Hi,
The drivers datasheet clearly shows the drivers response (in a box).
The simulator assumes the driver is resistive and Le causes a roll-off.
It also assumes a resistive driver has a flat response which it doesn't.
Most drivers clearly don't roll-off according to Re and Le, not much
to explain, the reasons are complicated but the evidence very plain.
Those reasons are not built into the simulator and "FR affected by Le"
an easy to implement but pretty useless function of the simulator.
rgds, sreten.
I can't believe sims lie that much.
My experience with a Peerless XXls 12" 830845
Le =3.6 mh The manufacturer shows a flat graph untill 400Hz and more
Simulators show a roll of as soon as 100hz or so.
Real life shows that sim is right and the manufacturer lies.
One more point.
Zaph|Audio - SB12.3 3-Way Tower
While the manufacturer shows a driver usable untill almost 1Khz , Zaph, after testing and because of an Le of 1.8mH chooses to crossover at 220Hz.
"I wouldn't use this woofer too high in frequency, and the highish inductance isn't going to let me anyway."
Scanspeak makes woofers with low Le and some others too.
My experience with a Peerless XXls 12" 830845
Le =3.6 mh The manufacturer shows a flat graph untill 400Hz and more
Simulators show a roll of as soon as 100hz or so.
Real life shows that sim is right and the manufacturer lies.
One more point.
Zaph|Audio - SB12.3 3-Way Tower
While the manufacturer shows a driver usable untill almost 1Khz , Zaph, after testing and because of an Le of 1.8mH chooses to crossover at 220Hz.
"I wouldn't use this woofer too high in frequency, and the highish inductance isn't going to let me anyway."
Scanspeak makes woofers with low Le and some others too.
Last edited:
http://www.madisound.com/store/manuals/12_SB34SWNRX.pdf
Hi,
The manufactures curve is worse than zaphs measurements,
and it certainly doesn't imply the driver is usable up to 1KHz.
Seas spec sheets are very accurate / consistent and can be trusted.
Simulators don't model all real drivers accurately, is the truth.
rgds, sreten.
Last edited:
In this sim I added 0,4 ohm series resistance total from passive filter, wiring and contact resistance.
It will raise the low end a bit so it works positive for this driver to have better subbas.
Hi,
True, but there is nothing misleading about the published specs.
rgds, sreten.
I looked at that Peerless, there is a serious disparity between
what you say and the published specifications, very unusual.
I am curious about what indicates, in the published SB frequency response, not to cross over above 220hz.
To me the high Le is responsible of this limitation.
An 1.95 mH inductance is nothing but a 6db/octave low pass that starts rolling off at 250Hz or so.
I think a low Le is desirable in any driver but a subwoofer.
For the peerless, i run it untill 250 hz (active) but with a serious equalisation.
And i am still looking for a 12" or even 10" woofer working easily untill 300Hz closed box 65 liters max.
The CA26RFX is a candidate.
I hesitate because Le.
To me the high Le is responsible of this limitation.
An 1.95 mH inductance is nothing but a 6db/octave low pass that starts rolling off at 250Hz or so.
I think a low Le is desirable in any driver but a subwoofer.
For the peerless, i run it untill 250 hz (active) but with a serious equalisation.
And i am still looking for a 12" or even 10" woofer working easily untill 300Hz closed box 65 liters max.
The CA26RFX is a candidate.
I hesitate because Le.
Last edited:
Hi,
I'd say the fact the response droops from around 200Hz
upwards its a lot easier to use this droop rather fight it.
Even so, I don't really understand Zaph's comment.
This clearly goes up to > 1KHz and is 3mH.
H1316-08 CA26RE4X
rgds, sreten.
Last edited:
to be honest that is correct and you are right
however i had a good feeling about my driver selection, so i will be creating these speakers
in my mind baised on the responce curves they will match up quite well.
however to be honest i was unaware of the crossover being such a big issue for a decent sound,
i was just assuming a simple 12db/octave at 500hz and 12/24db.octave at 3000hz would produce a good (but not perfect) result
and complex weeks of testing would produce a more perfect but never absolute, result
correct me if im wrong. speaker building is a quest for perfection, perfection cannot be achived. that is the perfection of imperfection, because it gives birth to the huge range of designs, and there individual carictorised "sounds" (coloring) i dont want perfect sound, it is unobtainable. i will be happy with good sound.
i will cross the crossover bridge when i come to it
all issues aside, this project will be a learning curve and much welcome experiance for me, and what ever result i get will be better than no result.
i welcome this challange
I started out building simple 2-way speakers. It was quite easy to get something decent by ear. Then after several 2-ways I decided to try three way. I had no measurement tools, and did everything from the books. A nice three-way tower with four 6" woofer, one 4½" mid and a tweeter. All units with well behaved frequency respons and not dead cheap products.
