Patching it up

May 23, 2021
 by Paul McGowan

In yesterday’s post I shared with you the fact we’ve made some changes/modifications in Music Room II.

I was quite surprised to find reactions to that news varying from outright betrayal, shock and horror, to ‘atta boy’ and everything in between.

Thank goodness no one in this crowd has any strong opinions. 🙂

The problem we were addressing in the IRS V is a well known one since the introduction of the speaker in 1985. The midrange planar ribbons called EMIM (Electro Magnetic Induction Midrange) can almost stretch as low as 100Hz, and it does, but down at that frequency there’s a ton of distortion as the driver has exceeded its practical limitation. IRS owners know to add the Infinity mandated external high-pass filter to their power amplifier inputs in an effort to reduce the problem.

Having on and off lived with these speakers for decades I am fully aware of this minor limitation and make mental allowances for it when listening.

But back to our story. A little recent history will help us understand what we did and why.

About a year ago our loudspeaker genius, Chris Brunhaver, convinced me it was time to retire the IRSV woofers in favor of more modern and better-designed woofers. Our chief engineer, Bob Stadtherr designed an entirely new woofer amplifying system that removed the IRSV servo and implemented a 1,000-watt amplifier for each woofer. We then used DSP to get the woofer system as flat as possible. Once finished, the changes were remarkable.

Of course, back in 1985 when the speakers were state-of-the-art there was no such thing as DSP nor were there woofers designed like the ones now used in the system.

Now it was time to fix the IRSV EMIM distortion problem and make yet another major improvement to this already wonderful loudspeaker system. Chris proposed designing a custom notch filter with a very specific Q, gentle enough not to be noticed, yet high enough to rid the EMIMs of their distortion without touching elsewhere in the frequency band. Once done we would have a hole in our response that would be filled in with the new woofers—drivers more than capable of excellent performance up to several hundreds of Hz ( a feat the old IRSV woofers were incapable of).

Here’s what the notch filter looks like:

With this network in place the distortion’s gone and the EMIMs have less to do thus helping them sound better overall.

Next was the task of moving the woofers closer (since they now operate at a higher frequency) and readjusting the DSP system. Here’s what they now look like relative to the wings:

How’s it sound? Wonderfull. Much punchier, cleaner, and (hard to believe) even more resolving than I’ve ever heard it (and that’s saying a lot).

As I wrote yesterday, weekends are great times for rethinking your system and I am glad we spent the time and engineering resources to take our reference system up yet another notch in performance.

Change is never easy and often requires a bit of bravery. Consider that when it comes to making fundamental changes to one of the world’s great reference systems, I’ve got to balance the outcome to the reaction: a combination of social shaming, catcalls, compliments and kudos.

Change is tough but if you move in the right direction it’s the right thing to do.

Subscribe to Paul's Posts

58 comments on “Patching it up”

  1. “outright betrayal”…hahahaha!!
    (I honestly though that this part 2 would’ve been called ‘Sunday Morning’)

    Paul, I have to admit that when I heard you talking about the EMIM’s in your
    IRSV’s having to play down to 100Hz, late last year, I was scratching my
    head thinking, “Hmm, that can’t be right…he’s drunk again” & I didn’t give it
    another thought because, after all, you’re Paul McGowan, so I was not
    surprised when you presented the information in yesterday’s post.
    I know how extremely (epoxy resin-like) an audio-nut 🙂 can get attached to
    his/her favourite long-term loudspeakers, so I’m very happy for you that you
    have Chris Brunhaver (the loudspeaker wiz-kid) to not only keep your beloved’s
    on life support but to be able to inject them with a healthy dose of adrenaline.

    Good for you!!

    This might also lift your spirits further.

    **PS Audio vs PS Audio**
    (Windom vs Sunlight)
    ‘A Canadian Perspective’
    (Main article from 15:30)

    1. …it’s probably easier to find the bravery to make the
      changes when you’ve got Excalibur (CB) by your side.

      Now the subs look like they’re playing ‘Peek-a-boo!’

  2. I really never understood the design ideas of the IRSVs. Real subwoofers should only play frequencies below 70 Hz. And in order to play time aligned with the mid range drivers the subwoofer should be positioned nearer to the listener or the satellites have to be delayed. Not to mention the lack of DSP-equalization. But as Paul mentioned: an audiophile accepts a lot of sound degradations in favor of getting his preferred spectacular (!) sound “effects”! As the cocktail-party effect shows our brain is the best filter! And look at the tiny wires of the notch filter. Why using thick monster snake like loudspeaker cables?

