I am interested in getting a good hi-if setup. I am looking for a very realistic soundstage with a rather flat frequency response. The music I would be listening to would be centred around piano solo and orchestral work.
Looking into this, I found out about open baffle speakers and they seem like this would be the best route to go. The room that I would be using these in is quite large with many diffusers on the walls and ceiling.
However, I want to build my own and as this is my first attempt at hifi diy (normally do musical instrument diy), I am at a bit of a loss of where to start.
What are the most popular/respected diy designs, and what differentiates them? I have read about the LX521.4, but I have absolutely no idea why it needs such strange type of amplification? I ideally wanted to use a tube amp, like a leak stereo 20 or similar.
When looking into designs, I found that to get good bass response, a high Q woofer is needed, but this can be quite undefined compared to a low Q woofer.
I am after a very detailed and defined sound with convincing 3D soundstage.
I am obviously not adept enough to design my own, so I am generally asking for some guidance on which already proven diy designs are out there, and what books/resources would be good for me to read to get a better understanding of this area.
22uf
Looking into this, I found out about open baffle speakers and they seem like this would be the best route to go. The room that I would be using these in is quite large with many diffusers on the walls and ceiling.
However, I want to build my own and as this is my first attempt at hifi diy (normally do musical instrument diy), I am at a bit of a loss of where to start.
What are the most popular/respected diy designs, and what differentiates them? I have read about the LX521.4, but I have absolutely no idea why it needs such strange type of amplification? I ideally wanted to use a tube amp, like a leak stereo 20 or similar.
When looking into designs, I found that to get good bass response, a high Q woofer is needed, but this can be quite undefined compared to a low Q woofer.
I am after a very detailed and defined sound with convincing 3D soundstage.
I am obviously not adept enough to design my own, so I am generally asking for some guidance on which already proven diy designs are out there, and what books/resources would be good for me to read to get a better understanding of this area.
22uf
Last edited:
Loads of good info on here - The Nao Note RS II is a good example and one that is not too expensive. However, the design is now no longer active so not sure where you would get the minidsp files for it from.
NaO Note II RS Details
Applying passive filters to an open baffle can be tricky which is why many use a DSP crossover solution with an amp per pair of drivers. For the NAO Note you could use an 8 channel AV amp, something like the Rotel RMB 1048 but this in probably under powererd for the bass driver so a 6 channel AV amp plus a higher power stereo amp for the bass (ideally with the same gain factor). Or just buy a load of cheap chip amps around the 50-100w range.
NaO Note II RS Details
Applying passive filters to an open baffle can be tricky which is why many use a DSP crossover solution with an amp per pair of drivers. For the NAO Note you could use an 8 channel AV amp, something like the Rotel RMB 1048 but this in probably under powererd for the bass driver so a 6 channel AV amp plus a higher power stereo amp for the bass (ideally with the same gain factor). Or just buy a load of cheap chip amps around the 50-100w range.
Having built and owned several of the speakers mentioned in posts 2 & 3, I can say they do very good 3D sound - if your room permits it.
If open baffle, how big can you go? Bigger generally means more efficient thus smaller amps needed. Open baffle isn't very efficient unless huge. For a big space and tube power, you're probably looking at two 15" or 18" woofers per side.
If open baffle, how big can you go? Bigger generally means more efficient thus smaller amps needed. Open baffle isn't very efficient unless huge. For a big space and tube power, you're probably looking at two 15" or 18" woofers per side.
In my experience, OBs don't really manage the "convincing soundstage" bit. Too much acoustic excitement of the listening room = mushy image. Others disagree, and I suspect psychological differences between listeners are to blame. My "mushy" might be another person's "spacious" etc etc.
I find horns to be better, and have settled on an 8" 2-way design with a 1" compression driver which covers 950Hz upwards.
FWIW, the LX521.4 requires extensive amplification because active crossovers and EQ are an important part of the design. A passive crossover for those speakers would be immensely difficult to pull off, and I wouldn't attempt it unless I had a big amplifier on-hand in order to make up for the inevitable component losses.
If you're near Sheffield, give me a shout when the virus is more under control.
Chris
I find horns to be better, and have settled on an 8" 2-way design with a 1" compression driver which covers 950Hz upwards.
FWIW, the LX521.4 requires extensive amplification because active crossovers and EQ are an important part of the design. A passive crossover for those speakers would be immensely difficult to pull off, and I wouldn't attempt it unless I had a big amplifier on-hand in order to make up for the inevitable component losses.
If you're near Sheffield, give me a shout when the virus is more under control.
Chris
There are 32 videos in this Tuesday Tech Talk series by Danny Richie. Those on open baffle speakers are a fairly decent source of info, given the caveat of course that he is trying to sell something.
Tuesday Tech Talk Episode 1: Soundstage - YouTube
Here's the GR Research link, although it might get too expensive for shipping to your neck of the woods?
Speaker kits, loudspeaker design services, sub-woofers and amplifiers. - NX Series open baffle designs
Tuesday Tech Talk Episode 1: Soundstage - YouTube
Here's the GR Research link, although it might get too expensive for shipping to your neck of the woods?
Speaker kits, loudspeaker design services, sub-woofers and amplifiers. - NX Series open baffle designs
My OB's cast a very nice soundstage. However I suspect the choice of driver - the one or ones that carry most of the music - has a lot to do with it. You didnt mention if you're after an OB using a full range speaker ("WAW"), or a 2-way / 3-way driver set mounted on just a baffle.
It's possible to first choose a driver, then design your system around it and its specifics. In today's world climate, unsure how you'd get to actually audition different drivers. You'll have to trust opinions and impressions written by others.
"WAW" generally means a low-ish crossover point, which I believe effects "convincing 3D soundstage" significantly - in the positive direction. That's just IMHO and something I'd assert for even closed cabinet designs.
One thing you left out of your requirements is how loud, in db SPL, you'd like your system to play - given you prefer to use the Leak stereo 20. Given that - plus the large room you have, would help with suggestions.
It's possible to first choose a driver, then design your system around it and its specifics. In today's world climate, unsure how you'd get to actually audition different drivers. You'll have to trust opinions and impressions written by others.
"WAW" generally means a low-ish crossover point, which I believe effects "convincing 3D soundstage" significantly - in the positive direction. That's just IMHO and something I'd assert for even closed cabinet designs.
One thing you left out of your requirements is how loud, in db SPL, you'd like your system to play - given you prefer to use the Leak stereo 20. Given that - plus the large room you have, would help with suggestions.
I'm a fan of OB speakers, but a single 20W amplifier in a large room seems a tall order.
I can only echo liunine's suggestion - a Manzanita or the Ultra version. Sourcing the bass driver may be problematic in GB.
Maybe you could consider an active-crossover OB, using the Leak in the midrange only?
I can only echo liunine's suggestion - a Manzanita or the Ultra version. Sourcing the bass driver may be problematic in GB.
Maybe you could consider an active-crossover OB, using the Leak in the midrange only?
The room always has a huge effect. With OB even more so. Although I do prefer horn speakers for the ultimate in realistic imaging, Open Baffle gets a long way there with much less effort. 😉
I've heard and built a number of OB that image better than anything short of large horns. In a large room with good diffusion like the OP speaks of, they can be very convincing.
This requires a dipole which many OB speakers are far from. Linkwitz Lab site explains pretty much all you need to know.
Have a look at this thread too In Pursuit of a 20-20k Dipole Loudspeaker
+1 on LX521
Assuming you have the space to get the speakers out away from the front/side walls, you might try just building yourself an LX521. There are CAD drawings available for the enclosure, or you can buy a flatpack.
There is a new hybrid version of it that reportedly comes very close to the .4 version. It requires only 4 channels of amplification total (two identical gain stereo 50 W amps, one per side). All versions described on the OPLUG forum at linkwitzlab.com
Disclosure - I built and own an LX521.4 setup (but with a slightly different open baffle bass approach) and I love it compared to all the many speakers I have tried in my years of this hobby.
