That sure seems about right, which I think pretty much takes care of the OP's contention it wouldn't work.... it's nice to see there is agreement between the math and the simulations. The assumption I was making in the manual calculations was that the full cubic foot volume would be available, which would then place the Fs of the box near the 49-50Hz range, and with Qts being about 0.71 the F3 would be right near 49-50Hz as well.
But another thing has me curious now, when comparing stuffing vs none, the sim is not showing any diff in the f3, but is it showing any change in Fs?
In practice the stuffing does bring Fs down, while also reducing Qts - so it's entirely possible F3 is going to stay close to the same...
BTW.... I notice the curious lack of response of the OP, once it looked like his contention has been well and truly shot down... moral of the story: OP, looks like you had your math wrong, perhaps you owe Sreten an apology. It's pretty clear the RS225 would fit the OP's original parameters, and fit them well.
Moral #2: Rather than spouting off about how one is an "engineer" with loads of experience, and clearly accusing others of BS'ing when they try to help and then being arrogant and intractable, and holding onto an erroneous notion even when clearly proven wrong... perhaps one should go back and re-check one's math.
Moral #3: MANY of the people in this forum have engineering degrees, or equivalent and in some cases more extensive experience. It can be a pretty tough crowd if you assail their technical proficiency before you double check your numbers. The donkey-kicking you have been receiving may well be warranted ! eek:
I used to work for a building company. Costain, as it goes. Our visitors often used to remark what a nice time they had calling in on us in Westminster Bridge Road. We gave them free coffee and a paper whilst they were waiting to get going for their meeting.
It was a tired old joke, but I often used to explain that we are "Civil" engineers.
TBH, some people round here could learn a thing or two.
A few weeks ago I tried to use a speaker simulation program to illustrate a point about how stuffing has an effect on both damping and the compression process (shifting it from adiabatic towards isothermal and making the box appear bigger). I was surprised to find only the former was included in the simulation. I don't know which simulation programs use a more complete model (anyone?) but I can see it contributing to online confusion.
I did one more thing just this morning. Since I modeled originally with the 26-liter box filled at 0.5 lb/ft3, I used the ROT that this effectively enlarges the box by 20%. I then ran the math again but with Vb 20% larger, and calculated the Qtc would end up at 0.71. Of course the box really ought to be built larger to compensate for volumes consumed by drivers, braces and crossover assembly, and I'm guessing 28-30 liters ought to be just about right.
I didn't really pay much attention to where the woofer's impedance peak occurred in frequency. My understanding is that adding stuffing essentially only lowers Qtc and does not actually lower Fs or F3 (but I won't argue with you), and the modeling I did with and without stuffing seems to bear that out. However, I've not built a sealed box for a woofer since the last half of the '90s; I build only TLs for personal use (plus model lots and lots for other DIYers).
Regarding the OP, there are people like him that, as I said in a previous post, already have an answer to a question, but still ask it, apparently expecting affirmation that their answer is correct, then don't react very well when it's not.
Paul
I didn't really pay much attention to where the woofer's impedance peak occurred in frequency. My understanding is that adding stuffing essentially only lowers Qtc and does not actually lower Fs or F3 (but I won't argue with you), and the modeling I did with and without stuffing seems to bear that out. However, I've not built a sealed box for a woofer since the last half of the '90s; I build only TLs for personal use (plus model lots and lots for other DIYers).
Regarding the OP, there are people like him that, as I said in a previous post, already have an answer to a question, but still ask it, apparently expecting affirmation that their answer is correct, then don't react very well when it's not.
Paul
Last edited:
I'm pretty sure I remember you posting something about compression effects. I'll be the first to admit I'm not qualified to address this, but perhaps the lack of it being considered in simulation programs indicates it's either not an issue at all, or its effects are minor to insignificant?
Paul
Paul
It can give an apparent increase in box size of about 30% or so. Not huge but enough to make ones measurements and theory differ by more than would be comfortable for many. I was about the provide a link to wikipedia in support but the relevant sentence reads:
"The effective volume increase can be as much as 40% and is due primarily to a reduction in the speed of sound, and not to the popular misconception of a change in operating conditions from adiabatic to isothermal."
