Review: Belkin WeMo Smart LED Globe & Test w/OSRAM LIGHTIFY

When it comes to Zigbee-enabled smart lighting devices in Australia, the one predominant vendor at this moment is Belkin with their WeMo system. While the WeMo system itself has its share of mixed feedback, it was reported by Belkin that OSRAM LIGHTIFY globes were usable with the WeMo system. As the WeMo globes have been available on the market for some time, I wondered if they too were compatible with the OSRAM LIGHTIFY hub.

Having recently reviewed the OSRAM LIGHTIFY system, I felt it was my duty to do some testing and find out whether the compatibility goes both ways. Thanks to an eBay 20% off selected tech sale, I was able to pick up a Belkin WeMo Smart LED Globe which normally sells at AU$59 for AU$47.20, plus AU$10 in shipping. For a light globe, this can be considered somewhat expensive, but this appears to be the norm when it comes to “smart” devices.


20151201-1052-0398 20151201-1051-0396

The product arrived in a slightly crushed box, but somewhat flashy box, as is expected of a premium product. This box features a transparent window where the actual globe can be seen through the packaging. This package is just the globe, and clearly reminds purchasers that a WeMo Link is required to control the globes (which you get from the WeMo starter kit). Of course, I’m not particularly interested in running a complete WeMo system, so I didn’t actually purchase a WeMo Link. The unit claims a 60W incandescent equivalency with a brightness of 800 lumens, just 10 shy of that claimed by OSRAMs globes, and likely to be an insignificant difference.

20151201-1052-0399 20151201-1051-0397 20151201-1052-0400

The side of the package tells you some of the features of the WeMo Link system, featuring the Zigbee logo. The power consumption is claimed to be 9.5W, with an A19 shape with compatibility with Zigbee Home Automation 1.2 Profile. The top of the box features a plastic hanger tab for stock display purposes.


The underside provides the full specifications, including a 2-year warranty and regulatory compliance marks for Australia. The nominal lifetime is claimed to be 25,000 hours which is 5,000 hours more than OSRAM’s claim of 20,000 hours, although I’d take that with a grain of salt. The switching cycle life is claimed to be greater than 12,500, whereas OSRAM claims 100,000. No information on CRI is provided.


20151201-1055-0404 20151201-1055-0405

20151201-1056-0406Inside the box, you get the globe itself, which is about 1.5cm longer than the OSRAM globes, which can make it a tight squeeze in smaller luminaires. It is Made in China and appears to be a recent stock from Week 3 of 2015.

When handling the globe, it is apparent that the globe is much lighter than the OSRAM counterparts weighing in at 81gm (about 40% less), which is normally a negative attribute, but looks can be deceiving. More on this in the teardown section.


You also get a one sided quick install guide and a declaration leaflet as well.


The big moment has arrived – will the globe work with the Lightify system? I plug it in, turn it on and just like the OSRAM globes, it comes on at full brightness. I open the app, and try adding a device …

lightify-wemo-scan lightify-wemo-adjust

… and as promised, the globe appears with a name of Light 01 and is correctly sensed as a dimming-only globe with colour temperature and colour dials locked out. The globe is a little different, as it does not turn immediately off when commanded off and instead slowly fades out to darkness. In this period, the globe becomes unresponsive to further on/off commands until its fading is completed. That’s a gentle touch but could be annoying if you wanted to rapidly turn the lights back on, and would look a little out of place if globes from OSRAM were mixed together with WeMo globes.

One thing you will probably miss out on is any firmware updates for the globes because they are unlikely to be distributed through the competitor’s hubs – although if you do pair it with its “native” hub, it is possible to do the update and then reset and re-pair it with the intended system.


Electrical Parameters


The globe shows a very wide voltage tolerance, with regulated operation down to 50Vrms (connectivity was not tested), where the power drops off linearly down to an impressive 10Vrms. It seems the globe really doesn’t care much about the power input at all. In the rated 220-240v operation band, the globe consumed just 8.9-9.0W, which is 0.5W less than claimed on the packet. The power factor was a disappointing 0.52-0.55, which is considered low, but of no great consequence to residential consumers.

Similar to other globes, there is a buzzing from the globe when in operation and when dimmed. For this particular globe, it was quieter than the OSRAM Lightify RGB, but slightly noisier than the OSRAM Lightify Tunable White which was imperceptible. It is mainly high frequency noise. The recording was made in my room with an air conditioner, four computers and a solid-tone emitting switch-mode charger running, and starts with the globe off. As the globe is dimmed, the “stepping” can be heard as the hub updates the brightness value with the globe.

