Osram vs. Gauss, or how I looked for a halogen replacement — updated a year later: it’s all bad

I got tired of replacing the capsule halogen bulbs in my desk lamp. I tried both branded and no-name ones, and the main difference was brightness. The lamp has also been on for 10–12 hours a day lately, so I decided to save some money and install something more modern—and take a look at what’s available.


I bought them at a brick-and-mortar store, Epicentr in Kharkiv. When choosing, I immediately ruled out Feron and similar brands and focused on better-known names. I also needed a bulb bright enough to replace the original 40 W halogen.

First I looked at the Osram bulbs they had in stock, rather than available to order. I picked the Osram Star 3.5 W, rated at 400 lm and advertised as equivalent to a 40 W incandescent bulb. It cost a little over $4, though the 4000 K color temperature wasn’t quite what I wanted.


As an alternative, I was also thinking of trying a Svitokomplekt bulb. It costs less, has a higher power rating, is advertised as equivalent to a 60 W incandescent bulb, but it’s longer and has a 4500 K color temperature.


But when I got to the store, I found out that:
1. They had one Svitokomplekt bulb. I only needed one anyway, so I picked it up.
2. They didn’t have the 3.5 W Osram bulbs, even though the website said they did.
3. They did have a 3.8 W, 2700 K Osram bulb, which wasn’t listed on the website.
4. They had 4 W, 2700 K Gauss bulbs, which the sales assistant recommended. I said I’d heard Gauss “wasn’t what it used to be,” but she replied that it was all “competitors’ dirty tricks” and that all the online reviews and tests were paid for. I could only smile at that (hi, Lamptest :).

I picked up all three bulbs—Svitokomplekt, Osram, and Gauss—and asked the sales assistant to turn them on so I could check for flicker. “And then there were two”: the Svitokomplekt was out, since its flicker was clearly visible on my phone camera.

I ended up bringing two bulbs home: the 3.8 W Osram for about $6 (145 UAH) and the Gauss, which cost one and a half times less, about $4 (99 UAH).

In principle, everything looks fine: the Osram is rated at 3.8 W and 470 lm, while the Gauss is 4 W and 400 lm. A bit odd...
But here I should add one more caveat. In the last photo, I highlighted the power rating and the equivalent incandescent bulb wattage.
The Osram honestly says: 3.8 W = 470 lm = a 40 W regular bulb.
The Gauss is a different story: 4 W = 400 lm = a 40 W halogen bulb.



Let me explain what bothered me so much about the Gauss bulb’s specifications.
Halogen bulbs are more efficient, at around 12 lm/W, while regular incandescent bulbs manage only 10 lm/W (that’s for 40 W bulbs; higher-wattage ones are more efficient).
So a 40 W halogen bulb produces 480 lm, which is close to what Osram claims, while a regular bulb produces only 400 lm, matching the Gauss figures. But Osram doesn’t say it’s equivalent to a halogen bulb, while Gauss does. As they say, in the latter case, “the deception is obvious.” Osram, on the other hand, has actually understated the equivalent wattage a little—or at least been honest about it (if they meant halogen bulbs).


I also later found out that the Osram G9 pin 3.8 W is sold at other stores for about 30 UAH (a little over a dollar) more, and comes in different packaging. More mysteries....


Anyway, let’s move on to the most interesting part: the inspection and tests.

First, a general view. You can immediately see that the Osram bulb is significantly larger than the Gauss. The Gauss is also a little larger than a regular halogen bulb, so keep that in mind when replacing one. It wasn’t an issue in my case, but it might not fit somewhere else.
The Osram bulb is about 60 mm long.


The Osram also has slightly shorter pins, which is why it wouldn’t light up in the test socket at the store. The sales assistant had to hold it in a regular socket to turn it on. I had no problems at home.


The Osram LEDs shine in only two directions. The Gauss LEDs are arranged not just around the bulb—in fact, on all six faces—but there’s also a pair on the end.


