versus dB sound at different distances - sengpielaudio Sengpiel Berlin
Tang Band makes some interesting small woofer drivers. This one is the most output/low-end extension for size that I have found:
http://www.tb-speaker.com/detail/1230_04/w2-1625sa.htm
This is a cool part from CUI. Although it has much higher efficiency than the Tang band woofer, it has a resonant frequency of 160Hz, which tends to define your lower cutoff point:
CUI Inc | Speakers | CMS0461KLX : 46 x 46mm, 2W, 22.8mm depth, 8Ω
Both of these drivers need to be paired with a tweeter. If you don't want to get very technical/fancy with this a truly practical person would just buy something from ihome and be done with it. I am approaching this from an absolute maximum performance perspective. The advice you received in an earlier post about there being no signal output available below 70Hz or something is pretty much bull. LF response is something that even the cheapest compressed digital audio systems can still do well simply because the signal speed is so low, so in all likelihood you have electrical response down to 20. The problem is you have no business using it! Trying to drive a speaker below resonance in this application an excercise in absolute waste and if you want to optimize overall performance you need to filter off below resonance, wherever you choose that to be. Higher resonances will typically give you increased efficiency for a given size. So already you have two crossover points required. I haven't had time to dive deep into the power amplifiers but I'm going to start that right now.
The battery is another issue. Really the Sours tin is probably not the smartest way to go if you're going to spend the time making something like this. You'd be better off to investigate the possibility of happening upon a collapsible/telescoping gig where you can increase the encosure volume to run and pack it down to carry.
Take a look at all this crap and let me know what you think.
I got your PM, and I really appreciate the help! I'm a COMPLETE noob, as I said before, but hopefully with help, I'll be able to pull this off. Any help you can offer is highly appreciated, so take all the time you need. I sent you a PM back with my main objectives in mind, so we can work at them.
I never said an ipod does not have electrical subbass capability. I just said that a fart is louder than the bass of its stock 'phones. And the same could be said about the stock phones on every device today, with the only exception being Archos.
As for driving a speaker below resonance, it's not that much of a sin as you think. But on a speaker this small you should stay above resonance, mainly because chances are response will be diminished before you even reach it. No reason to complicate and risk damaging the speaker. At resonant frequency the cone reaches maximum excursion, and with small drivers it's quite easy to damage them when you ask them to move too far.
Well you snapped me back to reality with your request for simple. I went off the deep end for a couple says looking at how to do extreme efficiency at low power, which is an entirely different design than simple and small. So anyway, you have a little power to burn in signal level terms since your source is designed to drive phones. A 12dB passive network can be used to high pass the amplified system at the lower cutoff. Where you put the cutoff will depend on the speakers you pick, and you may as well do that first.
You might want to consider these if you're still really interested in very small:
Parts-Express.com:Tang Band W1-1828S 1" Neodymium Full Range Driver | Tang Band W1-1828S 1" speaker Full Range Driver full range tb speakers neodymium driver line array point source midrange home theater computer speaker center channel set tube ampl
They list them as 1" but nothing about them is really one inch, even the coil is an 1-1/8. They make a smaller one with smoother and higher reaching top end, with higher Fs of course. Most of these tiny drivers with "bass" (midbass in hifi terms) need to sacrifice some top end performance to do it, which is why I recommended a tweeter. Most simple means you aren't going to bother with that, but decent tweeter frequencies require so little additional space it would seem a little off to forego it. Ideally you'd even port the enclosure to take advantage of the 6dB help at Fb, but this would be fiddly even for experienced builders to get the volume and tuning correct. The W1-1828S would require something like 5.5cu.in. ported (if I remember correctly), which is probably an entire regular Altoids can, the Sours jobbie might be a little smaller. After you get the battery and the speaker in there you may have too little space for even an optimum sealed alignment. In the amplifier section I still recommend class D for the woofer, or for the whole thing if you stay full range. The 90% efficiency is something that you really need. Analog Devices makes some 3 and 5 watt amps in tiny surface mount packages that you could conceivably hand wire with wire wrap wire. Don't worry that this is way over the average power that you'll likely have available, their efficiency holds and distortion improves far below max rated power. I'd turn them over and stick it to some metal with double-face film tape (not the thicker fabric based carpet tape, since you need thermal conductivity). With a low wattage or regulated iron and the right tip this is doable. You might not need lots of experience but you will need to be very careful and very patient. Other manufacturers make flip chip parts in that power class but assuming you could handle the size somehow I don't think the die (since that's all it is minus the solder balls) would take kindly to being repeatedly, instantly raised from 70 to 600F. Even the little AD parts would prefer you use a 15-20 watt iron or regulated set just above the liquidus temperature of your solder and a bit of dexterity. If all that's too much anyway then you might be SOL on the class D, since bigger parts are designed for much higher power and you'd have more in switching loss than output at low volume settings. A class B amp would be better. I looked for big parts that would allow you to turn the switching frequecy way down to run the woofer in a two-way but everything seems to be rated in a very limited freq. range. Trying to slow it down would possibly be troublesome and you'd probably get no support there from the manufacturer. National Semiconductor produced a big slow woofer amp chip but it went obsolete, and it was too big anyway.
Personally I'd look into using a lithium rechargeable battery for this but that will increase complexity a lot because you'll need to handle the charger as well, unless you find a way to buy that off the shelf too. In a simpler, more practical arrangement for test I'd probably run 3 or 4 AA alkaline batteries. I have a feeling you will want to just cob something together right away without great concern about what you'll settle upon until you get some of these diffferent arrangements assembled to see first hand what you can actually get out of it. The cost of the parts is so low that would probably be the best angle.
I didn't really look at the lower powered parts but I think you will have a bit of trouble trying to run below 2.2 volts and efficiency may suffer at 3 unless you messed around with higher load impedance. The best test performance from the AD parts comes at 5 volts, and the closest you can get to that with alkalines and no additional electronics is three cells.
You might want to consider these if you're still really interested in very small:
Parts-Express.com:Tang Band W1-1828S 1" Neodymium Full Range Driver | Tang Band W1-1828S 1" speaker Full Range Driver full range tb speakers neodymium driver line array point source midrange home theater computer speaker center channel set tube ampl
They list them as 1" but nothing about them is really one inch, even the coil is an 1-1/8. They make a smaller one with smoother and higher reaching top end, with higher Fs of course. Most of these tiny drivers with "bass" (midbass in hifi terms) need to sacrifice some top end performance to do it, which is why I recommended a tweeter. Most simple means you aren't going to bother with that, but decent tweeter frequencies require so little additional space it would seem a little off to forego it. Ideally you'd even port the enclosure to take advantage of the 6dB help at Fb, but this would be fiddly even for experienced builders to get the volume and tuning correct. The W1-1828S would require something like 5.5cu.in. ported (if I remember correctly), which is probably an entire regular Altoids can, the Sours jobbie might be a little smaller. After you get the battery and the speaker in there you may have too little space for even an optimum sealed alignment. In the amplifier section I still recommend class D for the woofer, or for the whole thing if you stay full range. The 90% efficiency is something that you really need. Analog Devices makes some 3 and 5 watt amps in tiny surface mount packages that you could conceivably hand wire with wire wrap wire. Don't worry that this is way over the average power that you'll likely have available, their efficiency holds and distortion improves far below max rated power. I'd turn them over and stick it to some metal with double-face film tape (not the thicker fabric based carpet tape, since you need thermal conductivity). With a low wattage or regulated iron and the right tip this is doable. You might not need lots of experience but you will need to be very careful and very patient. Other manufacturers make flip chip parts in that power class but assuming you could handle the size somehow I don't think the die (since that's all it is minus the solder balls) would take kindly to being repeatedly, instantly raised from 70 to 600F. Even the little AD parts would prefer you use a 15-20 watt iron or regulated set just above the liquidus temperature of your solder and a bit of dexterity. If all that's too much anyway then you might be SOL on the class D, since bigger parts are designed for much higher power and you'd have more in switching loss than output at low volume settings. A class B amp would be better. I looked for big parts that would allow you to turn the switching frequecy way down to run the woofer in a two-way but everything seems to be rated in a very limited freq. range. Trying to slow it down would possibly be troublesome and you'd probably get no support there from the manufacturer. National Semiconductor produced a big slow woofer amp chip but it went obsolete, and it was too big anyway.
Personally I'd look into using a lithium rechargeable battery for this but that will increase complexity a lot because you'll need to handle the charger as well, unless you find a way to buy that off the shelf too. In a simpler, more practical arrangement for test I'd probably run 3 or 4 AA alkaline batteries. I have a feeling you will want to just cob something together right away without great concern about what you'll settle upon until you get some of these diffferent arrangements assembled to see first hand what you can actually get out of it. The cost of the parts is so low that would probably be the best angle.
I didn't really look at the lower powered parts but I think you will have a bit of trouble trying to run below 2.2 volts and efficiency may suffer at 3 unless you messed around with higher load impedance. The best test performance from the AD parts comes at 5 volts, and the closest you can get to that with alkalines and no additional electronics is three cells.
