analysis epsilon - Page 13 - 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 15th January 2012, 02:40 PM   #121
diyAudio Member
 
Join Date: Jan 2007
Location: wigston leics england
Quote:
Originally Posted by harlee View Post
Hi Jamesbos,
You have some interesting speaker building there. I have about 300 pcs of 50x10x5 neos, and would be interested in building what you have. Some questions I would like to clarify if you don't mind.

1.By building the 60x40, how big is the sweet spot?
2.How wide is the frequency range?
3.How should I match the magnet pattern to the foil layout pattern?
4Should it have some through space like in the ESL?
5.What's the tension should be?
6.What do you think about a 80x30? Advantage and disadvantage?

Appreciated
Big enough,quite sufficient,depending how you place the speakers, from what I have read on the forum you could spend the rest of your life trying different positions heights toe in etc etc!! Just listen and after awhile the brain gets used to the sound.
30 to 20k nice tight bass and clear treble no boom or fizz. Of course you could experiment with resistors capacitors and inductors as some people do. I don't bother myself. I only use a resistor to make the impedance match the amp if required. My 60 x 40 work out at 6.3 ohms which is OK for my ONKYO receiver.
You will have to have 36 magnets in each column like I have to fit the layout as in the green diagram (POST 99) which has 36 magnets on it in 4 rows as you only have 300 magnets for the pair. I used 360 magnets as I used 5 rows.
Through space !??. Do you mean space between magnets and diaphragm? If so 3 mm is what I use.
I tension mine by hand as I trap the mylar between the mdf and the picture frame wood. I don't use any heat at all. Just go round using the screws to trap mylar under picture frame hole by hole.Until you can tap all round the periphery of the diaphragm and it feels like a drum.
I wouldn't reduce the width because this lowers the bass sensitivity.Having only 4 rows has already reduced the bass response as it is.
Best of luck with your effort, you will need a lot of patience putting the foil on, being easy to rip and tear. Patience is a virtue have several rests whilst doing it. Especially if you are old like me. Hard on the back! YOU WILL HAVE TO USE SELLOTAPE IF YOU DO IT IN ONE CONTINUOUS PIECE LIKE I DID. Otherwise you will have to do it in several parts as the tape has to cross after every section as in POST 98..
__________________
HENRY

Last edited by JAMESBOS; 15th January 2012 at 02:53 PM.
  Reply With Quote
Old 15th January 2012, 04:42 PM   #122
diyAudio Member
 
Join Date: Jan 2007
Location: wigston leics england
The only problem is the width of the tape, which should be3.5 mm maximum used with the 10 mm magnets unless you have a 100% accurate method of cutting it. With my 12.7 mm wide magnets I can use scissors to cut the 9 mm tape in half, which I use.!
__________________
HENRY
  Reply With Quote
Old 15th January 2012, 06:21 PM   #123
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Blog Entries: 1
What about 6'x2' or 7'x2'?

Each of the two epsilon patterns in post 98, for a 4'x2' layout, was 4.8 Ohms.

If I kept one pattern per side and just made them taller, then increasing the length to 6' would give roughly (slightly less than?) 7.2 Ohms per side and increasing to 7' would give roughly 8.4 Ohms per side.

Paralleling those would give 3.6 Ohms for 6' or 4.2 Ohms for 7' (or probably slightly less than those, due to less "overhead" percentage from the looping at the ends).

I wonder how the required power and SPL would change: Thinking out loud, the same current through a conductor would give the same membrane excursion, which would give the same SPL, PER UNIT OF AREA. So the same current would give something like 50% more SPL with the 6'x2' than with the 4'x2', and 75% more with the 7'x2'.

So to get the same SPL, the required current would be 67% as much for the 6'x2' and 57% as much for the 7'x2'. But the resistances would be different. So I wonder how that would affect things:

It turns out that the required power from the amp would be linearly proportional to the percentage of the 4'x2' 's current required, IF we considered only cases where the two patterns were connected in parallel.

Power from the amp would be iČR, or VČ/R. But the current for the same SPL scales down with panel size while the resistance scales up and it turns out that the voltage stays about the same and the power required from the amp for the same SPL scales down the same way the current does.

