Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Software Tools SPICE, PCB CAD, speaker design and measurement software, calculators

VituixCAD
VituixCAD
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 10th July 2019, 09:56 AM   #1331
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
^I guess it is Pavlov's dog?
__________________
Radikal aktivist AINOgradient speaker project
  Reply With Quote
Old 10th July 2019, 04:18 PM   #1332
kimmosto is offline kimmosto  Finland
diyAudio Member
 
kimmosto's Avatar
 
Join Date: Oct 2005
Location: Kuopio
Quote:
Originally Posted by ashok View Post
Can I have that puppy ? ( In your profile picture !)
Sorry. Puppy is not so puppy anymore (7 months now) and owner is my wife - who is not Pavlov as far as I know.
  Reply With Quote
Old 10th July 2019, 04:34 PM   #1333
kimmosto is offline kimmosto  Finland
diyAudio Member
 
kimmosto's Avatar
 
Join Date: Oct 2005
Location: Kuopio
Quote:
Originally Posted by Denri View Post
Where in merger or in impulse response converter these -20db plays role and should be set?
Nowhere if you're using ARTA, because measured level divided reference channel level stays constant in dual channel mode when volume is decreased with ARTA's signal generator. So magnitude scaling of both LF and HF part could be 0.0 dB despite of different signal levels.

My video lessons and instructions could have some details which are not totally compatible with ARTA because measurement data is usually captured with CLIO 10-12. Some mistakes are also possible, but I will try to find and fix them in the next revision.
  Reply With Quote
Old 10th July 2019, 06:23 PM   #1334
Denri is offline Denri  Russian Federation
diyAudio Member
 
Join Date: Feb 2019
Location: Moscow
Kimmo, thanks
Well I see the answer is possible only if the one understands theory of measurements.
I played with your programm few hours, looks smart and intelligent and it broken my dreams about FR for different filters and real speakers. I just imagine how much time it would save If send it to past. You done great thing.
  Reply With Quote
Old 16th July 2019, 09:24 PM   #1335
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Had a quick play with the enclosure simulator tool today for the first time and I have to say I'm a bit puzzled, and not sure if I've encountered a bug or two. I tried simulating a driver I actually have - a Visaton W300S 8 ohm which convieniently is already in the database.

With a 125 litre bass reflex enclosure and 25Hz port tuning I get the following:

VituixCAD-enclosure-tool-png

The thing puzzling me is the severe downwards slope in SPL with increasing frequency shown in the top left graph amounting to about 6dB down at 1Khz.

What is the SPL graph supposed to represent if not on axis response into half space ?

It looks almost as if the tool is naively using the voice coil inductance to attenuate the high frequency response assuming that the driver output is proportional to current at high frequencies - which of course it is not.

If I model the same driver in WinISD I get the expected result:

VituixCAD-winisd-png

I checked the documentation (just this once ) and note that page 28 of the english version shows an SPL graph that I would expect that is flat at high frequencies. (Given only T/S data and no high frequency measurement of the driver)

Any ideas what is going on ?

Also a second problem I've encountered is that after manipulating the values for the box/switching between closed/bass reflex etc for a couple of minutes the graphs have all "frozen" and no longer respond to any changes made on the GUI on the left side - I can even change it to closed box and it still shows the bass reflex alignment in the graphs....
Attached Images
File Type: png enclosure tool.png (202.6 KB, 316 views)
File Type: png winisd.png (72.2 KB, 316 views)
__________________
- Simon

Last edited by DBMandrake; 16th July 2019 at 09:29 PM.
  Reply With Quote
Old 17th July 2019, 05:21 AM   #1336
kimmosto is offline kimmosto  Finland
diyAudio Member
 
kimmosto's Avatar
 
Join Date: Oct 2005
Location: Kuopio
Quote:
Originally Posted by DBMandrake View Post
I checked the documentation (just this once )
User manual, in the middle of page 31:
"Voice coil inductance (with losses) is included in frequency response calculation by checking Show effect of inductance in context menu of SPL chart.

Flat SPL at HF is shown by unchecking, which might be expected by common citizen but not so much fact because inductance reduces current which reduces force and so on. Typical problem is that proper extended (lossy) inductance model parameters Le, Leb, Ke, Rss or Z1k and Z10k are not entered for the driver. Le alone is not good for SPL calculation. You can test SPL effect of advanced impedance model e.g. with Scan-Speak 10F/8424G00.
  Reply With Quote
Old 17th July 2019, 02:19 PM   #1337
kimmosto is offline kimmosto  Finland
diyAudio Member
 
kimmosto's Avatar
 
Join Date: Oct 2005
Location: Kuopio
Visaton W300S-8 seems to have Z1k=13 Ohms and Z10k=38 Ohms (instead of single Le) in driver database. Impedance curve with some semi-inductance is approximated with exponent function by those two values. Main part of Visaton data is imported from some German enclosure simulator "as is". SPL curve with simple Z model matches with that tool.
  Reply With Quote
Old 17th July 2019, 06:34 PM   #1338
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Quote:
Originally Posted by kimmosto View Post
User manual, in the middle of page 31:
"Voice coil inductance (with losses) is included in frequency response calculation by checking Show effect of inductance in context menu of SPL chart.
Whoops... I keep forgetting to check the context menus...
Quote:
Flat SPL at HF is shown by unchecking, which might be expected by common citizen but not so much fact because inductance reduces current which reduces force and so on.
I have to completely disagree here - the "show effect of voice coil inductance" setting is based entirely on a faulty premise and therefore doesn't give valid results.