    1. “cocktail-party effect”
      What’s that paulsquirrel?
      The ‘drunk with nostalgia’ filter??

      I have found that 1.5mm diameter (15 gauge) ‘solid core’
      copper wire will cure you of believing that “thick monster
      (vomit) snake (oil) like loudspeaker cables” are ‘da bomb’ 🙂

      1. And let us not forget to include the “servo activated stamping and seizing module that will keep a man punching until he’s drunk with power.” — FsT,yet again

  3. Sounds like the IVRS speakers are knackered. Lucky for Paul he’s got a massive skip down at Octave Records.

    Yesterday we demo’d an immersive sound and light system. Amazing. Another nail in the 2-channel coffin, during which my wife said “are we keeping the other speakers?”, referring to the Wilson she chose 6 months ago. We then bought a dishwasher (I like dishwashers), two OLED TVs, another camera, and then went to a recital of such transcendental magnificence that anyone who thinks home audio is even in the same ball park as live music is sorely mistaken.

      1. I thought I squeezed in an OUCH in both paragraphs.

        And so to bed, or as I used to tell my kids: Allez, allez, allez au lit, n’oubliez pas de faire pipi.

        1. I like to play the sound of a rainstorm to help lull me to sleep. Despite the title, this one sounds like an actual recording of rain and thunder, rather than poorly modulated noise from a synthesizer (either hardware or software). Looped, of course, but skillfully done.

          White noise / Bruit Blanc 8 heures pour dormir paisiblement. Pluie et Orage Sound Rain and Thunder
          Mr.P7 Mar 8, 2019

          Gorgeous cobalt blue sky, too.

  4. Proper job Paul & here I was congratulating myself on my 3 way Amp to speaker switch!
    2.5mm cables but only short runs inside the box & no doubt a few extra ohms but it seems to work well – my lazy man’s mood changer…. ; )
    Ever thought of PS considering doing similar accessories/cables etc only with top quality parts & labour?

  5. “Change is tough but if you move in the right direction it’s the right thing to do.”

    True on both counts, but what if change takes you in the wrong direction. I’m sure this is why we are cautious and reluctant about making changes especially to our precious systems. With apologies to Joni Mitchell…..

    Don’t it always seem to go
    That you don’t know what you’ve got
    Till it’s gone
    Not great sound
    But still you spent a lot

    Despite that don’t think I’m being negative or trying to discourage change. The hi-fi system is an evolutionary beast and I can’t immediately think of any change I have made that I regretted. Quite the opposite. Change encouraged me to listen more and provided ever increasing enjoyment.

  6. “The problem we were addressing in the IRS V is as well known one since the introduction of the speaker in 1985.”

    Maybe the reason for some of the reactions yesterday have to do with the statement above. What you assume is well known may not have been so “well known” to others. So speaking for myself it was a shock to hear about up to a 10% distortion in a certain range and then even more shocking to hear that for years it has been compensated for in the “mind”.

    So yesterday’s post was a bit of a shock, especially given all the previous gushing and promotion.

    What was never questioned is that once you were convinced the issue needed to be addressed it was. So congrats on the improvements and having the manpower and expertise at hand to address the weaknesses of the original design.

  7. I assume our going to put the home made notch filter into some kind of case so that the components are protected and not just lying on the floor. I hope you realize that only someone with your kind of resources could have this kind of love affair with a 35 year old room filling speaker system.

  8. Hmm… I wonder…is Paul McGowan the John Hammond and his company the INGEN of the audio world ?
    Bringing back to life these monsters from the past, recreate speakers that became extinct a very long time ago.
    Once they were king, but those days are long gone. These Jurassic juggernauts don’t fit in the modern world anymore.
    Too big, too slow. Who can and wants to have these behemoths in his or her living room.
    Maybe mr. McGowan if he could have it his way, but fortunately he married a very sensible woman.
    These Kong’s of the audio world are replaced nowadays by much smaller and “quicker” speaker-thoroughbreds.
    Like for instance the little “velociraptors” in the McGowan living room. Little “wireless” predators of our time, a.k.a. KEF LS50 (did anyone know LS stands for “Little Saurier” ?).
    These IRSV speakers have written R.I.P. all over them.
    So, again, is PmcG the Hammond of the audio “family”, OR the new Spielberg in his own movie : “Infinity, the lost world”. I’m also thinking of a sequel to “The Thing”.
    Time will tell.

