Win ISD help please.

Hi Guys, I finally got around to trying out the win ISD program and attempted to model the design I posted a while back. I guess one of the questions I have is what am I trying to achieve? What does a good plot curve look like compared to a bad one? I'm posting 2 pictures of the plot, one showing the port and one showing the box volume. Also if you have any idea on what the end correction thing means on the port window I'd be interested in knowing what that is too. I added the cabinet design so you know what I was talking about.
Thanks.

If I'm seeing what you see, there is really no difference. Scale at 20hz and at 50hz and also at 200hz are virtually identical. I don't know what this software is, but if is a curve to see what has been gained by porting the enclosure, there has been no change. Or the test method isnt set up correctly. Just my guess.

Quote:

---

Originally Posted by gvasale

If I'm seeing what you see, there is really no difference. Scale at 20hz and at 50hz and also at 200hz are virtually identical. I don't know what this software is, but if is a curve to see what has been gained by porting the enclosure, there has been no change. Or the test method isnt set up correctly. Just my guess.

---

It's a program where you put in your drivers parameters and your box parameters and your port size and the program plots out the performance of the three as a unit....I think. What your seeing is the same one I just switched windows so you could see the parameters of the box volume and the port settings. my question is what is an optimum looking curve and what is the significance of the -3db range? What would an optimum curve look like? or am I not trying to shoot for an optimum curve?

Good to see you are starting to take a more measured approach to this ( rather than just building boxes & hoping for the best ) .

Ultimately ( after playing with all the permutations & combinations / ie; alignments or tunings ), you're going to find that all 604 variants, generate the most extended bass response when used in a large enclosure ( ie; in a Bass Reflex box that means up to about 9.5 cu ft and typically no smaller than 7 cu ft ).


- Yes, you are looking for a "good" curve that extends well down into the LF areas .
- Your curve can be considered "bad" , if you are expecting reasonable bass response .
- OTOH, your curve can be considered "good" / if there's a "helper" woofer handling the bass duties .

I consider the following curve "good" / especially considering the size of box / as well as after factoring in a bit of LF room gain ;

http://www.hostboard.com/forums/hbmc.../2010/08/2.png


cheers

Why don't you use the on line CAD here Loudspeaker enclosure calculating with Thiele Small parameter
It's hard to see the graphs on your post, What speaker are you using? what is Vb?
....GC

Earl pretty much nailed it. Did you know that you can drag the curve (a point on the curve, actually) to have a better alignment? So.....if you want the rolloff of the bass to be steeper, you can choose the -3db point and drag that to the left side of the screen.

Two tips: Make sure you have the exact T/S parameters of your driver entered into the program, don't rely on the Winisd params. Try increasing or decreasing box sizes and alignment numbers to get a port that will will be friendly with your drivers and allow for the best rolloff characteristics. The 172" port length is not going to work.

Good luck with your enclosures.

So what Earl is saying is if I want good Bass response I want the curve to cross the -6db line at as low a frequency as it can? Right?... That is if I want the house to jump off of it's foundation...which I do. : )

Quote:

---

Originally Posted by westend9

Earl pretty much nailed it. Did you know that you can drag the curve (a point on the curve, actually) to have a better alignment? So.....if you want the rolloff of the bass to be steeper, you can choose the -3db point and drag that to the left side of the screen.

Two tips: Make sure you have the exact T/S parameters of your driver entered into the program, don't rely on the Winisd params. Try increasing or decreasing box sizes and alignment numbers to get a port that will will be friendly with your drivers and allow for the best rolloff characteristics. The 172" port length is not going to work.

Good luck with your enclosures.

---

Thanks Westend, They didn't have the 604 in the list so I had to put it in manually. I used the T/S parameters from the altec literature and filled in as many as I had. The only one that might be iffy is when I described the speaker magnet size I guesstimated it at 6 inches in dia and 7 inches long. I had read about the dragging of the curve but hadn't tried it, makes good sense though.

