Had some time to run the sims on MJK's double BLH worksheet with the MA CHR-70 driver, and as previously shown by Don Hills' AKABAK model, it does end up with quite a lumpy response, the problem seems to be the lack of HF attenuation in the horn, even fully stuffing the horn doesn't help enough.
Furthermore, it seems there's little loading on the driver by the horn.
The "compression chamber" needs redesigning so as to act as a proper Low pass filter for the higher frequencies, as well as the throat and mouth areas.
xrk971, would you mind remind me what the path lengths were (i worked on everything from 2 meters to 3 meters), and also if possible post your driver's T/S parameters (for the biggest horn)
I'll be modelling the FF125WK next.
On the bright side, this horn does provide very good gain on lower frequencies, more than 10db between 50hz and 150hz!!! a promising design for sure.
Time to get back to the worksheets.
Regards
Nick
Furthermore, it seems there's little loading on the driver by the horn.
The "compression chamber" needs redesigning so as to act as a proper Low pass filter for the higher frequencies, as well as the throat and mouth areas.
xrk971, would you mind remind me what the path lengths were (i worked on everything from 2 meters to 3 meters), and also if possible post your driver's T/S parameters (for the biggest horn)
I'll be modelling the FF125WK next.
On the bright side, this horn does provide very good gain on lower frequencies, more than 10db between 50hz and 150hz!!! a promising design for sure.
Time to get back to the worksheets.
Regards
Nick
Nick,
Thanks for running the sims, I think all your observations are correct, yet the cornu has a real special sound which must be the result of the mid bass emanating from the edges to fill the room and make it sound like a bigger speaker.
The standard cornu at 70 cm edge dimension will have a 2.2 m long horn and a 1.6 m long shorter segment. I think that playing with the compression chamber size and reducing the throat to allow horn loading and act as HF filter will help.
For my 20 in cornu, scale the lengths by (20/27.6) but I sized my internal channel throat and depth to match Sd (no compression). I have a Vifa Tc9fd - those TS params can be searched easily.
I ran some sims with HR and noticed that the first bass peak is lower relative to next 2 or 3 in low to mid bass region. Not sure what to do to balance out other than restricting horn mouth. Gain is huge though, I see up to 15 dB on mid bass. I think putting two horn lengths without bifurcation from driver chamber may work better and allow the peaks on one horn to fill out valleys on other horn.
Thanks for running the sims, I think all your observations are correct, yet the cornu has a real special sound which must be the result of the mid bass emanating from the edges to fill the room and make it sound like a bigger speaker.
The standard cornu at 70 cm edge dimension will have a 2.2 m long horn and a 1.6 m long shorter segment. I think that playing with the compression chamber size and reducing the throat to allow horn loading and act as HF filter will help.
For my 20 in cornu, scale the lengths by (20/27.6) but I sized my internal channel throat and depth to match Sd (no compression). I have a Vifa Tc9fd - those TS params can be searched easily.
I ran some sims with HR and noticed that the first bass peak is lower relative to next 2 or 3 in low to mid bass region. Not sure what to do to balance out other than restricting horn mouth. Gain is huge though, I see up to 15 dB on mid bass. I think putting two horn lengths without bifurcation from driver chamber may work better and allow the peaks on one horn to fill out valleys on other horn.
Looking at the W3-1053SC it appears to be about what we would look for to match the gain you just noted.
It appears to be flat to around 200hz with about a 10db drop between 50hz &150hz.
It also has a slightly rising response to 18k per the datasheet.
Unfortunately it has a QTS of .62 which seems a little high for this application though.
They are cheap at $15 to keep in line with the cost saving design intention.
The W3-1838 has always been a driver I'd like to work with but they are sooooo darned spendy for a 3".
It appears to be flat to around 200hz with about a 10db drop between 50hz &150hz.
It also has a slightly rising response to 18k per the datasheet.
Unfortunately it has a QTS of .62 which seems a little high for this application though.
They are cheap at $15 to keep in line with the cost saving design intention.
The W3-1838 has always been a driver I'd like to work with but they are sooooo darned spendy for a 3".
Last edited:
A very interesting driver indeed, although the ideal driver would have a rising response from 1khz to 20khz of about 10db.
xrk71, I'll look up the vifa specs and model it as soon as time permits.
Nick
P.S: the 10db gain btw is a solid 10 db over nominal driver efficiency all the way from 50 to 150hz, it's immense
Well I see y'all been right busy whilst I had to deal with year end stuff...
