I design/build with one aim only, to get something that conveys music the way i like in a purely subjectve way. To me there is nothing objective about the voice of Emmylou Harris or Gretchen Peters, it´s a transmission of real emotion when it works like i want it to.
There will allways be different opinions about for instance drivers, there are a lot of stuff out there i would never use, cant see any reason to blurt that out to someone who loves them. This is as it should be, we all like different things, otherwise there would be only one driver manufacturer.
Measurement to me are a way to try to find out why something works well or why i ended upp totally wrong.
The x-over issue has been rectified, will post new measurements when i´ve done them again, forgot to save them last time.
And yes i listen to all kinds of music, both country and western.
There will allways be different opinions about for instance drivers, there are a lot of stuff out there i would never use, cant see any reason to blurt that out to someone who loves them. This is as it should be, we all like different things, otherwise there would be only one driver manufacturer.
Measurement to me are a way to try to find out why something works well or why i ended upp totally wrong.
The x-over issue has been rectified, will post new measurements when i´ve done them again, forgot to save them last time.
And yes i listen to all kinds of music, both country and western.
I am sorry that this thread has become so unhelpfully critical. I see a very nice cabinet construction with very good drivers there. You are to be congratulated for this.
Naturally the crossover is a bit Mickey Mouse at this early stage. In fact it's straight out of the pages of sreten's beloved diyaudioandvideo:
It's OK to say it's good at simple country music and voices, but I suspect I would hate it for complex classical music where it'll probably become muddled and inaccurate and harsh. Tony Gee is a man who knows a thing or two about filters. He has an interesting article at SpeakerBuilding.com - The Proteus - To D' Appolito or not to D' Appolito, that is the question, Page 1 on D'Appolito designs with series crossovers. Hope it is of some interest, because the drivers are remarkably similar in inductance and frequency response. Enjoy.
Naturally the crossover is a bit Mickey Mouse at this early stage. In fact it's straight out of the pages of sreten's beloved diyaudioandvideo:
An externally hosted image should be here but it was not working when we last tested it.
It's OK to say it's good at simple country music and voices, but I suspect I would hate it for complex classical music where it'll probably become muddled and inaccurate and harsh. Tony Gee is a man who knows a thing or two about filters. He has an interesting article at SpeakerBuilding.com - The Proteus - To D' Appolito or not to D' Appolito, that is the question, Page 1 on D'Appolito designs with series crossovers. Hope it is of some interest, because the drivers are remarkably similar in inductance and frequency response. Enjoy.
Attachments
Hehe system7 you summed up my situation perfectly!! My MTM's started out with the mids running full range and just a 4.3uF cap on the tweeters. They sounded very good on some music and as you say, just not right on others. It wasn't till I saw a post by Lynn Olson on what he called girl and guitar speakers (and why they don't cope with complex music) that the penny finally dropped. At that stage they had transformed to a couple of notches on the mids and a cap and lpad on the tweeters).
The original crossover (with notches and cap on tweeter) was (acoustically) 2nd order bessel. The new one is acoustically 4th order bessel. The difference is quite large (and I believe the improvement is largely due to both the mids and the tweeters no longer struggling with frequencies they really weren't cut out for). They now sound even better on girl and guitar, and don't collapse when you throw more complex music at them either. There are still some horribly compressed stuff that just sounds bad though (which suprisingly can sound ok on the car stereo).
I'll be interested to see whether Ingvar's crossover continues to evolve or not. Time will no doubt tell
Tony.
The original crossover (with notches and cap on tweeter) was (acoustically) 2nd order bessel. The new one is acoustically 4th order bessel. The difference is quite large (and I believe the improvement is largely due to both the mids and the tweeters no longer struggling with frequencies they really weren't cut out for). They now sound even better on girl and guitar, and don't collapse when you throw more complex music at them either. There are still some horribly compressed stuff that just sounds bad though (which suprisingly can sound ok on the car stereo).
I'll be interested to see whether Ingvar's crossover continues to evolve or not. Time will no doubt tell
Tony.
Rullknuffs. Of course 2 x 6.5" drivers will have more dynamics than a single 4" full range. It doesn't mean this design can't be better.
