I see your point but we model and simulate a lot of things at work so why not speakers? I don't expect models or simulations to represent reality 100% (we don't need two realities, that's not what M&S are for) but I do expect them to lead the way. You know, a little bit more of this or a little bit less of that. It also helps me understand how things relate on a systems level. I actually enjoy it and would find randomly cutting wood loss of time, and materialI also agree that such theoretical fine tuning is complete loss of time. Whatever we might simulate, will always end with a result like this. So why box bother...
Last edited:
I see your point but we model and simulate a lot of things at work so why not speakers? I don't expect models or simulations to represent reality 100% (we don't need two realities, that's not what M&S are for) but I do expect them to lead the way. You know, a little bit more of this or a little bit less of that. It also helps me understand how things relate on a systems level. I actually enjoy it and would find randomly cutting wood loss of time, and material
Sure, but i think that weird loads as TLs or BLHs, dipoles, even BRs are not really worth considering. Closed box + made to measure eq just can fit any need, and it is very hard to make the wrong closed box, even choosing its size randomly. Question is that you will never make the right one. Without the right eq all boxes are plain wrong.
Last edited:
I used Martin King's software, the only modeling software I use. By using Zaph's T/S measurements for this driver, I knew from its highish Qts and lots and lots of TL modeling experience, the tuning frequency would likely need to be below 35 Hz. I started arbitrarily with a 60" line length and a 10:1 taper, which was close, but tweaked the taper a bit, then I adjusted the depth of the line to get an f3 of 40 Hz or a bit lower (going even lower would not be a good idea for only a single driver). I also located the driver at 20% of the line's length from the line's beginning because that's often preferred (but I have used the 33% location). With you wanting to emulate the Silent Speaker and place the driver towards the middle of the line, the results are not likely going to be nearly as good. I could reduce the line's volume while keeping the same length and taper ratio, but that will raise f3, and if the line's length is reduced, its taper ratio will need to be increased in order to achieve the same tuning frequency. I seldom even bother using published specs from manufacturers since so often they aren't highly representative of what you actually get (some are, however). Also, I don't even bother looking at how the line performs without any stuffing, and I know from experience that I will virtually always get good results with a stuffing density of 0.75 lb/ft3 (but I will vary from that as needed).
Okay, I quickly converted my model to a line length of 55" with the driver located at halfway down the line, and also reduced the line's volume to just over 40 liters. I increased the taper ratio a bit to ~13:1, with a starting area of 7.5" x 11" and an ending area of 7.5" x 0.875". System tuning is 30 Hz and the first 38" of the line is stuffed at 0.75 lb/ft3. I've attached the predicted system bass response and I think you'll agree it ain't pretty (I didn't spend much time trying to make it better, but it really won't matter all that much due to the driver's location in the line).
Paul
Okay, I quickly converted my model to a line length of 55" with the driver located at halfway down the line, and also reduced the line's volume to just over 40 liters. I increased the taper ratio a bit to ~13:1, with a starting area of 7.5" x 11" and an ending area of 7.5" x 0.875". System tuning is 30 Hz and the first 38" of the line is stuffed at 0.75 lb/ft3. I've attached the predicted system bass response and I think you'll agree it ain't pretty (I didn't spend much time trying to make it better, but it really won't matter all that much due to the driver's location in the line).
Paul
Attachments
Last edited:
How can I get hold of this software? I can't thank you enough for running my figures through your software. It seems like there is quite a diff between it and Hornresp (and I lack the experience to transpose modeled values into reality).
How important is Qts for a TL (pros and cons for Qts > 0.4 and Qts < 0.4)?
Is a driver positioned in the middle of the line doomed then? I am just surprised that speakers like the Silent Speaker and Heed Enigma seems to work well (or maybe they could work even better if the driver was re-positioned)?
Many thanks Paul for taking the time, it has been enlightening.
Last edited:
You may be right but this thread is not about building the perfect closed box. It is about me trying to understand a speaker like the Silent Speaker and learning something new in the process. "Weird loads" might not be the most straight forward solution but I find them intriguing from an engineering standpoint. I would also like to learn as much as possible about them because I have always stayed "within the box" of "normal" closed and bass reflex boxes before.
Du kommer från Sverige, hur är det möjligt att du inte känner Carlssons Sonab högtalarnas saga? Carlsson Ortho-Acoustic Loudspeakers: Design and Performance Principles (English) - CarlssonPlanet
The Swede Stig Carlsson designed many ortho-acoustic speakers during 1950's to 70's.
The Silent Speaker is jus a poor example of these sort of omnidirectional near-wall placed speakers. I have th classic OA-14 at my summer cabin. The sound is pleasing and much better than conventional measurements would tell.
The Swede Stig Carlsson designed many ortho-acoustic speakers during 1950's to 70's.
The Silent Speaker is jus a poor example of these sort of omnidirectional near-wall placed speakers. I have th classic OA-14 at my summer cabin. The sound is pleasing and much better than conventional measurements would tell.
Don't worry, I know my Carlsson history well, what HiFi interested Swede who grew up during the 70th doesn't? Note that I mentioned the Sonab OA-5 in my first post in this thread. It is also worth noting that Mr Carlsson started out all otoacoustic but it seems to me like his speakers became more and more like conventional speakers at the end of his career.