Sound is of course a compression of the air and it is the change of the compression process from largely adiabatic towards more isothermal that changes the speed of sound! Interestingly, it was precisely this that Newton got famously wrong when he first calculated the speed of sound as discussed here not on wikipedia. I learnt years ago not to waste my time attempting to correct things like this on wikipedia but if anyone has the stamina to argue with whatever fruitcake and his minions owns the page be my guest.
Getting a bass reflex right seems like a fairly big project by itself. Modeling programs can be helpful (I use SPICE for my circuit simulations), but when things get complicated (many variables) you may find that they are a bit off.
Getting a crossover right can also be a bit of a task. Especially a passive crossover. How much time does one want to spend getting everything right? For crossovers, I consider the varification process and final tweaking a significant part of the process.
The theoretical approach rarely gets things exact enough with bass reflex or passive crossover design (from my experience). Since one of the stated goals was to keep the cabinet small (if I remember correctly), I still vote for a closed box with active EQ to get the good low bass. Active electronics are very predictable, accurate, and may reduce the amount of reactance the power amp has to deal with. Box volume is significantly less critical with a closed box. If this speaker was going to be a production item, with many produced and marketed, I might have a different opinion.
As far as the abusive statements that some have contributed, my experience is that some of the best Engineers are often grouchy temperamental people in general. They got good by being on their own backs virtually continuously (very critical of them self). It became a habit. It carried over into the rest of their lives, and they may not have many good friends. It's not always easy to keep your psych in a good balance. I know I'm a bit that way myself.
Getting a crossover right can also be a bit of a task. Especially a passive crossover. How much time does one want to spend getting everything right? For crossovers, I consider the varification process and final tweaking a significant part of the process.
The theoretical approach rarely gets things exact enough with bass reflex or passive crossover design (from my experience). Since one of the stated goals was to keep the cabinet small (if I remember correctly), I still vote for a closed box with active EQ to get the good low bass. Active electronics are very predictable, accurate, and may reduce the amount of reactance the power amp has to deal with. Box volume is significantly less critical with a closed box. If this speaker was going to be a production item, with many produced and marketed, I might have a different opinion.
As far as the abusive statements that some have contributed, my experience is that some of the best Engineers are often grouchy temperamental people in general. They got good by being on their own backs virtually continuously (very critical of them self). It became a habit. It carried over into the rest of their lives, and they may not have many good friends. It's not always easy to keep your psych in a good balance. I know I'm a bit that way myself.
Without attempting to read anything on wikipedia, much less believe it, or argue with you or anyone else, if the quote is saying that adding fiber fill causes a reduction in the speed of sound (which in turn increases the effective volume by 40%), I'm not buying it.
Paul
Paul
Well you ought to because the speed of sound is given by sqrt(gamma.R.T) where gamma is the ratio of specific heats which varies from around 1.4 towards 1.0 as the stuffing increases the thermal mass changing the compression process from largely adiabatic towards more isothermal. The square root means the speed of sound only reduces by about 15%, the amount Newton was wrong. I cannot find a link to the actual article but post 4 here seems to contain a copy showing the effect of speaker stuffing.
There is apparently a slight decrease in the speed of sound as the sound passes through various material like poly fiber. A 15% reduction isn't all that much, unlike what many people have postulated happens, incorrectly concluding, for instance, the required line length for a particular 1/4-wavelength resonance could be reduced to half that for an unstuffed line. In all the TL modeling I've done using stuffing densities in the 0.5 to 1.0 lb/ft3 range and stuffing the first 50% to 67% of the lines, with F3s in the mid-20s to upper 30s, compared to no stuffing the lines' 1/4-wave resonance decreased by just a few Hz, which is on the order of 10%.
Paul
Paul
Hey y' all, I'm not so sure about the speed of sound argument either...
The subject of stuffing in T-Lines (and sealed boxes) is one which I find fascinating, perhaps because it seems like it's not yet fully understood... although MJK's models do seem to be quite close to reality where TL and ML-TQWTL are concerned. I've built three boxes based on those and found close agreement with the sim'd predictions.
There is still some argument about the reason for the changes in TL resonant frequencies, some hold it's because the speed of sound is being reduced, some hold it's due to the added resonant mass of the stuffing (lower Q because of a high resistive component). And some think that it's a little bit of each. And I think...well, who cares wot I think!
