Is there a free software out there to analyze crossover network?
What I want to do is to construct mathematical equations and solve an optimum value based on known constraints and requirements on a certain crossover network.
What I want to do is to construct mathematical equations and solve an optimum value based on known constraints and requirements on a certain crossover network.
crossovers
Do you really want to do what others have done - again?
You could save some time and use some of the freeware and shareware programs available on the net.
For a mathematical approach you could possibly use Mathcad or possibly even MS Excel.
Have ffun.
Do you really want to do what others have done - again?
You could save some time and use some of the freeware and shareware programs available on the net.
For a mathematical approach you could possibly use Mathcad or possibly even MS Excel.
Have ffun.
Hi ashok,
No No No... I'm not sure others have done this, that's why I asked. I did a quick check on the FRD website, and it seems that the circuit topologies used are fixed to the common ones. It is easy to understand that developing a software that accept any kind of circuit is a lot harder.
My crossovers had been built 100% by ears (based on my understanding of frequency behaviours in L/R/C). Of course I always choose series crossovers, or at least semi-series, or non-series but built based on series 🙂confused: ). The standard crossovers had been proven to conform very well with calculation. The non-standard ones, I don't know how to cross check, that's why I want to do this thing
. But I always do a side by side comparison, so what pleases my ears better should be kept, no matter what. Right? 
No No No... I'm not sure others have done this, that's why I asked. I did a quick check on the FRD website, and it seems that the circuit topologies used are fixed to the common ones. It is easy to understand that developing a software that accept any kind of circuit is a lot harder.
My crossovers had been built 100% by ears (based on my understanding of frequency behaviours in L/R/C). Of course I always choose series crossovers, or at least semi-series, or non-series but built based on series 🙂confused: ). The standard crossovers had been proven to conform very well with calculation. The non-standard ones, I don't know how to cross check, that's why I want to do this thing
. But I always do a side by side comparison, so what pleases my ears better should be kept, no matter what. Right? I got an 'evaluation' version of Microsim from here:
http://www.ee.mtu.edu/faculty/rzulinsk/pspice.htm
which is more than capable of simulating R/L/C circuits, if you can bear a rather big download...
Cheers
IH
http://www.ee.mtu.edu/faculty/rzulinsk/pspice.htm
which is more than capable of simulating R/L/C circuits, if you can bear a rather big download...
Cheers
IH
Thanks IH,
So such a software can be downloaded for free... But it seemed that everyone is trying to download!
I'll wait for half an hour..... 😎
So such a software can be downloaded for free... But it seemed that everyone is trying to download!
I'll wait for half an hour..... 😎
PSPICE is my favorite simulator, but it's incapable of handling real world impedance/acoustic measurements (last I heard).
Try Crossover Simulator, it is a general circuit simulator -- similar to SPICE. It's mentioned in the FRD Group site btw.
Enjoy,
Isaac
Try Crossover Simulator, it is a general circuit simulator -- similar to SPICE. It's mentioned in the FRD Group site btw.
Enjoy,
Isaac
or.....
Speakerworkshop at: http://www.speakerworkshop.com/
I've never used SPICE. The speakerworkshop download is much smaller and will take real world frequency, impedance and phase measurements (with the oscar winning Eric Walin impedance jig 🙂
Dave.
Speakerworkshop at: http://www.speakerworkshop.com/
I've never used SPICE. The speakerworkshop download is much smaller and will take real world frequency, impedance and phase measurements (with the oscar winning Eric Walin impedance jig 🙂
Dave.
f4ier said:
Dave said:
Hmmm, I think my original effort to model crossovers is gonna be useless. I've been building my crossovers by ears (with a little help from calculator). The crossovers is built with the drivers in the box and the box is on the speaker stand. IMO, unless using a very professional speaker design tools and expertise, designing by ears will still be better (at least for me). Accoustic properties of the enclosure just make the simple RLC formulas do not work!
Wonderfull designs I've seen were designed by ears. In my efforts, I used several 100n MKP-10 (sometimes down to 10n) to tune the correct capacitance. If I crossed the woofer at about 1K3 and crossed the tweeter at about 5K, and, to make things more complicated, compensated everything with non-standard procedure, but then I got a beautifull sound (perfect transient, no cancellation, all details are there, no fatigue, good soundstage, and more importantly, suit my taste), what can a PSPICE tell?
