By the way I have discussed the t-line theory with Martin . He agrees his designs are different then Fried's designs. There is another way to design t-lines by using flue pipe formulas that there are patents for. This design uses PA Drivers.
The man that devloped them use to hold the world record for creating the world's loudest sound. The formulas I gave to one forum user for t-line design he coudl not understand. I was told they had nothing to do t-line designs.
I have given AllenB the crossover plans to one design like I have stated numerous times.
It would seem many of his posts have been ignored. Look at the respone graphs.
Toroidal Whistle Prototype Test - YouTube
The man that devloped them use to hold the world record for creating the world's loudest sound. The formulas I gave to one forum user for t-line design he coudl not understand. I was told they had nothing to do t-line designs.
I have given AllenB the crossover plans to one design like I have stated numerous times.
It would seem many of his posts have been ignored. Look at the respone graphs.
Toroidal Whistle Prototype Test - YouTube
Yep. It is being elusive alright, and a certain someone is being immensely condescending in his rebuttals.
Later,
Wolf
A person would have to be a fool to post another companies legally owned designs on the internet. It does not matter if the patents run out. When a corporation claims a financial loss these are damages claimed. This is enough for a civil litigation claim . CAUSE of ACTION
Cause of action - Wikipedia, the free encyclopedia
I am no fan of the Bose Corporation.
If the electronic equalization circuits for their sub modules, wave radio and 901 speakers were posted on the internet you would spend years in court.
Cause of action - Wikipedia, the free encyclopedia
I am no fan of the Bose Corporation.
If the electronic equalization circuits for their sub modules, wave radio and 901 speakers were posted on the internet you would spend years in court.
Wolf-teeth you are condescending when some understands something you don't.
If a student makes fun of his teacher he will never learn anything. The same can be said in self defense training as an example.
All the conjecture on Fried's series crossovers on Parts Express are wrong.
If a student makes fun of his teacher he will never learn anything. The same can be said in self defense training as an example.
All the conjecture on Fried's series crossovers on Parts Express are wrong.
Last edited:
I have no problem if you copy stuff from my old site but please reference it, or better yet, provide the link to the full page: (Click Impatient)
http://web.archive.org/web/20021012193843/http://www.geocities.com/kreskovs/ar.html
After all, it is copyrighted, as noted at the bottom of the page.
IIRC that crossover had three drivers in series, electrically in phase. Mid and woofer had a capacitor across them and a coil leading to them from the input. Tweeter had one resistor across it and one resistor leading to it from the input.
But with series, what will it cost? The priority ought to be to blend the drivers' acoustic response into a good system.
Furthermore what good is flattening a driver's impedance before the crossover, when the acoustic response will not necessarily be flat? A filter can be designed which will produce the desired end result without going through this extra step.
Most amps aren't fussy but when they are...that's where Zobel correction is useful. I've had limited success correcting in particular the crossover regions of some speakers when used with particular amps.
But with series, what will it cost? The priority ought to be to blend the drivers' acoustic response into a good system.
There's no need to be concerned about a little reactivity. Particularly when it is minimum phase.
Furthermore what good is flattening a driver's impedance before the crossover, when the acoustic response will not necessarily be flat? A filter can be designed which will produce the desired end result without going through this extra step.
This thread got atarted with Sreten disagreeing with me about my opinion of the many AES papers on constant violtage network design theory. At least that is my opinion.
The aurhors of those papers did not use computers, They did Physics the old fashioned way. All crossover design basis comes from their works.
Streten's posts help my argument. Personally I think he just likes to argue.
The aurhors of those papers did not use computers, They did Physics the old fashioned way. All crossover design basis comes from their works.
Streten's posts help my argument. Personally I think he just likes to argue.
AllenB playing them witha Dynaco ST-400, 200 watt 8ohm amp compared to Paradound HCA-3500, 350 wats 8ohms amp is like night and day.
John K, a pleasure to meet you Sir. 🙂
Apologies if I didn't credit you. Frankly I've found this whole series crossover business a bit of a chore to research. Your own old sites lack many images, which leave them incomplete. I just grabbed some at random. 😕
Let me tell you I spent a frustrating time at college researching the group delay inherent in any real-time filter with steep slopes. I finally decided that it is actually quite straightforward to correct group delay by reversing the arrow of time, and filtering backwards in time. Sounds strange, well not really... You just filter the recording on time backwards playback. An idea familiar to tape recording enthusiasts. In an ideal World, the CDs we buy will incorporate group advance. 😎
I'll calculate the transfer function of what one might call second-order series crossovers, i.e. the simple circuit with supplementary second-order elements. Frankly, I'm expecting nothing really new. But insight always follows creative stuff. The series crossover has some real elegance. It certainly makes you consider some symettries otherwise overlooked. I think you will understand me better than most, in that I think we will find a Q of 0.5 is optimum with most components.