Yes, impedance rise with frequency caused by voice coil inductance reduces current and therefore SPL at higher frequencies compared to no inductance, that's simple ohms law - but the effect of the voice coil inductance on SPL is already taken into account in the design of the driver and the published frequency response curve of the driver which is based on the driver being voltage driven from a low impedance souce not current driven from an infinite impedance source...

Nearly all cone drivers that you would use in a box simulator are designed to give as flat as possible frequency response with voltage drive from a low impedance source - they aren't designed to be flat when current driven.

In a well designed cone driver the increasing directivity of the driver at higher frequencies compliments the reduction in drive current due to inductance to give an overall flat frequency response on axis when voltage driven which neither trends down or up until the final cutoff frequency.

In fact without this inductance many cone drivers would actually have a rising response at high frequencies. A good example is comparing the response of the Fostex FE207 and FE206E - both drivers have the same cone design but the FE206E has a shorting ring to dramatically reduce inductance while the FE207 does not. As a result the on axis high frequency response of the FE206E is tipped up by nearly 10dB due to the very low inductance, compared to the FE207 with a more normal inductance which has a more neutral high frequency response.

By calculating the effects of impedance and then applying it to the predicted frequency response you are effecively applying this correction twice resulting in a steep downwards SPL slope that simply doesn't exist in reality. (If the driver had zero inductance the high frequency response would have actually trended up at high frequencies, at least until the high frequency cutoff of the driver)

Look no further than the manufacturers published frequency response of the W300S 8 ohm to see that - it's not 6dB down at 1Khz, that's for sure. In fact it's remarkably flat up to 1Khz for a 12" woofer, (only on axis of course) one reason I like them as a driver.

The problem with trying to predict the effects of voice coil inductance at high frequencies is that the software doesn't have all the information it needs.

It only has T/S parameters including inductance - from that it can calculate low frequency response but it knows nothing about the high frequency response of the driver as it is not modelling things like shifts in directivity that counteract and compensate for the impedance change, or any other mechanical or accoustic properties that might cause the on axis response to increase with frequency with a constant current drive. (Progressive cone decoupling is another)

In other words, the "show effect of voice coil inductance" setting is simply guessing based on a lack of information, and guessing wrong. For that reason I don't understand its inclusion let alone it being enabled by default.

A box simulator that only has Theile/Small parameters available to work from should make the assumption that the high frequency response of the driver tends towards a flat 0dB response at high frequencies when voltage fed from a low impedance source. This is what all other software I have used before does, and it's the only logical thing to do without additional data being available.

To get any more detailed than that at high frequencies, actual accoustic measurements would need to be merged with the calculated alignment response.

Quote:
Typical problem is that proper extended (lossy) inductance model parameters Le, Leb, Ke, Rss or Z1k and Z10k are not entered for the driver. Le alone is not good for SPL calculation. You can test SPL effect of advanced impedance model e.g. with Scan-Speak 10F/8424G00.
Quote:
Originally Posted by kimmosto View Post
Visaton W300S-8 seems to have Z1k=13 Ohms and Z10k=38 Ohms (instead of single Le) in driver database. Impedance curve with some semi-inductance is approximated with exponent function by those two values. Main part of Visaton data is imported from some German enclosure simulator "as is". SPL curve with simple Z model matches with that tool.
The problem is not in the accuracy of the driver parameters - it's the whole underlying assumptions behind trying to "model" the effect of the voice coil inductance on the frequency response when the voice coil inductance effects are already included in the design of the driver and published frequency responses....
__________________
- Simon

Last edited by DBMandrake; 17th July 2019 at 06:46 PM.
  Reply With Quote
Old 17th July 2019, 07:48 PM   #1339
mbrennwa is offline mbrennwa  Switzerland
diyAudio Member
 
mbrennwa's Avatar
 
Join Date: Mar 2005
Location: Zurich, Switzerland
Quote:
Originally Posted by DBMandrake View Post
Nearly all cone drivers that you would use in a box simulator are designed to give as flat as possible frequency response with voltage drive from a low impedance source
There may be some drivers that are designed like this, but certainly not nearly all cone drivers. It really depends on what a driver is designed for. For instance, a bass driver does not need to have flat response higher than a few hundred Hz.
__________________
------
Group buy for augerpro waveguides for Scan Speak tweeters -- elliptical version -- circular version
  Reply With Quote
Old 17th July 2019, 07:51 PM   #1340
kimmosto is offline kimmosto  Finland
diyAudio Member
 
kimmosto's Avatar
 
Join Date: Oct 2005
Location: Kuopio
Quote:
Originally Posted by DBMandrake View Post
Yes, impedance rise with frequency caused by voice coil inductance reduces current and therefore SPL at higher frequencies compared to no inductance, that's simple ohms law - but the effect of the voice coil inductance on SPL is already taken into account in the design of the driver and the published frequency response curve of the driver which is based on the driver being voltage driven from a low impedance souce not current driven from an infinite impedance source...
Now you're talking about actual drivers in practice. Not ideal piston with current driven motor. Enclosure tool tries to play both of those, but user is responsible to select which SPL curve is shown and exported. For example I never look SPL curve with impedance effects because it's not very practical approach and useful for purpose of enclosure simulation (which is driver selection and box dimensioning). Basic simulators are not able to simulate HF response so that range is quite useless no matter is there some Z effects or not.
  Reply With Quote

Reply


VituixCADHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


New To Site? Need Help?

All times are GMT. The time now is 01:04 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki