Greg_a_tronic© Tapered TL : finally ive finished designing!
FIrstly, id like to thank PKITT and all the others that have helped me out with my little questions, whether the were stupid, or informed questions: and also to those that offerred critique, (even if i still ended up ignoring it), and also to the guy who did the excel sheet of MJK's alignment tables, and made my calculator work unneeded, and as a result, more designs to be considered. If not for him (and MJK) i would be more than likely STILL: designing, or possibly just building another SBB4 box. Instead im taking a voyage into new territory, QW land....
This have proved a challenge as although i am fairly competant at wood work, i am no master craftsman, and have only built cuboid boxes before. For this project i was to delve into non parallel (or less parallel) domain. so i chickened out, for want of a near perfect result, and not another botched cabinet! The plans are all done and just been sent to the CNC man, i just have the waiting and gluing, finishing before i can tune them and get used to the first finished speaker ive had in about 10 years. YES the others are about 10 years old, a first project, and STILL have wires hanging out of driver holes......
So here is a mock up pic of what the finished article should look like roughly to scale.
in case anyone is interested, the drivers are Visaton AL130 8, and g20sc tweeter. 2nd order elec/3rd acoustic HP, and 2nd order LP.
comments both good and bad welcomed
FIrstly, id like to thank PKITT and all the others that have helped me out with my little questions, whether the were stupid, or informed questions: and also to those that offerred critique, (even if i still ended up ignoring it), and also to the guy who did the excel sheet of MJK's alignment tables, and made my calculator work unneeded, and as a result, more designs to be considered. If not for him (and MJK) i would be more than likely STILL: designing, or possibly just building another SBB4 box. Instead im taking a voyage into new territory, QW land....
This have proved a challenge as although i am fairly competant at wood work, i am no master craftsman, and have only built cuboid boxes before. For this project i was to delve into non parallel (or less parallel) domain. so i chickened out, for want of a near perfect result, and not another botched cabinet! The plans are all done and just been sent to the CNC man, i just have the waiting and gluing, finishing before i can tune them and get used to the first finished speaker ive had in about 10 years. YES the others are about 10 years old, a first project, and STILL have wires hanging out of driver holes......
So here is a mock up pic of what the finished article should look like roughly to scale.
An externally hosted image should be here but it was not working when we last tested it.
in case anyone is interested, the drivers are Visaton AL130 8, and g20sc tweeter. 2nd order elec/3rd acoustic HP, and 2nd order LP.
comments both good and bad welcomed
Last edited:
ah ha! and there was me thinking i was being original in some way...ho hum
i cant say ive seen the fullrangers youre talking about, would be interested in seeing them though so i may have to search them and see what turns up!
and yes foolishly i thought they were different to may of the 'conventional' looking folded TQWP designs i had seen, but with so many people out there with ideas im not surprised i didnt think of it first IAH...i was trying to avoid emulating others designs but your praise of those designs makes it an easier cross to bear. haha. never theless, the aim was to eliminate a fold and any possible sectional resonances that this may introduce...modelled F3 is around 40Hz, and the taper ratio is 20:1, if memory serves me correctly(so many sims), pkitt did a few options for me on the MJK models since i dont have them, and didnt want to overdesign, what is a 1st QW project for me.
either way i expect it to be more natural sounding than the SBB4 i had planned previously......
pics as soon as i have the parts back from CNC man.
Thanks again for the input.
i cant say ive seen the fullrangers youre talking about, would be interested in seeing them though so i may have to search them and see what turns up!
and yes foolishly i thought they were different to may of the 'conventional' looking folded TQWP designs i had seen, but with so many people out there with ideas im not surprised i didnt think of it first IAH...i was trying to avoid emulating others designs but your praise of those designs makes it an easier cross to bear. haha. never theless, the aim was to eliminate a fold and any possible sectional resonances that this may introduce...modelled F3 is around 40Hz, and the taper ratio is 20:1, if memory serves me correctly(so many sims), pkitt did a few options for me on the MJK models since i dont have them, and didnt want to overdesign, what is a 1st QW project for me.
either way i expect it to be more natural sounding than the SBB4 i had planned previously......
pics as soon as i have the parts back from CNC man.
Thanks again for the input.
thanks! im pretty chuffed with the look of them now that theyre drawn out to scale too...its the sound ill be dying to hear though!
Very nice design!
