Inductance Thread - diyAudio
Go Back   Home > Forums > Loudspeakers > Subwoofers

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 14th June 2015, 07:35 AM   #1
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
Lightbulb Inductance Thread

I felt that this Important topic deserves it's own dedicated thread. I credit just a guy for initially arousing my interest in this, others for their inputs, & Josh Ricci @ DataBass for providing good tests & data. For those that havn't been following, the discussions recetnly sprung up in the 16 hz for church organ thread 16 hz for church organ

Out of curiosity, i simmed the same driver with the EXACT same data for both the driver & box, except for changing the inductance of one of them. To do this i only had to create a new instance of the driver file, so i could then load both into WinISD to compare them. WinISD has a very useful feature, whereby you can click on a tab to account for inductance, or not !

Interestingly, it certainly does appear to make a difference, in this sim not a great deal, but it's there. From this, i "expect" higher inductance drivers to exhibit an even greater difference, & starting even lower down the f scale.

Also by changing values in the Voice Coil Temp Rise tab in real time, which effects Re, allows you to see the combined differences both make together.

I look forward to more stimulating etc discussions & tests etc. Plus it would be great if for eg HornResponse etc could be made to alter Inductance in real time.
Attached Images
File Type: png Ind.png (6.7 KB, 152 views)
File Type: png Not.png (4.4 KB, 150 views)
  Reply With Quote
Old 14th June 2015, 07:43 AM   #2
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by Zero D View Post
Plus it would be great if for eg HornResponse etc could be made to alter Inductance in real time.
You mean like this?
Attached Images
File Type: png Attach_1.png (56.4 KB, 149 views)
__________________
www.hornresp.net
  Reply With Quote
Old 14th June 2015, 10:19 AM   #3
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
I posted a reply in the 16-Hz organ thread but I suppose it really belongs here, now that this thread has started.

I gather that that the Thiele-Small model starts to fall apart (to a small degree) when inductance, resistance, (and I suppose all other parameters) vary with frequency. And more surprised that inductive reactance, which is pretty small at low frequencies matters.

I have no trouble intuitively expecting mechanical parameters to vary although surprised to hear electrical inductance and resistance can vary (saturation or compression effects excepted). Other instances or devices where they vary?

Is it because the underlying theory of electrical analogies to mechanical and electrical parameters in speakers has no analogies where inductance, etc. vary with frequency? Or because, well, all models are a bit crude?

I find it scientifically distasteful to finagle parameter values until the resulting output kind of matches measured results in the absence of understanding just what is going on. Not a video game.

Ben
__________________
Current projects: motional feedback woofing and large ESL panels. HiFi aspirations since 1956
  Reply With Quote
Old 14th June 2015, 01:28 PM   #4
diyAudio Member
 
Join Date: May 2006
Quote:
Originally Posted by Zero D View Post
...
Inductance absolutely does make a big difference in a sim, especially if it is a large value for Le.

Some drivers can't be simulated accurately, that was the point of the discussion in the 16 hz thread. Until yesterday I thought the effect was solely due to inductance, but here's what I wrote in the 16 hz thread today.

When I'm wrong I admit it. When searching for the UM18 measurements on data-bass I saw the familiar "inductance hump", which shouldn't be there in such a low inductance driver. This means I may be wrong about inductance being the sole cause of the issue. The cause may just be very large and heavy voice coils, although at this point I don't want to speculate too much. However hopefully I can be excused for correlating this effect with inductance, since 99 percent of this type of ultra high excursion driver with very large and heavy voice coils also has ultra high inductance.

Anyway, the method still works on this type of driver. It works great to sim the UM18 with it's "inductance hump" too, even though it's not a high inductance driver and the hump doesn't look like it's caused by inductance.

Anyway, the correction method works great, but it's back to the drawing board to find the exact cause of the issue that makes this type of driver not sim right, which is useful in identifying them without seeing them measured first, which is one of the biggest advantages to the correction method. If Le:Re is higher than 1:1 you can bet it won't sim right, but now there are a couple of identified driver will Le:Re much lower than that, that also need correction.


Anyway, adding artificial Le to a sim to try to accurately sim these big drivers does not work, but as alluded to above, I do have a method that does work really well, 100 percent of the time, for all drivers of this type in all enclosure. The correction method is not perfect, it could use refining but I don't have time to play with it and it's a lot more accurate than just ignoring these effects and running the t/s parameters provided through a regular sim.
  Reply With Quote
Old 14th June 2015, 01:41 PM   #5
diyAudio Member
 
Join Date: May 2006
Quote:
Originally Posted by bentoronto View Post
I find it scientifically distasteful to finagle parameter values until the resulting output kind of matches measured results in the absence of understanding just what is going on. Not a video game.

Ben
The 3 point complex inductance model by Leach (and at least one other complex inductance model authored by somone else) has t/s parameters specifically to describe frequency dependent resistance and frequency dependent reactance.

No this is NOT a video game, what I am doing is forcing a simulator that cannot accept these parameters to consider them. I told you that already. It's not a perfect method but it works 100 percent of the time on all drivers of this type in all enclosures.

Science is the process of testing a theory and coming up with enough evidence that the theory works that it cannot be disproved. IF it is disproved, the theory is over and done. What I am doing is science, what you are doing is complaining. Disprove the theory if you like. Good luck with that.

Your hatred of simulators and simulations is well documented. Even though you don't use simulators and don't trust their results even when shown mountains of proof that sims match measurements. Simulators are the BEST tool we have to estimate performance of a driver and they are REALLY good at it. There are very few exceptions in which they fail at what they are made to do, this is one of them, and my method corrects that.

The problem, Ben is that you would like simulators to do things they are not made for, like estimating in room response or simulating the type of drivers we are talking about here without being able to accept complex parameters. Another problem is that you've heard bad designs (like max flat ported boxes) that sound terribly boomy in room, and just because the design was created in a simulator, you blame the simulator, not the bad design. Max flat is generally not appropriate for using in a room with any amount of room gain - especially when you prefer the sound of a rising response of a sealed box as you do. As I've told you several times, you can design that response curve shape into a ported box too, and if you did you wouldn't be able to tell which was sealed and which was ported in a blind test until you turn them up really loud and the sealed box can't keep up anymore. It's not resonances you hate when you hear a design, it a bad design that you hate, and unfortunately you haven't been able to distinguish bad designs from characteristics and mistakenly blame things like simulators and resonances as a whole when the problem is really just bad designs.

Since I've told you all this several times now, you either are not reading or not understanding what I am telling you, since you continue your all out assault of simulators, simulations, and resonances.

Last edited by just a guy; 14th June 2015 at 01:47 PM.
  Reply With Quote
Old 14th June 2015, 04:13 PM   #6
tb46 is offline tb46  United States
diyAudio Member
 
Join Date: Jan 2006
Location: Texas
Hi Zero D,

JAG has been beating on this for a while, here are some additional links,
on the avs forum: DIY Speakers and Subs - AVS | Home Theater Discussions And Reviews
and on diyaudio,: http://www.diyaudio.com/forums/subwo...urately-2.html

avsforum LTD02 picked up on this, and applied a reduction to Bl, at times in combination w/ an increase in Le. This looks like a promising approach as it addresses parameters that are known to change w/ frequency, and high power application.

The bad thing is that sofar there is no method that can be applied universally, so we are back to more extensive empirical testing to get the data for entry into the simulators.

Regards,
__________________
Oliver
  Reply With Quote
Old 14th June 2015, 05:29 PM   #7
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Quote:
Originally Posted by tb46 View Post
This looks like a promising approach as it addresses parameters that are known to change w/ frequency, and high power application.
Yes, that's scientific, for want of another word to describe acting rationally, rather than by guesswork.

In practice, you CAN use guesswork and accidentally get reasonable predictions some of the time because some of the parameters in the T-S model behave co-relatedly. That's you-know-whose method. But it has no merit for advancing understanding of what is going on with higher inductance drivers.

Ben
__________________
Current projects: motional feedback woofing and large ESL panels. HiFi aspirations since 1956
  Reply With Quote
Old 14th June 2015, 06:02 PM   #8
diyAudio Member
 
Join Date: May 2006
Quote:
Originally Posted by bentoronto View Post
Yes, that's scientific, for want of another word to describe acting rationally, rather than by guesswork.

In practice, you CAN use guesswork and accidentally get reasonable predictions some of the time because some of the parameters in the T-S model behave co-relatedly. That's you-know-whose method. But it has no merit for advancing understanding of what is going on with higher inductance drivers.

Ben
Ben, if you actually read the links you would see that this is not guesswork and it's not accidental and it doesn't work "sometimes", it works 100 percent of the time on all high inductance drivers in all enclosures. I noticed a few years ago already that the higher inductance drivers didn't measure the same as the sims predicted. I also noticed that the effect looked exactly like the effect of adding additional Re to the sim, as I am always doing to see the estimated effect of power compression. The response curve humps up around the impedance peak(s) and rolls off everywhere else. So I tried adding a bit of Re, then adding a bit of power to offset the added Re losses, and it worked. All the time. On all high inductance drivers. In all enclosures, sealed, ported, tapped horn, front loaded horn, every enclosure type I could get a high quality measurement and box dimensions on.

And coincidentally, at a glance (without knowing too much about the subject) this seems to be exactly the point of the 3 point complex inductance parameters and simulators that can accept them, frequency dependent Re and reactance are actual t/s parameters. I'm just forcing a simulator that won't accept these parameters to consider their effect.

I'm not a driver expert, and I really don't think I need to be. Sometimes you don't need to completely understand WHY a problem is happening to find a cure for it. Actually this happens all the time in science.

While it would be nice to know exactly what is causing the issue, there's plenty of data to guess which drivers will be affected (ones with very big and heavy voice coils, especially if their inductance/resistance ratio is 1:1 or higher), and the correction method works very well 100 percent of the time. What else do you need?

And seriously, why do you even care about any of this? You don't believe in the validity of simulations in the first place, despite being presented with mountains of evidence, so why is this particular issue so important to you? And if you want to disprove it so badly, why don't you provide some evidence of your own to show it doesn't work? I've shown plenty that shows it works very well. You spend most of your time on this forum attempting to discredit well known science (in the form of math done by simulators) and now you are complaining that I'm not scientific enough?

Last edited by just a guy; 14th June 2015 at 06:19 PM.
  Reply With Quote
Old 14th June 2015, 06:15 PM   #9
diyAudio Member
 
Join Date: May 2006
Quote:
Originally Posted by tb46 View Post
The bad thing is that sofar there is no method that can be applied universally, so we are back to more extensive empirical testing to get the data for entry into the simulators.

Regards,
Actually I do apply the correction method universally on any driver that obviously needs it.

To explain that better, the exact same tweak is applied to every driver that has Le:Re ratio of 1:1 or higher. The only problem here,which I just discovered yesterday, is that there are some drivers with considerably less Le:Re ratio of 1:1 that also need that also need to be corrected, as they suffer from the same issue. But so far, all the drivers I've identified that have this issue have really big, heavy voice coils, and almost (but not quite) 100 percent of them also have very high inductance.

There needs to be more work done to find out EXACTLY which drivers need the tweak just by looking at published specs and/or voice coil physical data (although we already know that high inductance drivers need the tweak 100 percent of the time, I've never seen a high inductance driver that didn't suffer this issue, but now I've found a couple of lower inductance drivers that are also afflicted). AND the method could be a bit further refined so it's not exactly the same simple tweak applied to all afflicted drivers, in other words, it could be improved to be a bit more accurate. But as it is it's quite good already and I've posted a bunch of examples showing this in the links you provided.
  Reply With Quote
Old 14th June 2015, 08:05 PM   #10
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
@ David McBean

Yes, exactly like that ! Sorry i forgot about that feature, as i havn't played with it lately. I tried it earlier & it sure works. Good job

@ tb46

Thanx for the info & links, which i'll check out. EDIT, the link to http://www.diyaudio.com/forums/subwo...urately-2.html is broken ? Also which LTD02 thread/s ? as there appear to be many !

@ just a guy

Adding Re & reducing Le etc, in Actual real time & with powered enclosures, sounds like a task that for eg AceBass could do

Last edited by Zero D; 14th June 2015 at 08:10 PM.
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
inductance measurements jneutron Parts 3 17th May 2013 03:15 PM
Should a person who starts a thread have quasi moderator status on their own thread? erin The Lounge 15 23rd October 2012 10:41 PM
Change inductance Lostcause Parts 4 7th February 2006 03:29 PM
inductance multiplier millwood Solid State 15 18th February 2004 05:30 PM


New To Site? Need Help?

All times are GMT. The time now is 10:44 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2