Differents results from differents simulation sofware

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi, I'm trying to understand, for my knowledge, how to do bass reflex sim. I've find formulas here: The Subwoofer DIY Page - Ported Systems: Frequency Response Calculations (Thank you Brian to share this)

I've put formulas into Excel and compare with others software. Excel gives me same results as BassBox PRO and WinISD. But if I compare to Hornresp and Akabak, I've seen some differences, sensitivity and fréquency response. Akabak and Hornresp gives similar results.

Here copy screen of Hornresp, Excel, WinISD. I've made perfect flat curve in Hornesp with Beyma Driver 18P1200Nd, ( http://profesional.beyma.com/ingles/pdf/18P1200NdN.pdf ) Vb 110 liters, Fb 50.14Hz, then same sim in WinISD and Excel (with Ql=7):

Hornresp sim:
An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.



Excel:
An externally hosted image should be here but it was not working when we last tested it.



WinISD:
An externally hosted image should be here but it was not working when we last tested it.


As you can see, differences in WinISD and Excel are : little bump , F3 is higher, sensitivity is lower by 3.5dB.

So, Which software say the truth?
 
Last edited:
Hi Bjorno, I have include the input screen of Hornresp in my first post, and repost all the frequency response, there was a little error, but nothing change: Akabak and Hornresp have the same flat response (or very similar), same F3, but WinISD, excel formula, and even Bassbox show little bump, higher F3, lower sensitivity.
 
Hi.

I would put more trust in software designed to create reflex cabinets if that is the goal than, software designed to create horns. Upon looking at the parameters required for horn response to draw to its conclusion, it is evident that it is not using the traditional parameters used to create a reflex cabinet.
 
Hi Djim, it's very simple and quick to makes a comparison between Hornresp and WinIsd, even with UNIBOX which gives similars results as WinISD and BassBox. So everybody could see the problem by himself, it will be easier.

I've tried to use SPL voltage in excel. It's improves the comparison a little, but not totally and not for all drivers.

I don't see why software like Hornresp or Akabak couldn't do bassreflex sims.

If David McBean read this thread and can tell us if his software can do Bassreflex sims with accuracy or not? I will be interesting also to know why these differences.
 
I would put more trust in software designed to create reflex cabinets if that is the goal than, software designed to create horns.

Hi OMNIFEX,

Hornresp uses a more sophisticated bass-reflex simulation model than WinISD.

While I don't know for sure, I would expect that in theory the Hornresp predictions should be the more accurate.

AkAbak and Hornresp use essentially the same model.

Kind regards,

David
 
Hi David.

I have a lot of experience with Eminence Designer that is a light version of Harristech’s Bass Box pro in which, sebDIY mentioned. I’ve compared Bass Box Pro to Eminence designer and the results are exactly the same.

The problem with Horn Response is that it does not offer the traditional parameters used for designing a reflex cabinet. If I need to convert the parameters in Eminence Designer in order for Horn Response to recognise the figures it is requesting, why would I bother with Horn Response to design a reflex cabinet?

Possibly, it would be better for Horn Response to have compatible parameters that reflect what manufactures offer when they sell raw drivers if Horn Response is looking to expand into reflex designs.

Cheers!
 
Hi OMNIFEX,

I’ve compared Bass Box Pro to Eminence designer and the results are exactly the same.

That's because they both use the same "simplified" simulation model :).

The problem with Horn Response is that it does not offer the traditional parameters used for designing a reflex cabinet.

The driver electro-mechanical parameters used by Hornresp are in fact the traditional ones. Thiele-Small parameters are relative new-comers in this regard :).

If I need to convert the parameters in Eminence Designer in order for Horn Response to recognise the figures it is requesting, why would I bother with Horn Response to design a reflex cabinet?

You probably wouldn't, unless you were interested in checking the predictions against those generated by your own software, and noting any significant differences - knowing that Hornresp (and AkAbak) use more advanced simulation models.

Rest assured, I am not trying to make a case for Hornresp to be used in lieu of other loudspeaker simulation programs to analyse bass-reflex designs. I am simply attempting to address the concerns expressed by sebDIY and yourself regarding the ability of Hornresp to effectively model bass-reflex loudspeakers.

Possibly, it would be better for Horn Response to have compatible parameters that reflect what manufactures offer when they sell raw drivers

In Edit mode, click on the Cms input box and see what happens :).

Kind regards,

David
 
Last edited:
Hi OMNIFEX,



That's because they both use the same "simplified" simulation model :).



The driver electro-mechanical parameters used by Hornresp are in fact the traditional ones. Thiele-Small parameters are relative new-comers in this regard :).

Fair enough!

Well let me rephrase it another way :D

If the majority of manufactures that sell raw loudspeakers to the masses provide parameters that reflect the Thiele-Small formula, which you consider is the new way of doing things, why can you not insert those figures in Horn Response and Horn Response calculates those figures into the old traditional way?