Nah, just kidding, that was a quick trip to the accountant and then onto the important stuff - speakers!
Long live the spiral horn! This time the 50cm.
I really do like the speed when you poke holes and do a rough hand trace. And you can see how many scorings I do for the looser corners, about twice that many for the inside near the chamber. She's all glued and waiting for tomorrow.
You can see I still seem to have some sheets leftover, hmmm... Mind you my wife gave that look when I told her what I was doing last night and today so I had to use the old - you're right honey, I'm gonna call up some buddies and go hang out at the strip bar instead. Suddenly she took a keen interest in speakers again. I love my honey
Nah, just kidding, that was a quick trip to the accountant and then onto the important stuff - speakers!
Long live the spiral horn! This time the 50cm.
I really do like the speed when you poke holes and do a rough hand trace. And you can see how many scorings I do for the looser corners, about twice that many for the inside near the chamber. She's all glued and waiting for tomorrow.
You can see I still seem to have some sheets leftover, hmmm... Mind you my wife gave that look when I told her what I was doing last night and today so I had to use the old - you're right honey, I'm gonna call up some buddies and go hang out at the strip bar instead. Suddenly she took a keen interest in speakers again. I love my honey
Attachments
Well guys, I hate to say it but I'm just not totally convinced yet. It's very possible something went wrong in my build, despite my best efforts. If this horn is supposed to produce an extra 10db over 50-150hz, I'm not hearing it! They sound thin compared to my DBRs. I've already tested and confirmed the DBRs as being pretty near to flat from 50-16k, so I don't think it's got anything to do with them being terribly bass-heavy. Suggestions? Maybe a larger driver?
I have always wondered what Jordan's would sound like in a cornu. Based on your channel depth and throat width I assume the Jordan's have Sd of 23 cm^2? That Sd may be a little small for driving a 50 cm as we may be seeing this effect on Chaz's build as he is using 30 cm square Sd on 24 inch cornu. The Vifa has 36.4 cm sq Sd. Which plan did you use - endia's curvy one?
Last edited:
If I measured correctly it's 27 cm^2 for the driver and who know how close I was on the actual throat openings. Tomorrow's honey-do-list looks clear until about noon so maybe I can give them a test.
Them there Jordan's have been gathering dust since the project below went south a couple years ago. Please note the Margarita machine in the background, even upside down it's huge when it comes to speaker inspiration.
Them there Jordan's have been gathering dust since the project below went south a couple years ago. Please note the Margarita machine in the background, even upside down it's huge when it comes to speaker inspiration.
Attachments
Some measurements.
Loudspeaker hung on the wall, 170 cm (5.6 ft) from the floor, 140 cm (4.6 ft) from the ceiling, microphone (not calibrated) at 1 m (3.3 ft), gate of about 6 ms.
The red graph should be MLS, green the sweep.
Frequency response is more irregular than that posted by Pano (here) but the same notches at 50 and 200 hz are presents.
I cannot exclude I did something wrong, I'm not an expert in measuring loudspeaker.
PS: happy new year
Loudspeaker hung on the wall, 170 cm (5.6 ft) from the floor, 140 cm (4.6 ft) from the ceiling, microphone (not calibrated) at 1 m (3.3 ft), gate of about 6 ms.
The red graph should be MLS, green the sweep.
Frequency response is more irregular than that posted by Pano (here) but the same notches at 50 and 200 hz are presents.
I cannot exclude I did something wrong, I'm not an expert in measuring loudspeaker.
PS: happy new year
Attachments
Yes, it shoudn't
I think it could be room reinforcement or environmental noise (or me doing something wrong).
PS: when I wrote my post it was a bit too late for me, in the graph posted by Pano there is no notch at 50 hz.
For those of you building the Cornu in foam, if you have an interest in building the FH3 in foam, I just posted build pics and first impressions on the foam core thread. It's a winner also.
http://www.diyaudio.com/forums/full-range/223313-foam-core-board-speaker-enclosures-104.html
http://www.diyaudio.com/forums/full-range/223313-foam-core-board-speaker-enclosures-104.html
Ah thanks! This is my little vintage setup which i love, the reel to reel sounds good and silence is silence...no cracks or noise. I must say I prefer my turntable but the Tape brings back good old memories.
The sound of the Cornu is still a little hollow but the longer I play the more the frs8 loosens up so I will wait with new stuffing.
I am curious at your horns, I will take a peak
The sound of the Cornu is still a little hollow but the longer I play the more the frs8 loosens up so I will wait with new stuffing.
I am curious at your horns, I will take a peak