This is an interesting twist. This air aroma unit has been completely redesigned to be more compact. I was expecting the circuitry to also have been minimised, but in reality it's similar to the original version with some interesting twists.
The power supply for the chip was a bit of a mind bender. I've not see this approach taken before.
The unit will operate down to about 3V so is suited for use with rechargeable AA cells.
It's been pointed out that the parasitic diode in the MOSFET will bridge the current limited microcontroller supply to the battery voltage with a typical diode voltage drop. That means the microcontroller will only see the full 5V when the batteries are new. That voltage nudge will still potentially extend the run time on a set of batteries though.
In hindsight, the crystal will also be providing accuracy to the 8 hours on - 16 hours off timer that this unit uses.
The image on the non-track side of the PCB was reversed to aid reverse engineering, and yes their code name for the design seems to be maltesers. Candy codenames are a good way to go.
Current draw while actively atomising is:-
400mA at 3V
340mA at 3.5V
280mA at 4V
250mA at 4.5V
Standby current at 4.5V is less than 70uA
Standby current at 3.5V is less than 30uA
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:-
http://www.bigclive.com/coffee.htm
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators

A while ago, i took a look at an airwick uh air aromatizer that uses bottles of aroma chemical with a wiccan top, and it's got an ultrasonic atomizer and every small phone will just power up and or just put a whiff of the stuff into the air. They've already released the mark ii version of it and i thought i'd get one of them to actually compare the circuitry inside. I didn't go into it too deeply when i took this one apart, because i was traveling at the time. So i'm going to go into this one, a lot deeper and reverse engineer, but i'm expecting the circuitry to be a lot simpler, but i shall show you the mechanism inside i'll actually turn it on and show you running.

So i've got a bit of card here. So you can actually see the haze as it comes out. You can see it there if i was to use a little light and shine it into it. You'd probably see it better.

Yeah, that's going to make it more visible, but it's basically ultrasonically atomized aroma, and i want to point out that when you initially power up, i've not tried this for a long period of time, but it will initially put out quite a lot, possibly just to prime The error with aroma when you first get it and it has a four positions - the switch it's got off: low, medium and high. Let's open it up, so the bottle and the bottles have changed. The bottle comes with a tamper proof cap to stop kids, opening and drinking, which is quite good. So it's got that clicky clicky thing that you have to push it down really hard to actually unscrew it.

It also takes batteries. I've got rechargeable batteries in this. In a different way to the previous ones, so the batteries in this one go inside in this compartment here, and that differs to the previous version which the top pops off, and then you get the battery compartment here and the aroma compartment here. This definitely has very much streamlined design.

I can remove this bit of card. I do not need this piece of card now, so the sequence. When you take the base off, you have to put batteries in first uh, because otherwise you wouldn't get the aroma in so they go in fairly easily. It's a quite a nice design.

Then the aroma bottle goes in well, let's use this one and stick it in it just pushes in and then you turn it on it does that whole sequence. Initially, when you turn it on the led in the end will glow blue initially and then it will do that sort of hazing out the aroma type thing which is doing right now, but less visible, because i've not got that black backdrop: okay enough waffle, let's uh Open it up, a provisional exploration shows that there are two screw holes down here, quite well recessed. I thought this was another screw hole and spent ages trying to get a screwdriver long enough to go down into this one. It turns out that is possibly just a electrical test point for the batteries, so they can actually put this in a jig and it will power up for testing the factory.

So there's one screw down here and there's another screw over here. It's just going to release it, i'm kind of expecting the circuit to be super simple now. Well, the whole thing comes out. That's quite neat and there's a circuit board which is presumably held in by the actual case itself.
It does look like it does hold it in the case um. It's a very bare circuit board. Oh, i lie it's not bare at all. It's actually quite a lot on it right.

Tell you what one moment please, while i reverse engineer this i'll, be back in a jiffy, the reverse engineering is complete. That took quite a time because this is a mind bender. I didn't reverse engineer the previous one, because i was traveling away from home and didn't have the facilities to blow up images and and take a closer look, so i kind of just assessed at that time. I don't think the circuitry has changed that much.

There's a few changes, but it's all more or less the same. So in this side of the circuit board, which is reversed uh for ease of reverse engineering, we have the switch which choose between off mid uh, medium and max. We've got a little inductor here, which turns out to be for generation, 12, volt, rail, and then we've got a larger inductor for the piezoelectric crystal note. The missing connector here is one of those annoying connectors that when you try pulling the plug off, it pulls the whole housing off.

The only other things on this side are another connector for the led on the atomizer section and a little crystal here, this uh 2021. This is a very recent march: 2021., oh ralph's, birthday, 24th of march 2021, no less interesting, uh. Here's the back of the circuit board! Well, the track side. Well, the component side for most the stuff.

Really, it's got a custom program, mic controller, i'm guessing. It has a little voltage boost circuit based on a b628 style chip. It's got a mosfet down here. That does some really weird stuff and it's got a mosfet over here for the atomizer section and initially, when i looked at this circuit board, i thought maybe i had the option for a remote control, because i thought that was a little infrared, remote control receiver.

