I have always thought that it was the rising response to resonance that made these cheao drivers ( high "Q" drivers ARE usually cheaper ) better in OB, they also model better if bass drivers are placed as close to the floor as possible
The graphs don't tell the whole story tho, sometimes systems have to be listened to before forming opinions on SQ, agreed that the drivers are definitely not the best quality but the performance is acceptable for the price and everything is a compromise isn't it.
Regards
Ted
The graphs don't tell the whole story tho, sometimes systems have to be listened to before forming opinions on SQ, agreed that the drivers are definitely not the best quality but the performance is acceptable for the price and everything is a compromise isn't it.
Regards
Ted
Hi Ted,
You're quite right, with the appropriately placed 2nd order low-pass, they do look OK (they develop a flat section when it's in the right place), just lacking in extension. I'm not sure why these relatively high Q drivers lack in extension so particularly badly when they have a fairly low Fs, too, but I can only assume that it's something to do with their small cone area.
All I was demonstrating was that:
a) Equalisation really does significantly contribute to bass extension
b) That it's possible to model the response of a Linkwitz transformed driver if one is willing to muck around enough
c) That by producing a nearly flat response, I ought to be able to use LR4 filters to cross without having to consider the true acoustical order of the slope.
Anyhow, who's looking for perfection? Some people just like to build things and hear them. I'm of a scientific mindset and broke, so I like to examine the numbers first......and yes, ultimately, these are worth a $5 each, but certainly not worth the $25 each that Jaycar still wants for them.
You're quite right, with the appropriately placed 2nd order low-pass, they do look OK (they develop a flat section when it's in the right place), just lacking in extension. I'm not sure why these relatively high Q drivers lack in extension so particularly badly when they have a fairly low Fs, too, but I can only assume that it's something to do with their small cone area.
All I was demonstrating was that:
a) Equalisation really does significantly contribute to bass extension
b) That it's possible to model the response of a Linkwitz transformed driver if one is willing to muck around enough
c) That by producing a nearly flat response, I ought to be able to use LR4 filters to cross without having to consider the true acoustical order of the slope.
Anyhow, who's looking for perfection? Some people just like to build things and hear them. I'm of a scientific mindset and broke, so I like to examine the numbers first......and yes, ultimately, these are worth a $5 each, but certainly not worth the $25 each that Jaycar still wants for them.
They should be cheaper than that, I thought this driver was onclearance like the rest of the cheaper high "Q" drivers.
Correct not worth $20 when the 12 inch is selling for $21 ( if you can find them.
I found a pair of Tandy 8's that would be worth crossing at 2.7/3k.
Would you consider a 2.5 way using the Jaycar as the .5??
Postage would cost more than the driver, merajsalek didn't get back to me; if you want them use email and I'll pack and send.
regards
Ted
Correct not worth $20 when the 12 inch is selling for $21 ( if you can find them.
I found a pair of Tandy 8's that would be worth crossing at 2.7/3k.
Would you consider a 2.5 way using the Jaycar as the .5??
Postage would cost more than the driver, merajsalek didn't get back to me; if you want them use email and I'll pack and send.
regards
Ted
It's possible that they are on clearance, I was reading from the 2008 catalogue, not the shelves of a real store. I've looked for those old very high Q 12's, but the local Jaycar didn't have any. Neither appear on their website any more.
So you reckon Jaycar 8's up to 200ish, Tandy 8's up to 2k7 then a proper tweet? Fair enough....don't suppose you've any of those? Failing that, one of the current Jaycar tweeters is the HiVi K1, which ought to work OK crossed that high.
I'll drop you an email to work it out.
So you reckon Jaycar 8's up to 200ish, Tandy 8's up to 2k7 then a proper tweet? Fair enough....don't suppose you've any of those? Failing that, one of the current Jaycar tweeters is the HiVi K1, which ought to work OK crossed that high.
I'll drop you an email to work it out.