I think it's great that you're going to build some new cabinets for your 604's--you'll need to build larger cabinets than the black ones you're currently using to get the most out of those drivers. From your previous posts I recall that cabinet size was a concern due to the amount of space there is in your listening room--sounded to me like you wouldn't be comfortable with cabinets that are larger than the Heathkit cabs you own (that are currently gathering dust?). I would recommend that you modify your Heathkit cabinets and compare their sound to the new cabs that you build--you might be surprised that the Heathkits sound better than the ones that are currently on the "drawing board" and it wouldn't be much work to "fix" those Heathkit cabs. You might even want to consider modifying the Heathkit cabinets first--and you may decide you don't need to build a new pair. From your previous posts it sounds like the tuning of your Heathkit cabinets is WAY off. If they were mine, I'd first remove the "shelves" that someone added--that would get them closer to their "stock" configuration. Then, I'd reduce the port size (most likely by quite a bit) either by using your computer program to determine the optimum port size and/or fine tuning by ear (or even designing an adjustable port). I think those Heathkit cabs, with the right port size and no "shelves" would sound amazing......

Thanks Vott....and yes they are collecting dust....in the kitchen. :( You wouldn't believe me...well you probably would but this morning I thought there was something wrong with one of the speakers so I tried adjusting the balance and realized that it was way over to one side....I'm not sure when that happened but I'm betting it was way back when I switched to the big square ports. I remember switching it back and forth between a big square port and the original round hole I had put in them. needless to say i think all this time it was off. I also noticed that ever since I switched the ports it didn't do any soundstaging like they used to...I think it was because I had the balance way off. I can be rather retarded at times....well most of the time. Now I'm gonna have to do more listening to see what I really have.....they do sound a lot better now though. :D

Quote:

---

Originally Posted by cradeldorf

Thanks Westend, They didn't have the 604 in the list so I had to put it in manually. I used the T/S parameters from the altec literature and filled in as many as I had. The only one that might be iffy is when I described the speaker magnet size I guesstimated it at 6 inches in dia and 7 inches long. I had read about the dragging of the curve but hadn't tried it, makes good sense though.

---

WinISD, unless an updated version has been introduced off late, does not take into account all of the T/S parameters except for Fs, Qt & Vas to give the projections. Ofcourse it uses Qes & Pe for the vent-mach calculations. But this way it just gives a rough idea of the enclosure size and vent sizes. But it really misses out on the sim-ed projection of the FR as it does not use Le. Its nowhere near the true one. So only for a rough Vb idea its good. One can play with as many drivers as possible and very quickly will find/learn the relationship between Vas, Qt, Fs and the different resultant cab sizes. WinPRO, its cousin, is far better, but is very unstable to use and also probabely not free (?)

Aditya

Quote:

---

Originally Posted by cradeldorf

Now I'm gonna have to do more listening to see what I really have.....they do sound a lot better now though. :D

---

Great! I don't know what the internal volume of those black cabinets is but of course the bottom line is how good do they sound to you/in the acoustical environment of your listening room. If, with the new ports (and proper balance setting) they sound great, you can always go by the old rule "if it ain't broke, don't fix it"! I'd still do my suggested tweaks to those Heathkit cabs and compare them to the black ones--it would be a quick fix and you might be blown away by the sound of those things with the "shelves" removed and the adjusted "correct" port size....

Quote:

---

Originally Posted by aditya

WinISD, unless an updated version has been introduced off late, does not take into account all of the T/S parameters except for Fs, Qt & Vas to give the projections. Ofcourse it uses Qes & Pe for the vent-mach calculations. But this way it just gives a rough idea of the enclosure size and vent sizes. But it really misses out on the sim-ed projection of the FR as it does not use Le. Its nowhere near the true one. So only for a rough Vb idea its good. One can play with as many drivers as possible and very quickly will find/learn the relationship between Vas, Qt, Fs and the different resultant cab sizes. WinPRO, its cousin, is far better, but is very unstable to use and also probabely not free (?)

