Nah, they're in Martin's gallery: http://www.quarter-wave.com/Gallery/Gallery.html
6 pictures down. Very impressive indeed.
6 pictures down. Very impressive indeed.
I acomplish the cabinet for the 15in.
RCA LC1 coaxial driver using modyfied MCD sheet
TLsections... Measurements on TS of the driver done by
LSP Lab...
Some changes are made, I was forced to change the tweeter from
dynaudio esotar T330 in exp-hyp horn to ribbon tweeter homemade by my friend...
even the almost equal db sound peasure, about 92-93db, I have constant lack of highs.
RCA LC1 have hugh energy at low and mid-low end, very linear but only to about 12KHz,
signed from the back as "extended range" driver
amp is 2x monoblock Quad II KT66 push-pull old amplifier...
RCA LC1 coaxial driver using modyfied MCD sheet
TLsections... Measurements on TS of the driver done by
LSP Lab...
Some changes are made, I was forced to change the tweeter from
dynaudio esotar T330 in exp-hyp horn to ribbon tweeter homemade by my friend...
even the almost equal db sound peasure, about 92-93db, I have constant lack of highs.
RCA LC1 have hugh energy at low and mid-low end, very linear but only to about 12KHz,
signed from the back as "extended range" driver
amp is 2x monoblock Quad II KT66 push-pull old amplifier...
so, that was a long read....
great news that MLK is updating his sheets.
on the new features: realy great addition, finally can I simulate a TL in the trunk
of course I have a wish list: the possibility to simulate 2 or more drivers in the line would be great, according many this is a great way to reduce the dips caused by the 3/4 and 5/4 modes + that an enclosure with e.g. 2 12" drivers has a much higher WAF than one 18"
would it be possible (maybe it already is?) to export calculated curve? I mean both the graphs I see (all I can do now in MC2k is prtscr) and the raw data (so I can get it to excel)
another one, which might already be in, but don't know if anyone has tried it: another way to get rid of the dips caused by the qw modes, is to use an internal helmholz absorber (a well stuffed compartment with a tube). writing this, this should be possible even with the current sheets...
great news that MLK is updating his sheets.
on the new features: realy great addition, finally can I simulate a TL in the trunk
of course I have a wish list: the possibility to simulate 2 or more drivers in the line would be great, according many this is a great way to reduce the dips caused by the 3/4 and 5/4 modes + that an enclosure with e.g. 2 12" drivers has a much higher WAF than one 18"
would it be possible (maybe it already is?) to export calculated curve? I mean both the graphs I see (all I can do now in MC2k is prtscr) and the raw data (so I can get it to excel)
another one, which might already be in, but don't know if anyone has tried it: another way to get rid of the dips caused by the qw modes, is to use an internal helmholz absorber (a well stuffed compartment with a tube). writing this, this should be possible even with the current sheets...
Be careful of assuming that by using 2 or more drivers alone, you will help sort the problem of ripple: according to Martin it doesn't work that way. A quote from a recent email discussion I had with him on this matter (the remark was made in the original article about the Seas Thor Kit: I wasn't convinced so I thought I'd deferr to a higher authority on QWR behaviour). I've mentioned it before elsewhere but it's useful here too:
‘I am not sure that this is correct. The standing waves in a TL are independent of driver position. The driver position just determines how much excitation is applied to each standing wave, for example a driver at 1/3 L in a straight TL will not excite the 3/4 mode and will still excite the 1/4 and the 5/4 modes but at a slightly reduced level. The most efficient excitation is at the closed end. My guess is that when I get the two driver TL worksheet running that there will be pluses and minuses for both one or two drivers. I don't see two drivers solving the ripple problem’.
Best
Scott
‘I am not sure that this is correct. The standing waves in a TL are independent of driver position. The driver position just determines how much excitation is applied to each standing wave, for example a driver at 1/3 L in a straight TL will not excite the 3/4 mode and will still excite the 1/4 and the 5/4 modes but at a slightly reduced level. The most efficient excitation is at the closed end. My guess is that when I get the two driver TL worksheet running that there will be pluses and minuses for both one or two drivers. I don't see two drivers solving the ripple problem’.
