I built some desktop OBs in the past, and this is to revisit the design using newly learnt stuff, mainly around polar response. One of the most promising of the previous built was this one. While it sounded ok at that time now I realise that running the full range down to lower frequency would result in not so smooth off-axis response.
In fact it should be the other way around. The larger woofer should have taken the lower frequency up to dipole peak.
So the idea here is:
- Seas L21RNXP from low freq - 800Hz, then TangBand FR for 800hz-20khz
- By crossing the alumunium cone L21 low enough the metal resonance may be eliminated completely.
- Still use thin baffle to obtain uniform response up to 2.5kHz
Some foreseen "problems" and compromises
- Excursion limit of 8" woofer. Can it provide satisfaction?
- Distance to the rear wall is 40-50cm max. Will it smear the sound?
The FR is TangBand W3-1364S. It's pretty and measures well. It has great opening at the back because of the small neo magnet.
Seas L21RNXP. From previous project I learnt to deal with the metal resonance, but crossed at 1.5kHz it does not dissapear completely.
In fact it should be the other way around. The larger woofer should have taken the lower frequency up to dipole peak.
So the idea here is:
- Seas L21RNXP from low freq - 800Hz, then TangBand FR for 800hz-20khz
- By crossing the alumunium cone L21 low enough the metal resonance may be eliminated completely.
- Still use thin baffle to obtain uniform response up to 2.5kHz
Some foreseen "problems" and compromises
- Excursion limit of 8" woofer. Can it provide satisfaction?
- Distance to the rear wall is 40-50cm max. Will it smear the sound?
The FR is TangBand W3-1364S. It's pretty and measures well. It has great opening at the back because of the small neo magnet.
Seas L21RNXP. From previous project I learnt to deal with the metal resonance, but crossed at 1.5kHz it does not dissapear completely.
Last edited:
Baffle
Pine wood is my favourite baffle material. It's not as dusty as MDF or chipboard, cheap, and smells nice. It's important to coat the wood otherwise it would change to yellowish colour. I use Estapol clear Satin.
With OB the rear chamber needs to be as open as possible.
Pine wood is my favourite baffle material. It's not as dusty as MDF or chipboard, cheap, and smells nice. It's important to coat the wood otherwise it would change to yellowish colour. I use Estapol clear Satin.
With OB the rear chamber needs to be as open as possible.
The drivers installed and tested. The Seas works very.. very hard 🙂. Rattles everywhere on the baffle and my desk. Most have been eliminated but there's more.
Two pieces of foam underneath the speakers seems effective enough to decouple them from the desk.
Two pieces of foam underneath the speakers seems effective enough to decouple them from the desk.
Looking good😀
TB's small magnet doesn't block the rear wave very much, so you can probably get a dipole response almost all the way to the top.
TB's small magnet doesn't block the rear wave very much, so you can probably get a dipole response almost all the way to the top.
XO
True. My the transition to that tangband is so smooth which makes me question my previous builds. I guess having 3 to 4 octave below xo frequency really helps. This warrants further investigation.
I've dialed the XO a bit but using the same thin baffle there's not much. It's still 1.5kHz and I wonder what kind of changes will bring by lowering it to 800hz.
The Seas is not too happy played at high SPL from 20-35hz. I think a 40hz or 30hz (but -6db) is a more realistic target. But I don't like non-20hz speakers 😀
Le audiophile xo with pure copper jumpers
True. My the transition to that tangband is so smooth which makes me question my previous builds. I guess having 3 to 4 octave below xo frequency really helps. This warrants further investigation.
I've dialed the XO a bit but using the same thin baffle there's not much. It's still 1.5kHz and I wonder what kind of changes will bring by lowering it to 800hz.
The Seas is not too happy played at high SPL from 20-35hz. I think a 40hz or 30hz (but -6db) is a more realistic target. But I don't like non-20hz speakers 😀
Le audiophile xo with pure copper jumpers
😆
40Hz seems about right for an 8 incher. I've played with 8" widerangers on OB for a while. When playing (relatively) loud bass, cone movement was easily seen, maybe 5~6mm p-p. It still sounded OK even with such excursion.
For a desktop application, I think this would be the over the top hi-end speakers😀
gainphile your experiments are always so interesting! thank you for sharing so in the details
perhaps what about adding an h-frame woofer under the desk? like a 2.1 system (you have amplifiers enough to do that 🙂 ) would it work? or would it mess up everything due to phase?
hope you'll find some good 8" FR to test on the bigger baffles somedays, perhaps they could be a good match to the make with 3 woofers per side you did some time ago
perhaps what about adding an h-frame woofer under the desk? like a 2.1 system (you have amplifiers enough to do that 🙂 ) would it work? or would it mess up everything due to phase?
hope you'll find some good 8" FR to test on the bigger baffles somedays, perhaps they could be a good match to the make with 3 woofers per side you did some time ago
Finished!