This whole business has made me think more deeply about the significance of complex conjugates, which lie at the bottom of the series crossover approach. Very nice. 🙂
Thanks John for the link. A very interesting analysis. It seems to me the series XO, like all other XO's has it's faults and advantages. Eh?
Some of the figures seem to be missing though. Just a box with a red X.
I kind of believe in a Universal Sod's Law with this sort of Voodoo audio. The cold, hard equations of Physics are rather inflexible. The series crossover is deeply interesting, but at the end of the day, we may simply discover you can blend a Fostex fullrange unit operating in the bass with a Fostex fullrange unit operating in the treble so that it's indistinguishable from a single driver. Does that help us build better speakers? I'm not sure at this stage...😕
And you can think what you want- you are welcome to do so.
I stand by my statement of transient response not being equated with Le. I've talked to Ken Kantor on numerous occasions on the former MAD board, and not once did he mention that relationship, IIRC.
I still think you can't just use your 'Fried taper TL' with many drivers and achieve decent results. There's more to it than that.
Oh- and just because you think that Fried was the be-all/end-all, and that he knew his stuff, does not mean it's the best/only way to do a series xover and get great results. LC, AR, DiAural, quasi-12dB, or 2 part 6dB can all work if the conditions are right.
I don't have a beef with you, and you started with the personally demeaning comments. I just mentioned it was the same old thing as usual.
I would like it if you would cease,
Wolf
This is what I mentioned before in one my posts to this thread. Someone had done this but digitally, where the original signal was split into two equal halves, then each half was filtered, one with a high pass, one with a low pass, then the two filtered halves were recombined and saved etc. The net effect was that you couldn't hear any difference between the filtered files and the original. If there were differences then they were so small that it isn't even really worth giving much attention.
It does help us design loudspeakers because it showed that the higher order slopes don't, in and of themselves, really = poorer fidelity (I think they only went up to 8th order). That is, if you need a 4th order to really make two drivers work well together, then use the 4th order.
Now lots of people say they prefer a well executed 2nd order design over a well executed 4th order design. This poses a problem because to compare apples with apples then you really need to have a loudspeaker that will function just as well with the 4th order crossover as it will with the second, with the xover set at the same frequency and without any kind of asymmetrical slopes being required for proper driver summation. But lets say there is some truth in the 2nd order sounding better then the 4th. The biggest reason I think for why this might be is that there's naturally a wider area of overlap between the two drivers and that as a result the drivers individual sonic signatures 'blend' together more convincingly.
For me it is sufficient to "prove" it using real world examples. Pick any successful 2nd order and pick any successful 4th order, they sound different. Drivers difference doesn't even bother me to draw "conclusion".
Actually many sites and people mention the physics behind everything (hint: THD). It is just that when you experience it by ears that you will not take it for granted.
Drivers blending depends on phase in a wide area, "not only" where there is response overlap (!), surely not only depends on the situation at the crossover point. But unfortunately nobody have made useful rules regarding to this (length of span, degree of attenuation, etc) so that it can be achieved without having to listen to the real speaker/prototype (well, the deep null helps a bit).
Phase overlap is difficult to achieve for even half an octave above and below crossover frequency. Wider area of (response) overlap is never a plus when the phase is never suitable for that.
Hi,
I like arguing with people who can argue, and
don't resort to the sort of stuff posted above.
Trying to pretend your always right means you lose
the ability be wrong, a fatally tedious mistake to
make for any allegedly enquiring / learning mind.
Its boring, it is waffle, its ego-centric, and I don't care.
rgds, sreten.
Wolf-teeth Ken will not discuss current design theories being used by customers of the companies he is associated with. You are wrong about transient response. Look at his comment on Audiokarma. Driver inductance is frequency related. Drivers with a wider bandwith have higher inductance.
TL vs Overdamped BR - Opinions? - Page 11 - AudioKarma.org Home Audio Stereo Discussion Forums
Did you ever hear of Buggtussel Loudspeakers. The PHD owner and designer of the former company was using similiar if not the same design parameters as Fried and Wright. He used foam for damping not Acousta stuf.
Richard Small was the first persons works that I know of that used driver inductance as a design parameter in series circuits for determining the crossover frequency.. Driver inductance used to design Zoebel circuits
Small's time constant formula being an inverse gave me the idea of using tweeter inductance being the basis for the crossover frequency. It explained to me the theory behind Fried's designs..
