So you can squeeze the 8in woofer into the waveguide to reduce CTC spacing? I would be using an RS225 vs Byrtt's SS beauty. Actually, maybe if you give me the basic 10F waveguide object I can blend it into mine in Solidworks. I may go to a dual 5.25in Aurum Cantus carbon fiber cone. That gets 90dB bass after baffle step. That will keep dia smaller for easier integration between woofer and tweeter.
Thanks, X. BTW, how come the low frequency reverberation on the jazz clip only appear in one of the drivers [B80?]??? That's not fair
The lows should be equal for all as the XO is at 350Hz. But the fact is only B80 get that low frequency. So I think the Harsch XO has contribution to this anomaly.
Last edited:
Round 4 Reveal
Thank you all who participated and voted. Special thanks to those who contributed drivers for testing. This thread would not have been possible without you. Biggest thanks goes to Byrtt, whom to date has supplied the majority of drivers tested and provided me the means to make all this possible.
I hope you all had fun doing this!
Cheers,
X
****
The time for the reveal is here... drum roll please.
My secret surprise entrant was the Dayton PS95-8: a $24 alloy framed, paper coned, inverted surround, full range driver with a phase plug. It is a bit bright on-axis, so I did a passive "EQ" by pointing it 30 deg away from the mic to balance the response. It has received some votes and I think in general the votes seem to indicate that these are ALL great drivers and you can make a great sounding speaker with them.
Here are the measured frequency responses (4 ms gated and 1/48th oct smoothing, 10dB shifts to allow viewing):
Corresponding Impulse responses
PS95-8 at 30 deg off axis:
TG9FD:
TC9FD:
10F/8424:
B80:
***
It seems that there was a clear preference for the B80, 10F/8424, and the PS95-8. There is a similarity between the B80 and PS95 construction: phase plug, paper cone, large cast vented baskets. But the PS95-8 is something we can all afford, and more importantly, FIND in the U.S. easily. The B80 is almost impossible to get in the U.S. Although not getting many votes, technically, I think the TG9FD has the flattest frequency response and I would not hesitate to use it. The fact that the $12 TC9FD even made it into a competition against a $100 ScanSpeak says volumes about how capable it is. Don't discount it because it did not get a lot of votes. Finally, I hope everyone got to experience the sound of these drivers virtually and that it was useful.
Thank you all who participated and voted. Special thanks to those who contributed drivers for testing. This thread would not have been possible without you. Biggest thanks goes to Byrtt, whom to date has supplied the majority of drivers tested and provided me the means to make all this possible.
I hope you all had fun doing this!
Cheers,
X
****
The time for the reveal is here... drum roll please.
Code:
Round 4 Driver Key
xrk971
14:01GMT
Sept 15, 2015
A = Dayton PS95-8 at 30 deg off axis
B = Vifa TG9FD10-08
D = Vifa TC9FD18-08
E = ScanSpeak 10F/8424G00
F = Visaton B80-8ohmMy secret surprise entrant was the Dayton PS95-8: a $24 alloy framed, paper coned, inverted surround, full range driver with a phase plug. It is a bit bright on-axis, so I did a passive "EQ" by pointing it 30 deg away from the mic to balance the response. It has received some votes and I think in general the votes seem to indicate that these are ALL great drivers and you can make a great sounding speaker with them.
Here are the measured frequency responses (4 ms gated and 1/48th oct smoothing, 10dB shifts to allow viewing):
Corresponding Impulse responses
PS95-8 at 30 deg off axis:
TG9FD:
TC9FD:
10F/8424:
B80:
***
It seems that there was a clear preference for the B80, 10F/8424, and the PS95-8. There is a similarity between the B80 and PS95 construction: phase plug, paper cone, large cast vented baskets. But the PS95-8 is something we can all afford, and more importantly, FIND in the U.S. easily. The B80 is almost impossible to get in the U.S. Although not getting many votes, technically, I think the TG9FD has the flattest frequency response and I would not hesitate to use it. The fact that the $12 TC9FD even made it into a competition against a $100 ScanSpeak says volumes about how capable it is. Don't discount it because it did not get a lot of votes. Finally, I hope everyone got to experience the sound of these drivers virtually and that it was useful.
