hi folks,
Finally got a very rough prototype built no measurements yet - have to get my mic and amp set up. Sounds very good plenty of bass - mids sound fine - tweeter not wired or dialed in yet.
Yes they are as ugly as sin and thrown together but they definately seem to have potential.
Finally got a very rough prototype built no measurements yet - have to get my mic and amp set up. Sounds very good plenty of bass - mids sound fine - tweeter not wired or dialed in yet.
Yes they are as ugly as sin and thrown together but they definately seem to have potential.
Attachments
Don't worry! This is your prototype. Once you get everything dialed in and your happy with where you are going, then worry about putting together a finished product. Enjoy the ride and experience you gain with playing around with this. And remember: Sinners always get more laughs than Saints;-)
With all due respect to Sreten sometimes it just "HAS" to be built, then listened to, then rebuild.
But this is supposed to be fun and so I say "just build it!!"
This is just my non-expert opinion and I wouldn't waste good drivers by not modeling, just that experiments and tinkering are part of the fun of DIY
But this is supposed to be fun and so I say "just build it!!"
This is just my non-expert opinion and I wouldn't waste good drivers by not modeling, just that experiments and tinkering are part of the fun of DIY
Update
Got some decent measurements
1/3 octave smoothing at 1m on axis from the front this time 😀
+/- about 2db from 30hz to 20khz
Here are some pics,
Can someone explain if this is a good waterwall plot or not and it the spectrogram is good or bad - i dont know what i'm looking for.
Output from audiotrak prodigy sound card running active crossovers
input from Behringer ECM800 and ART Tube amp to Sound blaster prodigy ZS
Amps are a mixture thrown together for testing
sure 100w digital amp - subs, 20w digital amp mids, 30w nad 3020b tweeters
crossovers
sub 300hz 8th order 0.4 q to tame rising response 30hz 4th order 1.5q
mid 300hz 8th order very high q may change to higher crossover point between sub and mids.
mid to tweeter 5khz 8th order
all crossovers are using linear phase
Please note the frequency graph covers a 3db range only from 30hz to 20khz
Got some decent measurements
1/3 octave smoothing at 1m on axis from the front this time 😀
+/- about 2db from 30hz to 20khz
Here are some pics,
Can someone explain if this is a good waterwall plot or not and it the spectrogram is good or bad - i dont know what i'm looking for.
Output from audiotrak prodigy sound card running active crossovers
input from Behringer ECM800 and ART Tube amp to Sound blaster prodigy ZS
Amps are a mixture thrown together for testing
sure 100w digital amp - subs, 20w digital amp mids, 30w nad 3020b tweeters
crossovers
sub 300hz 8th order 0.4 q to tame rising response 30hz 4th order 1.5q
mid 300hz 8th order very high q may change to higher crossover point between sub and mids.
mid to tweeter 5khz 8th order
all crossovers are using linear phase
Please note the frequency graph covers a 3db range only from 30hz to 20khz
Attachments
That's a very impressive frequency response! How much EQ is in use here, if any?
Would be interesting to see off-axis measurements as well.
Would be interesting to see off-axis measurements as well.
Thanks stig,
As i'm a complete noob i'm not sure what looks good and what doesn't
I'm not quite sure how to measure off axis. Would i just move the mic to approx 45 degrees and keep the distance at 1m? If so ill see what i can do.
I'm not totally happy with the mids and as you suggested earlier a crossover in the 500hz range may be a better option.
I'll have to do a bit of testing - this is just my preliminary results.
Yes there has been a fair bit of equalisation.
I've lowered the Q of the sub to mid crossover to compensate for a rising response so that it reads flat from 30 to 300 this also involved upping the Q of the sub low cut to compensate for the bass rollof - please bear in mind that they are basically flat baffles for the subs with no H's or U's involved.
I'm running a very h Q midrange low cut as the little 4" speakers cant quite handle the 300hz crossover. The tweeter also have a high Q high pass filter.
