OK, lets look at the enclosure first.
I will first try to explain some of my choices.
Wide baffle - push baffle step down to 2-300hz, and place the xo-point there. with an active xo and biamping this would hopefully make the BSC easier.
Wide baffle - from 2-300hz and up I'll get reflections from the baffle itself. This makes the placement of the speaker easier. The distance from the back wall will make little difference in mid and tweet.
Mid enclosure will be a part of the enclosures bracing, and also makes a room for the passive xo. The wall between the mid enclosure and the xo box, will have an angle to reduce standing waves.
The bracing will hold the bassdrivers magnet, and make the assembly of the curved front easier.
I will first try to explain some of my choices.
Wide baffle - push baffle step down to 2-300hz, and place the xo-point there. with an active xo and biamping this would hopefully make the BSC easier.
Wide baffle - from 2-300hz and up I'll get reflections from the baffle itself. This makes the placement of the speaker easier. The distance from the back wall will make little difference in mid and tweet.
Mid enclosure will be a part of the enclosures bracing, and also makes a room for the passive xo. The wall between the mid enclosure and the xo box, will have an angle to reduce standing waves.
The bracing will hold the bassdrivers magnet, and make the assembly of the curved front easier.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Hi Norcad,
Looking good! I like the magnet support for the woofer. Is that a 3" diameter port? if both ends are flared should be large enough (otherwise I know a 4" port will be too long for the cabinet - even with elbows)
Have you had a chance to measure inbox response? Any braces being off center is a good idea.
Cheers,
David
Looking good! I like the magnet support for the woofer. Is that a 3" diameter port? if both ends are flared should be large enough (otherwise I know a 4" port will be too long for the cabinet - even with elbows)
Have you had a chance to measure inbox response? Any braces being off center is a good idea.
Cheers,
David
Yes its a 3" port and flared in just one end. But with low Fb this is not a problem when listening to music.
And yes I have a lot of measurements!
This is a learning project for me and therefore I'am measuring and testing a lot of things, just for the learning process.
Heres the inbox freq resp of the drivers and port measured nearfield.
Red=27tdfc
Blue=ca18rnx
Black=ca26rnx
Green=port
And yes I have a lot of measurements!
This is a learning project for me and therefore I'am measuring and testing a lot of things, just for the learning process.
Heres the inbox freq resp of the drivers and port measured nearfield.
Red=27tdfc
Blue=ca18rnx
Black=ca26rnx
Green=port
An externally hosted image should be here but it was not working when we last tested it.
XO
I started up with a 2.order LR, and to get the phase right I had to add a resistor to the mid xo, that make it a 1.order with a zobel actually. The result was a 2.order roll off for the mid and 3.order for the tweet, with a rather good reverse null. I testet several xo point with this topology from 1,8k to 3khz but the sound wasnt to good.
I always ended up with a sharpness that I (after long time) located to be around 4khz. And a "fat" midtone with the highest peak around 700hz.
The last one was a combo of the ca 18' freq resp, resonate from the backplate of the mid enclosure and some standing waves from the same. Bad luck!
I'm thinking of redesign the midenclosure, but hasnt tested it yet.
The sound became much better when I redesigned the xo, to get 2.order rolloff on mid, and 3.order on tweet. I also added a notch filter at 700hz.
And this is the result, measured onaxis at 1m, with 3ms gating:
I started up with a 2.order LR, and to get the phase right I had to add a resistor to the mid xo, that make it a 1.order with a zobel actually. The result was a 2.order roll off for the mid and 3.order for the tweet, with a rather good reverse null. I testet several xo point with this topology from 1,8k to 3khz but the sound wasnt to good.
I always ended up with a sharpness that I (after long time) located to be around 4khz. And a "fat" midtone with the highest peak around 700hz.
The last one was a combo of the ca 18' freq resp, resonate from the backplate of the mid enclosure and some standing waves from the same. Bad luck!
I'm thinking of redesign the midenclosure, but hasnt tested it yet.
The sound became much better when I redesigned the xo, to get 2.order rolloff on mid, and 3.order on tweet. I also added a notch filter at 700hz.
And this is the result, measured onaxis at 1m, with 3ms gating:
An externally hosted image should be here but it was not working when we last tested it.
You are on the way.
Don't forget to take at least some horizontal off-axis measurements (15 and 30 degrees as a minimum) with each iteration.
Also have a look at the voicing preferences of people like Zaph / Jay in their FR curves. I know this can vary depending on distortion profiles and other factors.
The bump around 1800Hz (in relation to the rest of the spectrum) may be audible (?) Probably too late, but maybe some baffle placement or moving mid/tweeter xo points could settle it a bit.
Cheers,
David.
Don't forget to take at least some horizontal off-axis measurements (15 and 30 degrees as a minimum) with each iteration.
Also have a look at the voicing preferences of people like Zaph / Jay in their FR curves. I know this can vary depending on distortion profiles and other factors.
The bump around 1800Hz (in relation to the rest of the spectrum) may be audible (?) Probably too late, but maybe some baffle placement or moving mid/tweeter xo points could settle it a bit.
Cheers,
David.
After a lot of measuring, listening and tweaking of the xo, I have found my voicing preference, at least in my listening room.
