Hi, I quickly prototyped some open baffle line array speakers using cheap 8ohm 120Hz - 16Khz "full range" 88dB/W 7.5cm dia drivers. They are wired in parallel series strings to present an 8ohm load. The problem is that at anything above moderate spl they distort quite nastily. I would describe the distortion as asif hearing a very loud sound (so loud that the ear produces a distirtion). I would guess standing waves of some sort as that is what causes the distortion in the ear at very high sound levels. Just to clarify the speakers are not outputting anything near that level before the distortion kicks in.
Anyway as a solution I was thinking of increasing the width of the baffle and the distance between the top and the bottom of the speakers and the top and the bottom of the panel (increasing overall height). I think also reducing the legnth of the longer stiffening strip may also help. Anyway any advice is apriciated as I am very novice at all this.
I am planning to use a pair of the improved version of this speaker with a couple of subs so thought the 120Hz point would be low enough for the crossover (poor student not everything can be hi-fi), however they apear to be cutting off way before that with the narrow baffle. Putting a sheet of MDF against the baffle side apeared to fix that though therfore increasing baffle width should help that aswell.
Anyway as a solution I was thinking of increasing the width of the baffle and the distance between the top and the bottom of the speakers and the top and the bottom of the panel (increasing overall height). I think also reducing the legnth of the longer stiffening strip may also help. Anyway any advice is apriciated as I am very novice at all this.
I am planning to use a pair of the improved version of this speaker with a couple of subs so thought the 120Hz point would be low enough for the crossover (poor student not everything can be hi-fi), however they apear to be cutting off way before that with the narrow baffle. Putting a sheet of MDF against the baffle side apeared to fix that though therfore increasing baffle width should help that aswell.
Attachments
A 7.5cm dia driver is rather small and even if you have 8 of them it does not total up a large radiating area. Plus if the drivers are very cheap then I am not surprised that they distort at higher sound levels.
I would find out the driver details, such as frequency response, impedance, Xmax, break-up regions, etc, before rebuilding the baffles.
Regards,
Bill
I would find out the driver details, such as frequency response, impedance, Xmax, break-up regions, etc, before rebuilding the baffles.
Regards,
Bill
I am surprised they are distorting at moderate levels. Do a quick check of the DCR with your multimeter to make sure you're not overdriving the amp. Those should be able to go quite loud before you hear the speaker distortion. And perhaps do as Peter says with the high pass cap. Also triple check your wiring to make sure there's no phase cancellation.
I dont know
The idea is that cheap drivers are ok, if theres enough of them
Only, that rather "simple" way of looking at it mostly relates to woofers
Maybe you could try crossing them at about 300hz
Anyway, if you have a rising response curve, and no BSC, it could easily sound like its distorted
Phase issues could cause distortion too
And all those voice coils may take more power than you imagine
The idea is that cheap drivers are ok, if theres enough of them
Only, that rather "simple" way of looking at it mostly relates to woofers
Maybe you could try crossing them at about 300hz
Anyway, if you have a rising response curve, and no BSC, it could easily sound like its distorted
Phase issues could cause distortion too
And all those voice coils may take more power than you imagine
its more off a high frequancy distortion problem, I have tried cutting the bass out and there was very little improvment the dc resistance is fine. the drivers are connected like this:
top
1---<+
2
3--->-
1---<+
2
3--->-
1---<+
2
3--->-
bottom
so 3 parallel strings of series drivers. Unfortunatly don't have any more detialed specs. Pretty sure the amp isn't overdriven as the same output level is fine when pluged into my 8ohm monitors.
top
1---<+
2
3--->-
1---<+
2
3--->-
1---<+
2
3--->-
bottom
so 3 parallel strings of series drivers. Unfortunatly don't have any more detialed specs. Pretty sure the amp isn't overdriven as the same output level is fine when pluged into my 8ohm monitors.
I suspect dome problem aswell, if I equalise out everything >1.6Khz with a single driver I can reach pretty crazy spl from just that one driver in free air before significant distortion. However this makes them rather useless speakers for me. If I could remove the dome then could I equalise the extra octave out of them considering that without that hf distortion they are capable of quite a loud volume?
Trying to run a line of drivers full range will always result in the high frequencies falling apart at a point based on the center-to-center spacing of the drivers. I'm not surprised that you were able to get a single driver to sound better. Even if you mounted the drivers with the frames touching they should be XOd by 4500Hz to a line of cheap tweeters as close together as possible(both the drivers and the lines).
Hi,
did you measure frequency response?
The drivers for sure need to be equalized, no matter what frequency response they have when measured alone. Most likely what you hear is not distortion but a very loud midrange (where these drivers couple as a line) as boydon_lepasci said.
Take a look at this excel sheet: http://www.pvconsultants.com/audio/radiation/vpr.htm to see what your "line array" will propably do.
-Micha
did you measure frequency response?