I have a feeling i may end up going for a single large midrange driver but at the minute i'm working with what i've got.
btw can you explain the other two pictures i posted and what i should be looking for - I don't really understand what they are showing 😱
As i'm a complete noob i'm not sure what looks good and what doesn't
I'm not quite sure how to measure off axis. Would i just move the mic to approx 45 degrees and keep the distance at 1m? If so ill see what i can do.
I'm not totally happy with the mids and as you suggested earlier a crossover in the 500hz range may be a better option.
I'll have to do a bit of testing - this is just my preliminary results.
Yes there has been a fair bit of equalisation.
I've lowered the Q of the sub to mid crossover to compensate for a rising response so that it reads flat from 30 to 300 this also involved upping the Q of the sub low cut to compensate for the bass rollof - please bear in mind that they are basically flat baffles for the subs with no H's or U's involved.
I'm running a very h Q midrange low cut as the little 4" speakers cant quite handle the 300hz crossover. The tweeter also have a high Q high pass filter.
I have a feeling i may end up going for a single large midrange driver but at the minute i'm working with what i've got.
btw can you explain the other two pictures i posted and what i should be looking for - I don't really understand what they are showing 😱
Off-axis measurement is done the way you descibe it... do measurements at 15-30-45-60-75-90 degrees
I was wondering if you had any EQ to flatten small peaks and dips... but if you havent, this is a very very good result I think.
I would keep the small mids, as its easier to get good response on the upper end of the midrange. 5 kHz is a bit high in my opinion, and 5 kHz with a larger driver simply wont work.
I was wondering if you had any EQ to flatten small peaks and dips... but if you havent, this is a very very good result I think.
I would keep the small mids, as its easier to get good response on the upper end of the midrange. 5 kHz is a bit high in my opinion, and 5 kHz with a larger driver simply wont work.
Thanks Stig,
I'll give your suggestions a go once i have another free day (very few of them atm) I also want to try a 2khz crossover for the mids - combing should in theory kick in around 1800hz so if the tweets can play well enough down to 2k i'll be happy.
As for baffleless i'm going to give it a shot as the baffle i have atm is all but pointless i think i'm also going to try a single driver and have plans for front and rear facing tweeter - im going to be removing most of the tweeter faceplate.
Finally - as you will see from earlier pics the rear magnet of these mids can be partially removed leading to a low qts and better highs - i might try running one or two of them without the magnets and do some measurements.
I'll give your suggestions a go once i have another free day (very few of them atm) I also want to try a 2khz crossover for the mids - combing should in theory kick in around 1800hz so if the tweets can play well enough down to 2k i'll be happy.
As for baffleless i'm going to give it a shot as the baffle i have atm is all but pointless i think i'm also going to try a single driver and have plans for front and rear facing tweeter - im going to be removing most of the tweeter faceplate.
Finally - as you will see from earlier pics the rear magnet of these mids can be partially removed leading to a low qts and better highs - i might try running one or two of them without the magnets and do some measurements.
A few suggestions: 1) mount (or hang) the mids alternating the way you've one with the woofers to get better front to back wave symmetry and lower distortion, 2) try varying the number of pairs of mids in use, and 3) try lowering the upper crossover of the mids from 5kHz.
How's your crossover implemented? I've found it advantageous to use +20dB, Q = 0.5 peak biquad EQ patches for dipole equalization to better manage excursion. They also have useful properties when interacting with Arbitrator.
How's your crossover implemented? I've found it advantageous to use +20dB, Q = 0.5 peak biquad EQ patches for dipole equalization to better manage excursion. They also have useful properties when interacting with Arbitrator.
UPDATE
Hi Folks,
It's been a while sorry for the delay but its getting time at home to actually do some measurements. I've been doing some tweaking today and have done a few off axis measurements. Its not perfect and will have to do something with the off axis response around the 100hz mark. I've done a bit of equalisation not too much but i am getting a major drop on the woofers at 70hz it almost looks like they are cancelling each other out at the frequency - i've equalised this out in the graph.
The measurements were taken from 0 degrees through to 90 degrees off axis up to 60 degrees things looks quite good since this is my first speaker.