For example, 2 Amps with 2.4 Ohms (the two 4.8 Ohm patterns in parallel) in the 4'x2' panel would require 9.6 Watts. The same SPL should result from 2x0.67 = 1.34 Amps with 3.6 Ohms in the 6'x2' panel, requiring 6.46 Watts (which is also 9.6 W x 0.67), and 2x0.57 = 1.14 Amps with 4.2 Ohms in the 7'x2' panel should give the same SPL and would require 5.46 Watts (which is also 9.6W x 0.57). The amplifier's output voltage would be about 4.8 in each case. For comparing to 4 amps through the 4'x2' 's 2.4 Ohms, the numbers become 4/2.68/2.28 Amps, 38.4/25.8/21.8 Watts, and 9.6 Volts.

So, the required current for the same SPL goes down as the panel gets taller, but the resistance increases by exactly (roughly, anyway) the same proportion, meaning that the voltage from the amp, i x R, stays roughly the same. So the power from the amp, vČ/R, decreases as the panel gets taller and R increases.

That brings me to Henry's series connection of the two 4.8 Ohm patterns, giving 9.6 Ohms:

Henry, I think that you should consider paralleling them to get 2.4 Ohms and then adding a resistor if necessary. That way, (if you added a 2.4 Ohm R) you would only need HALF as much power from your amp, to get the same SPL you get now (or you could get 2X the SPL for the same power).

Take the above example with 2 Amps of current: With 9.6 Ohms for the speaker, it would need 38.4 Watts from your amplifier. If your amplifier could drive 2.4 Ohms, the same 2 Amps would only require 9.6 Watts! And if you added a 2.4-Ohm resistor (just to make the math easier, here), your amp would be driving 4.8 Ohms and the same 2 Amps would require only 19.2 Watts. So if your amplifier can safely drive 4.8 Ohms, you might want to try that.

If you add a 2.4 Ohm resistor, I would use a resistor that's rated at what your amp's rated for at 4.8 Ohms, to give a 100% margin so it won't get too hot, or use at least half of that if you never turn it up too loud. You might have to parallel some resistors, so you can use lower-rated ones. For example, four 10-Ohm resistors in parallel would give 2.5 Ohms and each would only need 1/4th the power rating that a single resistor would need.

Wirewound resistors should be fine. A little inductance (up to 5 uH, say) would be good for keeping most of the RF from feeding back into your amplifier's output. These types of speakers make excellent antennas! The RF will find its way around any global feedback loops and get into the inputs. Any PN semiconductor junction might rectify it. It might get demodulated or it might just become DC-like and cause mischief like changing transistor operating points, or cause audible pops if it's from a broadband burst (like something switching on or off might cause). Actually, the late, great Al Sekela, over at the Planar Speaker Asylum, advocated heavily for always using 5 uH of inductance in planar magnetic speaker connections.

Disclaimer/Question: I'm not sure how the SPL varies with the current's magnitude. I know that the force on a conductor in a magnetic field varies in direct proportion to the current's value, if the B-field is parallel to the membrane and perpendicular to the conductor, as it should mostly be in this case. So the acceleration of the conductor (and the membrane) would be proportional to the current. But does SPL depend on the membrane's acceleration or its velocity, or what? Can anyone shed a little light on that, please?

Last edited by gootee; 15th January 2012 at 06:45 PM.
  Reply With Quote
Old 15th January 2012, 07:03 PM   #124
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Blog Entries: 1
Another disclaimer: The "SPL" concept that I tried to use, above, was not well-defined. I was just trying to use it to derive current and power requirements. It might be related to the SPL that would be perceived at a long distance from the speakers, or the total energy emitted by the speakers. But it probably isn't the same as what would be perceived from a listening position close to the speakers, since I was assuming that the amplitude of the sound produced "per unit of area" was staying the same. That would mean that the portion of the speaker that was closest to you (the center or ear-level part, say) would still produce the same loudness, but the additional area of the taller parts, which might be farther away and not directed at your ears, would produce additional sound. I don't know exactly how the actual perceived SPL at the listening position would change, for the different heights of speakers considered above, for a given amplifier output power level.

Last edited by gootee; 15th January 2012 at 07:20 PM.
  Reply With Quote
Old 16th January 2012, 07:59 AM   #125
diyAudio Member
 
Join Date: Jan 2007
Location: wigston leics england
Thanks Gootee for your suggestions, I'm alright at building planars but not very good at the maths, I can't be bothered at my time of life now. Will try your suggestion with my 2 tape layout before I build it's mate.See how things go!! I won't be going any bigger tho' A bit too heavy with ferrites and too costly with neos.
__________________
HENRY

Last edited by JAMESBOS; 16th January 2012 at 08:03 AM.
  Reply With Quote
Old 16th January 2012, 07:13 PM   #126
gootee is offline gootee  United States
diyAudio Member
 
Join Date: Nov 2006
Location: Indiana
Blog Entries: 1
Understood. And quite understandble. Please let us know what happens, if you try them in parallel.