Hi!! merajsalek WAF wins again?? Yeah OB; even desk top ones need a big bit of wood ( substute favourite baffle material here)
Seeker
I'd go even lower with the paper 8's if you have an inductor big enuff 5.6mH from J******* should give you 230 apro but there is a 9mH avaliable from them, big problem is the high DCR catalogue says 0.9 Ohms mine read 1.1, I think that cuts 2dB but not sure.
How cheap a tweeter do you need?? I rebuilt a pair of those DickSmith bookshelf speakers a while ago and I got no takers on ebay starting at $3.50 a pair yours for that price and I'll throw in 4 off 2uF caps to play with.
The Tandy poly cones have Fs of 30Hz but Qts is 0.33 so they roll off really fast, thinking about it you may be OK with a first order at 230, I'll let you play with those figures, would $20 be too much for your budget for the Tandy's??
Regards
Ted
Seeker
I'd go even lower with the paper 8's if you have an inductor big enuff 5.6mH from J******* should give you 230 apro but there is a 9mH avaliable from them, big problem is the high DCR catalogue says 0.9 Ohms mine read 1.1, I think that cuts 2dB but not sure.
How cheap a tweeter do you need?? I rebuilt a pair of those DickSmith bookshelf speakers a while ago and I got no takers on ebay starting at $3.50 a pair yours for that price and I'll throw in 4 off 2uF caps to play with.
The Tandy poly cones have Fs of 30Hz but Qts is 0.33 so they roll off really fast, thinking about it you may be OK with a first order at 230, I'll let you play with those figures, would $20 be too much for your budget for the Tandy's??
Regards
Ted
So, $33.50 + shipping for the lot? Sounds pretty cool, so long as shipping doesn't some to some outrageous number, guess it'll come down to weight....can you weigh them and give me the numbers to plug into the Auspost calculator? (see? I like my numbers!)
Remember that I'm using active filters, so the crossover can go anywhere I want it to....I'm just thinking that low Q might be rolling off by that point. I'll go and fudge the numbers, and see what I come up with. But that's just details at this point, I'm sure I can get something to work out.
Remember that I'm using active filters, so the crossover can go anywhere I want it to....I'm just thinking that low Q might be rolling off by that point. I'll go and fudge the numbers, and see what I come up with. But that's just details at this point, I'm sure I can get something to work out.
I think I'll give this a try...thanks for sharing Ion.
Ted:
Not sure yet, the Vifas I have in mind are coaxial. Steve (apexjr) likes their sound in a 4 litre ported box....no available T/S parameters tho....but since they came from M&K Sound after the company shut down, I'm guessing they can't be all that bad. The price is good too.
I'll play by ear on this one. The source is my laptop and I'll play a mixture of Lossless and MP3s. Not much point in beating myself up trying to make it 'perfect'
"Lucky you", he said, as his face turned a shade of green with envy...
Seeker while I tend to actively power most of my projects is it possible to do a 2.5 system actively>?? This may be one time that a passive implementation might just work better??
I just thought that the main reason for the .5 is purely to cover baffle step and to boost low bass.
I need some help on this one, somebody please jump in and help us answer that query
I just thought that the main reason for the .5 is purely to cover baffle step and to boost low bass.
I need some help on this one, somebody please jump in and help us answer that query
Actually, I was thinking of a full 3-way....back to the equalisation of the bottom end idea. I don't really understand how .5 ways work, so I was planning to just EQ the woofer and midrange to relatively flat (dipole 6dB / oct from 300hz) then cross LR4 as low as I can get the Tandy 8's to go (mucking around with the sim on that now), say it might be 150-200hz. The response after the dipole boost should look quite flat, so it shouldn't interact with the crossover. Similarly, highs would be crossed either LR2 or LR4 (haven't really thought it out yet, does it make much difference when you're crossing as high as 3k?).
If you can show me that the 2.5 way would have benefits rather than doing it this way, I'm interested to learn.
So to show the signal chain as I plan to use it:
Source > Dipole Correction > 2 way LR4 at 200hz
Highs > 2 way LR2 at 2.7k
Treble > Tweeter
Mids > 8" Tandy Woofer
Lows > Linkwitz Transform (Optional) > 8" Jaycar Woofer
If you can show me that the 2.5 way would have benefits rather than doing it this way, I'm interested to learn.
