AVS Forum banner

41 - 60 of 86 Posts

·
Registered
Joined
·
8,456 Posts
Eh, depends on your filters and hardware you're using. I'm not experiencing this issue, but I'm using 16k taps per channel with 1024 delay. That said, if you can match phase without issues at your xover points, there isn't a huge benefit to going FIR. I still want to test this comparing minimum phase to linear phase at some point, but I still haven't found the time.



What are you using to do all your processing? That is a lot of taps!
 

·
Registered
Joined
·
10,479 Posts
What are you using to do all your processing? That is a lot of taps!
A very tiny fairly powerful pc running brutefir with usb audio interfaces ;)

The audio interfaces being an RME UFX+ with a MOTU 8a connected to it via adat for extra io. The UFX+ is pricey, but it works with linux in class compat mode very nicely. They came out with a cheaper version shortly after, the UFX II which is pretty much the same thing with some features like Dante removed. I think the MOTU might work as well in linux, but it did not play nice with my hypex nc400s.

That impulse response was from in room measurements btw, not electrical.
 

·
Registered
Joined
·
4,340 Posts
with a ribbon like the beyma, lets say it was crossed over at around 1200hz, is it then possible to cross it over with a midrange driver that can get us down to about 60-80hz?

What about the 10000hz+ range? Is the natural rolloff good or is it better to have yet another tweeter for that range?

I have limited understanding of speaker design/philosophy and its hard for me to decide how much time as a hobby to spend on figuring it all out. I know I will enjoy the process, but not sure if it is all "worth it" for one pair of speakers. If I go that route to build a pair, I really would like them to function really well :).

I feel like a simple project with perhaps a single midrange driver and a single ribbon might be a good speaker design to try. I do have a table saw now but I don't have anything to cut circles well into wood. I can make a crappy jigsawcut and surface mount the speaker I suppose.

A passive crossover might make a small speaker a lot more versatile to use than to have a multichannel amp for each?

I see comparisons for the RAAL vs BEYMA tweeters. besides the higher price on RAAL, what is your impressions between these two and pro/cons from a speaker design perspective.
 

·
Registered
Joined
·
11,479 Posts
Without the filtering the raw components look like this, windowed and averaged around the listening position

The large room, controlled directivity speakers, and the treated reflection points go a long way into having a nice response at the listening position, though.
Those are almost flawless FR charts, you must be happy with the sound at this point.
 

·
Registered
Joined
·
794 Posts
I'd say 1200 is the low end for the tpl150h. It loses pattern control rapidly below there.

IMO Build a 2-way with 2453H-SL CD on STX825 waveguide. It blows away any of the CD/WG combos in the DIYSG line. It can pair well with 12s or 15s. See link in my sig for 15" 2-way build. An MTM would be ideal to help match vertical directivity better if you can manage the space/budget.
Your builds are awesome I’ve checked them out :). I’m not sure I want a 15” woofer because I’m still ‘negotiating’ where in the house I can set up a two channel space. If I do it somewhere not fully dedicated, boxes that fit 15” woofers gonna be tough. Even 12” is pushing it. Also, I really want to hear what a good tweeter/amt can do just for my own education, so am torn (I’m also curious if I can build something for 1-2k per pair that beats 30k ‘hifi two channel’ speakers). Even if that means building a 3 way so I can cross the tweeter higher up. I wouldn’t mind a nice 3 way tower for two channel listening to be honest but also want a 2 way for smaller spaces.
 

·
Registered
Joined
·
1,053 Posts
I know this might be a long shot but I do see that lots of people on here have used these Beyma TPL-150's. Does anyone have any measurement files to share with ? I'm "thinking" about buying a pair and building a pair of speakers with them. Id be using the Satori WO24P-8 9.5" woofer in a 2 way..
.
I think you mentioned that you already got the Satori which looks fine but I thought I'd chime in about the TD10M from AE. I use it with a Seos 10 /Radian beryllium compression driver on a smaller system. I ran it 70hz up to 1400hz. The TD10m is 97db as a 4 ohm and doesn't have the odd top end. I have the apollo version but without it, the TD10M is around $349.