It's just the facility to put another mosfet, a through-hole one on top of the circuit board, but they've allowed the functionality of either putting it in this side or the other side um, and that's more or less it anything else worth mentioning this side. Notable this little zener diode here for capping the voltage across this for a very odd reason: okay, that's more or less it! Let's go straight to the schematic, here's the schematic! Let me zoom down a little bit more, so i've divided into sections, because it's quite a complicated circuit. The person who designed this did a very good job of actually fitting it all onto a single sided circuit board, with just one link here. They did very well.

So we have the stack of three double-a cells. It's worth mentioning that this unit will operate down to three volts, so it will actually, you can run up with nicometro hydride cells, decoupling capacitor across that, and that gives you the supply rail of roughly 3 to 4.5 volts. The microcontroller supply rail can be fed from two sources. It can either be coupled directly to that rail by this mosfet.
It's worth mentioning that the mosfet has a pull-down resistor, it's a p-channel mosfet, which means that by default, even before the microcontroller is activated uh. This mosfet will turn on and it will basically connect the battery connection to the microcontroller supply line and it will get whatever the battery voltage is at that time. However, when the microcontroller decides to actually get active, um and turn the uh piezo electric disc, on to actually atomize the aroma, it puts a 12-volt enable out, but that 12-volt enable also goes to this mosfet, which actually because it this is going positive. It effectively turns a mosfet off, but it turns a 12 volt supply on that.

Then uh powers this via a current limiting resistor, and there is a zener diode to cap that at five volts so effectively powers the microcontroller at the full five volts. This will do two things: it will make it more stable if the battery voltage is dipping with the extra load of the boost for the piezoelectric device, because it makes it independent the battery voltage that can theoretically down to about two volts and this, the circuitry that Boosts it up will still get up to 12 volts, but it also provides a nice stable, five volts, which might be quite important because it's got a sense line coming into that to sense the activity of the piezoelectric other things. Here we've got the crystal with its two load: resistors we've got the little switch that goes between off low medium high. It's worth mentioning that off isn't actually off, it's always active, but when you turn it into the off position, it signals to the microcontroller it's off, but it just goes into a low current standby mode and we've got the blue led in the tip of the atomizer.

Here that makes it glow, which is powered directly from that rail and via a 1k resistor. So initially when this starts up the led, if the batteries were running low, this led would be a little bit dimmer. But when it actually does it's a ramp up thing and does the atomization it's then powered via the 12-volt supply for the piezoelectric and that resistor, so it's going to be brighter. It's an odd arrangement.

Okay, that's that side covered. Let's go to the next page, so the three outputs, the mic controller 12, will enable the piezoelectric drive and the sense back from piezo when it wants to activate the piezo. It sends the enable signal to this little b628 bush chip, which then uses this inductor pulsed to the zero volt rail to put an inductive charge on the inductor. Then, when it turns off, it goes through this shortcut diode to charge up these capacitors and there's two resistors across here that are used to set the voltage, because this is a programmable uh voltage regulator boost regulator.
In this case the choice of resistors they've chosen. Uh will give it roughly 12 volts out a couple of capacitors, a small one, a large one, just the small one for dealing with high frequencies and the larger one just as a buffer. But that feeds the piezoelectric circuitry then, which is a larger inductor and an n n-channel mosfet, with a one ohm resistor in its source. Is that right? I think it's the source um and that is used for sensing, the current flowing through the inductor.

So when it turns the piezo crystal on it, uh turns this mosfet on and when it turns inductron uh. It turns the mosfet on current flows through as the current the magnetic field is built up. It uh the current increases slightly until the voltage across this one. Ohm resistor, which is sensed by the microcontroller via this 100 ohm resistor, and a little filtering cap.

It detects when a certain threshold has been reached and we'll turn it off and then, as the magnetic field collapses, it will potentially put out quite a high voltage. Now. This is an odd arrangement here. They have two capacitors in series couldn't measure them because they were in the circuit effectively and the piezoelectric crystal across that um.

I'm not not quite sure how they're what they're actually doing here, but the the circuitry is such that it. It's actually quite pert, it's driving it quite powerfully. The crystal itself is a disc with the wick pushed against it and it's a stainless steel disc. Well, this thing has just decided to atomize aroma right now, but it's got a a little disk.

Will you you'll be able to see the little disc with the? If i zoom down this, you can see the little dimple in that disc, okay. So it's a disc of the little dimple raised dimple with tiny little holes perforated in it, and then a circular ring of piezoelectric material with one connection being the disc itself and then there's another electrode on top that couples to other side, the piezoelectric material. And when that uh, when you put the aroma cartridge in this, is actually pushed up slightly because there's a spring and a spacer pushing down to protect the surface of the crystal and also provide the correct pressure. So, as you put it in, it pushes up against the spring and it puts a suitable pressure against the top of the wick.

