An Afternoon With TheJoumal is pleased to publish another in this series of interviews conducted by the AES Los Angeles Section. These "Afternoon with ..." encounters offer spontaneous conversational portrayals of the careers and experiences of major figures in audio history, followed by informal, open-ended exchange with attending guests. On May 4, 1980, Peter Sutheim moderated this dialogue with Paul Klipsch. Mr. Klipsch, born in 1904 in Elkhart, Indiana, has written papers and holds patents in fields as diverse as geophysics, acoustics, and firearms. He is the recipient of the Audio Engineering Society Silver Medal (1978) for his contributions to loudspeaker design and for the measurement of distortion. In this excerpt from the transcript, Mr. Klipsch offers a retrospective glimpse of a profound, enduring, and distinctive audio career. utheim: Welcome to Paul S: Thereby proving himself one of the old telephone b u ~ l d ~ n g the Prov~ng at Klipsch, president of Klipsch first audiophiles. Ground, wh~chhad about 3000 square and Associates in Hope. feet of space. The basement was 12 feet Arkansas, manufacturers of K: Yes, indeed. He had a TRF tuner so h~gh, we had plenty of room to swlng loudspeakers. Paul is,a Fellow of the feeding an M I 4 Jensen loudspeaker in an e~ght-foot board around or feed ~t AES and the IEEE. a member of the a real baffle-strange beast in those through a saw. Lloqd used to put thrngs Acoustical Society of America and of days. It sounded less like a radio than I together, and I tned to be a salesman In Tau Beta Pi and Sigma XI. and he is. list- had. ever heard before. The symphony 1948, we produced 26 Kl~pschhorns. ed in Wlo's Who it? Et1gincri.iizg. Paul is was plkying that aftemobn. I had never one every two weeks. In 1978, some- well known for his classic folded horn heard symphonic music, but I decided I thrng over 2000 Khpsch horn4 were pro- loudspeaker, for research in the acous- liked it, and I put together some boards duced, and that accounted for only ten tics of loudspeaker radiation. for driver and a 12-inch loudspeaker fed by a re- percent of our total volume. We had design, and for AES publications on built radio. many other products-the Heresy, for modulation distortion in loudspeakers. Much later, in Houston. my wife said. example. u h ~ c htooh care of almost 50 He is an advocate.of comer placement "You've been wanting to build loud- percent of our dollar volume. of loudspeakers and the use of a derived speakers. Why don't you use the back center channel for center fill. room for a laboratory?"-words like a S: Paul, how did the Heresy get its Paul, can you tell us how you became red flag to a bull. and the beginning of name? interested in loudspeakers'? the end, I guess. I began with an X-l woofer and an X-2 squawker. World K: Back in 1955, I wanted to revive the Klipsch: I was a radio bug in 1920. My War I1 came, Uncle Sam punched my Bell Telephone Laboratory idea of a first loudspeaker was a cardboard tube ticket. and I went off to the Southwest bridged center speaker. We were mak- attached to a Brandes earphone. It didn't Proving Ground. ing only corner loudspeakers, and we work very well: it was years later that I After the war. 1 rented a little tin shed needed a noncomer speaker. I was sit- found out about exponential horns. At behind the laundry where I learned to ting there drawing pictures on the backs Stanford University one of my class- mold parts vie no longer make-the X-5 of envelopes, and my then Philadelphia mates invited me to his house one Sun- and X-5 horns-and hired my first em- sales rep, leaning over my shoulder, day afternoon, and he said, "You want ployee in 1948. One of us had to back said, "That's not a comer speaker. Paul." to hear the radio?" I said no, and he out of that barn before the other could I said no, and told him why. He shook turned in on anyway.. . tum around. But we moved out to the his head and said. "Paul. coming from J Audio Eng. Soc., Vcl. 40. No. 6, 1992 June An Afternoon With Paul Klipsch you that would be heresy." The lights K: If you want another dB or two of came on and I said, "That's what we'll power output, I guess that is the way to call it." go. But I still like passive crossovers. S: The current preference for direct-radi- Q: Doesn't that c h o i c e also have ator loudspeakers seems to have something to do with intermodulation eclipsed the horn approach except where distortion? very large sound pressures are needed. I'm curious to know the reasons behind K: Yes, but with present amplifiers, I that drift away from horns. Vented box- think overall intermodulation is going to es have come back in the last five years be well below audibility as long as you or so, but horns are not popular for stay below clipping. We do it. We don't home use. offer biamplification as an option, but the customer can buy our professional K: True, they never will be. But the models without crossover networks and trend is toward horns. not away from biamplify or triamplify as he chooses. them. We are growing faster than many The legendary "Heresy" loudspeaker, of the other manufacturers. We are one designed in 1955 by Paul Klipsch. S: Could we talk for a moment about of the few manufacturers of ham-loaded comer placement and your reasons for treble systems. Doppler effect and was reminded that it it? That recommendation puts you at might occur in loudspeakers. I made an odds with the advice of other manufac- S: I'd like to know more about frequen- eccentric capstan for an old tape ma- turers. The standard nowadays is up off cy-modulation distortion, an aspect that chine. It produced about one percent the floor and away from the walls. you seem to have pursued more or less peak FM, and with a 2 kHz tone made alone. Occasionally I talk to designers of an odd sound. So I knew that FM could K: Let's look at a comer speaker, l ~ o t direct-radiator systems, and they tend to produce audible effects. Beers and Belar necessarily a comer design, just a speak- pooh-pooh its importance. It's an acous- had published their 1943 paper on FM er in a comer. We have three mirror im- tical phenomenon, not an electrical one, distortion in loudspeakers, and their ages, and if we rub out the walls and re- and if it is indeed audible-in music, comment was that the distortion was place each image with a real speaker, say-is there any way around it? Is horn hard to describe. then as far as frequency response is con- loading the only way to prevent it? I One commentator said that it doesn't cerned, there is no difference between know it shows up in your measure- exist. and even if it did, it wouldn't be this array and the one vpeaker in the cor- ments. but how audible is it*? audible. I think it is hard to identify the ner. For all the frequencies for which particular oddity of sound from a n these are good walls, the efficiency dif- K: Perhaps the first coaxial speaker was .overloaded direct radiator a s d u e ference between the comer speaker and described in Elec.tronic,s, about 1938. specifically to FM or AM distortion. the four-speaker, no-wall array, will be The objective was to make the woofer But if you listen to a typical 12-inch di- very small. I have found, as I've put it and tweeter sounds coincident, as nearly 'tect r5diator at 100 d B SPL and com- before, that bass response can be im- as possible. We rigged one of our speak- pare it with a good horn at the same proved out to 10000 Hz. If you analyze ers with a tweeter mounted in the hollow output level but with a tenth of the it this way, the comer begins to make of the woofer cone. The tweeter was fed power input, there is a recognizable sense, and the folks opposed to it appear a little over 9000 Hz, the woofer 50 Hz. cleanness in the horn. In our measure- to be a little prejudiced. If you don't be- Someone said, "They are different dia- ments, we find that AM distortion usu- lieve the comer will help a box loud- phragms; why should any modulation ally exceeds FM in direct radiators. In speaker, try it and see. You may have to distortion show up'?" Think about it. The horns, the FM distortion predominates, tum the bass down to get it back in bal- woofer sound pressure forces the tweet- but it's an order of magnitude lower ance. but you w ~ l l then find the distor- er diaphragm, perhaps generating a little than in direct radiators. I haven't found tion down by about the same amount. cross-modulation there. And the tweeter out how to conduct a satisfactory listen- sound diffracts around the edges of its ing test to correlate listening impres- Q: When you have two corner loud- motor and bounces off the moving sions with measurements. In other speakers and the third speaker halfway woofer: you could expect frequency words, I cannot really answer your between. ir it fed the sum of the two modulation, and we found some, down question. I do know that there is some channels or the difference'? 27 dB. Next, we moved the tweeter to sort of correlation. and people who lis- its normal location, about a foot away ten to good sound over horns recognize K: The sum. In 1957 1published a paper from the woofer center, and the distor- the difference. where I said it made no difference. I tion dropped 13 dB. now have to disagree with the author. In the 1950s I was trying to fasten on Question: What is your feeling about The program material I had played up to why the horn sounded better than the di- multiamplification compared with pas- that time didn't make a difference. But rect radiator, and I knew about the sive dividing networks? then I got a particular Nat King Cole J. Audto Eng. Soc., Vol. 40. No. 6, 1992 June four-track. On one side, the three-speak- 300, 600, 1000 watts input. It's output is polar pattern. There are times when er setup sounded just fine. Playing the that matters. Bragging about input pow- you want to project the sound tightly, other side, the soloist in the center disap- er is a little like bragging about how but usually you want to cover a fairly peared. Obviously, phases were differ- much gasoline your car can bum. The wide area. Last in importance is ampli- ent on the two sides. second important characteristic is distor- tude-frequency response. tion. The two go together; distortion (to M: Any parting advice? a first approximation) is proportional to John Prose: Ixt me introduce myself. I power input in any loudspeaker. It fol- am chairman of the section, and I want K: When you buy a loudspeaker, it is lows that distortion is inversely propor- to thank you very much for coming and supposed to put out acoustic power. tional to efficiency. Horn speakers are for this particular opportunity we have Speakers are advertised that will handle inherently more efficient. And the third had to be with you.