Beyond the Ariel

Administrator
Joined 2004
Paid Member
Re: cinema "sound" reality check

mige0 said:
To give an even more correct explanation – the inverse square law IS valid also with lenses in use. At double the distance you always get ¼ the illumination given the SAME lens applied. The size of the picture gets 4 times bigger then of course. (no cylinder wave front with light !)

Yes, right! With the same lens. Here I was giving Lenard a hard time and I was even less clear. :xeye:

I was thinking more in practical terms. If you need to fill a 30 foot screen, it doesn't matter if the projector is 40 or 150 feet from the screen. Why? Because you'd use a different lens. The closer to the screen, the wider the angle of the lens needs to be, and visa versa. A lot of people will look at the long throw distance and talk about how much light is lost because of the "inverse square law." But of course it isn't lost, because the longer focal length lens is squeezing the light into a tighter beam. So the same amount of light hits the screen either way, minus some small air losses.

The image will look a little different, though. I suppose because the image coming from a long distance has rays that are more parallel than the wider angle lens up close. There is less hot spotting on the screen and more even off axis viewing with the long throw distance.

This may be what Lenard is getting at with his large sound source. How flat the wavefront is may effect how large we perceive the source to be. I certainly prefer large sources. Small speakers always sound small to me, no matter how loud.
 
Somehow I doubt that the shape of the wavefront has something to do with perception of size. The mechanism that gives us the apparent source width of a low frequency source according to conventional wisdom, AFAIK is the envelope of the LF signal. In other words it is the higher frequency part of a real world signal, and its spread-out arrival time to our ears if the source is large, that conveys the size.

This also explains why stereo sounds larger than mono (directional clues nonwithstanding) and why line sources may seem to sound larger than life. Both have to do with large arrival time differences, in stereo it's L vs. R, in a line source it's the higher vs. the lower part of the line. And even a mono source that is physically large should sound "larger" than a small mono source.

So why would horns sound "large" then? Same thing: mouth diffractions from a physically large mouth, generating arrival time differences for largeness perception. An artefact in other words.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
MBK said:
Michael,

re: the 3 dB/decade falloff in line arrays, I never properly understood where it comes from. I suspect cancellations from comb filtering.


To my perception the theory of a cylindrical wavefront is not the explanation. Its the geometry of the array's dispersion. When you squeeze your energy vertically to fit it almost only in a horizontal plane in the near field, don't you just boost it by 3dB? You skip a dimension of expansion. Here you go. Then you start losing control vertically when the distance gets longer and you cross to the far field, where you get square law -6dB per doubling of distance spherical expansion business as usual.
The lower the frequency the shorter the -3dB per doubling of distance near field definition. λ divides the near field array length. So you lose 3dB per octave going lower in frequency at a given distance within the near field. About 9-10dB SPL less per decade. Allow for +6dB gain when forming a large enough front baffle after setting the array (+3B addition of sources, + 3dB vertical baffle boundary reinforcement) . You end up to -3dB per decade. Is it coincidental that the rule of thumb is 2 subs per array satellite cabinet in reinforcement practice? Thats +3dB of lows. No wonder all line arrays sound tipped up in balance when playing alone, and mountains of subwoofers are there to support their mid-high long throw avoiding the use of excessive EQ not to lose precious top available SPL. Music for the masses...

See a single EAW KF-730 satellite measurement. It is crossed and processed for fitting in a usually no less than 8 cabinet array.
 

Attachments

  • screenshot27.jpg
    screenshot27.jpg
    41.5 KB · Views: 1,311
I can only partially follow.

- The nearfield-farfield transition vs. frequency, good point, but still in a domestic floor to ceiling array the line is supposed to be of "infinite" length (including floor and ceiling mirror images, and those work best for the LF), and all listening at all relevant frequencies is supposed to be in the line array nearfield.
- the +6dB from forming the array in the first place is a baffle step and source addition effect that only occurs once, and likely spread out over a single octave that yields those 6 dB - not a gradual effect over two decades.

It seems more likely that the nearfield-farfield transition is very gradual, and that the "infinite" assumption only partially holds, ending up with -3 dB/decade in practice. But maybe the line array specialists here can give us the official theory on this...
 
diyAudio Chief Moderator
Joined 2002
Paid Member
MBK said:
I can only partially follow.