Attachments
Last edited:
@Wesayso:
When X sent me back my alpairs and a couple of waveguide adapter plates I had drawn, the charges at the post office were a rather unpleasant surprise. It would be cheaper to work out a design in a group effort and then print it locally. Only after X had shipped the package (which I am still waiting for), I discovered 3D hubs:
https://www.3dhubs.com/
It turns out that even my small village has a couple of geeks willing to print for a reasonable fee.
@X: thanks for the disclosure. I felt that the TC9 and the TG9 were very close soundwise, and that the 10F was just a hair better. The B80 and the PS95 were clearly different, where the PS95's somewhat brighter presentation sounded particularly good to me with the rock clip. I also liked the B80. But a rock clip with a somewhat stingier tonal balance might work out very badly with the PS95, and tip the scale towards the TC9/TG9 as cost effective options and the 10F as an upgrade option. So I was not entirely sure of my vote, and I am still not. I have TC9's and TG9's (both 4 and 8 Ohms) and I can easily listen to all of them.
When X sent me back my alpairs and a couple of waveguide adapter plates I had drawn, the charges at the post office were a rather unpleasant surprise. It would be cheaper to work out a design in a group effort and then print it locally. Only after X had shipped the package (which I am still waiting for), I discovered 3D hubs:
https://www.3dhubs.com/
It turns out that even my small village has a couple of geeks willing to print for a reasonable fee.
@X: thanks for the disclosure. I felt that the TC9 and the TG9 were very close soundwise, and that the 10F was just a hair better. The B80 and the PS95 were clearly different, where the PS95's somewhat brighter presentation sounded particularly good to me with the rock clip. I also liked the B80. But a rock clip with a somewhat stingier tonal balance might work out very badly with the PS95, and tip the scale towards the TC9/TG9 as cost effective options and the 10F as an upgrade option. So I was not entirely sure of my vote, and I am still not. I have TC9's and TG9's (both 4 and 8 Ohms) and I can easily listen to all of them.
Clip 2. There are many LF reverberation information in clip 2. Easily heard with 10 inches woofer. This extra low frequency makes B80 a better driver. I had to focus on mid high so not to fall into this "trap".
How about Solen, do they ship to U.S. ?
http://solen.ca/products/speakers/home-speakers/fullranges/b80/
Brilliant comparison. Thanks X! And thanks for selecting an interesting and affordable mystery driver. I am going to buy it and compare in person. I liked it least and the cheap TC9 best but all of the drivers in this round are top notch.
I'm off work today waiting to pick up my two teenage boys. In the meantime Jazz is streaming thru TC9s, the coffee is freshly ground and I'm smiling ear to ear.
I'm off work today waiting to pick up my two teenage boys. In the meantime Jazz is streaming thru TC9s, the coffee is freshly ground and I'm smiling ear to ear.
Good point about cost of shipping. If under 4lbs can do it more reasonably as first class mail. But I agree that a local 3d place is great if available.
So I did prefer the paper cones after all. I chose the B80 as number one and the TC9 as second! I didn't like the PS95 sound and now I know why. On the Barracuda clip I didn't feel the upper octave boost of the B80 as unpleasant. Though it didn't seem to reveal the most details. I liked F (B80), D (TC9) and E (10F) best of these, can't figure out why I had less love for the TG9.
I did only use the Barracuda track to judge, maybe I should have given the other tracks more time.
Thanks for a fun round X!
There are a couple of online services as well that I have used. Lot's of materials these days but more expensive. We could start a thread on this though, coming up with a viable shallow guide for the 10F, which would also fit the TG9 and TC9.
I'd love to pair it with the big one, BYRTT... 8" is so much more fun!
I did only use the Barracuda track to judge, maybe I should have given the other tracks more time.
Thanks for a fun round X!
There are a couple of online services as well that I have used. Lot's of materials these days but more expensive. We could start a thread on this though, coming up with a viable shallow guide for the 10F, which would also fit the TG9 and TC9.
I'd love to pair it with the big one, BYRTT... 8" is so much more fun!
xrk971,
Thanks for fun interesting round to squeeze data from.
Below summary from all rounds it seems baffle/woofer-system from round 2 verse round 4 change how its F10 or TG9FD that win flattest frq response on axis, at round 2 10F was micro flatter than TG9FD but reversed here at round 4 .
Round 4:
Round 3:
Round 2:
Round 1:
Thanks for fun interesting round to squeeze data from
.Below summary from all rounds it seems baffle/woofer-system from round 2 verse round 4 change how its F10 or TG9FD that win flattest frq response on axis, at round 2 10F was micro flatter than TG9FD but reversed here at round 4 .
