Hi Captain Jack
Your "Family Dollar" floor polish coating sounds like the ticket... easy and cheap is a good thing! Please tell me; do you dilute it or add anything to it?
Please be aware, that some newly made Saran wrap may be low density polyethylene instead of original polyvinylidenedichloride. Then the coating would be a definite problem. Window film could be an option as well.
Alex
(moving this back on topic 
)
An LC delay line such as used in the Quad 63's would be very hard to implement. IIRC Peter Walker used air core inductors of several Henries. You won't believe how much (ultra thin) wire goes into such a coil.
RC lines are a much more realistic alternative. There is a paper in the oct. 2009 issue of the journal of the AES, which describes a null-free polar response using an RC delay line (the C being formed by the diaphragm and portions of the stator, of course).
Finally I want to stress that to reduce beaming is a tradeoff, and therefore not inherently something to strive for. The sweet spot widens, but perfect imaging becomes harder to attain.
'night,
Kenneth
)An LC delay line such as used in the Quad 63's would be very hard to implement. IIRC Peter Walker used air core inductors of several Henries. You won't believe how much (ultra thin) wire goes into such a coil.
RC lines are a much more realistic alternative. There is a paper in the oct. 2009 issue of the journal of the AES, which describes a null-free polar response using an RC delay line (the C being formed by the diaphragm and portions of the stator, of course).
Finally I want to stress that to reduce beaming is a tradeoff, and therefore not inherently something to strive for. The sweet spot widens, but perfect imaging becomes harder to attain.
'night,
Kenneth
Hi,
imho with such LC delay-lines You gain nothing over the simpler RC-line.
For most users a slightly widened sweet spot is more practical than the extreme beaming character of a flat panel.
There have been measurements of mid-sized flat panels which showed amplitude deviations of ~10dB within just 1° change of listening position.
That is certainly non-practical. So either a curved panel or a distribution shaping design is helpful. A single curved panel beeing superior to an array of angled panels (with regard to ditribution character and in case of a hybrid design) and a RC-line driven segmented flat panel allowing for fullrange use.
Besides the difficulty to design or obtain the right inductances, these coils form resonant circuits -i.e energy storages. Care must be taken that those resonances remain of low Q. Still though I dislike the idea to introduce such energy storage mechanisms into a system that is otherwise nearly free from those and whiches sonic superiority over dynamic drivers results imo -at least in parts- from this freeness of energy storing mechanisms.
jauu
Calvin
imho with such LC delay-lines You gain nothing over the simpler RC-line.
For most users a slightly widened sweet spot is more practical than the extreme beaming character of a flat panel.
There have been measurements of mid-sized flat panels which showed amplitude deviations of ~10dB within just 1° change of listening position.
That is certainly non-practical. So either a curved panel or a distribution shaping design is helpful. A single curved panel beeing superior to an array of angled panels (with regard to ditribution character and in case of a hybrid design) and a RC-line driven segmented flat panel allowing for fullrange use.
Besides the difficulty to design or obtain the right inductances, these coils form resonant circuits -i.e energy storages. Care must be taken that those resonances remain of low Q. Still though I dislike the idea to introduce such energy storage mechanisms into a system that is otherwise nearly free from those and whiches sonic superiority over dynamic drivers results imo -at least in parts- from this freeness of energy storing mechanisms.
jauu
Calvin
Hello, all,
This is a great thread -- glad to see some recent input!
I am contemplating a nearfield application which will use a single Acoustat 8" panel in an infinite baffle and want to try RC segmentation. I have the Acoustat schematics for the Spectra series so I can work out R values, but I haven't a clue what minimum power or voltage ratings the resistors need to have. Anybody ideas/guesses/suggestions? Unfortunately, Acoustat didn't publish these specs, but if anyone has an old Spectra interface they could look at, I would be very grateful. Also, I think Audiostatic may have used this segmentation method -- can someone confirm?
This is a great thread -- glad to see some recent input!
I am contemplating a nearfield application which will use a single Acoustat 8" panel in an infinite baffle and want to try RC segmentation. I have the Acoustat schematics for the Spectra series so I can work out R values, but I haven't a clue what minimum power or voltage ratings the resistors need to have. Anybody ideas/guesses/suggestions? Unfortunately, Acoustat didn't publish these specs, but if anyone has an old Spectra interface they could look at, I would be very grateful. Also, I think Audiostatic may have used this segmentation method -- can someone confirm?
take a look at the Dutch ESL site check out the large assortment of projects under the heading "Projecten" there are a number of RC delay line designs there.
