Improve fullrange PC Speakers? Genius SP-HF350X

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello. I have just bought these active PC speakers. They sounded a lot more consistent than everything else that was available, yet they have one strong resonance, and i would like to have some recommendations on how to get an improvement.

Unfortunately, i cannot find their specs in english. Here are the original (german) ones: http://www.genius-europe.com/produktdetail.php?ID2=71&ID=29&ID3=352

They are "wooden" (apparently something like 1cm thick hard paper), approx. 14x21x16cm large, and have a single 3.5" fullrange driver. The cabinet has no wool in it.

The BR is placed at the top back, with only 4cm(1.5") axial offset from the speaker. The BR is 1" diameter and 7cm (2,75") long. According to BOXSIM, resonance is at fb=73 Hz.

By playing around with generating sine waves on the PC, i can say that the speakers really play from 45Hz upwards quite consistently, up to 110 Hz. That's where the problem starts - an awful, evil resonance around 130 Hz! It's also a great pity, since the sound in the middle and high frequency range is flawless, or at least miles better than any PC speakers i have heard so far. They are inexpensive too, just 30 EUR, which suits me quite well since i'm on low budget atm and have no amps or speakers otherwise. :dead:

How does this resonance happen? Is there any way to reduce it, while not taking away too much lower bass?

Thanks!
 
Hello and thanks a lot for the reply! :worship:

using software EQ does seem to remedy the situation, though it doesn't seem to be possible to pipe all the sound output through the EQ. So i would be interested in building a bandstop filter in hardware.

If i would go for a passive design, i could measure the impedance and amplitude curves using soundcard, schematics and software from german Linux-Magazin, and load that into boxsim to design the filter. Yet an active solution on the amp input would seem more simple and efficient to me. However, i have no idea on how to go about designing active filters. Are there any materials on this? I only have as much background on electronics as digital equipment is conserned. :xeye:

What would be disadvantages of an active filter compared to the digital EQ? Wouldn't perhaps a resonance circuit of the bandstop filter model the resonance more correctly than an EQ?

How high does the driver Q generally need to get so that it results in such an awfully audible resonance?

My parents seem to have a pair of these drivers: http://www.visaton.de/en/chassis_zubehoer/breitband/frs8_8.html
Would this kind of resonance happen with them, if they are placed in a similar box? The one i had thrown together in boxsim would be 4-5l, with 80-90Hz BR tube, and it didn't predict any major resonant peak... is it missing anything?

Cheers!
 
Hi,

4 to 5 l with a driver of 0.9 L Fs is pretty much a waste of time.
The driver has a Qts of 1.0 which is too high for reflexing.

The best way to use all small drivers is to roll them off and add
a subwoofer for some real bass. Speakers cannot produce any
bass below port frequencies, your 45hz claim is highly dubious
and is probably muddy harmonics.

:)/sreten.
 
3yE said:
Hello and thanks a lot for the reply! :worship:

using software EQ does seem to remedy the situation, though it doesn't seem to be possible to pipe all the sound output through the EQ. So i would be interested in building a bandstop filter in hardware.

If you're using a Mac, you can process all system audio with effects using Audio Hijack Pro. I use it a lot to correct the response of my headphones, my built-in speaker and even as a notch filter for my fullrange units.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.