Standby Power Consumption

Full compliance standby test was carried out using the Tektronix PA1000, with a result of 0.55981W consumption, which is a hair less than the OSRAM globes, resulting in an energy consumption of 4.90kWh/year and $1.23/year in energy bills assuming $0.25/kWh. This is 1c less per year compared to the OSRAM tunable white globe. The full report is here.

Photometric Quality


The globe doesn’t appear to be anything out of the ordinary, comprised of standard white LEDs. When dimmed, the dimming curve appears steeper than that of the OSRAM globes, reducing in brightness more rapidly. The dimming range to minimum brightness is relatively similar.

wemo-dimming-cri-cctResults from the spectrometer imply a CCT of 3263.5K which is slightly cooler than implied but within the margin of error, with a CRI of 78.7 which is very similar to the OSRAM Tunable White. The absolute value is probably slightly better, as the calibration and accuracy of the (loaned) spectrometer is not perfect.


Dimming on this globe was met with very interesting results, with a residual AC visible in the form of 100Hz “waves” with a faster PWM wave riding on top of this.


At a shorter timebase, this dimming seems to take effect at a rate of about 868Hz, and is not a hard modulated “on-off” PWM as you may expect, instead, resulting in a sawtooth pattern (as if filtered by capacitor). This results in less obvious flicker of objects in motion compared to hard-switching.


Barely hours after receiving a fairly expensive globe, what do I go and do with it? Yes, that’s right – I break it apart!

This particular unit is a pain to take apart because an excessive amount of rubberized silicone glue is used in securing the dome to the base. As a result, almost 30 minutes of prying and cursing was necessary to take the dome off, and afterwards, I ended up with a pretty gnawed up bulb.


The globe may be light, but its strength is evident upon teardown. The outer collar of the base is actually made of aluminium and is directly connected to the heatsink, thus dissipates heat very effectively and becomes quite hot to the touch during operation. This is unlike many other globes including OSRAMs which have a plastic coat over the heatsink which reduces its efficiency and adds weight.


The globe itself also has an additional feature of having a physical antenna placed inside the “dome” section for better range and reception, as the metal heatsink would often interfere with proper radio reception and transmission. This may affect the light distribution slightly but it was not noticeable in regular operation.

The PCB is marked L-A880S-A1R-01 V2.2 with a date of 14/02/2014. This implies the product is indeed made by Leedarson Lighting and is a close cousin to the OSRAM Tunable White globe.


The top solder joint in white is quite a suspect one, and it popped off during further photographing of the globe, necessitating a re-solder to restore operation. It wasn’t particularly easy as there are very close clearances, and the MCPCB wicks away heat necessary to make the joint. It also seems the MCPCB was slightly bent during installation as it did not quite fit the registration holes perfectly during assembly.

The LEDs are arranged in a 4-parallel x 6-series configuration for a total of 24 SMD units. The quad-parallel arrangement seems to be pushing the limits of sensibility in paralleling LEDs which may not share current evenly due to component differences, leading to possible premature failure. I would much prefer long series strings with no paralleling to avoid this mode of potential failure although I suppose the main reason for going with this route is to reduce the LED string voltage which may make the control electronics easier and cheaper to implement.


The Belkin WeMo globes can interoperate with the OSRAM LIGHTIFY system without any major issues which is a big win for consumers, as it proves that the Zigbee LightLink/Home Automation profiles are being implemented properly across vendors. While the globe was interoperable, differences in configuration were found – where OSRAM globes turn off “immediately” on command, WeMo globes “fade” out and are unresponsive to commands during the fade process. The dimming profile is also different, with a “steeper” response than the OSRAM globes, although the dimming range is similar. Mixing and matching globes within the same room/scene may result in obvious brightness differences.

The WeMo globes were from the same OEM as the Tunable White globes and thus, the compatibility might not be so surprising. The standby power consumption was similar to the OSRAM LIGHTIFY Tunable White globe, although the physical globe construction was different with a large discrete antenna for better range and an aluminium collar not covered in plastic for better heat dissipation. Construction quality was mixed, as a poor solder joint was unearthed as soon as the globe was disassembled and required repair. The LED arrangement was also somewhat worrying due to the quad-paralleling of LED packages.

The globe itself can be considered somewhat expensive, and only comes with dimming as the only feature. Its fixed 3000K colour temperature may suit homes at night, but is not as flexible as the tunable white offered by the OSRAM globes especially for mixed daylighting applications. It remains to be seen how much of a premium OSRAM globes will command although I suspect it may be well justified given the added flexibility and similar pricing in the US.

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 Lighting and tagged , , , , , , , . Bookmark the permalink.

Error: Comment is Missing!