During the inspection, I found that the Osram has a disassemblable design, which I naturally took advantage of.
Inside is a board with 34 LEDs, 17 on each side, and an aluminum heatsink. And here’s another mystery: the Osram 3.8 W shown above, but in blister packaging, has 40 LEDs, 20 on each side.
Under the heatsink is an electrolytic capacitor that looks suspiciously like one from Samwha’s budget series. Presumably 1 µF.
The whole thing is controlled by the BP5151 driver.


Another difference with the Osram is that its housing acts as a diffuser, while the Gauss just has clear plastic. By the way, it has 32 LEDs.


After inspecting them, I was already leaning more toward the Osram, but I still wanted to compare them in operation.

First, the Osram
The lamp says its current is 35 mA, and the driver's datasheet specifies 40 mA, but this can be changed by replacing the current-sense resistor.
In reality, the meter showed a current of about 29 mA, which fluctuated between 29 and 29.5 mA during operation.
The lamp's power, measured with a wattmeter, was 4.4 W, with a current of 30 mA and a power factor of 0.62. The mains voltage was around 233–235 V.

Of course, a wattmeter isn't very accurate at such low power levels, but so far it looks like the lamp's actual power is about 10–15% higher than specified.


Gauss
The current is 27 mA and fluctuates a little too. The wattmeter shows a power of 3.4 W, clearly short of the specified 4 W.
The power factor is 0.72, which usually means the filter capacitor has a lower capacitance than in the previous lamp.
After warming up, the wattmeter reading dropped to 3.3 W, so we're looking at an under-delivery of about 20%.


I also measured the temperature after 15 minutes of operation on the desk. It was similarly high for both lamps, exceeding 80 degrees. I think the Osram's aluminum heatsink saves it: the temperature is the same, but its power is 25% higher.
In any case, I had to remove the protective glass from my desk lamp so the new bulbs wouldn't suffocate.


For the test setup, I used my usual desk lamp. The distance from the bulb to the desk surface was about 40 cm. Neither the lamp nor the smartphone I used as a lux meter was moved during the test.


Now let's look at illuminance and color rendering.
1, 2. First, with the halogen bulb installed, I calibrated the white balance using a sheet of white paper. I also found that the illuminance fluctuates a little. A smartphone is obviously not a measuring instrument, but it's perfectly adequate for a rough estimate, especially a relative one rather than an absolute one.
3. The Osram lamp: illuminance dropped by half, and the color has a slight green cast.
4. The Gauss lamp: illuminance is really poor, more than twice as low as with the halogen bulb. But the color is slightly cooler and has less of a green cast.


And now, subjectively.
Just looking at the light, the Osram isn't much dimmer than a 40 W halogen bulb—by eye, about 30% dimmer—and the green tint isn't noticeable. In practice, its light is very similar to the halogen's, and you could even mistake one for the other.
The Gauss has slightly less pleasant light and is noticeably dimmer. This may partly be due to its different design, but it's hard to say anything definite.

And objectively.
The Osram lamp draws more power than specified (4.4 W versus 3.8 W), but its illuminance is lower than that of a 40 W halogen bulb—more like a regular 40 W incandescent bulb.
The Gauss is rated at 4 W but actually draws just 3.3 W. It is also noticeably dimmer, perhaps even matching its specified 400 lm, but both its design and specifications are clearly inferior to the Osram lamp.

I'm hoping to save money based on the fact that my old bulb was 40 W and was on for, say, 11 hours a day, for a total of 440 Wh. The new one is 4.4 W, or 50 Wh per day. That's a difference of 0.39 kWh per day, which at a rate of 1.7 UAH per kWh works out to savings of 0.663 UAH per day. The lamp will pay for itself if it lasts 145�.66330=7.3 months. The only question is whether it will last that long... Oh, and I almost forgot: as for how long they'll last, Gauss offers a 1-year warranty, while Osram offers 3 years.