- The nearfield-farfield transition vs. frequency, good point, but still in a domestic floor to ceiling array the line is supposed to be of "infinite" length (including floor and ceiling mirror images, and those work best for the LF), and all listening at all relevant frequencies is supposed to be in the line array nearfield.
- the +6dB from forming the array in the first place is a baffle step and source addition effect that only occurs once, and likely spread out over a single octave that yields those 6 dB - not a gradual effect over two decades.

It seems more likely that the nearfield-farfield transition is very gradual, and that the "infinite" assumption only partially holds, ending up with -3 dB/decade in practice. But maybe the line array specialists here can give us the official theory on this...

My point is that the near field shortens the lower you go in frequency and this correlates with your (right by my reckoning) assumption that the transition is gradual in reality and not infinite. In practice I have never experienced anything infinite or anything mathematically absolute in nature. All stuff averages out in certain tendencies and no absolutes or infinity.
 
parts quality at line vs speaker level

Lynn Olson said:


12 years later, when I was designing the Ariel and its 5 dB greater sensitivity, I discovered that the crossover was exquisitely sensitive to parts quality, especially the capacitors in the HF section. The Sprague 730P and Hovland were the only ones of the day that seemed transparent enough, and the Hovland irritatingly enough required a crossover adjustment to maintain the same subjective balance. I was not happy to make this discovery - parts tweaking is one of my most annoying and least rewarding hifi activities, something I'd rather not do at all.

Nowadays we have many more choices but cap coloration is still an awkward business in a high-resolution speaker - and I found out the hard way that the higher the efficiency, the less forgiving the speaker is of mediocre crossover parts. You can throw junk at a 86 dB/metre speaker and it isn't that audible. 92 dB/metre and up is a different ballgame. I'm expecting 97~100 dB/metre to be completely unforgiving of crossover parts quality - a severe incentive to keep complexity down. These drivers are expensive - the idea of wasting that much money on crossover parts isn't attractive at all.

Do you find the sensitivity to parts quality less of a factor in line-level electronics?
I would be interested, Lynn, in your views on the trade-offs inherent in active crossover implementation vs passive.

Emerald Audio seems to be doing well with the DBX unit used this way, presumably stock :
http://www.dbxpro.com/PA/PA.htm
I wonder whether a little tweaking of the analog stages would produce a better result than more expensive components in a passive EQ.
 
Hi

Michael,

re: the 3 dB/decade falloff in line arrays, I never properly understood where it comes from. I suspect cancellations from comb filtering.



This page gives me some answers. I read it some time ago but forgot. Shame on me!

http://meyersound.com/support/papers/line_array_theory.htm
"http://meyersound.com/support/papers/line_array_theory.htm"



The pro audio ELS just made me thinking abut some side aspects of highly asymmetric directivity loudspeakers like line arrays, small long ribbons / AMT's or the multiple speakers of the current OB project radiating within the same frequency band.

One of the effects here is that the SPL does not seem to fall at –6 dB with double distance but only around –3 dB du to the cylinder shape of the wave front.

At least for "line arrays ....and the multiple speakers of the current OB project radiating within the same frequency band" I went completely wrong as it seems ! We need quite some height compared to wave length AND very low separation distance between drivers to get that cylinder waveform behaviour.
For long or closely stacked ribbons and AMTs it might work in the upper frequency range.



An externally hosted image should be here but it was not working when we last tested it.




Greetings
Michael
 
Interesting article indeed, although they also have their priorities: as a prosound company they are interested in very long throw systems, and there, especially without boundary reinforcement (stadiums...) the assumptions break down.

In domestic line arrays some of the assumptions may hold "close enough" up to 4-6 m listening distances (as their own table shows: the bass doesn't hold up but at 250 Hz+ the dB loss is close enough to 3 dB / distance doubling). But yes, from all calculations I have seen the line does have to be floor-to-ceiling, with a practical limit of at least 75% floor-to-ceiling if one assumes good reflection by the boundaries, at least for the bass and mids. For the Highs something like 24-36" line length might be enough at 1.5 kHz+, from memory, to preserve line character up to 4 m listening distance.

But all this matters most for achieving the mythical 3 dB/distance doubling SPL loss. If only naturalness and correct source size perception are concerned, then simply making the source larger for lower frequencies, might be good enough.
 
Re: parts quality at line vs speaker level

Russell Dawkins said:


Do you find the sensitivity to parts quality less of a factor in line-level electronics?
I would be interested, Lynn, in your views on the trade-offs inherent in active crossover implementation vs passive.

Emerald Audio seems to be doing well with the DBX unit used this way, presumably stock :
http://www.dbxpro.com/PA/PA.htm
I wonder whether a little tweaking of the analog stages would produce a better result than more expensive components in a passive EQ.

Well, this is where individual perceptions enter in. Earl Geddes, who is a talented speaker designer and sharp theoretician, believes Costco-quality electronics are good enough to exhibit at the RMAF, and anyone who thinks otherwise is fooling themselves, or worse, trying to sell snake oil to the gullible.

Siegfried Linkwitz is firmly in the solid-state camp, and from his perspective, the better-grade opamps are transparent sounding enough to use in many different locations in the active crossover and equalizer.

I'm pretty much on the other side of debate - and trust my perceptions, having designed audio gear for some thirty-plus years. I find the addition, or substitution, of a single part in the signal chain quite audible. Sometimes the substitution makes the difference between high-quality sound and being completely unacceptable for high-fidelity use.

When I first built the Amity power amplifier in 1997, I was quite disappointed by nearly every linestage I auditioned. It was simple to test whether they were altering the signal, since I had the choice between the DAC output going right into the power amp (without attenuation) and the linestage under test. The Jeff Rowland was one of the very few not to grossly degrade transparency and add annoying tonal colorations.

Most transistor linestages were grainy and two-dimensional, and most tube linestages were bass-heavy with lots of tubey jukebox colorations. Again, this was by flipping a switch and doing a direct A/B comparison - nothing involving multi-hour listening sessions with a long checklist. It was obvious in seconds or at most a minute or two.

It took me a while to come to terms with the brutal transparency of the Amity amplifier - it was forgiving of musical styles, and brought out the best in many speakers, but was utterly unforgiving of poor linestages or substandard DACs (mine was a heavily modified PCM-63K based unit). We're talking of very brief yes-no comparisons here - a linestage was either listenable, or not, nothing too complicated subjectively.

So I'm on the other side of the fence in terms of electronics. I find most electronics unlistenable, and that certainly includes just about all high-priced audiophile products. I didn't design the Amity, Raven, Aurora, and Karna for fun - I didn't care for the sound of commercially available electronics, and was forced to find my own way.

Given how dissatisfied I am with linestages, active crossovers sound worse, much worse. The tube-based ones have gross design errors in the 12AX7 tube section, with tubes operated at far too low of a current to drive the reactive loads of the crossover. The opamps in the solid-state ones are trash, the kind of thing that belongs in boomboxes. I find these things unlistenable, enough so I can't honestly assess the improvements from the EQ and time-alignment.

This seems to be a matter of individual perception. Some people are completely happy with generic opamps with 5V/uSec slew rates and quasi-complementary Class AB output sections. To me, they limit the sound to no better than home-theater receiver quality - barely hifi, down at the MP3 quality level.

When I see slow, generic opamps, with electrolytic coupling caps, I don't usually even bother to turn it on. And unfortunately, that's what you see in $5000 SACD players and "audiophile" linestages, equalizers, and crossovers. Some of us are sensitive to this, some of us aren't. More power to you if opamps and solid-state works for you - you have a lot more choices open to you.

The perception of parts degradation in passive crossovers vs active crossovers is also individual. There's been a big fad in Europe over the last several years of using the Behringer DCX, either modified or stock, in high-end time-equalized horn systems. At the 2004 ETF, I heard some very high-quality horn systems, and when doing a direct comparison between passive and DCX crossovers, I felt the DCX ruined the sound, flattening out all the subtle details and giving an unpleasant metallic coloration to everything. With the passive crossover, at least things sounded more natural.

Now most of the colorations I objected to were at mid and higher frequencies - at lower frequencies, such as 200 Hz or lower, the gizmo sounded fairly benign, and not too offensive. But from 200 Hz on up, it sounded no better than an iPod - and I own one, and know what they sound like.