I've also been wondering if changing the shape of the panel might make it image even better. I've read somewhere that "taller than they are wide" should be better.

With basically the same amount of materials as one of your 4'x2' panels, an 8'x1' panel could be made. That might be a bit extreme. But 6'x1' (~7 Ohms) or even 7'x1' (~8 Ohms) might work well. Have you ever tried any "narrower and taller" shapes?

I also wonder about the tensioning of the mylar. I know how you said you do it but do you ever try to measure the tension so you can get it the same again if desired (maybe with a standard weight on the center of a horizontal panel and measure the sag distance, for example)? Do you tension large panels to different tensions than smaller ones? Have you noticed differences in performance or sound quality with different tensioning?
  Reply With Quote
Old 17th January 2012, 04:24 AM   #127
harlee is offline harlee  Singapore
diyAudio Member
 
Join Date: Nov 2006
Hi Jamesbos,
Thanks for the fast reply and advice, some more questions to help me understand more if you dont mind,
your 60x40 is the outer dimension of your speaker?
What would be your membrane dimension?
The magnet width will cover 3 spacing of 1mm and 2 trace of 3.5mm?
What's the gap you have between each column of magnets?

Thank you very much
  Reply With Quote
Old 17th January 2012, 07:58 AM   #128
diyAudio Member
 
Join Date: Jan 2007
Location: wigston leics england
Quote:
Originally Posted by harlee View Post
Hi Jamesbos,
Thanks for the fast reply and advice, some more questions to help me understand more if you dont mind,
your 60x40 is the outer dimension of your speaker?
What would be your membrane dimension?
The magnet width will cover 3 spacing of 1mm and 2 trace of 3.5mm?
What's the gap you have between each column of magnets?

Thank you very much
Yes it is the outer dimension
Membrane slightly larger than this say 65 x 45 cms. So I can pull to tension the 23 um mylar. Afterwards you can cut it of to suite yourself!
Depends how big you make the cut out for the magnets and the width and length of your magnets., and the width of your tape.I would use tape between 3 and 3.5 mm, this would leave enough room between each piece. I usually use a piece of wood or aluminium rod 1mm wide, fixed to a piece of wood big enough for the perforated sheet to sit on.So I can keep them shipshape. I usually put them on the perforated sheet using wood or aluminium at one end to get and keep the magnets in line important for marking the diaphragm.My perforated sheet is 50 x 33 cm. My gap is approximately 1 cm. Remember patience is the key, those size magnets can do a lot of damage to your fingers.
__________________
HENRY
  Reply With Quote
Old 17th January 2012, 03:10 PM   #129
diyAudio Member
 
Join Date: Jan 2007
Location: wigston leics england
Quote:
Originally Posted by gootee View Post
What about 6'x2' or 7'x2'?

Each of the two epsilon patterns in post 98, for a 4'x2' layout, was 4.8 Ohms.

If I kept one pattern per side and just made them taller, then increasing the length to 6' would give roughly (slightly less than?) 7.2 Ohms per side and increasing to 7' would give roughly 8.4 Ohms per side.

Paralleling those would give 3.6 Ohms for 6' or 4.2 Ohms for 7' (or probably slightly less than those, due to less "overhead" percentage from the looping at the ends).

I wonder how the required power and SPL would change: Thinking out loud, the same current through a conductor would give the same membrane excursion, which would give the same SPL, PER UNIT OF AREA. So the same current would give something like 50% more SPL with the 6'x2' than with the 4'x2', and 75% more with the 7'x2'.

So to get the same SPL, the required current would be 67% as much for the 6'x2' and 57% as much for the 7'x2'. But the resistances would be different. So I wonder how that would affect things:

It turns out that the required power from the amp would be linearly proportional to the percentage of the 4'x2' 's current required, IF we considered only cases where the two patterns were connected in parallel.

Power from the amp would be iČR, or VČ/R. But the current for the same SPL scales down with panel size while the resistance scales up and it turns out that the voltage stays about the same and the power required from the amp for the same SPL scales down the same way the current does.