I agree with mr Gravesen that a small tilt in the freq resp sounds better than flat. 1-3dB down from 200 to10k. But from around 8-10k I want a peaking curve, to compensate for my old and tired ears!
And of course the power response also must be considered.
So here is the best sounding fr curve I have come up with till now.
As you can see, I have added another notch filter to the tweet.
Red= tweet and mid at 1m, 3ms gating
Pink = tweet alone
Thin dark red = simulated 3.order@2500hz
It may not look perfect, but now its voiced to my room, and both power response and CMD is a little better this way.
Theres still some issues I wish to get better.
The midtone enclosure as mentioned before.
And I lack some in the midbass/upper bass. But I havent started to check this out yet. Maybe its my room that sucks out this freq?
But I know that the mid roll off a little higher than expected. I assume that it is a combo of Fb (100hz - 5L closed) and Baffle step?
Maybe both problems would be solved with a bigger volume for the mid.
I agree with mr Gravesen that a small tilt in the freq resp sounds better than flat. 1-3dB down from 200 to10k. But from around 8-10k I want a peaking curve, to compensate for my old and tired ears!
And of course the power response also must be considered.
So here is the best sounding fr curve I have come up with till now.
As you can see, I have added another notch filter to the tweet.
Red= tweet and mid at 1m, 3ms gating
Pink = tweet alone
Thin dark red = simulated 3.order@2500hz
An externally hosted image should be here but it was not working when we last tested it.
It may not look perfect, but now its voiced to my room, and both power response and CMD is a little better this way.
Theres still some issues I wish to get better.
The midtone enclosure as mentioned before.
And I lack some in the midbass/upper bass. But I havent started to check this out yet. Maybe its my room that sucks out this freq?
But I know that the mid roll off a little higher than expected. I assume that it is a combo of Fb (100hz - 5L closed) and Baffle step?
Maybe both problems would be solved with a bigger volume for the mid.
how interesting can this get!!!
i was searching for a interaction between the drivers mentioned (exactly the same!!!) and kabammm... this appeared! i think i can consider myself lucky.
the thing is, i would like to build an TMW loudspeaker, TM in an open-baffle structure and W in a transmission line enclosure (despite the low Qts), either closed or open (so far i built only opened ones, so i'd like to try the closed type... but an half wave resonator is just too humongous!!
).
i selected the tweeter and the woofer by the pleasure i had with them, in different loudspeakers (meaning that i never heard them together). i also want the loudspeaker to go as low as possible, so a very low Fs (in this case, 20Hz) associated with the stiffness of aluminium are just in my mind.
the thing is that both my father and i have some experience in wood (and plaster) works, so the bo wouldnt be the problem, but crossovers are always the messy part.
i'm using xover pro, but the database is not complete, and it requires some knowledge that im willing to learn, but for what i have not time now (studying for the final university exam, 1 year from now).
therefore, i'd like to ask you if you could share your crossover scheme (for TMW) or any sugestions leading me to a successful project!
thank you very much, and congratulations for your project!
i was searching for a interaction between the drivers mentioned (exactly the same!!!) and kabammm... this appeared! i think i can consider myself lucky.
the thing is, i would like to build an TMW loudspeaker, TM in an open-baffle structure and W in a transmission line enclosure (despite the low Qts), either closed or open (so far i built only opened ones, so i'd like to try the closed type... but an half wave resonator is just too humongous!!
).i selected the tweeter and the woofer by the pleasure i had with them, in different loudspeakers (meaning that i never heard them together). i also want the loudspeaker to go as low as possible, so a very low Fs (in this case, 20Hz) associated with the stiffness of aluminium are just in my mind.
the thing is that both my father and i have some experience in wood (and plaster) works, so the bo wouldnt be the problem, but crossovers are always the messy part.
i'm using xover pro, but the database is not complete, and it requires some knowledge that im willing to learn, but for what i have not time now (studying for the final university exam, 1 year from now).
therefore, i'd like to ask you if you could share your crossover scheme (for TMW) or any sugestions leading me to a successful project!
thank you very much, and congratulations for your project!
hello Norcad
i am very interested to learn how your final speakers sounded
i currently have the same bass, and midrange you used in your design, however i have the 27tdc tweeter
i was planning to put the c826rfx into a vented enclosure of 80.5liters, tuned to 24hz
it is a nice coincidance for me to have come across a design using the same woofer and midrange, and i have learnt alot from the previous posts
all i can say about the tweeter, is it is pretty much the same as yours but plays abit lower
if you happen to have any more infomation that would be useful to me, (especially regarding your crossover...) i would be very gratefull
my email address is teasugermilk@gmail.com
i am very interested to learn how your final speakers sounded
i currently have the same bass, and midrange you used in your design, however i have the 27tdc tweeter
i was planning to put the c826rfx into a vented enclosure of 80.5liters, tuned to 24hz
it is a nice coincidance for me to have come across a design using the same woofer and midrange, and i have learnt alot from the previous posts
all i can say about the tweeter, is it is pretty much the same as yours but plays abit lower
if you happen to have any more infomation that would be useful to me, (especially regarding your crossover...
) i would be very gratefull my email address is teasugermilk@gmail.com
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.