EMUSIC-L Digest Volume 42, Issue 11 This issue's topics: Reproduction (no, not *that* kind...) Eno "third speaker" (3 messages) How Many Speakers fit on an Amp? (10 messages) Your EMUSIC-L Digest moderator is Joe McMahon . You may subscribe to EMUSIC-L by sending mail to listserv@american.edu with the line "SUB EMUSIC-L your name" as the text. The EMUSIC-L archive is a service of SunSite (sunsite.unc.edu) at the University of North Carolina. ------------------------------------------------------------------------ Date: Thu, 30 Jul 1992 11:40:36 CDT From: John C Edwards Subject: Eno "third speaker" ...has anyone experimented with the setup described by Eno on one of his Ambient albums? It consisted of a "difference" speaker which was connected to the two positive terminals and arranged in a triangle with the left and right speakers. Long ago in the Phonograph Days when I tried this...the "difference" signal had a magnifying effect on clicks/pops and rumble... Could music be engineered that would sound vastly different (voices, other tracks) by hiding the information in phase differences which would show up on such a speaker arrangement? I think that Eno indicated that this was partially possible...Also, would this setup have any detrimental effects to the amplifier used? I can see Channel/Channel separation getting worse from the through leakage. J. ------------------------------ Date: Thu, 30 Jul 1992 14:30:55 EST From: Jon Crystal Subject: Re: In Query On the Eno 3rd speaker experiment: I hooked one up almost a year ago, after the last round of discussions here. The impact varies considerably with the nature of music being played: virtually none from mono signal, to what seems almost l ike a third distinct track onsome recordings (phase differences?). Overall I li ke the effect - it is in fact similar to early surround sound efforts - and the re has been no evidence of negative impact on the amp. ------------------------------ Date: Thu, 30 Jul 1992 13:30:29 EDT From: The Radio Gnome Subject: Re: In Query On Thu, 30 Jul 1992 11:40:36 CDT John C Edwards said: >...has anyone experimented with the setup described by Eno on one of his > Ambient >albums? It consisted of a "difference" speaker which was connected to the two >positive terminals and arranged in a triangle with the left and right speakers. I tried this way back in highschool and got some pretty bad sound. Then again all I had to go on was my then limited collection of top-40 music. Maybe I should try it again with some Klaus Schulze or John Adams material... >Could music be engineered that would sound vastly different (voices, other >tracks) by hiding the information in phase differences which would show up on >such a speaker arrangement? This brings up an interesting observation/accusation in Stereo Review that too many sound engineers neglect to preserve a lot of phase encoded information in a recording in order to maintain decent mono reproduction. This is apparently more prevalent in pop production than in jazz/classical. >Also, would this setup have any detrimental effects to the amplifier >used? I can see Channel/Channel separation getting worse from the through >leakage. It depends on how much out of phase information is present in the signal. An ideal mono signal should produce NO output fron the third speaker. One with a lot of out of phase info could result in large amounts of current flowing between the channels, in essence operating the amp in 'partial' bridge mode. Very few amps are capable of operating in bridge mode since the impedance 'seen' by the amp is half of what it sees in standard mode. For example, driving 8 ohms bridged makes each channel 'see' 4 ohms. If your amp is very robust or is explicitly designed to operate in bridge mode, then I wouldn't see too much harm in using the third speaker hookup. Some early 'surround sound' systems would tap a signal off of the two hot speaker terminals and feed it to a separate mono amp to drive the third speaker. Hope this helps! Andy Wing ------------------------------ Date: Thu, 30 Jul 1992 07:57:40 EST From: Jon Crystal Subject: How Many Speakers fit on an Amp? Can anyone offer advice on the wisdom (or lack thereof) of connecting more than one set of speakers to the same output of a "consumer" amplifier? I don't mean using the "remote speaker" connections, but doubling the main ones for a total of four speakers. The amp is rated for 8-16 ohms: is that additive? Will this simply reduce the signal going to each - and at what cost? Thanks for the advic e. ------------------------------ Date: Thu, 30 Jul 1992 10:25:21 EDT From: ronin Subject: Re: How Many Speakers fit on an Amp? just to clarify martin's posting a little.... Serial Connection: † + ------> + SPEAKER 1 - -------> + SPEAKER 2 - ------† AMP † † † - <--------------------------------------------------† Parallel Connection: †------------------------------† † + †-----> + SPEAKER 1 - -----† †---> + SPEAKER 2 - ---† AMP † † † † - <--------------------------†--------------------------† what