Hi all,
I recently got a sample of Bellsing 260C90021 (short BRC260 hereafter) which are kinda like clones of the well known dual driver module Knowles GQ-30783-000 and I thought it would be interesting to make a direct comprison. I subsequently performed a number of side-by-side tests in order to investigate similarities.
I thought you guys would be interested in a short report:
1)
Appearance
Both the GQ and the BRC260 look like they are the same. Look at the side by side picture and you can conclude that there are virtually no differences to be seen by eye. Low frequency driver width as measured using a precision caliper confirms this observation, the dimensions are the same to the first digit after comma. All BRC260 had the same width, demonstrating substantial consistency.
The only observable differences are
- The PCB
While the layout is exactly the same (serial resistor 0.2 ohm on mid low driver, a cap serial on the high frequency driver, did not measure). On the BRC260 the PCB looks kind of lower quality with lots of flux residues and open vias. The GQ PCB look much better, very clean with gold plated solder pads, tented vias and no flux anywhere. Interestingly if you carefully on the backside of the PCBs, the high frequency driver is connected in reverse polariy on the BS versus the GQ. Bellsing says their drivers incl PCBs are ROHS compliant, I have to accept that because you’d have to perform ICP-OES or something like that to test whether any heavy metals impurities are present.
- Sound port, looks slightly different but nothing to complain about here.
2) Measurements
One interesting nondestructive test to probe inner workings of balanced armature drivers are impedance spectra. I have measured impedance of both, a group of 4 GQs and one group of 4 BRC260s and is an direct comparison of those (all with the same setup, x-Axis log-impedance in Ohms, y-axis: frequency in Hz; not much going on below 300Hz hence not shown):
The yellowish curves are the BRC260, the darker traces with average higher impedance are GQs. Its evident that among each driver type group there is very good consistency (apart from the one grey curve which is one of the four BRC260 drivers that really looks odd, don’t know why). There are also quite some differences in impedance between GQ and BRC260 in general pointing towards difference in construction (materials, coils) inside the package. RMS impedance of the GQs from this data are 56 ohms versus 49 ohms in the BRC260s. So, BRC260 appear to have overall slightly lower impedance with less amplitude (GQs peak at slighlty below 3 kHz at above 140 ohms and 190ohms at ca 6.7 Hz) which could be a tiny advantage for lower-powered amplifiers such as those in most mobile phones. The one BRC260 driver that has a markedly different impedance spectrum worries me a bit. I haven’t tested that one in full but this could point to issues. If I have time I may update on that later.
Onto acoustic frequency response, I just compare one of each drivers with each other as I had no time to solder up all of the BRC260s. termination was a 1cmx1.5mmID acoustic tube to my setup (For those raising fingers now: I have a DIY setup, don’t bother to ask for IEC / DIN accuracy or references etc; what counts here are relative results).
Relative to each other, I found the two drivers to nearly identical within margins of measurement error typically observed in my setup, with the BRC260 have a bit of a deeper dip at ca 9 kHz. And thats it more or less. Its not really objective, but I also listened to both driver modules at the same time (in one ear BRC260 in the other GQ), I could not hear any difference in a nearly identical setup but that’s only anecdotal evidence.
3) Conclusion
Actually I don’t know who invented this dual driver module with integrated crossover concept was first, but I do believe that Knowles came up with the very successful GQ module in first place several years ago. So if Bellsing came in second afterwards, they either did a meticulous reengineering job or they get some kind of OEM from the same factory where Knowles produces: Both chassis and sonic performance are virtually identical. I am pretty sure the untrained operator would not be able to distinguish the drivers in the shop if you don’t have a look at bottom markings (so beware when purchasing GQs from other than official sources..).
Impedance hints to slight internal built differences which apparently have no impact on sonic performance.
The GQ is a very successful module with many listeners appreciating it out of the box. It may be quite obvious to make an exact replacement with identical dimensions that fits already existing earphone platforms so manufacturers could make an easy switch with no one noticing. Why should you switch? Well current retail price of the GQ is at (on Mouser / Digikey you get them for 27 – 35 USD depending on quantity and the BRC260 are on average 10USD cheaper per piece, moving 20 USD from your COGs to your margin which may be quite a percentage depending on your pricing strategy and translate to quite some earnings on mid / high volume sales).
In the end what will also be important is reliability and quality. As said above impedance could hint to different materials used for the internals with unknown longevity track record. You don’t want customers returning your IEMs after a few months for some malfunctioning drivers. Here I don’t have any experience yet with BRC260s (I am sure the Sampler I got was handpicked, don’t know what the variability of the BRC260 is when you order and test like 50 – 100 of ‘em). Anyone has any "statistically robust" experience with Bellsing drivers?