Is the hole on front the terminus or is it a mass loaded design with a port?
Kinda reminds me of my 10:1 TL's. They are for wall mounting so no base plates are needed.
Wish you all luck with the building!
Regards H@kan
(attaching a image of my 10:1 TL's)
Is the hole on front the terminus or is it a mass loaded design with a port?
Kinda reminds me of my 10:1 TL's. They are for wall mounting so no base plates are needed.
Wish you all luck with the building!
Regards H@kan
(attaching a image of my 10:1 TL's)
Attachments
yes it looks similar to your wall mounted design. The hole in the front is just the terminus, its a simple 20:1 tapered QW tube, not a MLTL design. I considered a slot for the terminus but the 'convential' looking port 'hole' seemed easier to me to implement, so i chose to do that instead.
ive just finished the simulation of the design, using Boxsim. I modelled the baffle size and effect on response, and crossover mods to the overall FR. As there is no TL option, or that i know of(such is the level of my german reading), i substituted a BR box instead. In this way the FR below about 100hz isnt accurate, and the FR as a whole is still going to be subject to room modes, which boxsim doesnt calculate. All in all, im pretty satisfied with the predicted results.
An externally hosted image should be here but it was not working when we last tested it.
I know you've already started on this project, but for future ones...
http://www.diyaudio.com/forums/software-tools/158298-boxsim.html
I've been contemplating messing around with the al130 for a while, but when you're a teenager, such drivers are considered expensive.
Hope all goes well.
Chris
http://www.diyaudio.com/forums/software-tools/158298-boxsim.html
I've been contemplating messing around with the al130 for a while, but when you're a teenager, such drivers are considered expensive.
Hope all goes well.
Chris
lol thanks chris but if you look above your post you'll see that i HAVE used boxsim already.....and thanks for the informative post about its use, i also read that already, in fact i have it subscribed for reference.........lol
i used it merely to sim baffle effects and tweak some XO values to optimise things a bit, and i have had the program for ages, so the tutorial was helpful. Thanks
Last edited:
yea i guess the al130 is a tad expensive for a teenager. my first teenage project used Audax drivers, paper coned, also HDA 130mm, and cheap(about £15), they were good. they werent HIGH end in respect to detail and quality, but they worked real good, and impressed all but the most critical listeners. The HDA series were even better in detail and also cheap. They were even better when i got the XO right, in the end! these AL130's blow them all out of the water though, and at the price they should. i seriously doubt spending more then £200, (verses £60 for the audax); on drivers would make the same level of difference that i noticed when upgrading, so i respect the lower cost approach. just dont go TOO cheap!! lol
using the w130 with the same XO seemed to yield similar results as far as FR goes, with only minor adjustments, and i know it is considerably cheaper. however the ceramic dome visaton do(is it the Ke20?) slotted into the same XO and was even better regards flatness of FR, albeit with the sacrifice of increased beaming about 8k.
Last edited:
original LP curve from the vib130tl x-over, showing the hump i didnt like much. This would be the al cone breakup peak. could have designed a notch to remove this but i like the partial BSC provided by the original xover.
playing around with some component values didn't really improve things, so i added a coil and resistor to the network, strategically and managed, almost entirely by accident to almost perfectly nullify the peak!!! even though the LP attenuates the peak by some 20dB it still influence the HF response.
Comparing the next 2 screenshots, im happy to see a reasonable improvement in the flatness of the FR.
from this(the original):
to this(modified):
with the small mod the HF from 8-15k is much flatter, within +/-2dB across the entire bandwidth from around 50Hz. There is a 3dB rise around 17k-20k but this is due to the baffle effects and not too important IMO as i hear to about 17k anyway
all in all im quite pleased with the 'slight' improvement
An externally hosted image should be here but it was not working when we last tested it.
playing around with some component values didn't really improve things, so i added a coil and resistor to the network, strategically and managed, almost entirely by accident to almost perfectly nullify the peak!!! even though the LP attenuates the peak by some 20dB it still influence the HF response.
An externally hosted image should be here but it was not working when we last tested it.