  9. The only real constant is change. You’ll never know how good you have it till you change it – but then maybe the new will be better than the old . . .

  10. Paul, I too was a little surprised at how you’ve tolerated this apparently minor flaw for so long and ‘fessing up to it! 😮 However it makes me wonder, did Chris or any of you have this mod in mind when he redesigned the woofer system or was the solution more serendipitous?

    1. No speaker system no matter how well designed or thought of is without flaws. And most speaker systems have big flaws—at least “big” relative to electronics. These are the accepted compromises one accepts when you get into building a state-of-the-art system. Tradeoffs.

      I got the IRS system because at the time, and still to this day, there are no other systems I am aware of that can do what this system does. And I/we need what it does so well.

      So, for example, one of the system I might have chosen would be the big Wilsons. They are extremely revealing and would have served our purposes of aiding in design and being an ultimate reference. But they too have their shortcomings. (No, I am not going into them since I get scolded when I do that).

      Then there’s the other reference system I considered, the big KEF Blades. Again, they too would have been an excellent choice but….

      In the end we choose what gives us the fewest compromises in exchange for the greatest benefits. The IRSV more easily and quickly show off everything in the music (and I mean everything). No details escape them. And lastly, when the recording/equipment gets it right, the reward from this system is so wonderful that they make you glad you spent the time to get it right.

      I know of no other speaker system in the world that can lay claim to all that in the same measure as this system.

      1. Paul, I understand you must have a hard time with some of us (me included) posting guesses (not even opinions) about the vintage IRSV, they mostly never heard and which for sure sound far better than what we all have at home. The fact you since some time (and not just after Chris joined in) work on optimizations, for sure alone is proof for your awareness of weaknesses, how ever minor or major they are. It’s now interesting to guess if/how an existing speaker can be completely redesigned successfully with many variables fixed, but you have the right know how resources available.

        From your post above I got that you still count (maybe even the vintage original) better in sum than all or most of what else you know. And you have more experience than all of us here together. So there’s nothing to really argue and I respect that.

        When I now psychologically analyze and treat myself and part of the crowd, I think what remains is maybe still a bit of skepticism about vintage hifi when it comes to such speakers (although I know how good some of them still can sound, just thinking about some Apogee and considering that e.g. the big Tannoys are sold and seen as reference systems since decades) but mainly I think it’s that bit of relativation and orientation missing for the confused, after the huge IRSV, then the small IRSV killers, generally servo control, then not, active bass, then not, no passive chassis, then yes, the first, second and third generation of PSA speaker design drafts etc. were the latest sh*t on earth with the described big improvement between them. If the origin (IRSV vintage status) already was one of the worlds best reference systems til today and each of the multiple development incarnations of their successors were major improvements (at least in many characteristics, size set aside), then it’s difficult to understand why the IRSV had only very minor shortcomings. If we then take into account that even your final new speaker with all those superior qualities will have to face competition on the market (and not all competitors will probably agree with that upcoming superiority), then I think the doubts mentioned about the original IRSV’s just minor shortcomings, are mainly based on the will to get both stories together…the (from your descriptions) perceived big quality difference between it and countless optimizations of the new designs (which will quite sure still have serious competition)…and the one of the still absolute reference quality IRS with even today very minor shortcomings. That’s also why I mentioned your a gorgeous sales man because we want to believe averything, we just struggle a bit at times, sorting it all out 😉

        1. Well said and thank you. I think the thing that’s maybe missing is that I am not a steady rock. I move back and forth between what’s wonderful and working to what can be and might someday be. And then I discover something unexpected and exciting and I share that with this community. And that something might just contradict the last thing.

          I have never thought of myself as a salesman. I don’t try and sell anything. Rather I try and share my excitement and enthusiasm as genuinely and expressively as I know how – and at the same time, I too share my disappointments and mistakes. And I think this often gets people worried or feeling insecure.

          I am just like the rest of you.

          1. Yes, maybe I should have said „you can tell sories well, people like to listen to what you state and want to take it serious“. Although that’s also a major strength of a sales man, the term alone might be too negatively handicapped.

