Why are the Infinity IRS, Reference 1b, Beta ect speakers passive on the mids and highs?
I would like to know why all of the classic Infinity's and most other brands use passive crossovers for the mid to high transition? I don't think it was for cost and that level. Is passive better? Has anyone compared both to know which is better?
I think it's a practical matter. How many amplifiers do you want to have in your system? While it makes practical sense to separate the bass frequencies from the rest of the speaker system - especially when the manufacturer has already committed to using separate bass towers, as in the RS-1 and various IRS systems - there's less advantage to be gained by separating the mid from the high frequencies.
Separating the bass and using an active x-over also allows use of a servo system for those frequencies, which Infinity did with great success. I don't think it's feasible to use a servo on higher frequencies, however.
Why are the Infinity IRS, Reference 1b, Beta ect speakers passive on the mids and highs?
I think they found as I did. (with both ways bi-amped) I found good passive 18db xovers on the upper bass/ mids and highs of my Martin Logan ESL Monoliths III sounds more natural than any active crossover I tried, they all had a sterility to sound. But on the 12" bass 150hz down I use and my own 24db active.
I don't think adding another amp would be too much in cost or space but it might be that the sound is actually better passive. Has anyone compared both ways all other factors being equal? Marc
No, Passive is not better.... I think the real issue is that these are all older speakers. When these came out, small amplifiers were not abundant as they are today. At Marcof, we were experimenting with building chip amps for just this purpose. If Infinity revamped these today, I would not be surprised to see the drivers individually crossed and amped.
The quality of the passive xovers in those higher end Infinity’s were beyond reproach, and for the mids and highs far better than listening through the half a dozen opamps per channel of an active xover.
But for the bass, yes by all means use an active xover as then it can be direct coupled, which is always better for bass.
Hi Marc, Well, I can at least speak to the RS1B.... On the top end, we found that carefully matching the sensitivity of parts and fixing their output rather than using the pots was noticeable improvement. We never really changed the design, but a few key parts changes also made a difference. No doubt, they were well thought out...Not sure about beyond reproach. I'd rather listen to them matched with a good electronic crossover and amp any day. No comparison. If anyone truly multi amped these speakers with quality product, then take the time to properly adjust them out, t I would be surprised if you liked the passive crossovers better.
An active electronic x/o need not necessarily contain opamps, or ic’s as well. As an example, my First Watt B4 is free of both, and it’s use in a bi-amped speaker system can make for a better product than almost all non-bi-amped speakers using speaker-level filters.
I am a former Infinity RS-1b speaker owner (high end dealer Brooks Berdans personal pair), and I consider my current Eminent Technology LFT-8b/GR Research OB/Dipole Sub speaker combo, bi-amped via the B4 (sort of---the Sub is self-powered), to be considerably better than the RS-1b, and they were the special-order Cardas-wired version. The RS-1b outboard x/o was, claims to the contrary, rather shoddy, as was the structure of the panel, ridiculous at their price. The wooden panel wing needed a brace added from its top edge to its base to keep it from swaying in the breeze---a few wood screws held the wing to the base, sort of. Ridiculous! The Emit and Emin drivers were good, but the way they were employed left something to be desired; different frequency ranges appeared at different heights, for one thing. The best part of the RS-1b was its servo-feedback bass system, each enclosure containing six 8" servo-controlled woofers. Really punchy! But the GR Research OB/Dipole Sub, also a servo-feedback bass system, with two 12" woofers mounted in an H-frame, is a whole lot better. IMO, of course!
Having this idea allowed buyers to use boutique tube amps for the mid/highs, letting them feel part of the process of creating the sound.
Certainly when these speakers came out, no one wanted an all-active speaker in the high end market. That is pretty much still the case (sadly). There are almost no fully active speakers in the "high end" while they thrive in the pro world.
Don't shoot me, I know active speakers are out there, I'm just saying they have not yet caught on, and may never.
@bdp24 "The Emit and Emin drivers were good, but the way they were employed left something to be desired; different frequency ranges appeared at different heights, for one thing". This is corrected when you hand match the drivers for exact output and hardwired the pad. The pots were poor.