The drivers for sure need to be equalized, no matter what frequency response they have when measured alone. Most likely what you hear is not distortion but a very loud midrange (where these drivers couple as a line) as boydon_lepasci said.
Take a look at this excel sheet: http://www.pvconsultants.com/audio/radiation/vpr.htm to see what your "line array" will propably do.
-Micha
There will be a 6 - 8db dipole peak at around 700Hz for your size of baffle. Is this what you're hearing?
If you have measurement tools it would be easy to find out. Perhaps there is a cone breakup somewhere.
Increasing baffle width will only make the dipole peak lower, which is good and bad. Good because you get more bass, bad because then the peaks and dips will be more in the vocal range.
If you have measurement tools it would be easy to find out. Perhaps there is a cone breakup somewhere.
Increasing baffle width will only make the dipole peak lower, which is good and bad. Good because you get more bass, bad because then the peaks and dips will be more in the vocal range.
fantastic, I tried alot of EQ settings and your spot on at the dipole peak at 700Hz. Sound ok with the attached EQ. I will build myself a mesurment mic, does anyone know a uk source of mesurment mic capsules as us postage is very steep?
anyway from some googling based on what people have said I think I have found a solution for the comb filtering, basicly if I run the drivers edge to edge I raise the frequancy where it onsets and if I also low pass filter the drivers such that at high frequancies only the center driver is used and then active EQ the combined amplifier and speaker system so that it is flat again. Unfortunatly this means 9 drivers in parallel but I am an electronic engineering under graduate so I think I can deal with this although its going to lead to some very thick PCB traces and other dirty things like paraleled transformers.
I also need to raise the baffle off the ground aparently the avoid an "acustic short circuit" (visatons site). Widening the baffle and rounding the edges instead of having a lip there will help with my lack of bass issue (at the moment even by ear I can tell the subs will need crossing way too high).
Also I think switching to a central spine support will help as the panel seems to be vibrating a fair bit and is a little wobbley with the current mounting.
Anyway thanks for all the suggestions they have been most usefull.
anyway from some googling based on what people have said I think I have found a solution for the comb filtering, basicly if I run the drivers edge to edge I raise the frequancy where it onsets and if I also low pass filter the drivers such that at high frequancies only the center driver is used and then active EQ the combined amplifier and speaker system so that it is flat again. Unfortunatly this means 9 drivers in parallel but I am an electronic engineering under graduate so I think I can deal with this although its going to lead to some very thick PCB traces and other dirty things like paraleled transformers.
I also need to raise the baffle off the ground aparently the avoid an "acustic short circuit" (visatons site). Widening the baffle and rounding the edges instead of having a lip there will help with my lack of bass issue (at the moment even by ear I can tell the subs will need crossing way too high).
Also I think switching to a central spine support will help as the panel seems to be vibrating a fair bit and is a little wobbley with the current mounting.
Anyway thanks for all the suggestions they have been most usefull.
Attachments
ok new version built with much wider baffle (30cm), tighter driver grouping (82mm) and a vent at the bottom (used by vistion in their open baffle designs). The vent at the bottom is definatly doing something as quite abit of air is moved through it, the rear radiation of the speaker is reduced and bass response of the speaker also improved with the vent open.
Overall a massive massive improvment, the comb filtering should have been pushed to about 4.1KHz by the tighter spacing and there is now respectable bass exstension. They are now perfectly usable as standalone speakers just lacking in the <90Hz bass department and obviously comb filtering is still a problem for harmonic content giving them a slight unatural tone. I will quantify this all when I get round to that mesurment mic.
They are also capable of insane volume before any noticable distortion kicks in.
Anyway this leads me to what I hope is my final solution, to use these speakers as mids from 100Hz - 4Khz which they seem to do well and to stick with the original subs for 20 - 100Hz and to use tweeters for 4Khz - 20Khz. The system will have seperate amplifiers for bass, mid and treble as I want to impliment 4th order filters for the crossover and want exstensive eq (afterall I am only using cheap speakers). My budget is low are there any good value tweeters that would fit my requirments available in the uk?
Overall a massive massive improvment, the comb filtering should have been pushed to about 4.1KHz by the tighter spacing and there is now respectable bass exstension. They are now perfectly usable as standalone speakers just lacking in the <90Hz bass department and obviously comb filtering is still a problem for harmonic content giving them a slight unatural tone. I will quantify this all when I get round to that mesurment mic.
They are also capable of insane volume before any noticable distortion kicks in.
Anyway this leads me to what I hope is my final solution, to use these speakers as mids from 100Hz - 4Khz which they seem to do well and to stick with the original subs for 20 - 100Hz and to use tweeters for 4Khz - 20Khz. The system will have seperate amplifiers for bass, mid and treble as I want to impliment 4th order filters for the crossover and want exstensive eq (afterall I am only using cheap speakers). My budget is low are there any good value tweeters that would fit my requirments available in the uk?
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.