has to be watched out for is the parallel condition. as martin points out, serial connection is additive with respect to impedance (measured in ohms), where parallel is inverse multiplicative (divisive?). example: (again, somewhat redundant to martin) two 8 ohm speakers connected in series = 8 + 8 = 16 ohms load on the amplifier. two 8 ohms speakers connected in parallel = 1/(1/8 + 1/8) = 4 ohms. the trouble comes when applying power to the speakers. the basic equation is ohm's law, which is V=IR, or Volts = Amps times Resistance (resistance and impedance are interchangable in this context). now, let's say your amp puts out 16 volts at some point. flipping the equation for I=V/R, we get a current draw of 2 amps for the serial connection, and 4 amps for the parallel. and there's the lesson. parallel connections draw more power. this is why, unless your amp has open-circuit protection (which most actually do), you do not turn up the volume with no speakers attached. no speaker = inifinite resistance = infinite current draw. that's where the fuses go. anyway, the upshot of all this is that 1) hooking up speakers in serial is sort of ok... it'll just reduce the power output of your amp, perhaps forcing you to crank it a little harder to get the sound you want (if it's a cheap amp, it may distort) and 2) hooking up speakers in parallel is ok if you keep it down. turning it up will draw power twice as fast as you think. oh, one other little caveat. be extra careful with the parallel hookup if you're playing bass heavy music. very low frequencies reduce the speaker's impedance further, and draw yet more power. -----------< Cognitive Dissonance is a 20th Century Art Form >----------- Eric Harnden (Ronin) or The American University Physics Dept. 4400 Mass. Ave. NW, Washington, DC, 20016-8058 (202) 885-2748 ---------------------< Join the Cognitive Dissidents >------------------- ------------------------------ Date: Thu, 30 Jul 1992 11:11:14 EDT From: "Simon Weatherill" Subject: Re: How Many Speakers fit on an Amp? > your amp has open-circuit protection (which most actually do), you do not > turn up the volume with no speakers attached. no speaker = inifinite resistance > = infinite current draw. that's where the fuses go. anyway, the upshot of all Infinite resistance will actually cause no current - therefore no problem (I=V/R: 1 over infinity is 0). If you were to place just a wire across the terminals, you'd get infinite current. Simon ------------------------------ Date: Thu, 30 Jul 1992 11:14:00 EDT From: RAY BROHINSKY Subject: Re: How Many Speakers fit on an Amp? Where as a lot of what eric just said is fine and ok, this one statement: >you do not turn up the volume with no speakers attatched. no >speaker = infinite resistance = infinite current draw. >that's where the fuses go. is right out of the twilite zone! Infinite resistance means no current draw _through that path!_ The fuses go for other reasons if they go at all. If infinite resistance meant infinite current, we'd have some pretty hot plastics! raybro ------------------------------ Date: Thu, 30 Jul 1992 11:19:00 EDT From: wbf@ALUX2.ATT.COM Subject: How Many Speakers fit on an Amp? Date: Thu, 30 Jul 1992 07:57:40 EST To: Multiple recipients of list EMUSIC-L > Can anyone offer advice on the wisdom (or lack thereof) of connecting more than > one set of speakers to the same output of a "consumer" amplifier? I don't mean > using the "remote speaker" connections, but doubling the main ones for a total > of four speakers. The amp is rated for 8-16 ohms: is that additive? Will this > simply reduce the signal going to each - and at what cost? Thanks for the advic > e. Hi Jon, If your amplifier is rated at 8-16 ohms, that means that it is designed tosafely drive any load from 8 ohms to 16 ohms. That means if you use two 8 ohms in series, for a total of 16 ohms, you can drive them with your amp. ---------- /| | | ----/ | | +|-------------------|+ | | | | |----|- | | | | | ----\ | | | | \| | Left | | Duplicate this circuit for | Channel| | /| the right channel. | | | ----/ | | | |----|+ | | | -|-------------------|- | | | | ----\ | ---------- \| Bill Fox ------------------------------ Date: Thu, 30 Jul 1992 13:16:59 EDT From: The Radio Gnome Subject: Speaker Hookups Hi, Some thoughts on multiple speakers to one channel: 1) Unless you have a four channel amp, the 'remote' connections for speakers simply place the second pair in parallel with the main pair (unless of course the switch will only select one pair at a time.) This is probably why the ratings are for 8/16 ohms, meaning 8 ohms for one pair and 16 ohms for two pairs in parallel... the amp still sees a minimum of 8 ohms. 2) If the speakers have different efficiencies, then their relative volumes will be different for the same power input. 