OKAY, LET'S GET STARTED - EL CHEAPO (BILAVIDEO 1)
Quote:
INTRO
My wife is giving me grief about the mess of parts and tools sprawling across our dining table a/k/a my "lab." The WAF of the mess rates to about 0, so while I'm waiting for the crossover caps and some more TWFKs, I'll go ahead and get started with the simple stuff. |
I'm going to be building a single-driver IEM using the cheapest drivers available, the BK26824, which is available through Mouser for $18.84/driver
knowles acoustics, from Digikey for $18 even
SPEAKER Z=29OHMS SS PORT ON END - BK-26824-000 and from IHI Electronics
IHI Electronics - Parts Catalog - Part Detail - BK-26824-000 (price unknown), Global Spec
Part Number BK-26824-000 from Knowles Acoustics and Newark
KNOWLES ACOUSTICS|BK-26824-000|Speaker | Newark.com for the amazingly low price of $13.23 (if you're willing to wait the lag time for them to order it from KA).
Whatever the case, this is a very inexpensive driver. I got a pair from Mouser, which is often the least expensive source, but I could have saved a buck by buying from Digikey. I'm not suggesting it's the best driver you can buy, just that it's the cheapest and a great place to start for getting your hands dirty and your feet wet.
KA has a dozen drivers in the BK series, the last five of which (including the 26824) don't have a datasheet with specs. Like most of the BK drivers, the 26824 has a sensitivity of 118. It has a nominal impedence of 16 ohms, but its specific impedence depends on the frequency. It's 16 ohms at 500 Hz but 29 ohms at 1 kHz. The frequency response graph looks identical for the seven BK drivers listed. Here's what it looks like:
http://www.knowles.com/search/prods_...-21600-000.pdf
This is going to be a very loud driver, one with very forward mids, at least undamped. Here are some quick stats:
200 Hz - 106 dB
500 Hz - 107 dB
900 Hz - 115 dB
1.5kHz - 125 dB
2.0kHz - 117 dB
2.5kHz - 123 dB
3.0kHz - 115 dB
3.5kHz - 122 dB
4.0kHz - 105 dB
5.0kHz - 107 dB
6 kHz - 10 kHz - nothing
There are three frequency peaks, one at 1.5kHz (125 dB), one at 2.5kHz (123 dB) and a third at 3.5 dB (122 dB). Given the almost 20 dB difference between bass and the midrange peak, as well as the 14 dB drop from 3.5 kHz to 5 kHz, this is clearly a driver that likes the midrange, which is where you'd expect a driver for a hearing aid, given the frequency pitch of human speech.
But what if we could match it with the right filter? Would it help? Could we take the cheapest, crappiest, driver on the market and turn it into something halfway decent?
It's time to find out. For the sake of convenience, I'm going to:
1) Solder these bad boys to the ends of a Westone replacement cable;
2) Add a dab of hot glue to help protect the leads from getting yanked off; and
3) Pop these INTO a pair of medium cushions (my size) for deep inner-ear fit UNDAMPED.
4) Experiment with the following filters:[indent]680 ohms - white[indent]1000 ohms - brown[indent]1500 ohms - green[indent]2200 ohms - red[indent]3000 ohms - orange[indent]4700 ohms - yellow
5) Report and document the results. It may well be that with the right filter, this could be a fun single driver. It may also be that with the right filter, this could at least be a useful driver in a multi-driver configuration.
Pictures will follow.