Well that was a quick project which turns out nice.
The speakers are really small. I attached a pic with my notebook as scale. Very pleasing.
About the bass... the optimum compromise between distortion, listening level, and excursion is 40Hz. That's pretty deep, dipole or not. The woofers are working reaaaaallly hard. Some 30Hz can be felt but not respectable. Below 30Hz is just asking for trouble. These metal cone makes horrible sound when bottomed 😀. It's simply just physics and there is no way I could get more, except putting again a pair of 15" on my desk.
The problem down low is also related to materials. LP rips have those pesky subsonic rumble which drives the cone to excursion to maximum a lot quicker than the program material. A 2nd order highpass at 17Hz fixes that effectively at the cost of group delay. I could not hear it.
Aside from that the speakers makes really great bass, but you already know what dipole bass is all about. I think it's like a good espresso. Small in quantity but what comes out is pure. I have a Yamaha sub below the desk and while it can provide the extension to last octave, it degrades the quality of the whole system. I turn it off and perhaps will use it to watch movies.
The distance to rear wall is about 35cm. I did could not hear any faults. Comparing to my living room dipoles (which is 1m from rear wall), maybe a bit less soundstage, but that's really nitpicking. Phantom image is very scary, almost like the speakers making no sound and the music comes out of the wall. Listening to some classical recording I got goosebumps. If there is one best thing that stands out from them, it's the midrange.
The TangBand as tweeter, well this is new to me. It actually makes great tweeter and I can't hear lack of high. Even better, I think the transition from w-t is smoother than normal dome tweeter that I have been using. This warrants another investigation, perhaps using the TangBand as my livingroom tweeters. But I listen nearfield with these (1m - 2m) so maybe that's why it's ok. I did not lower the xo point as my listening so far can't hear any faults with 1.5kHz.
Thanks all for reading/comments. Now I'm going to listen to some music. 😎
Well that was a quick project which turns out nice.
The speakers are really small. I attached a pic with my notebook as scale. Very pleasing.
About the bass... the optimum compromise between distortion, listening level, and excursion is 40Hz. That's pretty deep, dipole or not. The woofers are working reaaaaallly hard. Some 30Hz can be felt but not respectable. Below 30Hz is just asking for trouble. These metal cone makes horrible sound when bottomed 😀. It's simply just physics and there is no way I could get more, except putting again a pair of 15" on my desk.
The problem down low is also related to materials. LP rips have those pesky subsonic rumble which drives the cone to excursion to maximum a lot quicker than the program material. A 2nd order highpass at 17Hz fixes that effectively at the cost of group delay. I could not hear it.
Aside from that the speakers makes really great bass, but you already know what dipole bass is all about. I think it's like a good espresso. Small in quantity but what comes out is pure. I have a Yamaha sub below the desk and while it can provide the extension to last octave, it degrades the quality of the whole system. I turn it off and perhaps will use it to watch movies.
The distance to rear wall is about 35cm. I did could not hear any faults. Comparing to my living room dipoles (which is 1m from rear wall), maybe a bit less soundstage, but that's really nitpicking. Phantom image is very scary, almost like the speakers making no sound and the music comes out of the wall. Listening to some classical recording I got goosebumps. If there is one best thing that stands out from them, it's the midrange.
The TangBand as tweeter, well this is new to me. It actually makes great tweeter and I can't hear lack of high. Even better, I think the transition from w-t is smoother than normal dome tweeter that I have been using. This warrants another investigation, perhaps using the TangBand as my livingroom tweeters. But I listen nearfield with these (1m - 2m) so maybe that's why it's ok. I did not lower the xo point as my listening so far can't hear any faults with 1.5kHz.
Thanks all for reading/comments. Now I'm going to listen to some music. 😎
Last edited:
Wonderful ! 😀
Like I said, for a desktop system (or, for PC sound), this is just over the top! You're using active xover and 4 channels of amps, right? And you called that 'really small' ! 😆
Like I said, for a desktop system (or, for PC sound), this is just over the top! You're using active xover and 4 channels of amps, right? And you called that 'really small' ! 😆
Seas replacement
Very nice setup!
The Seas L21RNXP is no longer available. Any suggestions for an alternate driver?
Very nice setup!
The Seas L21RNXP is no longer available. Any suggestions for an alternate driver?
SEAS L22RNXP is the replacement of L21. Basically look for woofers with good response and distortion as usual, but with OB you want low Fs and high xMax.
I acquired another pair of L21 previously so I will see how an MTM would do.