Pots are always a problem when old in any speaker, you replace the pots with appropriate resistor L pad power resistors of the same values for your preferred setting.
I am learning here. Timlub, it sounds like you have experience on the RS1B. Is there anywhere I can get specifics on what to do to improve the RS1b. What made me start the thread was an experiment I did with the RS1B's I have. I disconnected the crossover and tri-amped the system. I know a lot changed because all of the EMIM's went to a midrange whereas passively 6 are 150-700 and 1 is 700-3500 but I did not like the sound as a simple 3 way. I have another system where I have 10 EMIM's and a ribbon tweeter from a magnepan 20.7 where the tri amp works pretty well but falls short in some ways to the RS1B. I am trying to figure it out. Marc
I have not triamped these, I've never owned them. I worked for Marcof/SpeakerCraft, we did alot of custom work and I got these to work on years later because of my time there. I worked on the crossovers only... Measured each ribbons output individually, then customized the pads within the crossover. So without having complete driver measurements, that's why the basic design wasn't changed... better caps and coil or 2 and hand matched pads. For me, as long as your amps can handle driving ribbons (some cannot) and you have a very good crossover, this is a no brainer. It will take a bit to get these matched, but I would expect excellent results.
partroysound: If your m/t panel wings move relative to the wings base as much as mine did, I would definitely attach a length of hardwood (or metal rod, like the Quad 2805 and 2905) to the backside of the m/t panel wing near its top, the other end secured to the wings base. A small hinge at each end would work fine. Those wood screws "securing" the wing to its base plate are woefully insufficient for the job. Then, an outrigger fitted to the bottom of the wings base plate will create a deeper footprint (the base plate is too shallow to provide front-to-back stability), locking the EMIM and EMIT drivers in space by preventing fore-and-aft swaying, for increased resolution and transparency. The bass enclosure can also use a wider footprint, again via an outrigger. These measures address and remedy the structural weaknesses of the RS-1b. The drivers and their filters is a whole 'nother matter.
Having seen a few fixes online, I have to say @timlub is most probably right.
These crossovers were done very "empirically" and I doubt any designer with modern measurement and simulation tools today would arrive at the same crossover designs. Most designers today would absolutely scratch their heads about the choices being made, as well as some of the impedance problems that resulted.
Active crossovers properly designed would be head and shoulders better than the originals, if only from being better measured and better tuned. :)
However, with good passive crossover design and good parts I think they could be brought up several levels today for use with your favorite amps.
I haven't finalized on the crossover points yet. On the tweeter 3500Hz to 4000 Hz is good, the low point on the mid is very tough. 150hz to 160hz leaves a gap to the bass modules. As I drop the frequency the sound doesn't change much until the EMIM's start to distort for being too low. I have tried a Tympani panel for mid bass 70hz to 135hz It sounds good but not a great look. Marc
If I recall correctly, they are at 125, 700, 3k & 8k.... Does your crossover have a choice of Bessel, Butterworth, Linkwitz Riley etc? Can you change slope? Meaning 12db per octave, 18, 24...or even 64db per octave? Please let me know... type of slope and amount of roll off effect phasing and blending, plus each type of slope sounds different. Once I get that, I'll give you some basic advice to try.
I can only change frequency. Slope is 18db/octave. I don't know Bessel, ect... You are correct on the crossover frequencies except I thought it was 140hz not 125hz.
I have a couple. I have 2 audio control units from a Richter scale and an Epicenter. I can swap crossover frequencies easily. I also have an Audio Research tube crossover ec-4 which is hard to change. I think it is 12db on bass and 6db on mid and high. I am using that on a set of Tympani 1D but would swap it out if it was a better choice. And lastly I have a Nakamichi ec-302 which is a 12 volt unit on a power supply. The Nak can select 10hz steps so it is great to experiment with. Marc
140Hz does seem a little low for the EMIM’s to go down to , my epsilons L-EMIM goes down to 140Hz but it is a large driver, would be nice if the woofers were able to go a little higher.