Good luck!
gabkw
Assuming you have the space to get the speakers out away from the front/side walls, you might try just building yourself an LX521. There are CAD drawings available for the enclosure, or you can buy a flatpack.
There is a new hybrid version of it that reportedly comes very close to the .4 version. It requires only 4 channels of amplification total (two identical gain stereo 50 W amps, one per side). All versions described on the OPLUG forum at linkwitzlab.com
Disclosure - I built and own an LX521.4 setup (but with a slightly different open baffle bass approach) and I love it compared to all the many speakers I have tried in my years of this hobby.
Good luck!
gabkw
Last edited:
Don't know how loud it needs to go or how big the room is but with piano / orchestral stuff maybe Electrostatics with a decent sub(s) might fit the bill.
I would suspect that a 3 way would have better bass response? However, I am open to the idea of having a separate subwoofer for under 120hz. I am not after anything super loud, although I do doubt the leak 20’s ability to produce meaningful bass, but this could be dealt with be using a separate sub?
How would I approach designing a OB cabinet around selected drivers?
Yes going big would be fine, but I am worried about using the 15 and 18 inch woofers as I think they could be slow in stopping, unlike multiple 6 or 8 inch speakers to deal with the bass?
I also want to avoid a dsp as I want to retain a 100% analogue signal path.
Thanks for your suggestions so far, I am going to take a close look at all the links tomorrow and get back with any questions.
I built the Martin King OB's with H-frame 18" woofer. I'm a novice, but they work very well. The Jordan full rangers are no longer available, but I think other options have been used in the forum. They need to be placed about 5-feet from the back wall. Good luck. HeyBill
That can be a problem with high Q drivers, but you don't have to use small magnet/low QTS woofers. They just help with the low end. You'll have trouble getting the multiple 6s or 8s to play very low. Maybe if you use dedicated subwoofers with low Fs you could get down into the 30s. IME twin 18's is a better way to go. I've also used a single 21" on each side and the bass was very clean and surprisingly nimble.
My first successful OB speaker were my "Lounge Lizards" They used a Selenium 15" which is no linger made* with a lowish Fs and a Qts of 0.65 Very nice indeed and I was super happy with the bass. Two woofers like that on the baffle would be even better and need less voltage.
The trick with OB woofers is to cross them very low, far below the point you want to blend them to the midrange or tweeter. Because the woofer is going to have a distinct rising response on the open baffle, setting your crossover very low will flatten it out. That's not something you'd do with a normal box speaker, but it's necessary for OB.
What driver do you plan to use above the woofers?
* It looks like PRV are now making a very similar driver called the 15W700. I don't know if you can get Brazilian woofers in the UK, but it's worth a look.
The trick with OB woofers is to cross them very low, far below the point you want to blend them to the midrange or tweeter. Because the woofer is going to have a distinct rising response on the open baffle, setting your crossover very low will flatten it out. That's not something you'd do with a normal box speaker, but it's necessary for OB.
What driver do you plan to use above the woofers?
* It looks like PRV are now making a very similar driver called the 15W700. I don't know if you can get Brazilian woofers in the UK, but it's worth a look.
Attachments
Larger drivers typically have much more powerful motors than smaller drivers, which means they can "start/stop" just fine. It's a myth that large woofers are "slow". Unfortunately, that myth is perpetuated even at the highest levels of HiFi.
In fact, once you consider that a 15" driver might have twice the Xmax of an 8" driver, you'll find you'd need 8x 8" drivers to match 1x 15" driver in terms of volume displacement. Matching cone area alone is insufficient.
Chris
Well, let's find out. I'll probably write this up properly at some point.
I set up a simple Hornresp simulation, of a Fostex FE166En in a ported box. I those that driver because it simulates reasonably flat across a wide frequency range. I suppose I could've applied some EQ, but never mind.
Here's the frequency response. I went for a 15L ported box tuned around 70Hz.
It looks okay for free-field use, but might be a bit boomy in-room.