The TPL150, TD10M has been used already with a huge amount of data available on the build:

Arcus from vapor audio
 

·
Registered
Joined
·
783 Posts
Discussion Starter #47
I think you mentioned that you already got the Satori which looks fine but I thought I'd chime in about the TD10M from AE. I use it with a Seos 10 /Radian beryllium compression driver on a smaller system. I ran it 70hz up to 1400hz. The TD10m is 97db as a 4 ohm and doesn't have the odd top end. I have the apollo version but without it, the TD10M is around $349.

The TPL150, TD10M has been used already with a huge amount of data available on the build:

Arcus from vapor audio
Yeah i was looking at that, the white pictures is what got me interested in thinking about this project. Most of said on here that it would not be that good of a system.. I don't have any drivers bought yet. I have a budget of 2000$ CND to build a pair of speakers.. FOR 2 channel stereo listening..

Currently listing to a pair of my latest speakers. Sbacoustics 5" with a Peerless Tweeter with waveguide. I think Jav mentioned that the directivity on the 150 was really narrow. I one thing i hated about Fullrange speakers if you move your head slightly when listening to them you loose the "image / Hi-Frequency. They are very beamy and IMO lame.. :(
 

Attachments

·
Registered
Joined
·
794 Posts
Yeah i was looking at that, the white pictures is what got me interested in thinking about this project. Most of said on here that it would not be that good of a system.. I don't have any drivers bought yet. I have a budget of 2000$ CND to build a pair of speakers.. FOR 2 channel stereo listening..

Currently listing to a pair of my latest speakers. Sbacoustics 5" with a Peerless Tweeter with waveguide. I think Jav mentioned that the directivity on the 150 was really narrow. I one thing i hated about Fullrange speakers if you move your head slightly when listening to them you loose the "image / Hi-Frequency. They are very beamy and IMO lame.. :(
It’s ‘narrow’ in terms of vertical dispersion but I think what Javs was saying was that at reasonable distance from the speaker you easily fall within the directivity/dispersion window unless you stand up.
 

·
Registered
Joined
·
783 Posts
Discussion Starter #49
It’s ‘narrow’ in terms of vertical dispersion but I think what Javs was saying was that at reasonable distance from the speaker you easily fall within the directivity/dispersion window unless you stand up.
My bad maybe i miss understood him. If i were to build these i wouldn't care about what it sounded like standing up. Id be building them to be listened to sitting down. The stands would be build to perfect height for ear level ( or in that range for directive listening )
 

Attachments

·
Registered
Joined
·
794 Posts
My bad maybe i miss understood him. If i were to build these i wouldn't care about what it sounded like standing up. Id be building them to be listened to sitting down. The stands would be build to perfect height for ear level ( or in that range for directive listening )
Agreed. I would too if I went with that tweeter. I think Javs is extremely happy with his choice though he uses these speakers for home theater rather than 2 channel. But I, like you, am considering options for 2 channel. I’ve never really had 3-ways in my house before, though I’ve auditioned lots - so I’m tempted to figure out a 3-way design just for experience sake. But I do like the compact nature of 2-ways so I think I ultimately want a 2-way design as well. I’ve been toying with just building the Maximus 12 but I really want to experiment with ribbons/amts since I’ve never had them in my house. That led me to the tpl150h.

I think I may start with the 3-way. The question is, given I’m new to all this, do I go with something like the tpl150h or start with something less expensive but with good reviews like the Fountek neox3 or arum Cantus g1 or something. Big range in price. But I’ve always been one to just go big or go home, even if I’m just messing around experimenting..

In your case, if you’re set on a 2-way, it seems like the tpl150h will limit your woofer choices given how it needs to be crossed over. So if you’re set on those woofers you found, sounds like perhaps time to look at other tweeters.
 

·
Registered
Joined
·
8,456 Posts
I definitely use mine for two way music all the time. For me it's really important that the system does that too.

My previous towers were great for music also, but these new ones are better..

And yes with the dispersion I was referring to vertical, but I don't listen to content standing up so it's completely not an issue in a half decent treated room IMO.