Elektrostaten ESL-club[/urll
Google translator does a fair job of turning this into English
Elektrostaten ESL-club[/urll
Google translator does a fair job of turning this into English
dave,I was thinking of doing that with the small panels I have been playing with.
As all it would take is a couple of quick swipes with a dremel to the screens and wa la ,segmented stator.
And filp them so that they are horizontal insead of vertical so they will fit in the cubby hole better and dampen the rear waves with some fiberglass insulation or polyfill to reduce backwave cancelation and refractions and run them slightly monopole.
just one is quite loud and can't wait to get them in stereo.
what do you think? jer
As all it would take is a couple of quick swipes with a dremel to the screens and wa la ,segmented stator.
And filp them so that they are horizontal insead of vertical so they will fit in the cubby hole better and dampen the rear waves with some fiberglass insulation or polyfill to reduce backwave cancelation and refractions and run them slightly monopole.
just one is quite loud and can't wait to get them in stereo.
what do you think? jer
Attachments
Thanks, Moray, this was just what I needed to see -- the recommendation is 2W metal film, BTW. Good -- cheap and easy to find!
Jer -- you lost me for a minute with Dremel reference -- until I "zoomed" the picture of your panel construction. If you are going to try RC ladder segmentation I have no formulas for you to calculate R values -- maybe others can help.
Jer -- you lost me for a minute with Dremel reference -- until I "zoomed" the picture of your panel construction. If you are going to try RC ladder segmentation I have no formulas for you to calculate R values -- maybe others can help.
the r values can be determined by using the simple rc lowpass filter calculations and the capacitance of each section.
I haven't decided two go with 3 or 5 sections starting from the middle of the panel.
the panels them selves cost only pennies to build should i screw one up or decide to just build a new one. jer
I haven't decided two go with 3 or 5 sections starting from the middle of the panel.
the panels them selves cost only pennies to build should i screw one up or decide to just build a new one. jer
Dave,no there not single ended.
They sound awesome and with no distortions when properly driven.
A very cean sound with no honkyness or colorations like a box speaker sounds like.
When I made them I used my uncle's meter to verify them against the formula and online capacitance calculators at about 40pf and maybe 70pf with the wires attached. jer
They sound awesome and with no distortions when properly driven.
A very cean sound with no honkyness or colorations like a box speaker sounds like.
When I made them I used my uncle's meter to verify them against the formula and online capacitance calculators at about 40pf and maybe 70pf with the wires attached. jer
Hi,
You could do a rough calculation to estimate the approximate power rating.
First calculate the capacitance value of the specific segment after the formula: C=88,54*A/d [pF] (the segment area A in cm² and the stator-stator-distance d in mm).
Now calculate the impedance value of this segment which is: Xc=1/(2pi*f*C) [Ohm]
The current flowing through this impedance calculates to: Ipeak=Upeak/(Xc+R) [A]
(as frequency f use the -3dB crossover frequency of the RC-filter; Upeak the peak signal voltage --> maximum peak signal voltage; R=resistance of the RC-filter resistor). Since Upeak shall not be higher than Upol, the polarizing voltage, You can calculate with Upol. Now Upeak*Ipeak gives the peak Power [W]. The resistor´s rating should be equal or greater than this value (calculating for the -3dB-frequency and the fact that the filter slope is just -6dB/oct this calculation already includes an ample headroom --> worstcasecalculation). The wattage rating will be quite small 1-2W but the voltage rating needs to be high. You should keep well away from the maximum ratings stated in the resistor´s datasheet. Real highvoltage resistors are not cheap and maybe difficult to source. You can series connect resistors of lower voltage and wattage rating instead. With the series connection the voltage and wattage ratings simply add up with the number of resistors. The typical size 0207 (250-300V/0.6W) metal films are a cheap and easy to source alternative. The same applies to 1-2W metal-oxide resistros.
jauu
Calvin
You could do a rough calculation to estimate the approximate power rating.
First calculate the capacitance value of the specific segment after the formula: C=88,54*A/d [pF] (the segment area A in cm² and the stator-stator-distance d in mm).
Now calculate the impedance value of this segment which is: Xc=1/(2pi*f*C) [Ohm]
The current flowing through this impedance calculates to: Ipeak=Upeak/(Xc+R) [A]
(as frequency f use the -3dB crossover frequency of the RC-filter; Upeak the peak signal voltage --> maximum peak signal voltage; R=resistance of the RC-filter resistor). Since Upeak shall not be higher than Upol, the polarizing voltage, You can calculate with Upol. Now Upeak*Ipeak gives the peak Power [W]. The resistor´s rating should be equal or greater than this value (calculating for the -3dB-frequency and the fact that the filter slope is just -6dB/oct this calculation already includes an ample headroom --> worstcasecalculation). The wattage rating will be quite small 1-2W but the voltage rating needs to be high. You should keep well away from the maximum ratings stated in the resistor´s datasheet. Real highvoltage resistors are not cheap and maybe difficult to source. You can series connect resistors of lower voltage and wattage rating instead. With the series connection the voltage and wattage ratings simply add up with the number of resistors. The typical size 0207 (250-300V/0.6W) metal films are a cheap and easy to source alternative. The same applies to 1-2W metal-oxide resistros.
jauu
Calvin
yes,basicaly.
in the recording industry nearfleid monitors are used because they are small and very very acurate.
they don't require alot of power as ear fatige will set in quite easly.