A year later

Yes, it's been almost a year since I started using the Osram bulb, so it's time to sum things up.
And the results aren't encouraging. After 10 months, almost to the day, the bulb suddenly dimmed—so much that at first I thought it had failed completely. But I found that if I jiggled it a little, it would come back on at full brightness and could even keep working for a while.

Judging by how it reacts to being flexed, and by the fact that the LEDs started fading slowly after being switched off, there was probably a break in the LED circuit, either in the wiring or in one of the LEDs.


In the photo above, it looks like the bulb is still shining quite brightly even when broken, but that's just the camera's auto-exposure compensating. In reality, it's much worse. The faint glow on the left is the Osram bulb after 10 months.
If I jiggle it a little, it shines normally again for a short while, and its illuminance has dropped by about 10% from the original level during that time, i.e. I wouldn't say the LEDs had degraded much.


I wanted to take it apart and see what was inside, but the plastic had become more brittle over time, so I had to abandon the teardown and just go to the store to try to get it replaced, since it came with a 3-year warranty.

I'd never returned a light bulb under warranty before. Come to think of it, I can't remember ever contacting anyone about a warranty repair for any of my equipment. I was most worried that we'd try turning it on at the store and it would work, but Osram didn't let me down—the bulb refused to light there, too :)
They had no problem issuing a full refund, since they didn't have that exact bulb in stock, and offered me a replacement. The only catch was that they asked for some kind of ID. I happened to find a photo of my passport on my phone, and that did the trick.


So, I got 145 UAH back and bought a new bulb for 89, so you could say I got to use one for free and even came out ahead :)


The new bulb is called LED STAR PIN 40. Once again, the store's website is confusing: for some reason, it lists it as an Osram 2.6 W clear G9 220 V 2700 K bulb, and the picture and packaging look a little different - link. The same store also carries a more expensive (about twice the price) dimmable Osram Superstar Pin 3.5 W 2700 K and a non-dimmable Osram Star 3.5 W 4000 K.

As we can see, its rated power is 3.5 W, its brightness is 400 lm, and it's marketed as a replacement for a 40 W bulb, though it doesn't say whether that's a halogen or a regular incandescent bulb.
What's curious is that the previous bulb was rated at 3.8 W and 470 lm, and it was also advertised as a replacement for a 40 W incandescent...


It's about the same size as the previous one. For obvious reasons, I couldn't photograph them side by side, so here's a comparison with the Gauss bulb that replaced it for about a month and a half, and a regular G9 halogen bulb.


1, 2. According to the markings on the bulb, it's rated at 3.5 W. The previous one was rated at 3.8 W, even though it was sold as a 3.5 W bulb.
3, 4. The bulb's housing is oval rather than round. It wouldn't come apart, but even so I could see that it has the same kind of capacitor and chip inside.
5, 6. This time there are only 16 LEDs, compared with 34 in the previous bulb.


I happened to find a light distribution diagram for bulbs in this series online, in case it's useful.


I couldn't make any meaningful comparisons, since I no longer have the halogen bulb I used to set the white balance at home, so I simply measured the illuminance.
When switched on, it measured 1023 lm. The previous bulb measured 1122, but after just half a minute the illuminance drops to 971 lm, and after 10 minutes it's down to 934 lm. I didn't check how much the illuminance dropped as the bulb warmed up a year ago, but I don't remember it changing this much. I think I would have noticed.



And now for the conclusions.
Based on my limited experience, this branded LED bulb lasted only about twice as long as halogen bulbs do. The LED bulb wins in power consumption and heat, but subjectively, it gives off worse light, and the difference is noticeable.
I can't really talk about saving electricity here. In practice, I used the bulb for 10 months for free, since I eventually got all my money back for it.
The new bulb is a little dimmer, but subjectively, its color temperature, color rendering, and flicker are exactly the same.


That's all from me. I hope you found this useful.