The funny thing with simple cross overs that "work" is just as winter mute said it. They can be improved. Who's to say this speaker doesn't sound good? Yet it's easy to say this speaker can be better. Sorry Invar, but I don't believe in single caps on tweeters or woofers. I just explained why in another thread and if you're interested I can link you. I won't right now because I probably already sound like an *** and making assumptions that I may be wrong about you.
The funny thing with simple cross overs that "work" is just as winter mute said it. They can be improved. Who's to say this speaker doesn't sound good? Yet it's easy to say this speaker can be better. Sorry Invar, but I don't believe in single caps on tweeters or woofers. I just explained why in another thread and if you're interested I can link you. I won't right now because I probably already sound like an *** and making assumptions that I may be wrong about you.
Actually i´m not only listening to country/americana, its just where i find the music i love, was trying to be amusing, doesn´t allways work.
I dont listen that much to classical or orchestral music but do use a couple of records for testing just what You fear can be a problem area, Carl Orff Catulli Carmina which works well, i hear a myriad of voices, separated, not just a wall of sound, Symphonie fantastique, Hector Berlioz on professor Johnsons Reference Recordings RR11, rather complex and violently dynamic.
This works but might be a bit aggressive at some cresendos, this i´ve mostly blamed the tracking of the Denon DL103R for.
I have cartridges that track a lot better but not tried on these speakers yet.
Of course it is, i´m old enough (actually a lot older than that) to know that i have a lot left to learn.
I do like serial filters, although even there 1:st order is my primary choice, did´nt try this from start here as i thoght the twin rather large wc oft the lf units would at high levels produce quite som back emf for which the hf unit is exposed in a serial filter.
Also don´t quite understand, as with so much else, how a 2:nd order serial is just that, to me it looks as a serial 1:st order followed by a parallell 1:st order.
No matter what, i try to aproach these things with an open mind and absorb others knowledge (mine is almost never enough), so, i´ll try a 2:nd order serial in the style You directed me and we´ll see what happens.
And, actually, thanks for interest and that You care about it.
I dont listen that much to classical or orchestral music but do use a couple of records for testing just what You fear can be a problem area, Carl Orff Catulli Carmina which works well, i hear a myriad of voices, separated, not just a wall of sound, Symphonie fantastique, Hector Berlioz on professor Johnsons Reference Recordings RR11, rather complex and violently dynamic.
This works but might be a bit aggressive at some cresendos, this i´ve mostly blamed the tracking of the Denon DL103R for.
I have cartridges that track a lot better but not tried on these speakers yet.
Of course it is, i´m old enough (actually a lot older than that) to know that i have a lot left to learn.
I do like serial filters, although even there 1:st order is my primary choice, did´nt try this from start here as i thoght the twin rather large wc oft the lf units would at high levels produce quite som back emf for which the hf unit is exposed in a serial filter.
Also don´t quite understand, as with so much else, how a 2:nd order serial is just that, to me it looks as a serial 1:st order followed by a parallell 1:st order.
No matter what, i try to aproach these things with an open mind and absorb others knowledge (mine is almost never enough), so, i´ll try a 2:nd order serial in the style You directed me and we´ll see what happens.
And, actually, thanks for interest and that You care about it.
I found wintermute's "Little_Girl_With_A_Guitar" reference.
http://www.diyaudio.com/forums/multi-way/190663-can-you-have-sparkling-treble-but-without-sibilance-2.html#post2604196
Nicely explained by Lynn, as ever.
A serial crossover might not have everything you need here, and are quite hard to modify without going haywire. My own inclination would be to work with Lynn Olson's Ariel parallel crossover design:
The Ariel, Part II
Seems to me you really only have to get the treble level right, because everything else is similar.
Lynn Olson said:
http://www.diyaudio.com/forums/multi-way/190663-can-you-have-sparkling-treble-but-without-sibilance-2.html#post2604196
Nicely explained by Lynn, as ever.
A serial crossover might not have everything you need here, and are quite hard to modify without going haywire. My own inclination would be to work with Lynn Olson's Ariel parallel crossover design:
The Ariel, Part II
Seems to me you really only have to get the treble level right, because everything else is similar.
Hi,
Can you test a 1st/second order ? I think you can improve the mid treble..
To do this it is very easy, to the tweeter add a 0.33mH inductor in parallel after the 6.8uF and add a RL in parallel after the 0.22mH on the woofers R=3.3 ohms, C=33uF could be electrolytic.