For example, 2 Amps with 2.4 Ohms (the two 4.8 Ohm patterns in parallel) in the 4'x2' panel would require 9.6 Watts. The same SPL should result from 2x0.67 = 1.34 Amps with 3.6 Ohms in the 6'x2' panel, requiring 6.46 Watts (which is also 9.6 W x 0.67), and 2x0.57 = 1.14 Amps with 4.2 Ohms in the 7'x2' panel should give the same SPL and would require 5.46 Watts (which is also 9.6W x 0.57). The amplifier's output voltage would be about 4.8 in each case. For comparing to 4 amps through the 4'x2' 's 2.4 Ohms, the numbers become 4/2.68/2.28 Amps, 38.4/25.8/21.8 Watts, and 9.6 Volts.

So, the required current for the same SPL goes down as the panel gets taller, but the resistance increases by exactly (roughly, anyway) the same proportion, meaning that the voltage from the amp, i x R, stays roughly the same. So the power from the amp, vČ/R, decreases as the panel gets taller and R increases.

That brings me to Henry's series connection of the two 4.8 Ohm patterns, giving 9.6 Ohms:

Henry, I think that you should consider paralleling them to get 2.4 Ohms and then adding a resistor if necessary. That way, (if you added a 2.4 Ohm R) you would only need HALF as much power from your amp, to get the same SPL you get now (or you could get 2X the SPL for the same power).

Take the above example with 2 Amps of current: With 9.6 Ohms for the speaker, it would need 38.4 Watts from your amplifier. If your amplifier could drive 2.4 Ohms, the same 2 Amps would only require 9.6 Watts! And if you added a 2.4-Ohm resistor (just to make the math easier, here), your amp would be driving 4.8 Ohms and the same 2 Amps would require only 19.2 Watts. So if your amplifier can safely drive 4.8 Ohms, you might want to try that.

If you add a 2.4 Ohm resistor, I would use a resistor that's rated at what your amp's rated for at 4.8 Ohms, to give a 100% margin so it won't get too hot, or use at least half of that if you never turn it up too loud. You might have to parallel some resistors, so you can use lower-rated ones. For example, four 10-Ohm resistors in parallel would give 2.5 Ohms and each would only need 1/4th the power rating that a single resistor would need.

Wirewound resistors should be fine. A little inductance (up to 5 uH, say) would be good for keeping most of the RF from feeding back into your amplifier's output. These types of speakers make excellent antennas! The RF will find its way around any global feedback loops and get into the inputs. Any PN semiconductor junction might rectify it. It might get demodulated or it might just become DC-like and cause mischief like changing transistor operating points, or cause audible pops if it's from a broadband burst (like something switching on or off might cause). Actually, the late, great Al Sekela, over at the Planar Speaker Asylum, advocated heavily for always using 5 uH of inductance in planar magnetic speaker connections.

Disclaimer/Question: I'm not sure how the SPL varies with the current's magnitude. I know that the force on a conductor in a magnetic field varies in direct proportion to the current's value, if the B-field is parallel to the membrane and perpendicular to the conductor, as it should mostly be in this case. So the acceleration of the conductor (and the membrane) would be proportional to the current. But does SPL depend on the membrane's acceleration or its velocity, or what? Can anyone shed a little light on that, please?
Tried it in parallel with 3.9 ohm resistor to get minimum of 6.3 ohms for the ONKYO Receiver.Not made any difference in sensitivity if anything slightly lower. So not really worth the resistor. So will be going back to in series. I am rebuilding one that used plastic magnets, which wasn't very loud. Hope the new layout will make it worth using more of the plastic magnets Hopefully!!
__________________
HENRY
  Reply With Quote
Old 17th January 2012, 11:44 PM   #130
LFM is offline LFM  Canada
diyAudio Member
 
Join Date: Oct 2004
Location: Sault
Default foil repair

Hello Jamesbos. I have a question about the aluminum tape and how to deal with a problem. How do you splice the foil if it is torn accidentally. Redoing the pattern may be the best method but can it be salvaged if you have the pattern nearly completed. I believe you mentioned a method using clear tape. I would like to hear what you recommend ...Albert
  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
epsilon lavardin amp help zerohead_ak47 Solid State 7 29th December 2008 05:02 PM
Midrange Analysis mbutzkies Multi-Way 4 25th May 2005 01:51 PM
noise analysis tenderland Tubes / Valves 5 29th March 2005 11:22 PM
cap analysis yldouright Parts 3 7th January 2005 01:45 AM
Need help in circuit analysis Fossil Tubes / Valves 6 23rd November 2004 07:13 AM


New To Site? Need Help?

All times are GMT. The time now is 09:29 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