The sound was absolutly horrible, and with no measurements possible, I simply couldn't make them sound good. Buliding a 3-way speaker is 100 times more complex, than building af 2-way. If you don't intend to measure and make multiple tests, then go for at testet system.
thank you for the concern,
i will however be building these 3 way speakers, i havent spent £450 on drive units just so i can let them collect dust in a corner...
maybe you where abit unlucky and chose a bad match of drivers or something...
have been reading all around the forums, there is nothing i wont be capable of, and i still dont regret my choice of drivers... they look like they will work fine
though i will be honest, there are many more factors comming into play than i first antisipated, although a tad irritating i guess its a good thing because i will have to learn more. ..
i plan to trust my intuition to a point, so i am going to design and post the final cabinet design, including the volume for the mid. then will come the time to work on the crossover..... if the final crossover isnt to my liking... then will come the time for mesurment setups and tweeking, and by that time i will know more about the electroncs, and speakers in general.
plenty of rainy days in england to sit and read about and learn how to design crossovers
i love music and speakers combine many fields of interest. computers sound electronics oh and making things...
yes it will be a bad feeling if they turned out sounding bad. but i really cannot see that being the case..
i will however be building these 3 way speakers, i havent spent £450 on drive units just so i can let them collect dust in a corner...
maybe you where abit unlucky and chose a bad match of drivers or something...
have been reading all around the forums, there is nothing i wont be capable of, and i still dont regret my choice of drivers... they look like they will work fine
though i will be honest, there are many more factors comming into play than i first antisipated, although a tad irritating i guess its a good thing because i will have to learn more. ..
i plan to trust my intuition to a point, so i am going to design and post the final cabinet design, including the volume for the mid. then will come the time to work on the crossover..... if the final crossover isnt to my liking... then will come the time for mesurment setups and tweeking, and by that time i will know more about the electroncs, and speakers in general.
plenty of rainy days in england to sit and read about and learn how to design crossovers
i love music and speakers combine many fields of interest. computers sound electronics oh and making things...
yes it will be a bad feeling if they turned out sounding bad. but i really cannot see that being the case..
i still havent quite made sence of all that le stuff and the driver losing responce, i would hope they would provide a accurate plot so people can chose drivers with that in mind, i kinda lost the thread abit, but just incase theres some confusion...
remeber the driver in question, is this one. H1305-08 CA26RFX
download Holmimpulse - read this holmimpulse redspade
A mic can be made of a long alu tube and a Panasonic WM61A capsule.
A measurement microphone for taking acoustic measurements using the Panasonic WM-61A electret.
A mic can be made of a long alu tube and a Panasonic WM61A capsule.
A measurement microphone for taking acoustic measurements using the Panasonic WM-61A electret.
Nope, but simple textbook formulas, simply don't work in real world.
If you have acurate T/S parameters and use a good cross over simulation, and include the cabinet dimension, then you can get decent sound.
Also try playing around with Tolvan edge Home of the Edge. You simply can't predict frequency behaviour without simulation software and/or measurements.
The graph for at given woofer also don't show the the baffle loss from a narrow baffle.
All three drivers are aprox 90 dB. With midrange a bit below. But the woofer will need some Baffle step correction. Between 3-6 dB depending on your roomgain. You wil now have a rising frequency curve, and most people prefer the opposite. With simple textbook filters w/o BSC and other compensating network, you will end up with a bright sounding speaker.
Either you have to eq. the freq resp in the crossover or with an external dedicated eq.
Either you have to eq. the freq resp in the crossover or with an external dedicated eq.
not really wanting a rising responce...
with a woofer mounted on the upper part of a 80cm high by 35cm wide cabinet, rounded off at the corners...
low frequencys will not be effected by the baffle, wat what frequencys does the baffle begin to add gain, surely not by much at 500hz
on another note thanks for the information, i guess the mid will be getting some baffle gain, as the mid box will be about 20cm wide.. so as to ensure all drivers are the same distance in from the sides of there boxes (mainly cosmetic) before this project i had never heard of baffle step's
as i only know basics of how crossover work,
and have no experiance in designing one... i will post the final cabinet design, before working on the crossover, then i will factor in the baffle step compensation. untill i have time to check what thickness of wood is avalable, and finalise my cabinet dementions... i wont be focusing too much on the cross over...
i know its best to factor in all things at once, but i simply dont have the experiance, so i will be doing this in the stages mentioned above
Sorry if i brought some confusion.
I am myself a bit confused now about the Le importance.
This came with the Peerless driver i spoke about.
Anyway Sreten is certainly of better advice than i and his answers make me reconsider the problem.
In your case, there is nothing to worry about as in the worst case a crossover point at 300 hz would be very good also.
I use the same midrange driver as you and like it.
I wish you will enjoy your building with these nice drivers.
what frequencys does the baffle begin to add gain, surely not by much at 500hz
........
i guess the mid will be getting some baffle gain, as the mid box will be about 20cm wide.. so as to ensure all drivers are the same distance in from the sides of there boxes (mainly cosmetic) before this project i had never heard of baffle step's
The baffle dont really add gain, only horn moutning wil add gain. But center placement will add som fluctuations in the frequency response, so linear tweeter suddenly will look anything but linear. So for tweeter and midrange the optimal placement is off center (no distance to sides should be equal)
The baffle width is the key element, and also placement of woofer. The narrower the baffle the higher the frequency where baffle loss begins. Typically you will begin to have a loss around 300 Hz. Passively you can only compensate by losing some efficiency above 300 Hz. So your 90,5 dB woofer will now be a 87 dB woofer or even lower.
You can also compensate by making the baffle wider or place the woofer close to the floor. Any reflector (floor and walls) will give you a 3 dB gain. Tha's why a woofer in the corner will have modt output.
A fully compensated woofer will sound boomy in the corner, but a non compensated woofer will sound fine.
You could also add another woofer to compensate for the baffle loss. You now have e 3½ way system since both woofer plays lowest freq's but only one play up to the midrange.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.