If you improve the sound let us know
Can you test a 1st/second order ? I think you can improve the mid treble..
To do this it is very easy, to the tweeter add a 0.33mH inductor in parallel after the 6.8uF and add a RL in parallel after the 0.22mH on the woofers R=3.3 ohms, C=33uF could be electrolytic.
If you improve the sound let us know
Hi,
Though Lynn Olson could talk for a continent of the subject of loudspeakers,
one thing for a fact is the Ariel steadfastly completely ignores the baffle step,
seems the loss of sensitivity is too much for the overall design philosophy.
Crossover design is complex, and there are no short cuts. Sometimes
though through more luck than anything else you can stumble onto
a good arrangement, but its far more likely you will be playing with
arrangements none of which resemble any sensible objective target.
Its my opinion the only way of designing a proper crossover
is subjectively comparing proper objective arrangements.
rgds, sreten.
Though Lynn Olson could talk for a continent of the subject of loudspeakers,
one thing for a fact is the Ariel steadfastly completely ignores the baffle step,
seems the loss of sensitivity is too much for the overall design philosophy.
Crossover design is complex, and there are no short cuts. Sometimes
though through more luck than anything else you can stumble onto
a good arrangement, but its far more likely you will be playing with
arrangements none of which resemble any sensible objective target.
Its my opinion the only way of designing a proper crossover
is subjectively comparing proper objective arrangements.
rgds, sreten.
Last edited:
Hi,
You can't design anything without some sort of objective method that
you think leads to your subjective goals. Poor method leads to not
being able to examine all your options objectively and subjectively.
The more rigorous your objective skills, the more you know what
you have a subjective opinion about. As stated by many speaker
designers, building speakers with expensive drivers and not much
know how really isn't the way to go for proper speaker design.
rgds, sreten.
Where did i say i dont use objective and proven methods for the design work?
I told what my aim for the end products are, whitout knowhow i dont think i would have been a succesful speaker manufacturer the last 7 years, supplying speaker to professional users who need their equipment to work in the desired way.
I have designed speakers for well over 40 Years, and, believe it or not, i´m not as stupid as my way with words sometimes suggests.
I know my skills, You thou, seems to be without par in the department of arrogance and rudeness, or do You simply enjoy it?
I told what my aim for the end products are, whitout knowhow i dont think i would have been a succesful speaker manufacturer the last 7 years, supplying speaker to professional users who need their equipment to work in the desired way.
I have designed speakers for well over 40 Years, and, believe it or not, i´m not as stupid as my way with words sometimes suggests.
I know my skills, You thou, seems to be without par in the department of arrogance and rudeness, or do You simply enjoy it?
Sreten, be cool
Sometimes design speaker is fun although you don't have the skills or the tools.
ingvar ahlberg is an adept of first order but seems open to discover other thing ? right ?
I was myself very fond of first order but i discovered phase tracking and i can design things very complicated and very simple, no matter.
What I am less fan is when people invent physical laws or ignore it.
Sometimes design speaker is fun although you don't have the skills or the tools.
ingvar ahlberg is an adept of first order but seems open to discover other thing ? right ?
I was myself very fond of first order but i discovered phase tracking and i can design things very complicated and very simple, no matter.
What I am less fan is when people invent physical laws or ignore it.
+1
1st order is interesting, but after trying for years to get one to work properly, I have to say that I would almost always revert to 2nd order filters.
A 1st order series component, plus a shunt connected notch can work well as well as a 2nd order filter, and in some cases better.
I dont wish to berate the project, I just find Morel too expensive, or perhaps not good value in the same way the mega £££ Seas/SS stuff is poor value.
Since you have a good measurement setup, I suggest you experiment with minor changes, measure, listen. Im certain there is more these drivers could offer. The main problem with odd orders is the phase separation through the crossover region. Id say that 3rd is probably easier to get right, than a 1st order filter. Unfortunately, sometimes the Idealist has to yield to the Realist in us all.
But most of all
Enjoy
1st order is interesting, but after trying for years to get one to work properly, I have to say that I would almost always revert to 2nd order filters.
A 1st order series component, plus a shunt connected notch can work well as well as a 2nd order filter, and in some cases better.
I dont wish to berate the project, I just find Morel too expensive, or perhaps not good value in the same way the mega £££ Seas/SS stuff is poor value.
Since you have a good measurement setup, I suggest you experiment with minor changes, measure, listen. Im certain there is more these drivers could offer. The main problem with odd orders is the phase separation through the crossover region. Id say that 3rd is probably easier to get right, than a 1st order filter. Unfortunately, sometimes the Idealist has to yield to the Realist in us all.
But most of all
Enjoy
I see this interesting thread drifting off into the usual diyaudio waffle! That's a shame because I like smooth polycones.
My own inclination would be to try a polycarbonate or mylar tweeter with these drivers for the complete smooth sound, but I simmed something similar using a visaton metal tweeter and the 1.2mH Visaton W170S drivers. Your soft dome Morel CAT 408 tweeter should be fine.
This is an MTM layout on a 60x20cm baffle, aligned for freestanding on 40L overall closed box. It worked much better than I expected. The 5kHz notch is a BEAUTY which dumps some cone breakup and aligns the phase (and hence imaging) very well.
Obviously the 0.6mH Morel CAW 638 units will behave slightly differently, but I see enough there to encourage you. I forced a steepish rolloff on the tweeter to give it an easy life.
My own inclination would be to try a polycarbonate or mylar tweeter with these drivers for the complete smooth sound, but I simmed something similar using a visaton metal tweeter and the 1.2mH Visaton W170S drivers. Your soft dome Morel CAT 408 tweeter should be fine.
This is an MTM layout on a 60x20cm baffle, aligned for freestanding on 40L overall closed box. It worked much better than I expected. The 5kHz notch is a BEAUTY which dumps some cone breakup and aligns the phase (and hence imaging) very well.
Obviously the 0.6mH Morel CAW 638 units will behave slightly differently, but I see enough there to encourage you. I forced a steepish rolloff on the tweeter to give it an easy life.
Attachments
Fiddling with x-over a little, tried reversing phase of hf etc, mostly listening thou.
I didn´t think this kind of drivers needed this much breakin´in time, i´m used to that with "pro" drivers but thought that drivers with softer suspensions worked better out straight of the box.
Not so, they have improved a lot the latest weeks, both in lf extension but also in imaging and recreating (creating?) ambience in recordings, this might be filterrelated too ofcourse.
Will do new measurements and post soon.
Still, they have grown to something to my ears eminently enjoyable for listening to music on.
I didn´t think this kind of drivers needed this much breakin´in time, i´m used to that with "pro" drivers but thought that drivers with softer suspensions worked better out straight of the box.
Not so, they have improved a lot the latest weeks, both in lf extension but also in imaging and recreating (creating?) ambience in recordings, this might be filterrelated too ofcourse.
Will do new measurements and post soon.
Still, they have grown to something to my ears eminently enjoyable for listening to music on.
I like them a lot, not really a fan of domes, spent the latest 20 years listening to hf units with controlled dispersion, cd/horn and amt drivers.
This thou is something i really can live with and works good in my room.
Will build another pair of these speakers, allmost identical internally but with the SCM 636 lf units instead and a Mundorf AMT23CM in between, mainly out of curiosity and to see if the rather expensive lf drivers are worth the extra money.
Hi,
a) You didn't say you did, you just stated marketing waffle.
b) So do I. But I'm not a person who will quote such guff
as I've missed out in the above quote as an argument.
http://guruproaudio.com
I really can't do with disingenuous representation. Having
worked with a Swedish company for years (Ericsson), I
have first hand experience of how you might think its OK.
It isn't, not by different cultural standards, and you'll
never get as far as you could in English speaking market
with a name like guruproaudio, its a total oxymoron.
rgds, sreten.
Last edited:
Good Morning, it is here at least.
Where? Is that to imply that i´ve something to do
with Guru Audio? If so, that is not the case, never has been.
If there are any oxymora in my writing they are unintentional, based on
substandard knowledge in English, or might be clumsy attempts at beeing funny.
Where? Is that to imply that i´ve something to do
with Guru Audio? If so, that is not the case, never has been.
If there are any oxymora in my writing they are unintentional, based on
substandard knowledge in English, or might be clumsy attempts at beeing funny.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.