Teardown: Glade Sense & Spray Automatic Air Freshener (SCJ-158)

DSC_8213Sometimes, it seems the posts around here have no rhyme or reason, and that’s exactly the way I like it. As a part of a sampling “club”, a while back, I received a Glade Sense and Spray Automatic Air Freshener. This rather interesting device claims to have “motion sensor technology”, featuring a small refill that lasts about a month.

Having tried mine, and emptied the refill, I figured it wasn’t really my thing. But it does seem rather amazing that such a product is in the consumer market. What’s inside? How does it work? These are merely some of the questions I had. Let the teardown begin!

What is it made of?

Well, the unit feels plastic-ey. The exterior of the unit is brown, with speckles, a sort of poor wood imitation that reminds me of tupperware. It is adorned by curves, with a tapered top.

DSC_8216

A look at the tapered top shows a hole where the inner mechanism shows through. The top of it contains a very small tapered hole, no doubt a nozzle that helps to further atomize the spray from the refill.

DSC_8214

The bottom of the unit shows the identification marks – S.C. Johnson and Son is probably the parent company, with this unit designated model number SCJ-158. There are numerous variations of this unit, some of the same size with different plastic exterior. It is likely the internals are very similar. As you can see from the design of the base, the unit hinges open by pushing on the sides.

DSC_8217

Inside, we see the main unit, which is the white plastic section. This is hinged at the top where the nozzle is, and clips into the rear brown segment. This suggests the different exterior coverings share a common internal unit.

The unit itself has the front panel push button for an immediate spray (termed boost), with a translucent white plastic for the LED to shine through, and a hole for the sensor. The unit is supplied with two Hi-Watt ER6X Super Heavy Duty carbon-zinc cells.

DSC_8218 DSC_8219

The canister itself deserves special mention. Aside from the fact it is marketed as Forbidden Berry in Australia and New Zealand, and Winter Romance Berry in Japan, the unit is a special dosed canister arrangement. The plastic nib at the top is depressed to release a specific dose of the liquid inside, in a similar way to asthma inhalers. This simplifies the timing of the mechanism quite a bit, but might increase the cost of the canisters themselves. The canister isn’t particularly big either.

The canister is installed into the unit simply by pushing it into place, and is removed with the skin-coloured plastic pull-tab. Some users have complained of leaks at the base of the sprayer, which seems rather unlikely as it is a pressurized spray canister. It, instead, seems likely that the canister was not fully pushed in or aligned with the nozzle shown above, instead resulting in the ejected liquid collecting in the plastic nozzle piece and eventually dripping to the bottom of the mechanism.

DSC_8220

This is how the unit looks like after being de-populated of batteries and refill canister. It’s also been removed from the rear plastic shell.

DSC_8221

It seems the internal section of the unit is marked much the same as the outside, with FCC and CE approval markings. The internal unit is held together with four screws. Inside, we can see …

DSC_8222

… quite a few gears. The motor is on the left, coated in lubricant. This drives two reduction gears in sequence (i.e. the big white geared wheels) which then drive a sector-shaped gear which helps to convert the rotational motion into linear motion, pulling down on the plastic nozzle piece and then returning it to its original position.

DSC_8223

Removing the middle reduction gear “breaks” the chain of processes, and allows you to see how the sector-shaped gear works. There are two limit stop posts, which limits the travel of the sector shaped gear, and by extension, the amount of pull-down the nozzle piece has.

This makes the system remarkably simple – it’s actually free of any feedback on this end. It literally runs the motor forward for a given time, letting the gearing hit the limit stop, then runs it in reverse for a given time, until it hits the other limit stop. Simplicity! Here, I actually ran it without any gear in the middle, but it still happily spun away.

DSC_8224

A closer examination of the motor shows it’s a Techni Micro Motor, likely an RT-310P (2.5v nominal, 2200rpm/0.06A/0.072W), which strangely enough is stated on their product information page as being targeted to products including air fresheners. I never knew there was a whole motor market for air fresheners!

DSC_8226

The main PCB isn’t adorned with much components, and is a simple single-sided board. It looks to be marked PB775B01M20-R Revision 0, dated 3rd August 2010 (a while back) but made Week 38 of 2013. There is an electrolytic capacitor, a ceramic capacitor, a micro-switch, an LED, what appears to be a phototransistor and a black plastic shroud (to limit the interference from the LED and limit the viewing angle).

It seems that their motion sensing is nothing more sophisticated than monitoring when the light level on the phototransistor falls – this is why they claim a maximum 5-feet range, and the need for adequate light. So, it’s more of a shadow detection system.

DSC_8227

The bulk of the work seems to be done by a daughter PCB that stands vertically on the main PCB. This one seems to probably have some sort of ASIC which is glob topped onto the PCB. There are three resistance jumper settings on the left – these are fixed on this model, but is likely to set the lock-out time, as the other model they have on their web page can be configured for 15-minute, 20-minute and 30-minute lock-outs.

DSC_8228

The rear of the PCB shows the way the other PCB is slotted in and utilizes its edge connector to be “solder bridged” to the main PCB. A rather sophisticated system, and it seems some of the connections are not even necessary. I wonder if this is indicative that the daughter PCB is actually a programmable microcontroller, that’s factory programmed/tested through the unused pins, or whether this controller can be utilized for more sophisticated air fresheners.

Conclusion

Well, there you go. A tour of the guts of a Glade Sense and Spray. Unfortunately, there’s not much in there that will give hobbyists something to salvage, but it was interesting to see how simple it is given the fact it uses a dosed canister (thus, the timing of the nozzle up/down is not as important) and the use of limit stops to limit the travel (thus, no positional feedback is required).