The wick delivers the liquid this disc. Then, when it's being agitated at ultrasonic frequencies, it smashes up and down imperceivably and atomizes the liquid. Through these little holes here stretching into tiny little droplets, which creates that plume of vapor it's a very odd circuitry um, just you can evaluate things again because it might make more sense now. So here's the supply for the microcontroller being fed from that 12-volt rail via the resistor and capped to 5 volts, but other other times when the 12 volt is off this mosfets.
On connecting the rail to the uh battery directly, the microcontroller has its own little crystal, which is most likely. The crystal is more for the providing a stable frequency to the piezo. I think it's, i don't think, there's any resonant frequency thing being done. I think it's just a fixed frequency match to that crystal.

I don't know if it's able to actually sense anything other than the current through the inductor um and when the 12 volt enable is enabled well, that's uh, bring in this voltage regulator to create 12 volts, which is normally just floating at roughly the battery level. Just because it's not uh, it's not being boosted by the inductor, the current will still find its way through the inductor and diode, but you'll get a slightly lower than battery level voltage, but when it's needed it just boosts up to that 12 volts. It's very unusual. It's quite a nice design, it's quite a complex design and very similar to the original design, which suggests that it was a successful product and certainly they've reduced the size.

It's got a nice. I like the flat top here. It's quite a pleasant finish, also notable compared to the other one that when you do turn it on it does prime that uh it does prime the output with a significant amount of the vapor initially, because i guess it's just basically building up because you've just turned It on it's just building the room up in there, but it's quite a neat design, a very clever design, actually, the software in that chip, if it's, if it is a microcontroller, i think it probably is, though, will be part of the magic of this. That there'll be a lot of work involved in that developing that software to match the hardware, but that's very unusual.

The power rail being fed either from the batteries for the via the mosfet or the 12 volt rail. Once it's boosted for the giving extra oomph to the crystal the piezoelectric crystal uh, that's quite clever! It's something i've not seen before, so that is it the air earwick aroma unit, oh worth mentioning on the back. It says: uh, rb hygiene, home france and uh rb. Uk hygiene, home commercial, limited united kingdom uh very odd.

I don't know what that means. I don't know uh if that's just another subset of the earwick brand, but it's neat. It's definitely an improvement. I mean the smaller size means they can fit more in a crate.

Ultimately, when they're packaging it, it also looks quite neat and i think they have refined the electronics a little bit too so pretty neat quite a nice little unit.

15 thoughts on “Teardown of new airwick essential mist unit”
  1. Avataaar/Circle Created with python_avatars Ian C says:

    I think I'm in the minority but these artificial scents smell like a chemical factory to me ! Some of them even make me choke like inhaling smoke from burning plastic, often they leave me with a shortness of breath that can last for hours or even a day or two !
    I cannot understand the attraction myself.

  2. Avataaar/Circle Created with python_avatars TheEightBitLink says:

    A long time ago they had one called the Wisp, that took proprietary bottles. There was a Make magazine article on hacking them to drive them yourself.

  3. Avataaar/Circle Created with python_avatars David Queen says:

    Add spark igniter and you got a neat conversation starter when you have company.

  4. Avataaar/Circle Created with python_avatars jt1993 says:

    take a drag of it lol

  5. Avataaar/Circle Created with python_avatars nobody important says:

    they changed the bottles to look more like those essential oil bottles that new age people like

  6. Avataaar/Circle Created with python_avatars Torchizard says:

    That 55 15 date code had me very confused for a minute…

  7. Avataaar/Circle Created with python_avatars mikenco says:

    Am going to order one of these for my office.. nice! Do you think there's a way to tune them for a smaller space?

  8. Avataaar/Circle Created with python_avatars Chris Bruce says:

    could "vape juice" based fluid be used to refil those bottles Clive ?? (maybe a possible video trying different viscocity of fluids to home made menthol air freshner)

  9. Avataaar/Circle Created with python_avatars g0hjq says:

    When the 12v supply is on, won't the internal diode of the power supply MOSFET clamp the microprocessor power to VBatt + 0.7v? Very Odd.

  10. Avataaar/Circle Created with python_avatars PedroDaGr8 says:

    Ralfs birthday AND the circuit board code name was "Maltesers"!

  11. Avataaar/Circle Created with python_avatars DC Allan says:

    Nice insides to this new one, very minimal in parts ect.

  12. Avataaar/Circle Created with python_avatars Keith Lambell says:

    It does seem strange to power the low voltage microcontroller off the 12V rail when active. Happy New Year Clive.

  13. Avataaar/Circle Created with python_avatars devttyUSB0 says:

    Very intereting "tear down"! It is amazing how much circuitry goes in one of these things. We used to just plug a resistive heater in a wall outlet with that wick near to it. Heheh. Those were fun 'will my house burn down or not'-devices. 😉

  14. Avataaar/Circle Created with python_avatars pyromaniac303 says:

    I thought with the cylindrical crystal it would have some kind of daily timer function. Maybe the resonant frequency of that piezo is just that critical? Even most of the basic microcontrollers now give you a 1% or 2% accuracy internal oscillator so seems an odd choice for a cost optimised product. Think the microcontroller is a Fremont Micro Devices part but I can't find it on their website so could be something unlisted / customised.

  15. Avataaar/Circle Created with python_avatars AMDRADEONRUBY says:

    Yay a new video . Happy new year again Clive

Leave a Reply to mikenco Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.