I'm in the third camp.
When it comes to sealed boxes with relatively small dimensions, I'm especially not sure of the speed of sound thing. And as it looks, the 40% gain of internal volume (as predicted by the theoretical change in adiabatic vs isothermal) does yet not seem to be fully attainable:
1) In the building of over 30 sealed enclosures, I have NEVER been able to get what looks like a 40% increase in box volume. The max I was able to achieve was about 20%. The resulting reductions in Fs and Qts were on the order of 10-11%.
A 20% increase in box volume translates into a reduction of about 10-12% in the Alpha, where Alpha = Sqrt (1+Vas/BoxVolume) so it holds up mathematically as well.
I verified this the hard way about 35 years ago when I built a woofer with a Peerless 12" (Fs of 19Hz, Qts 0.38) in a 2cub ft box. Without stuffing Fbox was 40Hz, and I was hoping to get a Fbox=32Hz box out of it. I was very disappointed to find the lowest I could get was Fbox=36Hz
'Twas a decent woofer tho', sounded good...
2) I speculate that the main reason for the results in #1 is that the stuffing material itself displaces some of the internal volume, i.e. the denser the stuffing, the more air is displaced. I believe I verified this notion, because when I really packed in the stuffing, after a while the Fs started going back up.
3) The above experimental observations seemed to hold whether the box was big or small.
4) Maybe with some of the newer more absorbent fibers we could push the apparent increase closer to 30% or more as long as the fibers displace very little air. Hmm, micro-tubule poly-fill perhaps?
So in practical terms, we can clearly get a 10% lower Fbox and Qts, (and yes, sometimes also a lower f3) by adding the right amount of stuffing, and this seems to hold true with any sealed box of nearly any volume. Also, as it becomes elongated the pipe resonances become more of a factor as shown in MJK's work.
I wish I could access my MJK simulator, but that's no longer available. Please Martin, bring it back!
There's no denying Physics. The speed of sound absolutely is different in different materials, faster in most (4757 ft/sec in water), slower in some (846 ft/sec in CO2) per my 50-year-old college Physics book. But is it really slowed down by stuffing like polyester fiber, and how much? Does a line's 1/4-wave resonant frequency decrease a bit from the speed of sound being slowed as it passes through the fiber, or is there something else at work? Frankly, as long I can model a good design, I'm happy.
Paul
Paul
40% improvement in Vb in theory, but Richard Small said that 25% is practical, and I suppose we're talking 4Lb of fibreglass:
Thiel_small_analysis
A heavily stuffed box has a distinct effect on the midrange too, though I use BAF myself. I'm not sure I like it. Sounds constrained and tight.
Thiel_small_analysis
A heavily stuffed box has a distinct effect on the midrange too, though I use BAF myself. I'm not sure I like it. Sounds constrained and tight.
Last edited:
I wouldn't say sreten is caustic he's just very direect, to the point and has little patience for wasting time on superfluous endeavours. I enjoy the fact that DIYaudio has people like him around, kind of like the good cop bad cop routine. He says what needs to be said in a lot of situations too and probably what some of us are actually thinking but usually don't say.
Friend sreten is good on box calculations, I think he has some piece of software that helps. And can always find a Zaph Audio project that works best.
I think it's nice to be clever, but it's cleverer to be nice. Most people here are comparative rookies, and I enjoy helping with the the basic cheapie stuff as much as the high end. It makes me think clearly. After all, if you can't explain it simply, you don't understand it. A journalist once told me it's MUCH harder to write for a tabloid than an intellectual broadsheet.
If there was a perfect speaker, we wouldn't even be interested anymore. So I'm still learning.
I always try to keep cabinet wooden panel damping separate in my mind from volume damping. They are different things really.
The science seems to say that volume damping with closed box, aka stuffing which you can't use with reflex beyond the usual eggcrate lining foam, is just a cheap trick to make a cabinet physically smaller. Acoustic Research used it to the max.
TBH, if I was the original poster doing 8" bass, I'd just make a 45L cabinet with a deeper depth. It'll still look good and opens up all sorts of tidier bass drivers and simpler crossovers than these metal horrors.