But I have to keep returning to individual perception. Some people, who I respect as engineers and designers, are completely happy with generic transistor gear. Most of this gear sounds to me like the Bose stereo in my Acura. There is transistor gear I've heard that I liked, but it tends to be rather exotic and well off the beaten path. I'm kind of picky about tube gear as well - Audio Research, Cary, Jadis, VTL - uh, no thanks.

In Europe I heard the DCX with completely replaced, non-solid-state analog sections, and all-new power supplies that don't use switching supplies. All of that work lifted it from iPod to generic CD player quality - OK for bass, most certainly not for full-range use. I've been pretty underwhelmed with the other digital signal benders as well - they sound like generic CD players, no better than that, making it very hard to assess the other sonic benefits. I'd rather listen to a Dynaco Stereo 70 driving an ordinary passive crossover.

By now, I'm sure I sound like a complete nut to people who listen to popular solid-state equipment and like it. What can I say. I've met the reviewers from the audiophile magazines, and heard some of their systems. That is not the direction I am going - but neither am I heading towards 3-watt SET amplifiers and Altec and JBL vintage speakers.

So I can't endorse any of the digital equalizers, at least not in the midband and higher frequencies. For signal conditioning for the bass array and subwoofers, yes, that's what I plan to do myself. The DBX unit looks like a good choice for that application. For the widerange driver and tweeter (either CD or ribbon), it's going to be transformer-coupled triodes from DAC/phono preamp to the 1.2~1.8 kHz passive crossover, just as it is now for the Ariels.
 
Hi



Lynn, it's a blessing and a burden to have such fine ears / perception !

For audio guys life is MUCH easier if you are not too much bothered by listening at signals that are processed by silicone or some less than perfect parts elsewhere on its way from recording to playback. Did you finally work out an acceptable solution for your line stage ? I found power supply concept ( stage by stage !!! ) and ground layout as some of the most crucial points especially with discrete circuits. I once changed a "unlistenable" (- well not quite - ) Naim pre to a really nicely flavoured gear with incredible deep image this way. Even with minimising stages, I ended up with >30 supply lines right to the power supply and even more ground lines to the star point within the pre.
I am very curious what's the outcome at the end of your pretty openly OB design process.
Congratulations to you and all, for a thread that exceeds 1000 very interesting posts.

As for the active versus passive crossover discussion, my findings are that passive crossovers - even build with high quality parts throughout - have kind of coloration that I for myself describe best as lack of some "directness".

MBK, thanks for throwing in your personal experience with line arrays in a domestic context ! I never had special interest or considered to try LAs until this thread and some of your postings.

As for the source size perception, I am not sure that I understand what you mean with "envelope of the low frequency". Is it rise time you are referring to?

My understanding about source size is somehow mixed with the perception of the room size, which means the less early reflections ( that you cant really avoid in listening rooms at home ) the bigger size you get – presumed the recordings you listen to provide reverberation accordingly.



A horn and an optical lens are not the same...

We should not over stress analogy, but in terms of concentrating energy to a solid angle its basically has the same effect I think.


Greetings
Michael
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Lynn wrote:

''The perception of parts degradation in passive crossovers vs active crossovers is also individual. There's been a big fad in Europe over the last several years of using the Behringer DCX, either modified or stock, in high-end time-equalized horn systems. At the 2004 ETF, I heard some very high-quality horn systems, and when doing a direct comparison between passive and DCX crossovers, I felt the DCX ruined the sound, flattening out all the subtle details and giving an unpleasant metallic coloration to everything. With the passive crossover, at least things sounded more natural.''

In some older threads I have maintained that the DCX and DBX PA level is completely lo-fi and I have suggested the XTA brand which can be found for about $2000 second hand in perfect condition.

As for line circuits discussed by Lynn, my best design so far is a triode strapped RCA 6V6G smoked glass STbottle running at 22mA 350V B+ with a gain of 7.

At this level of sonic honesty, to use digital and op amps (especially cheap) higher than 200Hz commands oodles of make believe to sound right.
 
Hi


In some older threads I have maintained that the DCX and DBX PA level is completely lo-fi and I have suggested the XTA brand which can be found for about $2000 second hand in perfect condition.


salas, could you point me to that threads please or give a short summary ?
You really mean $ 2000.- for SECOND HAND ( DP 446 ? ) ??