So to show the signal chain as I plan to use it:
Source > Dipole Correction > 2 way LR4 at 200hz
Highs > 2 way LR2 at 2.7k
Treble > Tweeter
Mids > 8" Tandy Woofer
Lows > Linkwitz Transform (Optional) > 8" Jaycar Woofer
The benefit of a .5 way speaker is that it just "Fills in" where the other driver rolls off.
The way I did it was to run my mid-bass driver ( a similar Tandy 8) full range crossing to a tweeter at 3500 ( a touch high )so I used a second order filter on the woofer at 3500 ~but on the fill-in driver I just used a first order low pass.
Bass below 155 got a 6dB boost, but because of the 6dB slope of the 1st order electrical on the boost woofer when the cross-over frequency is reached it is only the midbass that is working in that band, I then used a high pass on the tweeter of course, it turned out with the tweeter I used to sound better with a third order electrical and it needed a zobel as well.
I use a Behringer CX 2310 so I'm stuck with LR4, the tweeter's I'm including won't work at 2700 LR2 as their Fs is 1500Hz.
Full 3-way is GOOD, but wasted on the J***** paper woofer, the Tandy on the other hand would do justice to a reasonable 12, crossed low and handling from 250 -> 2700 quite well
The way I did it was to run my mid-bass driver ( a similar Tandy 8) full range crossing to a tweeter at 3500 ( a touch high )so I used a second order filter on the woofer at 3500 ~but on the fill-in driver I just used a first order low pass.
Bass below 155 got a 6dB boost, but because of the 6dB slope of the 1st order electrical on the boost woofer when the cross-over frequency is reached it is only the midbass that is working in that band, I then used a high pass on the tweeter of course, it turned out with the tweeter I used to sound better with a third order electrical and it needed a zobel as well.
I use a Behringer CX 2310 so I'm stuck with LR4, the tweeter's I'm including won't work at 2700 LR2 as their Fs is 1500Hz.
Full 3-way is GOOD, but wasted on the J***** paper woofer, the Tandy on the other hand would do justice to a reasonable 12, crossed low and handling from 250 -> 2700 quite well
Hmm....I might stick with 3 way, I'm only going to build up the circuits on protoboard.
I'm coming up with some weird results trying to sim an estimated model of the Tandy woofer......it's exhibiting a 3db/octave (quite precisely) rolloff beyond about 800hz, that looks even worse on the sims due to a 5ish dB hump that's appearing at about 300hz. Might just be an artefact of the process, I'll try it again in MJK's worksheets later tonight. Looking more closely, the 3db/octave rolloff appears to be the response of the woofer, so it's probably just a result of my partially made up numbers, while the hump is definitely part of the baffle response as simmed by Edge.
Anyone know anything about this behaviour?
I'm coming up with some weird results trying to sim an estimated model of the Tandy woofer......it's exhibiting a 3db/octave (quite precisely) rolloff beyond about 800hz, that looks even worse on the sims due to a 5ish dB hump that's appearing at about 300hz. Might just be an artefact of the process, I'll try it again in MJK's worksheets later tonight. Looking more closely, the 3db/octave rolloff appears to be the response of the woofer, so it's probably just a result of my partially made up numbers, while the hump is definitely part of the baffle response as simmed by Edge.
Anyone know anything about this behaviour?
OK, I've done some more messing around with the simulations - it appears that the high frequency behaviour of the midrange is actually related most closely to the Vas. Why this is, I don't know, and whether it will matter, I don't know yet.....it looks like "build, measure and adjust" will be the way to go. Since at those frequencies behaviour is essentially infinite baffle, yet Ted reckons they're good up to 3k, I'm going to assume that they're well behaved. The low end peak is more worrisome, but it can be notched out if I measure it in reality. This seems to be common to many OB's, but I've never really seen anyone worry about it.
I had better amend that statement, they were wel behaved when I used them in a small sealed box, to be honest it was just gut feeling that they would work reasonably well OB, I'll knock up a baffle later out of foamboard and check by ear, what baffle size do you want me to use and what offset??
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.