Round 4:
Round 3:
Round 2:
Round 1:
Those backorder 8" 22W/8534 is original plan for FAST system where those 18W/8434 is plan to be close range monitor at desktop computer.
Attachments
Last edited:
Byrtt,
Thanks for the summary. The change from the trapezoidal baffle to the rectangular baffle used in Round 4 probably increased ripple due to diffraction effects. Diffraction should theoretically be less with trapezoid and that seems to pan out. The Nautaloss spiral sealed TL is also 3 feet long vs 12in long Dagger so there is more room and more volume for the drivers to "breathe" with the earlier rounds. Perhaps I need to go back to Nautaloss and trapezoid but keep woofer sealed with RS225 like in Round 3? That configuration is indeed much more flexible as it can accommodate 5in drivers.
Thanks for the summary. The change from the trapezoidal baffle to the rectangular baffle used in Round 4 probably increased ripple due to diffraction effects. Diffraction should theoretically be less with trapezoid and that seems to pan out. The Nautaloss spiral sealed TL is also 3 feet long vs 12in long Dagger so there is more room and more volume for the drivers to "breathe" with the earlier rounds. Perhaps I need to go back to Nautaloss and trapezoid but keep woofer sealed with RS225 like in Round 3? That configuration is indeed much more flexible as it can accommodate 5in drivers.
I picked the B80 by a whisker. The 10F was right on par, and for that matter so was the PS95. Apparently this was about what everyone else thought too.
I liked the TG9FD the least.
Now for the critique: I thought all the drivers sounded pretty bad. They were quite congested and the sound was not very resolving or realistic. Presumably this was due to the cumulative effects of the playback electronics, the recording mic and electronics, and the mp3 compression. I suspect the latter was a large chunk of the problem, but compression would be tough to avoid for this sort of internet test. There's also the issues of cabinet cry and room reverberation. At least these things are equal for all the drivers.
When everything is sort of fuzzy, the most obvious difference is the frequency response, and it doesn't surprise me that ratings have sort of tracked FR in earlier tests. However, this is also the most easily corrected issue. I am mostly concerned about the ultimate potential of a driver. High distortion cannot be corrected but FR can.
The other issue is that sometimes a cheap and cheerful fuzzy driver can be flattering with a less than perfect source, whereas a super clean hifi driver will make it sound grating and horrible. IMO, the best fullranges, even the most resolving ones, balance these qualities somewhat.
That said, clearly this test still was useful (and fun). We certainly learned something, as people were able to consistently pick out the arguably better drivers. Also, it's better to do something than nothing, despite imperfection, which is unavoidable anyway...
So kudos to X for conducting it, and kudos for those who lent the drivers.
I liked the TG9FD the least.
Now for the critique: I thought all the drivers sounded pretty bad. They were quite congested and the sound was not very resolving or realistic. Presumably this was due to the cumulative effects of the playback electronics, the recording mic and electronics, and the mp3 compression. I suspect the latter was a large chunk of the problem, but compression would be tough to avoid for this sort of internet test. There's also the issues of cabinet cry and room reverberation. At least these things are equal for all the drivers.
When everything is sort of fuzzy, the most obvious difference is the frequency response, and it doesn't surprise me that ratings have sort of tracked FR in earlier tests. However, this is also the most easily corrected issue. I am mostly concerned about the ultimate potential of a driver. High distortion cannot be corrected but FR can.
The other issue is that sometimes a cheap and cheerful fuzzy driver can be flattering with a less than perfect source, whereas a super clean hifi driver will make it sound grating and horrible. IMO, the best fullranges, even the most resolving ones, balance these qualities somewhat.
That said, clearly this test still was useful (and fun). We certainly learned something, as people were able to consistently pick out the arguably better drivers. Also, it's better to do something than nothing, despite imperfection, which is unavoidable anyway...
So kudos to X for conducting it, and kudos for those who lent the drivers.
Last edited:
I wouldn't mind that if it cures the upper tilt that wasn't there in round 2 but did throw me off a bit in round 4. Like Jay I was searching for those lovely Nora Jones vocal sounds and I know I like Ann Wilson. But somehow that wasn't present in this round. Closest for me were the D and E drivers somehow but they did have too much on top. Seeing the frequency response explains it a bit.