The driver is a B & C 6mbx44 in 3L sealed enclosure lined with felt. The crossover applied. Bypassing the crossover seems to have no effect on the issue. This graph is of the driver alone and has 1/12th octave smoothing applied.
There is a dip at approximately 820hz with a distortion spike. There is a small corresponding phase spike as well (not pictured). I initially thought it might have been a diffraction artifact but it is in every measurement no matter the distance.
The datasheet shows a small peak at 800hz, rather than a dip. Does anyone have a name or explanation for this phenomenon?
Imgur: The most awesome images on the Internet
There is a dip at approximately 820hz with a distortion spike. There is a small corresponding phase spike as well (not pictured). I initially thought it might have been a diffraction artifact but it is in every measurement no matter the distance.
The datasheet shows a small peak at 800hz, rather than a dip. Does anyone have a name or explanation for this phenomenon?
An externally hosted image should be here but it was not working when we last tested it.
Imgur: The most awesome images on the Internet
You could check the impedance of driver in the enclosure to see if it's coming from a lack of internal damping. If there is a wiggle there at ~820 Hz, check the impedance of the driver outside of the enclosure and see if it's still there.
Just a suggestion... You can measure impedance curves with a known resistor and some wiring. Just check google.
Just a suggestion... You can measure impedance curves with a known resistor and some wiring. Just check google.
Thanks for the response. Yeah that's the step I haven't tried yet. I'm sure there is a wiggle there. There should be enough dampening since it's lined with thick felt. Maybe too much? Quick little wiggles are such a pain!
I could be wrong, but that looks like the basket/magnet reflection.
If it really bothers you, buy some 'closed cell' weather stripping. Your local hardware store carries it.
First: the cones surround is glued to the basket. If you feel from the underside of the surround, see if there's a big lip. I know it's a pain, but stick the weather stripping to this lip. Be sure not to impede the motion of the cone. Measure.
Second: the baskets fingers.
Third: wrap the drivers magnet.
Do these one step at a time. You want to identify the culprit.
If it really bothers you, buy some 'closed cell' weather stripping. Your local hardware store carries it.
First: the cones surround is glued to the basket. If you feel from the underside of the surround, see if there's a big lip. I know it's a pain, but stick the weather stripping to this lip. Be sure not to impede the motion of the cone. Measure.
Second: the baskets fingers.
Third: wrap the drivers magnet.
Do these one step at a time. You want to identify the culprit.
That's why I use impedance measurements while filling my enclosure. I've had to use 2 different damping materials to solve everything I needed to solve.
Once you look at what the damping does to the impedance curve you're not relying on blind luck anymore
. I used real wool felt on the walls while filling the center of the enclosure with (fluffy and itching) fiberglass insulation material, using a wool felt layer between the back of the driver and the fiberglass insulation. It worked better than using only one or the other, the impedance curve will tell you.Do check the driver by itself though. Outside of the enclosure, to make sure it isn't caused by the driver.
Last edited:
That's why I use impedance measurements while filling my enclosure. I've had to use 2 different damping materials to solve everything I needed to solve.
Once you look at what the damping does to the impedance curve you're not relying on blind luck anymore
Do check the driver by itself though. Outside of the enclosure, to make sure it isn't caused by the driver.
You've had to put a wool layer behind a driver? Oh man hahah.
Well it sounds like the imp curve is likely being affected by the enclosure there and causing this. I'll do all the measurements to figure it out! Thanks so much wesayso. Is there a name for this phenomenon?
I chose to do that to avoid the fiberglass fill fibres to enter the breathing hole in the back of the driver.
It wasn't just one driver though...
What name are you looking for? As it also shows up in the distortion it could be a resonance of the enclosure.
Here's a couple of early results I got with all kinds of variations and combinations of damping materials I tested. This was one single driver in a test box:
See how different damping schemes resulted in different impedance curves?
I went from this test forward to find the best combination I could come up with before committing to fill all of the chambers.
It wasn't just one driver though...
What name are you looking for? As it also shows up in the distortion it could be a resonance of the enclosure.
Here's a couple of early results I got with all kinds of variations and combinations of damping materials I tested. This was one single driver in a test box:
See how different damping schemes resulted in different impedance curves?