Greetings
Michael
 
What I find quite striking is just how different individual perceptions are. People who I admire and respect - my dear, departed old friend Bob Sickler of Audionics, Siegfried Linkwitz, who I've met and chatted with at some length, and Earl Geddes - are all pretty accepting of digital, opamps, and Class AB solid-state sound.

I was in that group too until about 1990, when I went the second meeting of Oregon Triode Society and heard a trivially modified Dyna Stereo 70 utterly crush a $3500 Audio Research Classic 60 amplifier. That was a shock - more of a shock when I found out the "modification" was nothing more than converting the ST70 from ultra-linear to triode, nothing more than moving two wires.

When I finished the Ariels a year or two later, I auditioned them on something like 30 different amplifiers. I was appalled at how different the amplifiers sounded, and worse, some of the newest, most expensive, and best-reviewed sounded the worst - and some of the rusty old classics sounded remarkably good. Two of the classics that stood out were the Citation II and the Scott LK-150 - the newer, audiophile versions quite commonly had weird colorations (which I now attribute to poor regulation schemes and low-grade circuit boards).

That was a quality of the Ariels that took some adjustment - they were much less tolerant of amplifier quality than 86~88 dB/metre speakers. Low-efficiency speakers just gobble up the watts and aren't too particular about the quality. Huge banks of paralleled Class AB MOSFET or bipolars transistors are just fine with these kinds of speakers - it seems to take 200 watts just to get them to wake up and play music.

The Ariel was quite different. The old standby Audionics CC-2, although a little grainy sounding, was overall pretty neutral and listenable. The simplicity of the output stage, along with the high slew rate and inherent stability, worked in its favor. Much to my surprise, the Crown Macro Reference sounded quite good as well, and had plenty of power too. For some reason, the Pass Aleph 3 didn't work at all, and I never figured out why.

On the tube side of things, most of the audiophile PP pentode amps had funny colorations, as if they were unsuccessfully trying to mimic the sound of old, worn-out vintage gear. I suspected these were deliberate colorations to mask the sound of screechy CD players and harsh-sounding solid-state preamps. Direct-heated triode SET's were all over the place too, although many of them were much, much more transparent than the murky-sounding PP pentode gear.

I found that aiming for the simple goals of low coloration and transparency put me in a fairly small group of audio designers - I guess most of the industry are trying to please the reviewers of the Big Two magazines, who like a hard-edged, unnatural sound. When the magazines say XYZ component is "accurate" the sound to me is mechanical, unnatural, and harsh, and nothing like live acoustical music at all. I think magazine reviewers stopped listening to live classical music at least 20~25 years ago, based on the kinds of equipment they favor in the reviews.

Now that I've met Mike Sanders of Quicksilver Audio, I'm pleased to discover he tunes his amplifiers for a natural sound as well - and he designs both triode and pentode amplifiers. There are a few of us who don't care for the mainstream sound, but aren't ready to start collecting Western Electric, Klangfilm, and Vitavox gear just yet - we're not quite that far out. (Although one of these days I'll indulge myself and get an EMT 930 - that is one nice-sounding turntable!)

Returning to the topic at hand, the new speaker is going to be another 6 dB jump in efficiency, compared to the Ariels. All the things the Ariels did that was different than conventional low-efficiency audiophile speakers, I'll be facing again. The new ones will be firmly on the doorstep of horn territory, and will undoubtedly be just as picky about amplification. This is a certainty - it's the one thing about the new system I know for sure.

P.S. Good quote here from Bill F: "my non-scientific rule of thumb with OBs run much below f-equal is to imagine the volume displacement of an equivalent sealed system and then quadruple it if you want to end up with dynamic headroom". By the way, the volume displacement of a single 21" driver is comparable to three 12" drivers, and it would fit on the baffles we've seen here.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Lynn, I dare to air my assumption that we audio people don't really perceive differently, but we perceive selectively. If I just set my mind to prove a speaker concept for instance there is a big danger I will seek fast solutions for drive and control. The primary urge, the focus, is proving a mental structure. Its about mental aesthetics. And if it principally works, wow! Thats the kick I really look for! Mind is a big big con, using us. Happens all the time. Aural sense is an 80% brain process psychoacoustics says. Hammer, anvil and stirrup move the same in my and next man's head, but the individual 'translator' con brain has ideas, education, memory and feeling of its own.
Each of us pursuing individual goals just zooms in a portion of the picture. But some still have the veteran character control not to forget that the goal must be the total picture (music) and not specific technical perfection. Its a twist. We don't seek music many times when making stuff. We seek technical achievement. We get to like the sound of 'technique'. Real danger is that we may get to feel that it is the real thing, and become addicted. We are perverts many times. Its part of the game, but at a point we can get over it. And we have a powerful tool for waking up. Boredom! Non natural may succeed feeling even great, but at a point we get sick. Its like eating in fast food chains only, for too long.
 