Aditya

---

The Program I have is win ISD pro. and it was free.

Win ISD & Win PRO are the two I knew. You can check its limitations by varying/ommiting the Le value, an see. If it is affected then it is the better one.

Aditya

Greets!

Ideally you want a response curve that most closely mirrors your room's acoustics that combined yields a flat response throughout the listening area over whatever BW you design to.

Obviously, without extensive room measurements one is left just picking a response that the driver can handle in the max acceptable bulk, so the main thing is to not under-damp it, i.e. have excessive peaking at Fb or require too large, long a vent if vented such as you posted. A good rule-of-thumb (ROT) for a max vent length: 13,560"/(20*Fb)


The vent's end correction refers to how it is terminated, i.e. well away from any boundary or flanged at one or both ends:

Flanged-Flanged: 0.850
Flanged-Free: 0.732
Free-Free: 0.613


Flanged is flush with an enclosure wall and free is not, so for most all vented cabs it should be 0.732.

The reason for it is due to the pipe opening having an elastic component to it that makes it act like an acoustic version of a passive radiator (PR). As such, the mass air plug in the vent extends outside it a small amount based on its diameter and whether or not there's a baffle around it to increase its acoustic efficiency, so the end correction calculation is used to shorten the vent's physical length. Indeed, most simming programs treat it as a PR with a perfectly linear excursion, though a few $$$ ones also models vent non-linearity if it's too small for the power input.

WinISD Pro tutorial: WinISD Pro Tutorial and Download (A detailed guide on how to use WinISD Pro) - Home Theater Forum and Systems - HomeTheaterShack.com

Additional WinISD Pro driver database downloads: Downloadable WinISD Pro Files - Home Theater Forum and Systems - HomeTheaterShack.com

GM

Quote:

---

Originally Posted by voice of the theater

Great! I don't know what the internal volume of those black cabinets is but of course the bottom line is how good do they sound to you/in the acoustical environment of your listening room. If, with the new ports (and proper balance setting) they sound great, you can always go by the old rule "if it ain't broke, don't fix it"! I'd still do my suggested tweaks to those Heathkit cabs and compare them to the black ones--it would be a quick fix and you might be blown away by the sound of those things with the "shelves" removed and the adjusted "correct" port size....

---

I figured the cabinet volume to be around 5.75 cf. I just put in the Pink Floyd Pulse DVD and I'm actually going to step out of my realm and describe it as serene, divine, titillating and absolutely mesmerizing. they have not sounded this good in a while. it was literally fantastic. I noticed right after putting the ports in that the volume had to be up more than it used to be....well to make a long story short, It doesn't need to anymore. :2thumbsup: The bass was just thundering at -11 volume.

Quote:

---

Originally Posted by aditya

Ofcourse it uses Qes & Pe for the vent-mach calculations. But this way it just gives a rough idea of the enclosure size and vent sizes. But it really misses out on the sim-ed projection of the FR as it does not use Le.

---

The pro version's vent calcs are more accurate than the basic WinISD, matching Hornresp's predictions using wave equations plus it sims Le's effect on response. Both fall a bit short of MJK's MathCad software though since it takes any damping into consideration. AkAbak is free and probably the most accurate of all, but can require lots of inputting till you get a database built up and the learning curve is ballistic steep for many folks.

GM

Quote:

---

Originally Posted by GM

Greets!

Ideally you want a response curve that most closely mirrors your room's acoustics that combined yields a flat response throughout the listening area over whatever BW you design to.

Obviously, without extensive room measurements one is left just picking a response that the driver can handle in the max acceptable bulk, so the main thing is to not under-damp it, i.e. have excessive peaking at Fb or require too large, long a vent if vented such as you posted. A good rule-of-thumb (ROT) for a max vent length: 13,560"/(20*Fb)


The vent's end correction refers to how it is terminated, i.e. well away from any boundary or flanged at one or both ends:

Flanged-Flanged: 0.850
Flanged-Free: 0.732
Free-Free: 0.613


Flanged is flush with an enclosure wall and free is not, so for most all vented cabs it should be 0.732.