As always if you can offer any suggestions for improvement it would be much appreciated.
Equalisation consists of an increase around 70hz , a + 1.5db low shelf on the mids and thats it so not too severe i think.
I've attached the picture.
P.S. I recently ordered a pair of linaeum tweeters from the states as i think the only way to get close to a true open baffle response from a tweeter may be to use something like this. Any of you go any experience with linaeums?
Oh an i've bought a pair of old 8" paper cone philips speakers which are also on the way i want to do a comparison running these as the midrange rather than the array.
Hi Folks,
It's been a while sorry for the delay but its getting time at home to actually do some measurements. I've been doing some tweaking today and have done a few off axis measurements. Its not perfect and will have to do something with the off axis response around the 100hz mark. I've done a bit of equalisation not too much but i am getting a major drop on the woofers at 70hz it almost looks like they are cancelling each other out at the frequency - i've equalised this out in the graph.
The measurements were taken from 0 degrees through to 90 degrees off axis up to 60 degrees things looks quite good since this is my first speaker.
As always if you can offer any suggestions for improvement it would be much appreciated.
Equalisation consists of an increase around 70hz , a + 1.5db low shelf on the mids and thats it so not too severe i think.
I've attached the picture.
P.S. I recently ordered a pair of linaeum tweeters from the states as i think the only way to get close to a true open baffle response from a tweeter may be to use something like this. Any of you go any experience with linaeums?
Oh an i've bought a pair of old 8" paper cone philips speakers which are also on the way i want to do a comparison running these as the midrange rather than the array.
Attachments
Are you sure 70Hz isn't a room mode?
Are those the two inch open back Aurasounds? If not, have a link? In my experience the more common solution is to take the back cup off a BG Neo3, which has a few advantages over the Aurasounds.
I've ordered these - just the tweeters though
Optimus Pro LX5 LX-5 Speaker Pair (Linaeum Tweeter) on eBay (end time 11-Oct-10 16:46:34 BST)
currently the tweeter are from the wharfedale diamond series just your standard 25mm tweeters. The mids were part of a bulk buy of 64 of them originally used in the wharfedale modus cube surround speakers.
Never though about room modes. Ill try move them to a different position and remeasure to see if its that i'll also measure a single woofer to see if there is any form of cancellation occurring due to the woofer spacing.
Optimus Pro LX5 LX-5 Speaker Pair (Linaeum Tweeter) on eBay (end time 11-Oct-10 16:46:34 BST)
currently the tweeter are from the wharfedale diamond series just your standard 25mm tweeters. The mids were part of a bulk buy of 64 of them originally used in the wharfedale modus cube surround speakers.
Never though about room modes. Ill try move them to a different position and remeasure to see if its that i'll also measure a single woofer to see if there is any form of cancellation occurring due to the woofer spacing.
For diagnosing room interactions I usually just compare close mic versus 1 metre measurements. Much less hassle to move the mic instead of the speakers, though I often mute the channels I'm not interested in (I've a biamp, going on triamp, so it's easy to enable or disable individual drivers).
Be interesting to see how the Linaeum ribbons perform. I haven't looked at ribbons much since the ones Zaph measured weren't so great.
Be interesting to see how the Linaeum ribbons perform. I haven't looked at ribbons much since the ones Zaph measured weren't so great.
havent checked on the room mode yet, ill take your advice and try different mic positions thanks twest.
as for the linaeums i've heard good and bad things but they are a bit of a rare item these days however, they are worth a tryout i can always use them for something else if i don't like the response.
Btw. I just received the 8" vintage philips drivers i bought form ebay for 5 pounds including postage 🙂
currently giving them a run in with no baffle crossed at 500hz - unbelievable highs to the point where i've had to drop the response from 2k upwards using a 2db shelving filter.
No need for a tweeter with these as they run right past 20khz. Max power is only 10w but tbh that gets me into the 90db range which is plenty high enough for me. I would love to know the t/s parmeters i'll get around to measuring them soon.
here's what i found out about them.
Google Translate
- Status
- Not open for further replies.