I went from this test forward to find the best combination I could come up with before committing to fill all of the chambers.
Last edited:
My gut feeling is that this is par for the course with this driver, not a manufacturing defect. Many other midwoofers exhibit the exact same type of behaviour, see the Peerless 830882 doing it at 450Hz here:
Zaph|Audio
The tell tale signs are a 'peak/dip combination' at the problem frequency in the frequency response, increase non-linear distortion and maybe a peak or roughness visible on the impedance. I see these issues in both the frequency response and impedance in the B&C datasheet.
If it's any consolation, the fact that non-linearity is only significant on the 2nd order harmonic means that it may not sound subjectively offensive.
All distortions going up at the same time means something is rattling (but every so slightly). Might be the leads, for example. Might also explain the coincident dip in the FR (phenomenon costs energy, which is detracted from cone movement and hence acoustic output).
I used the 830884 and saw the same thing (at a slightly lower freq). I was hoping to get away from it
The distortion isn't so bad by itself. I'm more worried about the frequency response, it dips and jumps with an error of about 5db. Not exactly close to the 1 or 2 db rise at 800hz depicted on the site.
Here's a freq response with no crossover
Imgur: The most awesome images on the Internet
Vacuphile - I'll have to investigate that, I hadn't thought of that.
Hi Spice house.
Having looked at frequency response with no xover it looks like that is factory "feature". Factory impedance measurement have the resonance at 850Hz and slight peak in frequency response - if we take into account factory toleances that's it.
I had similar problem with 6md38. It has a dip in 700Hz area. I've found 5 different independant measurements of it later and some show dip at 700Hz and peak at 1.5KHz, other do not. Bad batch i guess.
cheers
Having looked at frequency response with no xover it looks like that is factory "feature". Factory impedance measurement have the resonance at 850Hz and slight peak in frequency response - if we take into account factory toleances that's it.
I had similar problem with 6md38. It has a dip in 700Hz area. I've found 5 different independant measurements of it later and some show dip at 700Hz and peak at 1.5KHz, other do not. Bad batch i guess.
cheers
Damn, I just ordered a 6md38 in hopes that it would work better.
'Bad batch' is a unexpected thing from such a praised manufacturer.
This is the only driver I've ever worked with that had such significantly different specs than specified. I suppose I should have gotten the hint when my modeled crossover didn't work at all as expected from this driver, and yet worked exactly as modeled for every other driver.
We'll see what the manufacturer says tomorrow, hopefully they can provide a solution. If not, I'll be sticking with scan speak from now on
'Bad batch' is a unexpected thing from such a praised manufacturer.
This is the only driver I've ever worked with that had such significantly different specs than specified. I suppose I should have gotten the hint when my modeled crossover didn't work at all as expected from this driver, and yet worked exactly as modeled for every other driver.
We'll see what the manufacturer says tomorrow, hopefully they can provide a solution. If not, I'll be sticking with scan speak from now on
This measurement is most similar to mine:
There are measurements done by Zaph and Dan Neubeck. They both use Soundeasy. On their measurements it looks flat but on every other more or less resembled factory measurement. Only the thing is that dip at 700-800 hz i had was at least 4dB. Rest of the frequency response was like factory (more or less).
Scan Speak lacks in sensitivity since i wanted 93-96 dB out of it. There are few more to try (PHL and 18sound) but the price is higher than B&C.
There are measurements done by Zaph and Dan Neubeck. They both use Soundeasy. On their measurements it looks flat but on every other more or less resembled factory measurement. Only the thing is that dip at 700-800 hz i had was at least 4dB. Rest of the frequency response was like factory (more or less).
Scan Speak lacks in sensitivity since i wanted 93-96 dB out of it. There are few more to try (PHL and 18sound) but the price is higher than B&C.
Last edited:
Depending on xover frequency you want to achieve, you could try 18sound 6nd410 or 18sound 6nd430-4.
I truly hope you'll do better than me with 6md38.
I truly hope you'll do better than me with 6md38.
Last edited:
Thanks for the warning! Just switched my order to the 6ndl38. Found some 3rd party measurements that proved it's at least as flat as advertised up til 1k.
Concerning the 6mbx44, the rep for the manufacturer told me he would contact the engineering dpt to see what they thought. We'll see!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.