In terms of its operation, all it does is:

  1. Look for a significant dip in the phototransistor signal.
  2. Blink LED quickly for a few seconds to indicate an impending spray.
  3. Run the motor in one direction for a given time, which releases the spray.
  4. Run the motor in the other direction for a given time, which resets the spray.
  5. Go into lock-out mode for a given amount of time and blink LED slowly to indicate this while looking for a push-button signal – if there is, then go to 3.
  6. Turn LED off and go to 1.

I call this the air freshener algorithm (tongue in cheek).

About lui_gough

I’m a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!

This entry was posted in Electronics and tagged . Bookmark the permalink.

10 Responses to Teardown: Glade Sense & Spray Automatic Air Freshener (SCJ-158)

  1. Duke Ashvin says:

    Is there any way to hack to fit the 6 ounce bottle.
    or else is there any alternate method to fill this canister from a 6 ounce bottle ?

    • lui_gough says:

      Not that I’m aware of … the 6 oz bottles are probably without any “metering” system and spray continuously when the nozzle is depressed and hence is not suitable for direct usage even if mechanically adapted. The first series of Sense and Spray units were for the larger bottles, so you might be able to find them somewhere …

      I’m not sure if it’s possible to refill the canisters either.

      – Gough

  2. Jose says:

    Nice article!
    I grabbed one off the shelf today for the same reason, to see if there were a couple parts to play with with the arduino, then found your article.

    Could you expand on the photo transistor a bit? How is it likely to be operating in this circuit? Can the individual connection points be identified and functions described? Direction of current, etc? Does the direction matter?

    Finally, on the pcb there are smooth round metal dots scattered about marked T1, T2 etc. They don’t look like solder joints. What are they and are they important?

    Any input much appreciated!

    • lui_gough says:

      I suppose if you do a little digging, you can find out for yourself. Keep in mind that I don’t really have the time to answer every question that comes my way, but I’ll give your questions a go.

      The phototransistor is basically a component that works a little bit like a diode in reverse. Instead of emitting light, it is used to sense light. You can also think of it as a transistor where the base terminal is not wired out of the package and instead the base current is provided by the light that shines on the silicon (i.e. the photovoltaic effect). On the detection of light, a current is generated inside the silicon, which triggers the base, causing the transistor to turn on and provide a larger amplified current flow. This is described quite well by Radio-Electronics.com here: http://www.radio-electronics.com/info/data/semicond/phototransistor/photo_transistor.php

      The phototransistor only has two pins, one is anode and one is cathode – if you use a voltmeter or follow the traces, you should be able to work it out fairly easily which one is which.

      The solder points T1, T2 are regular features on many PCBs known as test points. They are used in the factory by automated or manual testers to probe signals on certain lines to ensure the operation of the unit as part of quality control, or to provide readings to narrow down problem areas during troubleshooting. They are not technically solder joints, and are just pads connected to lines. You could use them to take measurements, or even as places to solder on wires if you wish to abuse them in that sense.

      – Gough

  3. Jose says:

    Very helpful, thank you Gough.

    I read the link but am a little hazy still on a couple of details.

    As the level of light varies the current from the base varies in proportion, I think. But does the current between collector and emitter also fluctuate, or does it suddenly come on at a set level when a threshold is reached in the amount of base current?

    I’m trying to imagine how the glade unit would operate.
    I can understand if the collector emitter current was constantly in proportion to the base current, then a sudden change to the base current caused by shadows moving on the phototransistor would effect a change in the emitter collector current and be readable to the device (above a certain threshold presumably).
    But if that’s not how it operates, and there’s a sudden onset of collector emitter current when the base current goes above a certain threshold, I can’t see how the device could use that action to detect movement via changing light levels, as not all changes to light levels would reach that threshold.

    Am I thinking about this the wrong way?

    • lui_gough says:

      The device itself is a transistor, which means the collector current is the base current multiplied by a gain factor which is dependent on the device itself. Emitter current is the sum of collector and base. This means that the current itself would be proportional to light input through a range – notable exception is no light, where a small leakage occurs, known as dark current, and very bright, when the transistor is fully on and driven into saturation.

      If you have used the unit, it doesn’t work that well (not compared to a proper PIR) but they tried to improve it by using a black shroud to mask the detection area to a smaller field of view. It might have a fixed threshold detection, as under artificial lighting, it doesn’t reliably trigger, although it would be possible to make it more sophisticated (e.g. automatic gain control, looking for the derivative of current – fast rate of change).

      – Gough

  4. Felix says:

    Hi Gough,
    do you think if there a way to re-program or hack this device so it does not do the automatic spraying which it does every 30 minutes which is too much waste of scent and makes too heavy to bread the environment. I am trying to understand if there is a way to make it spray only when it did sense the movement. I did open mine which is from 2007 but didn’t manage to understand it due to luck of instruments and too much knowledge to test whatever.
    Thanks,
    Felix

    • lui_gough says:

      As far as I can tell, it’s a fixed function device, so there isn’t really any way to reprogram it as such. You can probably interrupt power to the unit or build your own controller if you wish – but the whole premise of motion detection is rather limited and flawed anyway – in many cases, without the right lighting conditions, the unit is essentially “blind” and operating on a timed principle.

      – Gough

  5. Jake S says:

    Teardown: Glade Sense & Spray Automatic Air Freshener (SCJ-158) I want to add a sound sensor to this device. how do I do this?

  6. Michael pearce says:

    How can I replace middle large gear so that I can be sure that it is in the correct position to ensure that it operates correctly

Error: Comment is Missing!