The reason for it is due to the pipe opening having an elastic component to it that makes it act like an acoustic version of a passive radiator (PR). As such, the mass air plug in the vent extends outside it a small amount based on its diameter and whether or not there's a baffle around it to increase its acoustic efficiency, so the end correction calculation is used to shorten the vent's physical length. Indeed, most simming programs treat it as a PR with a perfectly linear excursion, though a few $$$ ones also models vent non-linearity if it's too small for the power input.

WinISD Pro tutorial: WinISD Pro Tutorial and Download (A detailed guide on how to use WinISD Pro) - Home Theater Forum and Systems - HomeTheaterShack.com

Additional WinISD Pro driver database downloads: Downloadable WinISD Pro Files - Home Theater Forum and Systems - HomeTheaterShack.com

GM

---

Awesome Greg, Thanks for the explanation. I'll go through the links and get a better Idea of how it all plays together. :2thumbsup:

Quote:

---

Originally Posted by GM

The pro version's vent calcs are more accurate than the basic WinISD, matching Hornresp's predictions using wave equations plus it sims Le's effect on response. Both fall a bit short of MJK's MathCad software though since it takes any damping into consideration. AkAbak is free and probably the most accurate of all, but can require lots of inputting till you get a database built up and the learning curve is ballistic steep for many folks.

GM

---

Yes, the PRO version takes Le into account. But it was a little unstable with the simulation part & the data entry part. May be its time to check up again for the latest version. I have been using Bass Box PRO6 for quite sometime now. I have also used Hornresp, and yes it is good. I read about AkAbak from one of your earlier posts, and infact today only :) while browisng the forum's older threads, and was thinking of giving it a try. AkAbak used to cost around 700$ 4/5 yrs back, and now its free ! Will definitely try to learn to use it. It wiil pay back.

Thanks
Aditya

Hi GM,

I visited the WinISD Pro site, but my version seems to be the latest one. There are two types there:- WinISD (Beta)v.44, and WinISD Pro (Alpha) v.7.

About WinISD Pro the Linear site says this (just pasted from the site):
"NOTE This version is early ALPHA version for preview and test purposes only. If you are looking stable tool, please download WinISD 0.4x series instead. Because of almost complete rewrite of software, there is many things which do not work like they should. All comments about this _ALPHA_ version are very wellcome"

They actually recommend the basic version for stable performance. This is what I was pointing at.
And I just confirmed that it does not take Le into consideration. It just shows the projected FR flat upto 20K (or whatever one sets at).

I have downloaded AkAbak, and all its relevant documents (whatever were available). This is a MS-DOS 16bit thing, and I hope it will run on XP, as the testing was upto Win2k. It seems the learning curve is steep mostly due to the dos format of data entry. A bit of relief here for me, as I was seriously into programming for 2/3 yrs while trying to develop a software for my personal use, and as a result keyboard habbit:). Let me see......

Regards
Aditya

Greets!

Correct, there's only the one Pro version published AFAIK.

WRT its stability, the only thing I've noticed (and can be a little frustrating/irritating), is in either moving windows around or moving from the design window to plot window and vice versa, but thousands of folks around the world including me find this a small price to pay for an easy to use, accurate enough freeware program to make a basic sealed, etc. cab design, and with its basic filter designer, excluding the now freeware AkAbak, it's a cut above the rest.

Hmm, don't know why you're not seeing any change WRT Le, but mine displays it in all the appropriate plots and I've seen its impact on too many posted plots across many DIY forums for it to be a fluke.

Excepting pure DOS programs, AkAbak and all other designers I've tried seem to work fine for me with XP Pro SP2 with all security updates.

GM

Happy to know about the AkAbak-XP compatibility.
I will again check up the Le issue of my WinISD Pro. Now it is clear something is not right at my end. I will do a fresh installation.

Many Thanks
Aditya

It runs flawlessly on my win xp sp2. : /