When you say Horn Response offers more advanced simulation models, what makes it more advanced?

I focus primarily on Pro Audio. I do understand that there may be some things that might be beneficial for those building subs for their homes. Ron E mentioned inductance, which would be helpful under home audio conditions. However, since I will be using an electronic crossover and setting my own crossover points, the inductance would not matter under pro audio conditions since we tend to filter out anything above 100 Hz.

So it would nice if you can expand further on Horn Response offers more advanced simulation models when designing a reflex enclosure.

I will need to download your latest version for it appears the last version I've downloaded was from 2010. :D
 
Hi OMNIFEX,

If the majority of manufactures that sell raw loudspeakers to the masses provide parameters that reflect the Thiele-Small formula, which you consider is the new way of doing things, why can you not insert those figures in Horn Response and Horn Response calculates those figures into the old traditional way?

I could, but I am not going to :).

Being a traditionalist, my preference is to specify a driver using the standard fundamental lumped-element electro-mechanical parameters. The only concession I am prepared to make is to include the Calculate Parameter conversion tool in Hornresp.

The original Hornresp program was written before Messrs Thiele and Small came up with their filter-based parameters anyway :).

So it would nice if you can expand further on Horn Response offers more advanced simulation models when designing a reflex enclosure.

All the bass-reflex simulation programs that I am aware of consider the enclosure as a simple acoustic compliance, and the port tube as a simple acoustic mass.

With the "resonances unmasked" option selected in Hornresp, the required acoustical impedances, volume velocities and pressures are all calculated by undertaking a rigorous analysis of the complete electro-mechano-acoustical equivalent circuit, without making any simplifying assumptions.

Hornresp (and AkAbak) can show resonances due to reflections within the enclosure, where the "box" simulation programs that I have seen, can not.

Kind regards,

David
 
I find entering driver parameters in Hornresp relatively simple.
The worst sim program for entering driver parameters I have used is WinISD Pro which unless the help file procedure is closely followed fails to accept the driver parameters.:mad:
In my builds I have found Hornresp reflex port lengths to be closer to the measured response than WinISD Pro.:)
The feature I find lacking with Hornresp is the ability to simulate response and driver excursion with a hi pass filter. I think that this is a essental part of the design process when trying to build a high power cabinet, it is the one thing that keeps me using WinISD as well as Hornresp when simulating a reflex cab.
Regards
Martin (Xoc1)
 
Hi sebDIY,

Maybe the difference in output level is because winISD uses Watts and HornResp/Akabak uses Voltage. .
+1
One difference, the simple simulators do not model the effects of voice coil inductance, which will tend to cause a bump above tuning and a roll off above that.
+1
Hi OMNIFEX,
In Edit mode, click on the Cms input box and see what happens :).
Kind regards,
David
or after entering the hr parameters,double click "sd" ,then fs,qes,qms,vas and qts is presented for you to check .
The feature I find lacking with Hornresp is the ability to simulate response and driver excursion with a hi pass filter. I think that this is a essental part of the design process when trying to build a high power cabinet, it is the one thing that keeps me using WinISD as well as Hornresp when simulating a reflex cab.
Regards
Martin (Xoc1)
agreed ,but it seems bjorno has a solution for that.
exporting the hr wav file and importing it in to another program(i stil have to dive in to that):eek:
 
Hi OMNIFEX,



I could, but I am not going to :).

Being a traditionalist, my preference is to specify a driver using the standard fundamental lumped-element electro-mechanical parameters. The only concession I am prepared to make is to include the Calculate Parameter conversion tool in Hornresp.

The original Hornresp program was written before Messrs Thiele and Small came up with their filter-based parameters anyway :).



All the bass-reflex simulation programs that I am aware of consider the enclosure as a simple acoustic compliance, and the port tube as a simple acoustic mass.

With the "resonances unmasked" option selected in Hornresp, the required acoustical impedances, volume velocities and pressures are all calculated by undertaking a rigorous analysis of the complete electro-mechano-acoustical equivalent circuit, without making any simplifying assumptions.

Hornresp (and AkAbak) can show resonances due to reflections within the enclosure, where the "box" simulation programs that I have seen, can not.

Kind regards,

David

Cheers David.

I am going to give it a go and see how well horn response matches the measured results of cabinets I’ve designed years ago designed in Eminence Designer.

What is the resolution of the frequency graph? Is there a way to adjust the resolution so it is not so smooth when evaluating frequencies?

That is the biggest gripe I have with all simulators...they offer too much smoothing.
 
Last edited:
The feature I find lacking with Hornresp is the ability to simulate response and driver excursion with a hi pass filter. I think that this is a essental part of the design process when trying to build a high power cabinet, it is the one thing that keeps me using WinISD as well as Hornresp when simulating a reflex cab
+1, hint, hint ;)

Best regards Johan
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.