Best
Scott
Now I find that that very interesting. Obviously it's either an electrical or mechanical (I define mechanical as being the physical properties of the drivers and the enclosure they are mounted in) effect but which? I wouldn't like to hazard a guess without knowing more. What sort of enclosure were they mounted in? Any dimensions you can share? When you say the drivers were not synchronised, do you mean the respective output of the drivers alone, or the total response of the enclosure with the drivers mounted in them? Sorry for the barrage of questions, but it's obviously something worth exploring.
Best
Scott
Best
Scott
Those features are on my list of things I would like to add, but they will come later.
I already have this part covered and have provided it to a few individuals in the past. The problem is that file export and import is disabled in the free Explorer program. So the only way to accomplish this is to have a fully licensed version of MathCad. The percentage of the user base with a licensed copy of MathCad is very small.
you sort of have my reply to your 1st part in the 2nd: having 2 drivers on the right location will help the ripple. the "worst" ripples occur at the 3/4 and 5/4 frequencies, now place 2 woofers one at 1/3 and the other at 1/5 of the line, that should iron out the worst ripples. this has sort of become the 'signature' design for a TL for the German magazine Klang+Ton.Scottmoose said:
soongsc said:
It could be the artillery was out of sync...
(a bad joke, I know...)
IMO, as T/S parameters vary between drivers, there are differences that show esp. at high SPLs, and you saw the effect of one woofer transiently "pumping" the other one. I don't think that the transient model for an enclosure + driver(s) has to be the same as for low level signals.
Cheers
Gastón
soongsc said:
Greets!
Unless the drivers have perfectly identical specs, then when each 'feels' the need for high power, whichever one has the lowest impedance will draw the most power, hence one can excurse visibly more, so in high current apps it's best to optimize each driver in its own enclosure. When drivers are in series then, the lowest impedance driver can be reduced to being little more than an expensive fused link in the signal chain.
GM
OK, so lets say the drivers are located at 1/3 and 1/5 of the length. At the frequency of the 3/4 wavelength mode, which is probably around 150 Hz for a straight TL tuned to 50 Hz, the driver at 1/3 of the length will not see any influence on the back of the cone from the standing wave. The driver at 1/5 of the length will excite the 3/4 wavelength mode and will see a back pressure so its motion will be attenuated. The 3/4 mode in the enclosure is still excited, one driver will be playing unloaded as if it were in a baffle, and the the second driver will have an attenuated output typical of a TL.
I have not done the math or built such a system, but I guess I do not see any obvious significant attenuation of the ripple. In fact I would speculate (actually guess) that you could make a trade off between one and two drivers in a TL and find that neither configuration is free of problems. I don't see the ripple issue being much improved with two drivers. I am not sure I buy into the findings in Klang+Ton.
Not sure how big an enclosure you built, but if you were after bass for the 1812 cannons my guess is it is a good sized box. The mis-syncronization my be differences in the driver's physical properties as several others have proposed or it could be the box loading the drivers differently. If standing waves exist in the box the back pressure might be different for each driver yielding unequal motions at specific frequencies.
Lots of speculations tonight,
ghpicard said:
It could be the artillery was out of sync...
(a bad joke, I know...)
IMO, as T/S parameters vary between drivers, there are differences that show esp. at high SPLs, and you saw the effect of one woofer transiently "pumping" the other one. I don't think that the transient model for an enclosure + driver(s) has to be the same as for low level signals.
Cheers
Gastón
That was one possibility I was thinking too. But it seems no software tool addresses this.
MJK said:
It seems like a lot has to do with the short time before resonance starts to happen. For example the phase differences that occur when pressure wave from one driver reaches the other, there should be some cancelling (or pumping) effects wouldn't it? So we might be able to assume that one driver will create some kind of load on the other causing it to respond differently and thus cancel out some parts of the resonance modes?
I don't think that is the case. The mode is set by the geometry, the pressure or velocity will have a prescribed fixed profile. The drivers provide excitation to the mode. The modes response is large enough to almost stop the drivers from moving. I think you need to consider the drivers as providing a small amount of excitation and the mode as the strong over riding response. The standing wave in the enclosure dominates the driver's motion and all acoustic output from the system.
MJK said:
For simple sine waves that might be the case. But one dimension has many orders of modes, it makes a difference whether the driver is at the node of the mode or at the peak of a mode. Additionally, once the mode is excited, the phase relationship between the driver and the mode determines how the response will be. A different driver location will excite a different mode, and thus might cancel out some modes created by another driver. Does this make sense?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.