Hi Marc, Ok, I'm familiar with most of what you have... The Richter Scale is not really adjustable enough, I was made for a fairly low crossover point for a 2 way with an eq on top. The Epicenter and the Nakamichi are 12v products... The Nak would probably work fairly well. It one of the better units for car audio. As much as I'd like to see you use the Audio Research the 6db slopes on top will not work well with the ribbons. There are more slopes than besel, butterworth and linkwitz riley, but these are the main 3 used today. The act differently with how linear they are and how they handle phase. Depending on frequency, 1 can cause smearing, another will bring a rise at the crossover point and another will be flat at the crossover... Being able to mix slope, we can correct some errors... Mix as in 12 up and 18 down... Although most electronic crossovers today are linkwitz riley 24/24 and that is always a good choice... even though you may like the sound of another slope, when in doubt, Linkwitz Riley is the go to... So the Nak is 18db per octave and I don't know what type of slope it uses.... but you can't make any changes, so it is what it is. The Nak will allow you to experiment, but if you ever want to get the most that you can from these beautiful speakers, you'll need something more versatile. What does it sound like using the published crossover points of 700 & 8k? and of course you realize the real problem comes in level matching.
The first Watt B4 x/o offers 1st/2nd/3rd/4th order (6/12/18/24 dB/octave) filters in 25Hz increments from 25Hz to 6275Hz. It is about to be discontinued, so if you want it better grab one while you can!
I thought that I'd stop speculating and look these up:
Here are the spec's from the Stereophile review: Description: Five-way, two-chassis, dynamic, floorstanding loudspeaker system with four EMIT ribbon tweeters, seven EMIM ribbon midrange units, and six servo-controlled polyproylene-cone woofers. Electronic crossover frequencies: 125Hz, 700Hz, 3kHz, and 8kHz. Dimensions: LF section: 60" H by 14" W by 15" D. HF section. 60" H by 22" by 13" D. Price: $5295/system (1985); no longer available (2006).
Hi timlib The Richter scale crossover is changeable through a chip. I have lots of values for it. You can even make your own crossover point, they give a formula in the manual. The other audio control is not an epicenter, I made a mistake but it uses the same crossover as the richter so using the two I can get a 3 way at any point I want. I am glad you have some respect for the Nakamichi, I wasn't sure what you would think about mixing car audio .I haven't tried using 4 amps at the factory crossover points. If I do I will let you know. Setting levels isn't a problem as I have an Audio Control 3050A spectrum analyzer. I am still not sure using so much electronics in the signal path is the best route to go. I really appreciate your help. Marc
Not sure If I can add much to this discussion but active crossovers would have been more expensive at the time. The system was already expensive when you add the amp and other requirements. I own two sets of the RS1B woofer towers and the servo controller. These parts are getting old and my servo now has a low level hum that it didn't have before. I now use an active crossover without the servo system and get very satisfactory results. The RS1B was a great system but after listening to the Infinity Emits and Emims for the last 4 decades I have moved on.
One of my best friends on the planet is an old timer very respected figure in the audio world. He uses a 12v crossover with a passive sub, of course, he modded it with some nice parts and built a fully regulated 12v supply, but it does a great job. Years gone by, I worked for Orion. I've heard of few of their amps in home use and at one time I owned a stereo store, we did 12v to high end...and there, we tried a soundstream amp with some home gear... I've seen very respectable results mixing.... Never true hard core audiophile with the amps, but some truly respectable results.
Hi Timlub I too am in the car audio business 36 years now. I have a 100 amp Audio Authority power supply. Lots of fun listening to a lot of car audio amps through the different speakers I have. There is quite a bit of difference among brands and models. Marc
I am not familiar with the Triangle Celius, but the rest are awesome speakers. I also like the Magnepans alot. I am not sure which I would pick if I had to make a choice.
You must have a verified phone number and physical address in order to post in the Audiogon Forums. Please return to Audiogon.com and complete this step. If you have any questions please contact Support.
