WinISD and Bassbox Pro - different simulations

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Ive experienced over time now that WinISD and Bassbox Pro is generating different simulations for the very same drivers. Take the JBL 2226H for instance. Same driver, same closed box Qt below.

Which software should I trust?
 

Attachments

  • Untitled-1.jpg
    Untitled-1.jpg
    147.2 KB · Views: 681
  • Untitled-2.jpg
    Untitled-2.jpg
    54.7 KB · Views: 656
Given the VAS of that driver is quite high, I'd be rather sceptical of any software suggesting 15L.

My first thought is that maybe it's a regional settings issue - I think it looks weird using a comma for a decimal seperator, so maybe Bassbox does too ;)

Mind that the simulations above is for a closed box. But in a vented box, the two softwares gives different max spl, and vent length. Both in the exact same box size and vent diameter.
 
Well, the WinISD beta sim is essentially an undamped sim (the simple equation I give above). Bassbox uses a more comprehensive model, but if you don't choose the proper values for Qa and Ql (and the defaults in the program might be bad) you will get a box that is very small. For starters, try Qa=Ql=10 to 20.

For the 2226H, for example, if you stuffed the box you might be able to get a Qtc of 0.7 in 25 Liters, where if there were only lining on the walls, you would need 35. This is where an undamped sim gets you an answer of 41.6. I don't think there is much support for 14.6L...
 
Well, the WinISD beta sim is essentially an undamped sim (the simple equation I give above). Bassbox uses a more comprehensive model, but if you don't choose the proper values for Qa and Ql (and the defaults in the program might be bad) you will get a box that is very small. For starters, try Qa=Ql=10 to 20.

For the 2226H, for example, if you stuffed the box you might be able to get a Qtc of 0.7 in 25 Liters, where if there were only lining on the walls, you would need 35. This is where an undamped sim gets you an answer of 41.6. I don't think there is much support for 14.6L...

Didnt quite get all that. The equation I can do, but that would only give me the volume needed for a given Qt value. It would be great to have a software that could draw an accurate spl graph, cone displacement and so on... As it seems to me now, neither WinISD or Bassbox can be trusted :(
 
The more accurate the sim, the more input parameters you will need...:D

The way you have posted, we don't even know if you are using the same parameter data for each program. Set Qa=Ql=30 or so (or 9999) and see how Bassbox compares...

A sealed box has two loss terms - Qa (absorption due to stuffing, etc) and Ql (due to leakage). To get an accurate sim you need to measure these.
 
The more accurate the sim, the more input parameters you will need...:D

The way you have posted, we don't even know if you are using the same parameter data for each program. Set Qa=Ql=30 or so (or 9999) and see how Bassbox compares...

A sealed box has two loss terms - Qa (absorption due to stuffing, etc) and Ql (due to leakage). To get an accurate sim you need to measure these.

In bassbox I have typed in all the parameters in expert mode. In WinISD ive typed all that is possible. Identical to the parameters typed in Bassbox. Allthough Bassbox is a lot more comprehensive. Where do you set these Qa and Ql values? In Bassbox I can allready choose how much stuffing I want, but that doesnt change the fact that it gives wrong simulations with both sealed and ported enclosures with every driver Ive tried. The same goes for WinISD :(
 
Last edited:
In bassbox I have typed in all the parameters in expert mode...that doesnt change the fact that it gives wrong simulations with both sealed and ported enclosures with every driver Ive tried. The same goes for WinISD :(

If you don't show evidence for this, you are just apparently complaining about the discrepancy between the programs, which has already been explained.
 
If you don't show evidence for this, you are just apparently complaining about the discrepancy between the programs, which has already been explained.

Its all over the page 1... But I get it as WinISD gives in general an oversized volume, and that Bassbox must have its so called Qa or Ql values manually configured to function properly?

Is it this the right configuration window?
 

Attachments

  • Untitled-1.jpg
    Untitled-1.jpg
    182.4 KB · Views: 465
Its all over the page 1... But I get it as WinISD gives in general an oversized volume, and that Bassbox must have its so called Qa or Ql values manually configured to function properly?

Is it this the right configuration window?

What I would do is configure Ql so that it is 10 for less than 50Liters and 7 for >200

Then notice the configuration to the right of that, where you have normal fill and heavy fill, change Qa to 30 and 15 respectively. I would reduce the gamma number for normal fill to something more like 1.05 and for heavy to 1.1. Change gamma for minimal fill to something like 1.02.

This should straighten things out for you quite a bit. I would probably just make gamma equal to 1 for all settings if I were using the program...but start with what I describe above and see what you think.
 
What I would do is configure Ql so that it is 10 for less than 50Liters and 7 for >200

Then notice the configuration to the right of that, where you have normal fill and heavy fill, change Qa to 30 and 15 respectively. I would reduce the gamma number for normal fill to something more like 1.05 and for heavy to 1.1. Change gamma for minimal fill to something like 1.02.

This should straighten things out for you quite a bit. I would probably just make gamma equal to 1 for all settings if I were using the program...but start with what I describe above and see what you think.

Done :) The ported simulations now is coherent with both softwares. The closed simulations is better, but still not very good. The amplitude response variations is about +/- 3db. I assume WinISD dont take stuffing into account when calculating volume needed for a given Qtc?

Here are Eminence BP102 and JBL 2226H simulations with a Qtc of 0,7.
 

Attachments

  • Untitled-1.jpg
    Untitled-1.jpg
    84.7 KB · Views: 437
  • Untitled-2.jpg
    Untitled-2.jpg
    139 KB · Views: 419
Done :) I assume WinISD dont take stuffing into account when calculating volume needed for a given Qtc?

Correct, WinISD "beta" does not take damping or leakage into account - it uses the equation I posted. If you change the default for closed boxes on that configuration page to "no fill", the answer will be closer to WinISD, which is probably not a good thing. WinISD "pro" uses a similar model to bassbox.

You can tweek the Qa/Ql values in Bassbox or WinISD pro based on your own experience and fill techniques. The effects will be on the impedance curve, but you need to be careful about your measurements in that the driver should be seeing approximately the same acoustic environment when parameters are measured as it will in the box.
 
I dont get the same result as you with Bassbox. Even when I modeled a complete stuffed box, I got a higher Vb.

I dont trust WinISD, just because it fails if you try to simulate som "odd" drivers with parameters a little beside the "normal".

If you want to test this two against another "simulator" I personally like, check out the Unibox 4.07 (excel app).
 
I dont get the same result as you with Bassbox. Even when I modeled a complete stuffed box, I got a higher Vb.

I dont trust WinISD, just because it fails if you try to simulate som "odd" drivers with parameters a little beside the "normal".

If you want to test this two against another "simulator" I personally like, check out the Unibox 4.07 (excel app).

You dont get the same results pre or post the Qa/Ql tweaks?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.