I think there must be someone out there, an expert in speaker design who gets used to speaker design by ears, and able to incorporate the modern tools in to the process. IMO, the techniques used in the process itself will be more than interesting.
If wonderful, open, forward, sweet, nice, warm... sounding speakers is the design goal, then adjusting component values by ear is ok. But if accurate music reproduction is the design goal, you really have to employ SPICE-like software which can handle driver-in-a-box acoustical/impedance measurements.
Our ears do such a wonderful job of 'equalizing' the sound we hear that even cheap and nasty speakers would start sounding good after just a few minutes/hours of listening [to them]. Take BOSE fans for example, I bet they wouldn't be so excited about their speakers after they see lab measurements (of BOSE speakers).
You can see both time domain and frequency domain data. Tools like SoundEasy, LspCAD, Crossover Simulator, SpeakerWorkshop etc. can present both time domain and frequency domain data. Personally, I'd train my ears to accurate speakers rather than someone else's idea of good sounding speakers. I may have an idea of what sounds good, but still... The fact remains that one gets the idea from someone else's design(s).
If you've seen one of John Kreskovsky's Transient Perfect designs, you'll surely agree with me that there is no way that such crossovers could be designed by ear. Drivers-in-the-box impedance/acoustical measurements must be taken into consideration when designing filters... Again, if accuracy is paramount.
Cheers 🙂
Isaac
f4ier said:
I don't think that such softwares are freewares. Even so, it would be way above my capacity.
My goal has been always ENJOYABLE SOUND.
Oh! So many people dump good gears and use the one that they think is better just because it is expensive! And now because of measurements?
I think you hit the point there. I myself had always been sinical when I heard advices such as to do a listening test when buying amps or speakers. I just don't know how many many years of experience required to not to be fooled by ears.
Yes, there is such a first/fake impression of good sound. Tube amps for example. They are accurate (in term of the sound), good soundstage, etc. But never yet made me enjoy the music. For me, enjoyment is the true measure.
I have been spending my time with listening and tweaking of audio gears for years. As a top-to-bottom type of a guy, there had been important question I had tried to answer: WHAT IS IT MAKES US HUMAN ENJOY THE MUSIC? (Please note that good sound here is enjoyable sound).
There's an answer to the question above that to me year by year become more convincing: TRANSIENT (Sometimes I call it sonic or dynamics). Hence my direction is to go for perfect transient with acceptable sacrifices (Aren't we living in an imperfect world? And so is the audio world, especially for the unwealthy guys like me). The primary trade-off, but is very critical is the distortion.
May be you have owned or designed systems that do not have much trade-offs. But let's look into one of John K speaker designs, the one with Vifa P17WJ. Everyone who is familiar with this mid-woofer know that this driver have a bad reputation in the capability to reproduce ultra fine (transient) details. I don't think that first class audio systems can help much when a low class speaker is used within the system. I think you will agree that his crossover is very "accurate", but the 1mH8 and resistors and zobel, do not help for transient. What I agree is that the speaker has been designed with sealed enclosure, even thought the driver by the manufacturer had been intended for vented system, as what many others will likely to do.
Cheers 🙂
This may be true for some, but most BOSE owners that I've met/seen are really convinced that the sound is high-end.
JohnK's P17WJ/D35AG design was not transient perfect, that was an old design. A speaker can have flat frequency response, but it does not mean that it is transient perfect. It can be TP and it follows that it has a flat frequency response.
JohnK's Transient Perfect project is a recent one, but John decided to take down most of his online pages -- including the TP project details -- (for personal reasons). That was what I was refering to when I mentioned transient perfect. The accompanying filters of his TP design(s) was quite complex -- and it included 5 drivers/enclosure (if memory serves me well) -- more elaborate than the one in his old P17WJ/D35AG project. The complex filter on that one made for a very good square wave response and transient response -- curves that are impossible to "measure" by ear.
Regards 🙂
Isaac
- Status
- Not open for further replies.