Michael,

just to set the record straight - all my comments on line arrays come from studying the relevant papers and other people's axperiences, because at some point I wanted to build one. But I never did, sorry if it sounded like I had actual experience in line array building. I have plenty of hands-on experience in OB building, and I like to keep up to date with research if I can, that's all.

Apparent souce width (ASW): it's a complicated issue, and most research comes from either concert hall acoustics considerations or from people who'd like to simulate soundfields. All effects are frequency dependent and several effects mix together.

According to Lexicon's Griesinger, ASW is strongly influenced by rise time of a note: shorter rise times sound smaller, and we are talking of ranges starting at 10 ms. My inaccurate "envelope" comment should have referred to "rise time".

Then, in actual architecture, come reflections and reverberation, but they must fit in particular time windows to be useful. Too short, and it sounds closed-in (especially below 1.5 ms). 10-15 ms or more should be the goal. But reflections in the 50-150 ms range make intelligibility suffer, because this time window corresponds to the average gaps during speech.

According to soundfield simulation research, the inter-aural correlation coefficient is ultimately responsible for the size impression: decorrelated phase/time delays lead to a more spacious impression, easy to verify if you compare correlated and uncorrelated pink noise in stereo - the correlated one sounds like a small point source, the decorrelated one seems to fill the entire room.

Finally, in general, HF sources sound smaller than LF sources, and soft sources sound smaller than loud sources.

Unfortunately when I was reviewing this I realized that much of it applies more to architecture than speaker building. The delays involved, from 10 ms and up, can't (hopefully!) come from the speaker. We can only try to avoid time-delayed very early reflections at the speaker's edges, below 0.6 to 0.7 ms, ideally up to 1.5 ms but that would call for a very large speaker. On the simpler side, having low bass at all and in correct loudness, will already help with size perception.

I also have a nagging suspicion that it's the imperfections in the reproduction chain that improve space and size impression: say low channel separation and phase tracking, as presumably in LP's, might actually and unwittingly provide a certain amount of crossfeed and decorrelated time delays that make the sound spacious (!!). Same goes for the time delays involved when listening to line arrays, dues to the different distances to different parts of the line, or the delayed reflections from large diameter horn mouths.

Conversely by the way if any room-speaker-listening position combination sounds "large" the imaging shouldn't be so good (mucho reflections), and vice versa - say widerange drivers in the nearfield, perfect imaging but small-sounding.

In other words I suspect that truly accurate reproduction without reflections probably sounds just like headphones: flat and dull. Because that's what's on most recordings :rolleyes: .

FWIW I finally found the sketch on time delay perception, from a presentation by Begault et al., attached.

References:

David Griesinger on space perception
Ptard et al on source wideness
 

Attachments

  • delay and echo times begault 2002.jpg
    delay and echo times begault 2002.jpg
    61.5 KB · Views: 1,131
Lynn,

actually SL does make some references to the problems of SS amps, especially the crossover distortion that would not show up in normal THD plots, and vary with level as well. That might also be a prime candidate as to why high sensitivity drivers could be more picky of amps, they would amplify the crossover distortions because the overall power level is lower to begin with to achieve a given SPL.
 
18 and 21 inch drivers

Hi Lynn,

Looking at 18” and 21” drives (as you mentioned) I formulated the table below (click on image) considering the Mms, Sd and the BL factor, and also included the TT to compare.

I am no expert and just learning from these posts.

The 18” Beyma 18G50 has the best ratio’s
My question is that could it replace 2 x 12” drivers or even replace the 3 X 12” drivers and use one TT 12” Alnico for the WR driver and the Neo Pro 5i tweeter for the HF.

I’m considering of building a cheap version for now, any comments appreciated.

Frank
 

Attachments

  • speakers.pdf
    29.9 KB · Views: 208