Here's the excursion curve. Typical ported box.
Here's the velocity curve, which is where things get a little more interesting.
From that, you can read off the maximum velocity of the cone (which occurs as it passes through the rest position), and find that it's moving at up to 0.5m/s, which is pretty slow. It's required to move fastest at 30Hz and 110Hz-ish.
However, the velocity curve is of limited actual importance. What's much more interesting is the acceleration curve. Why? Because of Newton's second law of motion: F=ma. The acceleration of a body (in this case, the cone) is proportional to the sum of the forces acting on it, and inversely proportional to the mass of the body (cone).
Now, this is much more interesting. Even where the driver's simulated HF response is falling off (above 1kHz), the acceleration is still rising.
In order to maintain SPL, much more acceleration is needed towards the high-frequency region. Since the acceleration of a body is proportional to the force applied, we can conclude that more force is needed at HF than LF.
This explains a few things:
- Why tweeters require light diaphragms (reduces force required to undergo high accelerations)
- Why heavy-coned woofers typically have minimal output above a few hundred Hz, unless they have a really serious motor which can force the cone to accelerate that quickly.
The whole "big woofer = slow" thing seems to be perpetuated by those that sell HiFi speakers with 6" drivers. In fact, if you match the SPLs, a 15" cone will be undergoing smaller accelerations than the 6" cone, while producing the same level. The 15" cone would also likely have a much larger motor, which would likely offset the increased cone mass.
Further, would you rather listen to a 15" cone moving 2mm p/p, or a 6" cone moving 12.5mm p/p?
The smaller driver will almost certainly be exhibiting higher distortion levels, and will certainly have nowhere left to go if you want to push things a little louder.
The 15" would be happily loafing along, and wondering when you're going to drop some real power in there.
In conclusion, there's no replacement for displacement, and "speed" only matters at the top of the bandwidth.
Chris
I set up a simple Hornresp simulation, of a Fostex FE166En in a ported box. I those that driver because it simulates reasonably flat across a wide frequency range. I suppose I could've applied some EQ, but never mind.
Here's the frequency response. I went for a 15L ported box tuned around 70Hz.
It looks okay for free-field use, but might be a bit boomy in-room.
Here's the excursion curve. Typical ported box.
Here's the velocity curve, which is where things get a little more interesting.
From that, you can read off the maximum velocity of the cone (which occurs as it passes through the rest position), and find that it's moving at up to 0.5m/s, which is pretty slow. It's required to move fastest at 30Hz and 110Hz-ish.
However, the velocity curve is of limited actual importance. What's much more interesting is the acceleration curve. Why? Because of Newton's second law of motion: F=ma. The acceleration of a body (in this case, the cone) is proportional to the sum of the forces acting on it, and inversely proportional to the mass of the body (cone).
Now, this is much more interesting. Even where the driver's simulated HF response is falling off (above 1kHz), the acceleration is still rising.
In order to maintain SPL, much more acceleration is needed towards the high-frequency region. Since the acceleration of a body is proportional to the force applied, we can conclude that more force is needed at HF than LF.
This explains a few things:
- Why tweeters require light diaphragms (reduces force required to undergo high accelerations)
- Why heavy-coned woofers typically have minimal output above a few hundred Hz, unless they have a really serious motor which can force the cone to accelerate that quickly.
The whole "big woofer = slow" thing seems to be perpetuated by those that sell HiFi speakers with 6" drivers. In fact, if you match the SPLs, a 15" cone will be undergoing smaller accelerations than the 6" cone, while producing the same level. The 15" cone would also likely have a much larger motor, which would likely offset the increased cone mass.
Further, would you rather listen to a 15" cone moving 2mm p/p, or a 6" cone moving 12.5mm p/p?
The smaller driver will almost certainly be exhibiting higher distortion levels, and will certainly have nowhere left to go if you want to push things a little louder.
The 15" would be happily loafing along, and wondering when you're going to drop some real power in there.
In conclusion, there's no replacement for displacement, and "speed" only matters at the top of the bandwidth.
Chris