So you have actually heard the Silent Speaker and is able to compare it to the OA-14? Omnis work well in my living room (the SoundSpans already proven it). I know the value of classic Carlsson and Sonab speakers, but they are not for everyone or any listening space. I am personally looking for something along the lines of omnis or otoacoustics and that is what led me to the Silent Speaker and Heed Enigma in the first place. But I am also interested in how they work and I am not sure they sound any better or worse than Mr Carlssons speaker designs. I have not had the pleasure to listen to the Silent Speaker but I think I could find a Heed Enigma here in Stockholm. I will not jump to conclusions until I actually have had a chance to compare them to lets say an OA-14...
Last edited:
No, I haven't heard SS. But since it has just one tweeter, I am doubtful. Lots of wrinkles in midrange and questionable step response because of that TL (or like) bass. On the other hand I have not measured my Sonabs.
SS obviously has however something special in how it works (but homepage is crazy pseudoscientific babble) Atkinson's measurements tell a lot. (It was measured in free air, not next to wall)
"I listened to the Silent Speaker after performing and analyzing the measurements, so my expectations were not very high. However, I was pleasantly surprised when I set them up in my room and drove them with a pair of MBL 9007 monoblock amplifiers and my Mac mini feeding a dCS Debussy DAC via asynchronous USB. Other than the mellow highs, the balance was considerably more neutral and uncolored than I'd expected."
Read more at http://www.stereophile.com/content/...t-speaker-ii-measurements#3MOkTrBVhosLIkoG.99
SS obviously has however something special in how it works (but homepage is crazy pseudoscientific babble) Atkinson's measurements tell a lot. (It was measured in free air, not next to wall)
"I listened to the Silent Speaker after performing and analyzing the measurements, so my expectations were not very high. However, I was pleasantly surprised when I set them up in my room and drove them with a pair of MBL 9007 monoblock amplifiers and my Mac mini feeding a dCS Debussy DAC via asynchronous USB. Other than the mellow highs, the balance was considerably more neutral and uncolored than I'd expected."
Read more at http://www.stereophile.com/content/...t-speaker-ii-measurements#3MOkTrBVhosLIkoG.99
Last edited:
Thanks for making me go back and read the measurements part of that article in detail. Loads of information and hints about the internal workings of the Silent Speaker...
Martin used to provide his software to the DIY community, first as a freebie, then later with a small fee ($25 one time), but over the years commercial entities were using it for their manufactured speakers and without paying their one-time fee of $250, so Martin stopped providing it. Martin has a Group/Forum on Yahoo and has allowed some people who join up to get his software, but I don't know if he still is doing that.
Regarding Qts, I prefer working with drivers having a Qts in the range of 0.35 to 0.50, but lower and higher values sometimes work out acceptably to quite well (the CA18RLY using Zaph's measurements is a good example).
With a driver located in the middle of the line, usually the only type of TL that will perform decently is an ML-TQWT, but since it has an expanding area, it has to be far longer than other types of lines. But you really shouldn't have to locate the driver mid-line in order to come up with a good-performing TL that allows the outside of the box and the driver location to mimic the Silent Speaker. I'll get back to you later after I've done some more modeling.
Paul
Regarding Qts, I prefer working with drivers having a Qts in the range of 0.35 to 0.50, but lower and higher values sometimes work out acceptably to quite well (the CA18RLY using Zaph's measurements is a good example).
With a driver located in the middle of the line, usually the only type of TL that will perform decently is an ML-TQWT, but since it has an expanding area, it has to be far longer than other types of lines. But you really shouldn't have to locate the driver mid-line in order to come up with a good-performing TL that allows the outside of the box and the driver location to mimic the Silent Speaker. I'll get back to you later after I've done some more modeling.
Paul
Why not try to make a more modern take of one of Olle Mirsch:es speakers?
Like the OM72 or OM2000?
Probably the best selling speaker manufacturer i Sweden but not so famous.
Lots of information about his speakers and thoughts from Olle himself:
http://user.faktiskt.se/vee-eight/ollemirsch/Mirsch means music.pdf
(sid 13-18 kan tex vara väldigt informativ för dig)
faktiskt.se • Visa tråd - Mirschkopplingscheman och mått
Like the OM72 or OM2000?
Probably the best selling speaker manufacturer i Sweden but not so famous.
Lots of information about his speakers and thoughts from Olle himself:
http://user.faktiskt.se/vee-eight/ollemirsch/Mirsch means music.pdf
(sid 13-18 kan tex vara väldigt informativ för dig)
faktiskt.se • Visa tråd - Mirschkopplingscheman och mått
That is actually a very good idea! I guess you get a bit blind for familiar design close to you like Olle Mirsch and Stig Carlsson. I will add it to my todo-list but I am still intrigued by tapered t-lines and mass loading. I have to explore it first but I might pick up some vintage Mirsch speakers (like these beautiful OM 3-28), they would fit right in with the rest of my vintage gear.
If we were to follow your advice for other decisions, we'd all be driving the same make and model car, eating the same food, dressing alike, etc.
Paul
Don't let me be misunderstood: closed box is good, boxcad is son of a bitch...
https://www.youtube.com/watch?v=fNZnBk87nZo
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.