Comparing the next 2 screenshots, im happy to see a reasonable improvement in the flatness of the FR.
from this(the original):
An externally hosted image should be here but it was not working when we last tested it.
to this(modified):
with the small mod the HF from 8-15k is much flatter, within +/-2dB across the entire bandwidth from around 50Hz. There is a 3dB rise around 17k-20k but this is due to the baffle effects and not too important IMO as i hear to about 17k anyway
all in all im quite pleased with the 'slight' improvement
Last edited:
now after some more minor(very minor) tweaks to the vibtl130 xover i have got a simmed FR which is within less than+/-1dB from 150Hz to 15kHz. This is simulated on the baffle size i am using and does not account for room modes, which i cannot sim. I will attempt to tune the listening room a little in order to minimise the room effect, as far as it is practicable.
some techincal 'itches' have cropped up, being the cost of the CNC bill and a posible redesign of the external appearance of the cabinet, with resoect to making an easier build for myself and saving me gioving all my cash to a CNC programer!!!
An externally hosted image should be here but it was not working when we last tested it.
some techincal 'itches' have cropped up, being the cost of the CNC bill and a posible redesign of the external appearance of the cabinet, with resoect to making an easier build for myself and saving me gioving all my cash to a CNC programer!!!
it was a great help thanks Chris, without it i wouldve had to resort to using babelfish to translate the german! Although there are still a number of options, which i THINK are useful, and i THINK i know what they MAY be ....I have a feeling the user can also set the baffle depth displacement of the drivers too, in order to play with the delays to optimise response...
lol I might have to translate the german with babelfish after all!!!
As for the al130 design, its around 83dB/W on average(not peak lol), so the flatness is a little at the expense of efficiency. When its finished and x-over os finalised ill post the 'tweaks' on here, for the day you get the money to buy the Al130's...in the meantime i might mess around with the w130s, and see what tweets or FR go well with them....i have a feeling, and truely no offence meant at all, im sure with a little tinkering you could get the FR even better than you have right now... and im sure well into the +/-2db or less territory. the w130 is a good lil driver, cant remember what FR your using but maybe thats the culprit as far as wiggly lines go!
Chris,
Do keep in mind, to use Geddes' wordsbecause he says it so much better than me, "on-axis FR is almost next to useless at telling how good a speaker is".
It is impressively flat, it is just a sim thou... what is the directivity index?
dave
I have to say i agree, but only to a point. on axis is how i will be listening to these speakers, or 30-0-30 wise. Whilst i respect everyone has an opinion, which is right for themselves, I am not a believer of using increased toe-in to achieve a wider sweet spot. Although i know that the sweet spot AND flat FR will be smaller with a speaker with a flat on-axis response, Im not all that concerned since I will be using these focussed on one spot, and i have found 'over toe-in' to muddle the stereo imagery of every speaker i have ever heard, including the family LS3/5a's.
Yes it is JUST a sim, but the baffle effects are modelled really quite well from what I have read, and the sim IS based on manufacturer data. Normally this would make me more than nervous, but i have had good experience with Visatons TSPs as they seem to care about collecting a reasonable data sample every couple of months, and the deviation i have measured is small. This gives me a little more confidence in the general consistancy of the build and materials, and hence the general consistancy of response and sound.
And RE the directivity index, im not sure what you mean entirely Dave....
As far as a numerical INDEX i have no idea, although i CAN post the FR graphs with 0-30-60-90 and if i recall correctly 180 degree OFF-axis plots too if you would like to see them. From memory i think the HF is fairly good at 30 and 60 rolling off fairly smoothly and certainly NOT omni, or alternatively i can sim a polar plot at any frequency using a slider, so maybe i should post a simmed polar of say 100, 500, 1k, 2k, 4k, 8k, 16k, and the off axis graphs?
As i said, room modes arent allowed for. If i had a magic wand then Boxsim would include room mode calcs AND be written in english, and in an IDEAL world, a universal program would allow the uploading of all major vendor FR and polar data in the same way Boxsim does.
It certainly is NOT flawless, but its as good a crossover design program as ive used, intuitive considering its german, and i only wish more manufacturers would do the same.
I chose the W130SC over the W130S and the AL130. It just looks better. The W170S also looks nice.
Last edited:
yup the w130sc looks easier to cross 1st order maybe as its smooth on HF roll off, problem remains then, to find a tweeter that goes low enough to be able to cross woofer low enough to minimise the beaming effects, and keep a similar polar profile over both drivers.
w170 looks fairly good too, with similar provisions as for the w130sc.
as for the AL130: well though it has a nasty break up peak, the range between 500 and 2k is pretty flat, and with a decent size choke it really is pancake flat.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.