I acquired another pair of L21 previously so I will see how an MTM would do.
Hi gainphile,
maybe you remember my question about the W3 in your blog? Since then I made a pair of prototypes that play around half an hour now (so they are far from being "wobbled in").
As my miniDSP hasn't arrived yet, I play music over foobar's crossover plugin which works quite well. The amplifier is a modified Yamaha DSP-E1000 that isn't "DSP" anymore 😀 (a little modification I did). It is a capable and yet affordable amplifier.
I didn't ask you about the measures of the baffle you used but I think I came pretty close with some simulations in "Edge".
I wanted to thank you for the inspiration. Without your blog I wouldn't have tried an open baffle. It sounds real good now...and I hope it will be even better with the miniDSP and some measuring 🙂.
The current look of my desktop open baffle prototypes is attached.
Many thanks
Sven
maybe you remember my question about the W3 in your blog? Since then I made a pair of prototypes that play around half an hour now (so they are far from being "wobbled in").
As my miniDSP hasn't arrived yet, I play music over foobar's crossover plugin which works quite well. The amplifier is a modified Yamaha DSP-E1000 that isn't "DSP" anymore 😀 (a little modification I did). It is a capable and yet affordable amplifier.
I didn't ask you about the measures of the baffle you used but I think I came pretty close with some simulations in "Edge".
I wanted to thank you for the inspiration. Without your blog I wouldn't have tried an open baffle. It sounds real good now...and I hope it will be even better with the miniDSP and some measuring 🙂.
The current look of my desktop open baffle prototypes is attached.
Many thanks
Sven
Attachments
Looks great! ... I'm planning to redo the baffles on my desktop OB with proper wood actually. Will document and let you know the settings.
Do you not hear any issues with the right speakers so close to the wall? They need some space to breathe.
Do you not hear any issues with the right speakers so close to the wall? They need some space to breathe.
Actually I have limited space on my desk. I'm planning to move the printer to a different location as it takes about half of the available space on the desk. Without it, proper positioning should be possible.
That doesn't fully answer your question, right? Okay, I'm hearing some small issues that need to be addressed 😉. But I have to measure the speakers before I change something.
They really sound quite good as they are. I guess that is what you can call a "promising concept" 😀.
As soon as I have the whole system working (PC-miniDSP-Amp-Speaker) I'll start with the optimization. I'll keep you updated.
That doesn't fully answer your question, right? Okay, I'm hearing some small issues that need to be addressed 😉. But I have to measure the speakers before I change something.
They really sound quite good as they are. I guess that is what you can call a "promising concept" 😀.
As soon as I have the whole system working (PC-miniDSP-Amp-Speaker) I'll start with the optimization. I'll keep you updated.
Should be similar. Except that the Fostek would have higher sensitivity so you'd need to adjust accordingly.
In any case the xo should be kept between 900hz to 1.5khz to maintain best dispersion given by the 8".
But check out this ebay find. Can you imagine mating this with Seas W22 !! 😎
Aucharm 4 inch Fullrange Speaker
eBay Australia: Buy new & used fashion, electronics & home d?r
In any case the xo should be kept between 900hz to 1.5khz to maintain best dispersion given by the 8".
But check out this ebay find. Can you imagine mating this with Seas W22 !! 😎
Aucharm 4 inch Fullrange Speaker
eBay Australia: Buy new & used fashion, electronics & home d?r
Finally...
after some weird results I got more or less acceptable ones 🙂.
The attached picture shows the frequency response at hearing position with both speakers playing. I added a plot with an applied nearfield response (below 200 Hz).
It is not perfect yet but I keep trying. In fact it sounds better than before as I did wire the TB and the SEAS false on right channel 😱. Now it is correct and it makes me 😀.
Greets Sven
after some weird results I got more or less acceptable ones 🙂.
The attached picture shows the frequency response at hearing position with both speakers playing. I added a plot with an applied nearfield response (below 200 Hz).
It is not perfect yet but I keep trying. In fact it sounds better than before as I did wire the TB and the SEAS false on right channel 😱. Now it is correct and it makes me 😀.
Greets Sven
Attachments
I'm now 'migrating' from ASP to DSP. Hope to post settings soon and compare notes.
I observed the audibility of Seas metal cone breakup. At the previous xo of 1.5khz they are audible, even with notches. At 1khz they are not audible whatsoever, even with no notch.
I took the opportunity to reinforce the baffle using rear 3.5cm ribs. It makes them far more rigid.
I observed the audibility of Seas metal cone breakup. At the previous xo of 1.5khz they are audible, even with notches. At 1khz they are not audible whatsoever, even with no notch.
I took the opportunity to reinforce the baffle using rear 3.5cm ribs. It makes them far more rigid.
Last edited: