Segmented Wire Stator ESL simulator (esl_seg_ui) - Page 2 - diyAudio
Go Back   Home > Forums > Loudspeakers > Planars & Exotics

Planars & Exotics ESL's, planars, and alternative technologies

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 16th February 2012, 05:50 AM   #11
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Thanks,Bolserst,here is what I have came up with so far using the program for my selected panel size.
It has a total of 9 segments like I had wanted to do.
I will look more into increasing the number of segments as I get used to how the program works.

This will be great when I finally get around to finishing the 10" X 36" TIG rod stators that I have already constructed.
Segmenting those will be a breeze with a couple of cut off wheels.

So far this is looking very good and it is very close to to my pink noise measured efficiency of my last little panel even though this design has about 60% more area.

The last panel was at 89.7db with pink noise at no more than 10Vpeak and 87.5db for test tones at 5Vpeak with a 1:256 transformation ratio and 5.56Kv of bias.

I will look more into using inductors as well.
I would think that this would work good for such a smaller panel and might have a better power handling capability than using resistors.

These particular panels will be used in the nearfield mostly but they are the research panels for some 8' line source types to supplement my current woofer system to do away with the dynamic midranges and tweeters in which I keep blowing up any how.

But I can say that my old pair of Pyle PMD5's that I have been using are very good drivers and are the only ones that have survived so far.
Every thing else has gone kaput! He,he,he

The only difference in the two examples is the last resistors in the chain.
The first chart shows a 4.7meg and the second one uses a 10meg for the last resistor.

I will work with it some more but I don't think that I can get it much better than it is the way it is shown.

Thanks again,Cheers !!!

jer
Attached Images
File Type: jpg 5meg.jpg (174.5 KB, 400 views)
File Type: jpg 10meg.jpg (175.7 KB, 392 views)

Last edited by geraldfryjr; 16th February 2012 at 05:53 AM.
  Reply With Quote
Old 16th February 2012, 06:27 AM   #12
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
WoW,This is Even Better !!!


This is the same as the last post.
Only with just a few resistor tweaks.

I guess the only way to improve the 30 and 45 degree off axis is to use more segments maybe.

jer

P.S Write down your data as there is no save button!
Attached Images
File Type: jpg EVEN BETTER.jpg (167.4 KB, 332 views)

Last edited by geraldfryjr; 16th February 2012 at 06:32 AM.
  Reply With Quote
Old 18th February 2012, 10:40 PM   #13
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
I don't know how accurate this program is but this shows some incredible results so far.

This is about the best that I can do using 2 rods per section.
There are lots of combinations but that are good.
This one gives me one of the flattest so far within .5db to .75 db.

This is perfect for my requirements right now.
It shows the same results from 60 degrees to 180 degrees as well so this is where I question its accuracy.
But, I will go by these figures and if it does this at over 30 degrees off center than that would be awesome!

I have tried these same setting for some longer panels at 36 inches and the only difference is the values of the resistors are lower.

For this calculation the resistors are as follows for 38 rods per stator,

180k,3.9M,5M,5M,5M,5M,5M,10M,10M,10M.

I used all standard values of resistors as the 5M's will be two 10M's in parallel.

The the only thing I am wondering is how do you determine the wattage of the resistors.
I do plan on doing sweep tests on the panels after they are built and I don't need any failures.
I will try to plot this up in circuit maker and as well as LTSpice as I am just now learning to use that program.

If this works as planned then my planned 6' (2 X 3 foot peices stacked) line source will be perfect for my current system.

Any Suggestions?

jer
Attached Images
File Type: jpg BEST.jpg (171.4 KB, 297 views)

Last edited by geraldfryjr; 18th February 2012 at 10:44 PM.
  Reply With Quote
Old 18th February 2012, 11:24 PM   #14
diyAudio Member
 
Join Date: Jul 2007
Ger-
I may be a little off base here, but my reading has told me that inductance may not be a good thing here. I am planning on using non-inductive r's in my build. In the past I have used Caddock metal film resistors, as they are totally non-inductive and dissipate huge amounts of power. You can also heatsink them. What I don't know, is how high a value you gan get. Other drawback is that they can be expensive- I used them in series with the primary of my step-up trafos, and the 25 watt version cost me about 6-8 bucks ea. They come packaged in a TO-220 plastic transistor package. If you subscribe to the notion that you can "hear" resistors, Caddocks get very high marks.
One of the papers from the Dutch ESL club website stated that the builder used 9 watt rated units in his build.
You can also find non-inductive wire-wounds with relative ease at Mouser or Digi-Key.
Again, I may be off base here, but I would resist the temptation to use carbon units. The noise they generate is a function of the applied voltage as I understand it. By the time you get a 9-10 watt rated series parallel network of carbons, you're going to be out several dollars per value anyway and then have to figure out how to neatly package the things.
Jay
I doubt if you could hear it, but I try to cover all the theoretical bases I know about when I build, and don't mind spending a little more on parts if for no other reason than peace of mind.
If you don't have a Mouser catalog, get one. www.mouser.com- free.
  Reply With Quote
Old 19th February 2012, 12:06 AM   #15
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Yes, this is understood.

According to ohms laws with a 10kv peak voltage the whole resistor chain of 59.08meg ohms requires a dissipation of only 1.69 watts.

And in the circuitmaker simulation it shows that the 3.9meg ohm resistor requires it to be at least 1.12 watts and the rest are below this value down the chain.

Remember for this simulation this is a very small panel with approximately no more than 2.5pf per section.
So, The current demand is quite small.

This had me quite worried as well as I am well aware of the price of good resistors.

I haven't yet worked out the values of the longer panels as of yet.

But as you know doubling the voltage to 20Kv peak quadruples the power requirement.
But,I highly doubt that I can get my little panel to withstand a 40kv p-p signal and getting it to do 20kv p-p is quite a challange and is where I will set the limit for now. Even half of that at 5kv is sufficient and very loud within 1 meter.

I will show the sims after I study it some more and convert the grahpics to be posted.

Jer

Last edited by geraldfryjr; 19th February 2012 at 12:12 AM.
  Reply With Quote
Old 19th February 2012, 02:14 AM   #16
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Okay,Here are the circuit maker simulations on my next little panel for resistor power requirements.

Using the example in post #13,
The Diagphram size is 5" X 10" with a D/S at 1.8mm or about .070" (70mil).
This example has 19 separate sections with 2 rods per section and are equally spaced from the center within the 5" width.

The capacitance per section is an estimated 8.368 pf per section and in a real world test this will likely be less than that.

These photos show the test points for voltages on the resistors as well as power dissipation for each of the resistors at 10kv peak and at test frequency's of 20Hz,200hz,2Khz and 20khz.


Cheers !!!


jer
Attached Images
File Type: jpg test points.jpg (66.7 KB, 285 views)
File Type: jpg SEGMENTED ESL SIM 20Hz 10Kv.jpg (44.7 KB, 63 views)
File Type: jpg SEGMENTED ESL SIM 20hz resistor power.jpg (112.2 KB, 37 views)
File Type: jpg SEGMENTED ESL SIM 200Hz 10Kv.jpg (78.6 KB, 48 views)
File Type: jpg SEGMENTED ESL SIM 200hz resistor power.jpg (109.4 KB, 42 views)
File Type: jpg SEGMENTED ESL SIM 2kHz 10Kv.jpg (57.4 KB, 31 views)
File Type: jpg SEGMENTED ESL SIM 2khz resistor power.jpg (43.5 KB, 30 views)
File Type: jpg SEGMENTED ESL SIM 20kHz 10Kv.jpg (52.3 KB, 27 views)
File Type: jpg SEGMENTED ESL SIM 20khz resistor power.jpg (47.9 KB, 21 views)
  Reply With Quote
Old 19th February 2012, 03:29 AM   #17
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Quote:
Originally Posted by geraldfryjr View Post
Okay,Here are the circuit maker simulations on my next little panel for resistor power requirements.

Using the example in post #13,
The Diagphram size is 5" X 10" with a D/S at 1.8mm or about .070" (70mil).
This example has 19 separate sections with 2 rods per section and are equally spaced from the center within the 5" width.

The capacitance per section is an estimated 8.368 pf per section and in a real world test this will likely be less than that.

These photos show the test points for voltages on the resistors as well as power dissipation for each of the resistors at 10kv peak and at test frequency's of 20Hz,200hz,2Khz and 20khz.


Cheers !!!


jer


P.S. The capacitance per section was calculated relative the the diagphram so the actual capacitance per section is approximately half (and slightly less) of what was stated.
  Reply With Quote
Old 19th February 2012, 03:13 PM   #18
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by geraldfryjr View Post
I don't know how accurate this program is but this shows some incredible results so far.
The program is accurate if used with the calculation assumptions in mind. I'm guessing my description of the assumptions in post#1 is not understood. Also, from looking at your SPICE model, I'm thinking my description of the physical layout the program is modeling based on the input parameters wasn't understood either.

I should have some time later today or tomorrow to post some further clarification on the above two topics.



Your idea to use SPICE to model your finished design to determine power ratings for the ladder resistors is a good one. Rather than looking at just specific frequencies, it is best to do a frequency sweep because power dissipation for each ladder resistor peaks at a different frequency.

Also, don't forget to also consider the voltage ratings. In general, most 2W - 3W metal film or metal oxide resistors are rated for 300VAC - 500VAC. If you get enough of them in series to cover your required voltage rating between sections, you are usually covered for the required power dissipation as well.
  Reply With Quote
Old 19th February 2012, 08:25 PM   #19
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Yes,only the First 2 or 3 resistors in the chain will have to made up with a few resistors in series to compensate for the voltage coefficient and breakdown factor (working voltage) that resistors have.

All though this depends on a particular configuration.

But the cool thing is that it shows that only a few of them need to have a higher power rating thus reducing unneeded costs.

I will be looking for your clarifications later.

I do understand the differences in the change of the response curve with distance as was mentioned.

As my new wire mesh flat panel will be used in nearfield use on my desktop and the TIG rod version is to test this dispersion technique in the far field for some large line source panels as mentioned.

jer
  Reply With Quote
Old 19th February 2012, 10:51 PM   #20
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Default Segmented Wire Stator ESL Simulator (Line Source Assumptions)

The ESL Simulator software assumes that the dimensions and listening distance you enter for modeling describe a Line Source. If they don't, the plotted SPL curves will not predict what you will measure.

An ESL dipole line source by definition has a height that is much larger than the listening distance, and a width that is much smaller than the listening distance. If these two criteria are met, the response will have a slope of 3dB/octave over the whole of the frequency range.

Attachment 1
For example, a 20m high, 0.5cm wide ESL would behave like a line source as modeled.

Attachment 2
Change the height to 0.25m, and the sofware still shows the some response since it assumes line source behavior. I reality, a 0.25m x 0.5cm ESL would behave like a point source when viewed from 5m and would have a response sloping off twice as fast, 6dB/oct.

Basically, for line source behavior to be a valid assumption, the listening position must place you in the far-field relative to the width dimension, but in the near-field relative to the height. Some call this the intermediate field.
If the listening position is in the near-field relative to width & height, the response will be flat.
If the listening position is in the far-field relative to width & height, the response will rise at 6dB/oct.

The formula for determining near/far field transition relative to a panel dimension is:
F = c * r / d^2

where:
c = speed of sound
r = listening distance
d = panel dimension


Attachment 3
Here is a plot showing the field transition points for a 0.25m heigh x 12.16cm wide ESL listened to from 1m away. The range of frequencies over which line source behavior is a valid assumption is between the vertical red and yellow dashed lines. To the left of the red line you are in the far-field. So, if the modeling software showed a flat response down to 200 Hz, in reality it would be down -12dB or so.

Attachment 4
Here is a plot showing the field transition points for an ESL similar to an Acoustat. You can see that line source behavior is present down to below 200Hz . With appropriately chosen diaphragm resonance frequency and Q, the 3dB/oct line source slope was extended down to 55Hz or so.

I have attached a zipfile containg the Excel spreadsheet I used to create Attachments 3 & 4.
Perhaps you will find it useful to determine validity of the line source assumptions for the ESL panel you are wanting to model.
Attached Images
File Type: gif ESL_line_3dB_1.gif (57.5 KB, 66 views)
File Type: gif ESL_line_3dB_2.gif (107.6 KB, 63 views)
File Type: gif ESL_line_01.gif (84.8 KB, 77 views)
File Type: gif ESL_line_02.gif (82.2 KB, 81 views)
Attached Files
File Type: zip Line.zip (367.2 KB, 37 views)

Last edited by bolserst; 19th February 2012 at 10:55 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
ESL stator rho Planars & Exotics 13 4th January 2008 10:37 PM
ESL stator fixing 405man Planars & Exotics 9 29th August 2007 10:19 PM
Wire Insulation for Wire stator Audio_idiot Planars & Exotics 6 25th May 2007 11:42 AM
single stator esl. tade Planars & Exotics 2 4th July 2005 05:00 PM
angles for segmented ESL? Few Planars & Exotics 21 25th February 2005 03:30 PM


New To Site? Need Help?

All times are GMT. The time now is 09:52 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2