            I think the capability of the above and the maybe reduced focus on a possible virtual classification, scale, comparability or whatever we’d call it, is probably the challenge.

            And you’re right, we’re all like that. If we add all the small improvements we here and there describe from our optimizations, we must have reached nirvana a few times already. The lucky position of us others is, no one takes us too serious anyway 😉

  11. The fact the system now has better resolving power does not surprise me. After all, by ridding the EMITs of the burden of going down to 100Hz, you allow them to do what they do best. It’s similar to if Hussein Bolt had been running all these years with ankle weights, then he removes them. That would be a huge increase in performance.

    1. Hi Neil,
      I hope you’re well…well, as well as well can be 😉
      FYI: the mids are the EMIM’s & the tweeters are the EMIT’s…just for future reference.

      1. Hey Martin, why would you think I didn’t know about EMIM’s and EMIT’S. I am one of the few fortunate people that have had more than a few hours of listening experience on the IRSV system that began at Lyric HiFi the day after Arnie Nudell set them up there with Mark Levinson electronics. This all happen by coincidence as I used to frequent lyric on a monthly basis back then. I was indeed a lucky guy. I don’t think that I have heard anything like them nor will I ever. The experience was incredible then and I’m not exactly sure what Paul is Referring to in this post but it’s so many years ago and my ears might not have been that educated back then. The setup in room II is no joke. It needs tender loving care and who can do it better than Paul who was involved in some respects with the design of these behemoths.

        1. I was replying to ‘tarheelneil’ not to you ‘stimpy2′
          …what you smokin’ there my friend? 😉
          (Unless ‘tarheelneil’ & ‘stimpy2’ are one & the same)

          1. Sorry FR. I’m bouncing around the replies not reading every word so I threw my own personal experience with the IRSV’s into the mix seeing my name mentioned.

  12. I’d love to listen to some music from your IRS setup – someday. Maybe new PS Audio speakers will replace them in a couple years.
    What woofers did Chris use in the IRS?

  13. Paul, two reactions:

    First I agree with Mike and others about how you tolerated this real problem without mentioning it previously. At least I don’t recall your ever admitting it before, you only raved about how wonderful the IRS sounded.

    Second, I know you are acquainted with Jim Smith. In his book, “Get Better Sound”, he makes a point that every music system in his experience which is musically satisfying “must have a flat-to-slightly-elevated response curve in the critical region from approximately 192 Hz to 384 Hz.” Any system which is lean in this area will usually prove to be boring musically. You didn’t say how far above 100 Hz your correction impacted but it sounds like the problem may have fallen within the lower end of this range. So again, surprising that you previously raved about these speakers for so long.

  14. Another notch in Paul’s gun in the quest for superlative sound. Bravo…. for a moment there it was like the Perils of Pauline waiting for the next segment tied to the pugnacious past, time loves a Hero !

  15. Understand how you feel, Paul! Congrats on the Step-up in the right direction! 🙂

    Thought I had a “cats-meow” stereo set-up, but over time, little indications pointed to some things amidst.

    Over 2 years of incremental tweaks (including HP/LP to the monitors/sub, cleaning up speaker wires, replacement of inner connects, very fine tuning of monitor placements/toe-out and finally, monitor upgrades) have transformed the presentation into a detailed, well defined holographic huge wall of live sound (monitors completely disappear)!!

    It’s not perfect (it’s a stereo), but often music reproduction is mind-blowing, defying the laws-of-physics (.5cf monitors) with realism and presence that sustains suspension-of-disbelief! For me, I’ve found my re-worked IRSV! 😉


  16. Bravo! When I read yesterday’s post, I could hardly wait until today to read what exactly you had done to the Mighty IRS. It looks like another step closer to getting every last drop out of a recording.

  17. Congratulations in improving an already great speakers system. I endorse moving the bass columns nearer to the main panels. In my experience, the bass columns do contribute to the mid range sound.
    However, it is hard to imagine what the improvements sound like by just reading the words. How about posting the improved sound onto YouTube for us to enjoy?

  18. I haven’t read through all the comments, but my takeaway from this would be summarized as … if you fancy buying yourself an old pair of IRS Vs, buy yourself a fair-sized high-end audio company first 🙂

    1. I was told there are several people in the world, regularly buying exactly the same equipment Michael Fremer uses himself (changing from time to time).

      Not sure if Paul already has some interested parties following his upgrade path. The necessary money might sit where those speakers are.

  19. Hi Paul,

    Did you enlist external DSP or make changes internally? I am curious because I am thinking of trying an external DSP system and wondering what brand/model to consider.

    1. Joe,
      Show me a ‘normal’ midrange driver that comfortably & competently goes down to 100Hz given that it has to deal with going up to around 2.5kHz.

      1. Fat Rat a headphone driver could go down that low. In fact it can produce sounds below 20hz and flat too. It’s a matter of output when going that low and the only way the IRS midrange is able to go that low is due to the use of many of them to divide the output thus reducing distortion levels. If you took that same IRS midrange and installed it into a basic 3 way system there is no way Infinity could use that low of a crossover point.

    2. Well, yes, I suppose so but that might have been a bit too radical as we would have had to design and manufacture new planar midranges from scratch so they would fit. That’s a bit more of a challenge than we were up for.

      1. The notch filter is a Band-Aid. You might lose some transparency. If you ever did decide to take on that challenge I bet it would sound even better with or without the notch filter. Another IRS V owner fixed the problem by adding multiple bass couplers to the wings of the speakers. I don’t believe the tweeters will ever need to be upgraded. But the midranges are going into deep territory even if the watts are divided into many of them.

  20. The friend who first turned me onto your posts, when I told her how much I enjoyed them, suggested (but more delicately) “don’t tell me, tell him.” So:

    I much appreciate the engineering, but even more cherish the sheer joy with which you report your pursuit of perfection. And don’t be too troubled by folks who conflate “not the way I would do it” with “wrong” — they’ll be along shortly. It’s a mistake I catch myself making all the time. (Which sounds like I catch myself *every* time. I wish!)

    I thrill to the sound of your new reference room, even though I will never hear it: after a careless lifetime, my hearing rolls off at 4 kHz. And it doesn’t take a thousand watts of bass to knock me off my feet, either. (Would that be a “kilowoofer”?)

    There are reference rooms and reference rooms; your audience may vary. One of my favorites was a six-transistor radio, playing 1960s Top 40 on a dying 9v carbon-zinc battery, beachside on the Jersey shore.

    Anyway, pun most sincerely intended: thank you for listening.

  21. WHAT?? You changed the Holy Grail IRSVs? What a Pollywannacracker! Just goofin. As a halfassed speaker builder meself, I am all for tweeking and experimenting to improve. Not such a bad thing to have a Wunderkind on staff, is it?

    Keep improving.


  22. Hold on, let me make sure I understand this correctly. You guys “implemented a 1,000-watt amplifier for each woofer”… so, are you running like 12kW to the woofer towers?!?!

    Don’t get me wrong, that is obviously kick-ass. But, I just wanted to double check.

    1. Yep. 1kW per woofer X 12 woofers. Now while that sounds impressive the 6 woofer amps per sub are few dromn a single P20 (so 2 P20, one for each tower). Which means we can’t deliver more than a few thousand watts at any one time.

      Headroom baby. Make sure them amps are just loafing!

  23. May I add? I noted a long while back, that in the specs, one of the IRS xover settings puts the mids down to 70Hz! I found that a bit of a stretch myself. Also, Paul (was) calling the bass modules ‘SUB-woofers’. Clearly at the 100Hz xover point, they are really low mid/bass speakers.

    No one really talks about how (in my opinion) Martin Logan gets it right. The large electrostat panel on my Classic 9’s is crossed over at 280Hz., well out of the sensitive spot of our hearing, AND well above the distortions Paul noted with the Infinitiys. And the bass module at the bottom is just that, performing as woofers.

    Not sure why so much additional power for them, I would think that 2kw per channel would have provided plenty of head room/ dynamic range for concert hall levels, guess it is stated that it all CAN’T be utilized at once….. lol.

    I DID take Paul’s advice and recently added the ML 1600X subwoofer with Anthem room correction. It is crossed over at 35Hz.

    Thanks for reading (O:

  24. Hi Paul,
    In your mod on your infinity’s you showed a photo of the Notch filter. I know Fat Rat asked about where crossover point is? I have heavily modified Ref 4.5’s and have built new crossovers from original schematics. Could you provide a simple sketch of devices and values? I bought my speakers new in 1979 and have never looked back. Replaced all front end components many times over. Love my Infinity’s!

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram