Contemplating the possibility maybe of doing a very low cost hearing aid. But it's not something I've designed before. The musts: Use mass market off the shelf parts, so it will necessarily be large Decent battery life to keep costs down, not like the cheapskate ones you can get that only run a few days on coin cells. Using AAA or AA is possible. If a chip is used it must be a very common one - and I can't think of one that would do this. The 386 has min 4-5v, Iq 4-8mA, hardly ideal, not totally ruling it out but don't love it.
So it's either a 386, which would use a massive 4 cell battery pack. AAA 0.3-0.5Ah at 8mA = 50 hours, and that's just I_q. No use at all. Which leaves only discrete designs left. And I lack experience with very low voltage amp design or with sliding class A. Suggestions welcome... and not pursuing it remains an option.
You might want to first survey what's currently available. Most of todays hearing aids are digital. They also carry a wide variety of options, such at Bluetooth, computer interconnect, phone interconnect, feedback reduction (bode shifter), smartphone BT interconnect, computer gain/freq_response/compression/etc adjustment, directional control, hearing modes (voice, music, noisy party, etc, and background noise reduction. Also, FDA approval.
Hearing Aids|NXP - NXP Semiconductors
Introduction to Hearing Aids and Important Design Considerations
Digital hearing aid design: Fact vs. fantasy
Seven Things Hearing Aid Manufacturers Should Think About
More:
However, if you want to design something that's really "very low cost", I suggest that you buy some of these on eBay, take them apart, and see if they can be improved: I suggest you avoid anything that does NOT have a rechargeable battery as these tend to kill batteries and are therefore expensive to own.
Good luck.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Out of all those links the only 2 features I'm considering are compression/ agc and a volume knob. This needs to outcompete the $5 hearing aids on purc hase price & run time.
The only real question I'm looking at at this point is what approach to tak e with the amp. I may consider a CMOS invertor chip class D later, but want to start with something minimal. The first question is whether to go with B, AB or sliding A, then I'll look at how to implement it.
Sigh. You're hard to please. Maybe an ear trumpet will meet your price objective: or something less Wagnerian:
I bought a few of the non-rechargeable battery type eBay earphones for a friend who claimed he needed just a minimal boost. It worked, but the batteries were expensive and didn't last very long.
1.5v at 4ma from an AG3 (typically 30ma-hr) battery. With a class A amplifier, the 4ma drain is continuous, so a battery would last maybe
7 hrs. I suggest you include the cost of at least one non-rechargeable battery per day into your cost estimate.
When I went looking for rechargeable hearing aids, I found that they were all digital (probably Class D) amplifiers such as the one I previously mentioned: I haven't had the opportunity to tear one apart and see if it's an NiMH or LiPo battery or what they use for a DSP chip. Probably something like this inside, but I'm guessing:
3.7V 160ma-hr.
I like to design things starting with what the customer wants, and work back to what I think would be cool. Besides a $5 cost (cost to sales, not including battery, etc), what features does the customer really need? AGC and gain control are a good start, but things like battery savers, rechargeable battery power, filtering, tone control, etc are in demand. Those are easy with digital, not so easy with analog. I would start with digital and only fall back to analog if you can't make the price objective.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
I still think retro is the way to go. A box in your pocket and headphones, but now the box is a computer (smartphone) and the headphones can be blue tooth. Just seems to me the computing power in your smart phone can be put to good use. I'm sure at some point there will be a great free app, that will only hear beautiful women that think you are smart, handsome, desirable and want to have sex with you. Or at least let you hear better in a crowded room. Mikek
It is complete madness - unless it is targetted at the third world DIYer. Modern NHS digital hearing aids kept pace with my fathers increasing deafness resulting from him spending most evenings in WWII at firing position on antiaircraft guns. There was a brief period when he had to go private to get something that made speech intelligible and he could lipread someone facing him pretty well.
You want something nearer the 100uA or less driving a crystal earpiece if you really must do this but you will be building a 1960's design!
There are plenty of opamps that would do at the 20p unit price in bulk.
But why would you bother? It will be hopelessly outclassed by any digital filtered implementation. NHS aids last around a week on a single tiny air button cell provided that you don't leave them on at night screaming at each other because of acoustic feedback.
You are trying to solve a problem that doesn't exist with entirely the wrong tools.
Lol. I did look at those a while back wondering if ear trumpets could be an even cheaper option. Conclusion was that anything small enough to be pract ical wasn't upto the job.
What the customer wants is 'cheaper.' The customer would like any feature a vailable, but only if it comes at the cheaper price. Digital is a nonstarte r here, even later in better specced versions.
I have a preliminary rough plan. You're not going to like it. The one way I can get the cost way down and the run time way up - which is a major facto r in run cost - is to go retro & forsake the modern ear-based layout. I'm l ooking at a pocket device that runs 100s of hours on a PP3. It's a sacrific e but the upsides are major. A cheap LM358 looks like it could manage that with very high R headphones.
ion/agc and a volume knob. This needs to outcompete the $5 hearing aids on purchase price & run time.
take with the amp. I may consider a CMOS invertor chip class D later, but want to start with something minimal. The first question is whether to go w ith B, AB or sliding A, then I'll look at how to implement it.
Last time I looked at those they had poor power drain in linear mode, which sadly kills their usefulness. Unless you know something I don't.
tell that to all the cheap hearing aid sellers out there. They might disagree.
it's for developing world market.
old fashioned designs aren't impossible, but crystal earpieces are too expensive and of too low reliability nowadays. Not to mention how grim the sound is.
oh sure, but no way am I spending that much.
It certainly does exist. Huge numbers of people can't afford a current hearing aid of any type. AFAIK I have the right tools, what do you think I'm missing?
They're a fiver on ebay. I can easily undercut that. The Chinese ones flatten the coin cells in a few days, an unacceptable expense. My 1st rough draft runs for months on one battery.
I'm not seeing the advantage. It still leaves an overly expensive hearing a id, and one product that depends for its sale on another from another compa ny is seldom a good thing. Also their high current consumption would slash battery life, so downsides all round.
Microcontrollers are totally impractical in the way we have things set up. You're not the first to suggest basing products on them, but it's really unworkable. For a typical Chinese factory that could work.
Even the cheapest ear horn is more than $5. Well, maybe not. This one is $4.19:
Unless you have a very high sales volume, I don't think you can build anything for $5. I would seem that $5 is your selling price, which would make the cost to sales about $1.11. The last time I ran the calculations for a US based factory, it cost $70 to handle and ship an empty box. That's a product with zero cost, but that carries all the usual overhead of QA, QC, inspections, documentation, billing, inventory, advertising, shipping, taxes, bribes, golden parachute fund, etc. Today, it might easily be twice that.
That sounds familiar. The initial RFQ is for every feature that could possibly be thrown into the device. A few months downstream, when those features have blown the budget and schedule, the customers comes back and says something like "Just design me something that does the basics".
Ok, cheap junk. You came to the right place.
You're wrong. I like it.
You're describing a pendant, which has the major advantage of allowing the use of larger components and a much bigger battery. Cramming everything in the ear is probably a requirement for those who don't want to be seen using a hearing aid, but is wasted on those who don't care, or need to use external sound belchers (POTS, smartphone, game consoles, etc) that require attachment.
Anyway, I suggest you find a commodity hearing aid earpiece and bring out the leads. Grab a signal generator and measure how many milliwatts of driver power you'll need to be able to hear something from the earpiece. The transducer impedance doesn't matter. What you're measuring is how much power is required for the EP cone or diaphragm to move air. That will be how much power your LM358 or whatever will need to provide, which I suspect will be the bulk of your power budget.
Gotta run. I have appointment with the cardiologist to get his approval for drugging me senseless during kidney stone surgery.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Lol. Those are novelty products of course, not horns. You can make a crude horn from little more than a soda bottle. If it's large enough to actually be effective, few would want to carry it around.
The BOM is expected to be far under $1.
That's one of the upsides of developing world manufacture. Some of those don't even cost a dollar.
Yeah, I don't do that. Like many things this is minimal & cheapest first, improved versions later. Cost is key.
It needs to work & be reliable, but there won't be any DSP, so not suitable for everyone. There won't be anything spent that doesn't need to be.
can't get the run time from a coin cell with a linear amp. can't even get close.
I might use piezos, not sure if that's going to be practical yet. If it is that hugely reduces power requirement. If not then moving iron, which is still much higher efficiency than moving coil.
It would be interesting and informative if you could tell us more about the environment in which you work and what the people there have access to.
It seems no microcontrollers, but CMOS logic and old op-amps are ok. I'd like to understand this. Where do the parts come from, who is the product for, presumably this is charitable work?
You've posted several times about low-cost stuff for, I suppose, the third world. Let's have a bit more of a clue. You never know, someone might be able to help.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.