The horizontal dispersion is pretty good. It drops off rather quickly from 10-20k if you move out of the sweet spot, but I'm sure a lot of drivers do that.

I will post the polar plots shortly which show all the off axis measurements I did up to about 75 degrees off axis. Which is rather extreme.
 

·
Registered
Joined
·
794 Posts
I definitely use mine for two way music all the time. For me it's really important that the system does that too
Ok cool. I hadn’t recalled anything from your build threads about music so I wasn’t sure. Glad to hear they are great for that application too! I love the modular concept.
 

·
Registered
Joined
·
8,456 Posts
Ok cool. I hadn’t recalled anything from your build threads about music so I wasn’t sure. Glad to hear they are great for that application too! I love the modular concept.
I exclusively used music to test out the design as I was going. Its also why I struggled with them at first, I had to get the toe in right because initially for music I was too close to them to get a great soundstage due to them simply being so damned large, they were too big I couldn't toe them properly. I was sitting only 2.8m from them at the time. Now I sit 3.2m and they are toed in properly and it's awesome sound stage.
 

·
Registered
Joined
·
8,456 Posts
So here are some off axis and polar measurements I did a very long time ago. I really should do them again since I am better at measuring now, and I didnt do these outdoors, but it should give some idea. I do think the proper measurements would be alsightly better now, as in the off axis wont actually bloom like it seems to here on a couple of the measurements.

Anyway. Above 10k is where it quickly suffers. The rest seems to decline in a pretty predictable manner.



Normalised:





Not sure which of these you guys prefer to normalise to, so I did a bunch of them.

Polar Normalised to 0deg.



7.5Deg



15 Deg



22.5 Deg

 

·
Registered
Joined
·
783 Posts
Discussion Starter #55
I've been drooling over these big bad boys.

http://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Calpamos.pdf

Sensitivity • 93 dB / 2,83 volts
Impedance • 8 ohms nominal
Frequency response • 37 - 20.000 Hz (-3dB) / 33 - 22.000Hz (-6dB)
Crossover frequency • 650Hz / 12dB (2nd order parallel crossover)
Power rating • circa 200 watts RMS
Dimensions (W x H x D) • 486 x 900 x 600mm (without stand)
Weight (finished product) • circa 75kg each


 

Attachments

·
Registered
Joined
·
794 Posts
Looks nice. Much different than fusion 15? Looks like it extends a little lower at the expense of sensitivity. Probably better for two channel I guess but my gut says the top end on the f15 sounds smoother just based on my experience with the HTM12s

I've been drooling over these big bad boys.

http://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Calpamos.pdf

Sensitivity • 93 dB / 2,83 volts
Impedance • 8 ohms nominal
Frequency response • 37 - 20.000 Hz (-3dB) / 33 - 22.000Hz (-6dB)
Crossover frequency • 650Hz / 12dB (2nd order parallel crossover)
Power rating • circa 200 watts RMS
Dimensions (W x H x D) • 486 x 900 x 600mm (without stand)
Weight (finished product) • circa 75kg each


 

·
Registered
Joined
·
783 Posts
Discussion Starter #57
Looks nice. Much different than fusion 15? Looks like it extends a little lower at the expense of sensitivity. Probably better for two channel I guess but my gut says the top end on the f15 sounds smoother just based on my experience with the HTM12s
Yep.!

I'm doing some googling to see if i can make something like this but using a 12" driver and smaller CD & WG
 

·
Registered
Joined
·
794 Posts
Yep.!

I'm doing some googling to see if i can make something like this but using a 12" driver and smaller CD & WG
HTM12?? I have three in my theater. Love them. Haven’t used them for any serious 2 channel listening but when I have listened to music it sounded great.
 

·
Registered
Joined
·
8,456 Posts
Not sure what you mean, they're fully active speakers, so all crossover, shaping, and room correction is done with FIR filtering...

Without the filtering the raw components look like this, windowed and averaged around the listening position






What you quoted is the supposed output after building the filters. It measures about that in room though, as you can see in the spatially averaged graph I posted above. I just posted those so you could see the crossover curves.