They sit directly in front of you at the mixing console much like your desktop computer speakers (hence nearfield).
But they will get up there when pushed,when there is bass it is just there with no boomyness.
I even used to play my bass guitar through them while recording tracks,as we never had to bother with miking up a bass amp,it was always direct in.
Even the yamaha ns-10's that were considered "THE STANDERED" in nearfield monitors for recording.
We were going to buy those until I started poking at some buttons and the krk's popped on and changed our mind.
We were out to get a pair of tannoy's gold series originaly.
We paid like $500 for a pair of krk's back 1994 and the stereo imaging rivaled anything I have ever heard,even to this day!
except for the ml esl's and also including the ones i have built.
Even my 7.75" x 22.25" diy panels give our apogee dueta's a run for the money with the only advantage is size.
Listening to the apogee's were my first experience of listening to some really big headphones.
the panel you see in the picture is only 3.5" x 10" and seems to mate very well with the sony SS-MB150H 4.5" wooffer.
The pink noise response of the two at 18" was flat with the panel running fullrange except below 80 hz was cut off with about 2db to 3db of boost with 12khz highpass shelving eq to taste.
Even though I'm having amplifier issues at the moment, I'm in awe at sound sound of these things just as they sit and thats just one, as I don't them set up in stereo yet.
You can't beat the price per performance either,the sony's were $20 a pair on sale at rat shack and maybe $5 in material for the esl panels.
We know what amps and transformers cost although i got mine as a clearance sale for $20 a piece so I got 10 of them.
I wish I had gotten 10 or 20 more transformers when I had the chance. jer
in the recording industry nearfleid monitors are used because they are small and very very acurate.
they don't require alot of power as ear fatige will set in quite easly.
They sit directly in front of you at the mixing console much like your desktop computer speakers (hence nearfield).
But they will get up there when pushed,when there is bass it is just there with no boomyness.
I even used to play my bass guitar through them while recording tracks,as we never had to bother with miking up a bass amp,it was always direct in.
Even the yamaha ns-10's that were considered "THE STANDERED" in nearfield monitors for recording.
We were going to buy those until I started poking at some buttons and the krk's popped on and changed our mind.
We were out to get a pair of tannoy's gold series originaly.
We paid like $500 for a pair of krk's back 1994 and the stereo imaging rivaled anything I have ever heard,even to this day!
except for the ml esl's and also including the ones i have built.
Even my 7.75" x 22.25" diy panels give our apogee dueta's a run for the money with the only advantage is size.
Listening to the apogee's were my first experience of listening to some really big headphones.
the panel you see in the picture is only 3.5" x 10" and seems to mate very well with the sony SS-MB150H 4.5" wooffer.
The pink noise response of the two at 18" was flat with the panel running fullrange except below 80 hz was cut off with about 2db to 3db of boost with 12khz highpass shelving eq to taste.
Even though I'm having amplifier issues at the moment, I'm in awe at sound sound of these things just as they sit and thats just one, as I don't them set up in stereo yet.
You can't beat the price per performance either,the sony's were $20 a pair on sale at rat shack and maybe $5 in material for the esl panels.
We know what amps and transformers cost although i got mine as a clearance sale for $20 a piece so I got 10 of them.
I wish I had gotten 10 or 20 more transformers when I had the chance. jer
Another really big virtue of near-field listening is that it allows for minimal "room sound" corruption. My application will be using AmbioDSP which is a very sophisticated cross-talk cancellation algorithm that produces headphone-like immediacy and detail but with sounds coming from in front with great depth and accurate ambiance cues -- if the recording is a good one. With a quasi-line source like the segmented 48" tall Acoustat panels, I should be able to get the speakers about 6' away from my listening position and still be in the near-field.
My plan is to use a pair of active NHT subs below about 100 Hz, so the panels will be relieved of deep bass duty.
Best to all,
David
My plan is to use a pair of active NHT subs below about 100 Hz, so the panels will be relieved of deep bass duty.
Best to all,
David
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.