Sreten I do not pretend to be right I usually am and have the facts to prove it. You might think I am ego centric I am not . There are many unfortunate people on this planet. If people can't be polite and get along on forums how can the world peacefully coexist. I feed a lot of needy people giving away deer meat that I harvest for crop damage under the supervision of the PA Game Commission.
I stated facts that AllenB verified are correct by his simulation. I like to discuss theory.
I hope your opinion of Fried changed . I started building his speakers as a teenager.
If you dont' like my posts ignore them or take up a more challenging hobby. I told Wolf_teeth this before. Most people don't get enough exercise or eat healthy.
This is just math.
TL vs Overdamped BR - Opinions? - Page 11 - AudioKarma.org Home Audio Stereo Discussion Forums
Did you ever hear of Buggtussel Loudspeakers. The PHD owner and designer of the former company was using similiar if not the same design parameters as Fried and Wright. He used foam for damping not Acousta stuf.
Richard Small was the first persons works that I know of that used driver inductance as a design parameter in series circuits for determining the crossover frequency.. Driver inductance used to design Zoebel circuits
Small's time constant formula being an inverse gave me the idea of using tweeter inductance being the basis for the crossover frequency. It explained to me the theory behind Fried's designs..
Sreten I do not pretend to be right I usually am and have the facts to prove it. You might think I am ego centric I am not . There are many unfortunate people on this planet. If people can't be polite and get along on forums how can the world peacefully coexist. I feed a lot of needy people giving away deer meat that I harvest for crop damage under the supervision of the PA Game Commission.
I stated facts that AllenB verified are correct by his simulation. I like to discuss theory.
I hope your opinion of Fried changed . I started building his speakers as a teenager.
If you dont' like my posts ignore them or take up a more challenging hobby. I told Wolf_teeth this before. Most people don't get enough exercise or eat healthy.
This is just math.
Then no one will take you seriously and heck I wouldn't even take myself seriously given that situation. How do I know if what I prefer is because of the crossover 'order' or something rather more important.
The only way to see if you prefer a 2nd order vs a 4th order design is to use a DSP based crossover where you have perfect acoustic 2nd order and 4th order slopes on each driver. Then you use time delay to phase align them both. The drivers used have to be equally capable with both the 2nd and 4th order slopes, ie you're not going to get a THD improvement by swapping to the 4th order. Then to compare the two you have to be able to switch between each set of filters, either 2nd or 4th order, instantly, at the push of a button from the comfort of your listening chair. ( Note - I have done this)
Now of course that isn't even perfect, but what you're trying to do is remove as many things from the equation as is possible so that what you're comparing is the difference between the filter order only and not something else that changing the order makes happen.
Well I have experienced it and the difference I heard was simply down to the different vertical off axis responses putting a little bit more energy into the presence band, other then that it sounded identical. Both the second and forth order also summed flat and were as close to identical as you're going to get on axis. This is a design with a wave-guide where the directivity of the wg is matched to the mid, the horizontal off axis was essentially the same for each filter - ie no holes off axis and CD after the wave-guide took over. The fact the two filters sounded different isn't a surprise because the C2C spacing is a somewhat high given the xover point, so that you're going to get comparatively larger differences in the lobing between the 2nd and 4th order. But in this case I wasn't technically hearing the difference between the 2nd and 4th order filters, I was hearing the difference in the polar plot.
Some would argue that what I was hearing was because of the filter and to a degree I agree. I preferred the 2nd order filter in that design precisely because of the way it affects the vertical off axis and therefore the 2nd order is better. This isn't quite the same thing as saying the 2nd order filter is better simply because it's a 2nd order filter though.
Either you have a design that phase tracks well or it doesn't. If you can't make it work with one type of filter then naturally you're forced to try something else.
5th element I have yet to hear a high order crossover that sounds natural they sound artificial to me including the B&W Nautilus 801.
I have ruler flat hearing according to the audiogram done by my audiologist even though I am 50.
I have ruler flat hearing according to the audiogram done by my audiologist even though I am 50.
Well the original 801 Nautilus did use high order filters that probably ended up giving them 4th order acoustic slopes. The current 800 series use very simple filters mind you and are a classic example of less is not always more, they leave a rather prominent peak in the FST unit completely un-attenuated, which is not my idea of trying to obtain perfection in audio reproduction. Not to mention the off axis problems that series suffers from.
No one has ruler flat hearing, our ears don't work like that and the overall frequency response of our ears will differ too from person to person. What your audiologist probably meant was that you've still got good hearing high up in frequency.
No one has ruler flat hearing, our ears don't work like that and the overall frequency response of our ears will differ too from person to person. What your audiologist probably meant was that you've still got good hearing high up in frequency.
- Status
- Not open for further replies.