The large room, controlled directivity speakers, and the treated reflection points go a long way into having a nice response at the listening position, though.

T30


Impulse ETC


(and before ceiling treatments, you can see how that reflection, though minor, shows at 3 and 4 ms)

Yellow is the close mic measurement (2ft) of only the HF section with 1/20 octave frequency dependent window. Red is a vector average of 4 measurements taken with the mic placed in front of the couch pointed directly at the speaker with 1/12 octave fdw. Green is a vector average of 4 measurements taken over the couch, mic pointed straight up essentially in positions your head would be, with the same windowing. Calibration files specific to the mic orientation were used. Distance at the listening position is about 4 meters (13-14ft). On the yellow nearfield trace, you can see the rolloff from the xover at 800hz on the left side and the attenuation (or lack of) of the HF frequencies over distance on the right side. Otherwise, they overlay nicely.

Hey, I found your data really interesting, Just from a pure interest point of view, I had a go at completely redoing my Active calibration yesterday this time using the raw driver measurements average from 6pts at the listening position as the initial PEQ, rather than using outdoor measurements. I am not really sure which is ultimately better to do regarding active. I thought I might be able to come up with a cleaner final response. Its pretty close to the old response at the end of the day really, not sure which I prefer. Any reason why you choose to do in room measurements for the raw drivers rather than outdoor, then separately calibrating in room?

Here is the raw drivers.







And then PEQ on the raw measurements, and then measured as one:



Here is the final response after PEQ on the above measurement for the other PEQ banks.



And other stuff. I am actually not super sure how to interpret these ones. It looks like the room could use better treatments though, I know that much, and the room itself is not symmetrical especially regarding speaker placement (I use half of a long rectangular room), so I am excited to be able to put all this stuff in a good properly built from the ground up room.





For the impulse, yours looks incredible, but what's vector average?

This is my normal one:



Then Min Phase (is that even relevant)



How is yours so flat after the initial impulse? That looks anechoic? Is that actually in room?
 

·
Registered
Joined
·
10,479 Posts
Hey, I found your data really interesting, Just from a pure interest point of view, I had a go at completely redoing my Active calibration yesterday this time using the raw driver measurements average from 6pts at the listening position as the initial PEQ, rather than using outdoor measurements. I am not really sure which is ultimately better to do regarding active. I thought I might be able to come up with a cleaner final response. Its pretty close to the old response at the end of the day really, not sure which I prefer. Any reason why you choose to do in room measurements for the raw drivers rather than outdoor, then separately calibrating in room?

Here is the raw drivers.







And then PEQ on the raw measurements, and then measured as one:



Here is the final response after PEQ on the above measurement for the other PEQ banks.



And other stuff. I am actually not super sure how to interpret these ones. It looks like the room could use better treatments though, I know that much, and the room itself is not symmetrical especially regarding speaker placement (I use half of a long rectangular room), so I am excited to be able to put all this stuff in a good properly built from the ground up room.





For the impulse, yours looks incredible, but what's vector average?

This is my normal one:



Then Min Phase (is that even relevant)



How is yours so flat after the initial impulse? That looks anechoic? Is that actually in room?
Final response looks great. I just figured if I was going to be correcting in room, might as well do it all at once. Gotta be careful on what not to correct. Some stuff will look OK in an average, but if you have all the spatial stuff displayed, you can see there may be some areas that aren't too pretty.

Here's an example of something that I over corrected around 400hz and possibly other places


And then after fixing... The small dip is preferable to a peak which may stand out more. You can see some notches in the target response in the earlier posts showing spots I let be.


Vector average (found in controls under all measurements) will average magnitude and phase. That's why my impulse looks cleaner, as when averaging multiple locations, it removes some of the room effects from the measurements.

Then once you're nicely eq'd and time aligned, your step will look like this:



Your T30 is a bit low if anything (ideal supposedly .25 - .35), your room would not likely benefit much from treatments. There are no big reflections in your ETC either. I see one on the right channel at around 5ms in, but not too bad. This is likely due to the closeness of the listening position to the speakers compared to the boundaries.
 
41 - 60 of 86 Posts
Top