Last edited:
Well, that seems extremely complicated. Mostly what I want here is help with the designing of the circuit - I know so extremely little about how to actually make it work, or designing it, that if you could design something that is reasonably small, and what parts I'd need for it (preferably from digikey, or somewhere cheaper, I just don't want to waste $ on shipping) then I can actually do the building/prototyping.
I just have absolutely 0 clue how to design the circuit in the first place - I'm sure you have the required experience to design it, and I'll be able to build it, but I'll need a parts list and a design to base it off of.
I'd rather not go lithium on this, because while it would definitely increase usability, it would complicate it to the point where I'd be hopeless. So, run down:
Small, simple
Runs off 9v
Energy efficient as possible without compromising size/simplicity
Loud as possible without compromising size/simplicity
Once again, sorry for my little experience, and thank you for your help.
I just have absolutely 0 clue how to design the circuit in the first place - I'm sure you have the required experience to design it, and I'll be able to build it, but I'll need a parts list and a design to base it off of.
I'd rather not go lithium on this, because while it would definitely increase usability, it would complicate it to the point where I'd be hopeless. So, run down:
Small, simple
Runs off 9v
Energy efficient as possible without compromising size/simplicity
Loud as possible without compromising size/simplicity
Once again, sorry for my little experience, and thank you for your help.
A lot of what I threw at you is information that may not be important depending on what you'd find after a little testing. Use the 386 and a nine volt for test. Pick a speaker. Pop it in a box. The thing is the commerically available designs are going to outperform your project because they use the class D amps and filters and lithium batteries, so unless you're willing to dig into this pretty seriously I'm not sure any goals you have will be satisfied other than to build something simple. The input filter is not complex. Here's the deal: Passive Line-Level Crossover
I made a mistake. The 90% efficiency 3 watt amps are not available except in flip chip. But just for giggles look at the SSM2315, it hits 93%!!
You can use this:
Digi-Key - SSM2301RMZ-REEL7CT-ND (Manufacturer - SSM2301RMZ-REEL7)
It's tiny and only 85% max efficient, but there IS a plastic carrier and leads. Look at the datasheet. The required circuit is no more complex than the 386. You can add output inductors for some improvement in efficiency, but barebones it will still blow away the 386. A nine volt is really the wrong battery for anything that will draw more than 10mA, in my opinion.
You can use this:
Digi-Key - SSM2301RMZ-REEL7CT-ND (Manufacturer - SSM2301RMZ-REEL7)
It's tiny and only 85% max efficient, but there IS a plastic carrier and leads. Look at the datasheet. The required circuit is no more complex than the 386. You can add output inductors for some improvement in efficiency, but barebones it will still blow away the 386. A nine volt is really the wrong battery for anything that will draw more than 10mA, in my opinion.
Last edited:
Hmm... that looks pretty nice, and a really simple circuit. Couple questions, though - it looks like it could be powered with 3AA's, does this seem like a better idea than the 9v? Also, I understand everything on the example circuit except for Vdd and shutdown - what are they, what are they for, how do I use them?
Alrighty, VDD is the supply input pin. You connect the + end of your 3 seriesed AA's here.
The shutdown pin is explained in the text on the upper right hand corner of the first page:
"The SSM2301 has a micropower shutdown mode with a maximum shutdown current of 30 nA. Shutdown is enabled by applying a logic low to the SD pin."
Disabling it means connecting that pin to VDD.
I'm sensitive to the fact that you're green, as it were, but maybe next time you go to the loo you could drag a printout of the datasheet with you and give it a good stare. Eventually you'll learn all sorts of neat things about the part. Take some time with it. Take a look at the 386 graphs and specs and see how it compares. You should realize that a real Electrical Engineer doesn't know much about these parts either until they read the datasheet.
Also notice the section about layout. You'll be trying to make all this as compact as possible, so some things that are important will happen sort of "automatically".
The shutdown pin is explained in the text on the upper right hand corner of the first page:
"The SSM2301 has a micropower shutdown mode with a maximum shutdown current of 30 nA. Shutdown is enabled by applying a logic low to the SD pin."
Disabling it means connecting that pin to VDD.
I'm sensitive to the fact that you're green, as it were, but maybe next time you go to the loo you could drag a printout of the datasheet with you and give it a good stare. Eventually you'll learn all sorts of neat things about the part. Take some time with it. Take a look at the 386 graphs and specs and see how it compares. You should realize that a real Electrical Engineer doesn't know much about these parts either until they read the datasheet.
Also notice the section about layout. You'll be trying to make all this as compact as possible, so some things that are important will happen sort of "automatically".
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.