3) By driving more than one speaker, you've increased the complexity of the load that the amp sees. This could result in some unwanted sonic effects. Your mileage will vary :-) Andy Wing ------------------------------ Date: Thu, 30 Jul 1992 18:17:50 EDT From: ronin Subject: speaker loads ok, ok, so we all flip out every now and then... but i still maintain that putting signal through an unloaded amp with its output power raised places it at risk of being fried. i'm reasonably sure that i have not hallucinated this phenomenon. -----------< Cognitive Dissonance is a 20th Century Art Form >----------- Eric Harnden (Ronin) or The American University Physics Dept. 4400 Mass. Ave. NW, Washington, DC, 20016-8058 (202) 885-2748 ---------------------< Join the Cognitive Dissidents >------------------- ------------------------------ Date: Thu, 30 Jul 1992 18:32:50 EDT From: ronin Subject: speaker loading cont'd sorry for the quick, repetitive posts here, but i just had a conversation with someone here and... what used to happen when you increased the gain on an unloaded amp was that you were cranking the base current flow of the last power output transistor stage. the emmiter/base part of the circuit, being open, provided no compensatory load or drain on the flow, so the base/collector junction would blow. my memory of the fuses comes from when the solution to this problem was to fuse the bias line so that an extraordinary flow would take the fuse out, protecting the transistor. newer amps used transformer output stages. when the second side of a transformer is unloaded, increased current at the primary does nothing... it just flows around and dissipates. this was open circuit protection. now, i believe that this is done differently, with high power fet output stages. of this i'm not sure. does that clear this up? -----------< Cognitive Dissonance is a 20th Century Art Form >----------- Eric Harnden (Ronin) or The American University Physics Dept. 4400 Mass. Ave. NW, Washington, DC, 20016-8058 (202) 885-2748 ---------------------< Join the Cognitive Dissidents >------------------- ------------------------------ Date: Thu, 30 Jul 1992 15:14:19 +0100 From: Martin Roth Subject: Re: How Many Speakers fit on an Amp? Hi, you can hook up two speakers to the same output either PARALLEL (Y-like cables; both speakers are connected to the two outputs of the amp) or SERIAL (a line from amp(+) to the first speaker(+), one from first speaker(-) to second speaker(+) and another from second speaker(-) to amp(-)). [a drawing would be nice, but ASCII-Art is too much work :-)]. If you connect two speakers of 8 Ohms each in PARALLEL, overall impedance will be 4 Ohms (!!!ATTENTION!!!), if you connect it in SERIES, you'll get an over-all 16 Ohms impedance (no problem). No problems occur if resulting impedance is higher than the impedance of the amplifier; only you'll get LESS power if the impedance is too high. If the speaker impedance is too low (lower than amp impedance), you can damage the speaker and/or the amp when running at full power. Just to make that clear: you allways get the _same_ power out of your amp (which has say 8 Ohm impedance) if you connnect a) two 4-Ohm speakers in series, b) one 8-Ohm speaker or c) two 16 Ohm speaker in parallel. With your case (amp 8-16 Ohms), connect speakers SERIALLY if they have 8 Ohms each (total of 16 Ohm), or connect them in PARALLEL if the habe 16 Ohms each (total of 8 Ohm). Hope this helps -Martin _______________________________________________________________________ _ Martin Roth Martin Roth ETHZ, ips, RZ F16 |\ /|_) Mail: roth@ips.id.ethz.ch Sandacker 14 g 01/256 55 68 | \/ | \ (Student of Computer CH-8154 Oberglatt p 01/850 32 75 Science / Engineering) Switzerland (F-)emails welcome! ----------------------------------------------------------------------- ------------------------------ Date: Fri, 31 Jul 1992 09:29:00 EDT From: RAY BROHINSKY Subject: Re: speaker loads Eric writes: but i still maintain that putting signal through an unloaded amp with its output power raised places it at risk of being fried. i'm reasonably sure that i have not hallucinated this phenomenon. (end quote) No, you're not hallucinating. In Tube amps that use the speaker as the sole load for the output stage, an open speaker connection can result in the grid being used as a quasi-plate. Grids are meant for controlling, not sinking lots of current, and they burn up when they take on current. Open-circuit protection was, I believe, pretty simple: supply a low-current path to keep the plate positive with respect to the grid, or ensure that the grid never, ever, ever, gets more positive than the cathode. In transistors, something similar can happen if the collector is open and the base-emitter junction gets forward biased, but again, there are safeguards. Short-circuit protection is a bit more complicated to do right, since you have to sense the current and determine what is a bad level, and then choke off the supply to keep current flow in a reasonable range. Still, this isn't hard to do, and there are even IC's on the market that do it without putting unacceptable noise in the system (if it's designed right). raybro ------------------------------ End of the EMUSIC-L Digest ******************************