Page 5 of 15 FirstFirst 1234567891011121314 ... LastLast
Results 81 to 100 of 296

Thread: New Harbeth P3ESR

  1. #81
    Join Date
    May 2009
    Location
    England
    Posts
    847

    Default Re: New Harbeth P3ESR

    Respect !!!

    And respect for making some truly excellent speakers. I'm just annoyed with myself for not discovering them years ago. I've been wasting my time all these years.

    David

  2. #82
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default P3ESR - measurements @ BBC anechoic chamber

    I attach pictures of the prototype P3ESR being measured. I was undecided as to which direction to steer the woofer's characteristics and in such a situation the only option is to make a selection of variations,and then take them to an environment well away from reflective surfaces. Then it's the fiddly task of installing them one by one into the reference cabinet, calibrating, putting them in position and taking some acoustic measurements. Working entirely alone and unaided, and with the pressure of $2000/day hire charge, one has to have a very clear measurement strategy when one arrives to get the most value from the session. There is not much time for analysis* - that can be done later back off-site. Every effort must be made to make absolutely reliable, repeatable measurements or subsequent man-months of computer simulation will be garbage.

    I should add, in answer to an earlier comment about the 3/5a, that the BBC were extremely fortuitous that the off-the-shelf KEF B110 (midrange) driver had enough bass to be used in what became the LS3/5a. In my view, any speaker brand with pretensions of designing a really high quality mini-monitor must design and manufacture the bass/mid unit themselves. This is the only way to optimise the parameters of sensitivity, extension, power handling, distortion etc.. A bought-in woofer from a far flung supplier who has no in-depth understanding of or more important, empathy with the magic of the BBC mini monitor concept cannot hope to balance the technical parameters suitable for the 'British shoe box monitor'. There are very few of us woofer manufacturers amongst the last remaining 'British' brands.

    *I wish it were this simple. The video camera was running during some of this session. On playback, I was surprised how many minutes I stood unmoving staring at the measurement computer's screen undecided whether woofer A, B or C was best. Weighing up in my mind the balance of characteristics as displayed and whether a further ad hoc adjustment could be made on site to make the best candidate even better.

    >
    Attached Images Attached Images
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  3. #83
    Join Date
    May 2009
    Location
    England
    Posts
    847

    Default Re: New Harbeth P3ESR

    There are countless 'box stuffers' out there but the only way to design and make a truly great speaker is to have the skill, knowledge and facilities to design and manufacture your own drivers. Without this ability, a design can only go so far.

  4. #84
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Skills needed to design at this standard ....

    The primary qualification needed to design good monitor-grade loudspeakers is a reliable internal memory of live sound. All other skills can be acquired if you have a natural curiosity about audio equipment and are willing to learn, to actually roll up your sleeves and make and listen to a physical loudspeaker rather than a fancy mathematical model in a laboratory. And then to mentally deconstruct what you hear into frequency bands, and from frequency bands into a correlation with the physical loudspeaker parts. No general engineering training can give you those skills.

    There are two other prerequisites: First, the designer must have an immense respect for the BBC's loudspeaker R&D heritage (and a real understanding of what they achieved, why they achieved it and how ). I was immensely lucky to be an unpaid Saturday volunteer at the local BBC radio station when I was in my early-mid teens at exactly the moment the LS3/5a appeared in the studio. I was hooked.

    Second, there is no place for the self-delusion that the designer could somehow, with limited R&D resources, develop a revolutionary new loudspeaker that the BBC with their huge resources couldn't. In other words, any progress that we at Harbeth are likely to make will never be to the core-concept of the 'BBC monitor' but to the detailed execution of that concept. The RADIAL2? cone is the perfect example of how we took our founder's* pioneering polypropylene cone and continued the development where he left off. In audio engineering a formal engineering background is not needed, and may prove to be a handicap. The BBC's greatest director of engineering - H.L. Kirke - had a penetrating insight into engineering problems and like Faraday and many other great engineers was entirely self-taught.

    Read about the BBC's H.L. Kirke and Harbeth's founder BBC engineer Dudley Harwood* here. Without these men and their audio legacy, the Harbeth company wouldn't exist.
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  5. #85
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Re: Easy load - yes of course it is

    Results from a user who reports excellent amplifier compatibility (as I would expect) with P3ESR. Read here and here.
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  6. #86
    Join Date
    May 2009
    Location
    England
    Posts
    847

    Default Re: New Harbeth P3ESR

    I have spent the past couple of afternoons in the company of customers, listening to the amazing P3ESR. We used a Rega Saturn for the CD player and various Nottingham Analogue turntables with assorted arms for the analogue side. Amplifiers were Rega Elicit, Naim XS-2, Naim Supernait and the new Croft pre/hybrid power. Nothing too exotic.

    First observation is that these little speakers sound huge !!! I had to smile when one casual listener, after a couple of minutes, said 'can I hear the small ones now ?'. He thought the large speakers in the corners were playing. That;'s the second time this week I have had that happen.

    Second observation is that these diddley little speakers produce real bass. Not the usual one note, ever present colouration that passes as bass from most small speakers but deep, tight, punchy, communicative bass that plays tunes. Amazing for such a tiny box.

    Third. These little fellas make perfectly balanced music at low volume levels. I like my music loud and am fortunate in having a separate building in which to play my music but most are not so fortunate. Unfortunately, very few speakers make much sense at low volume settings but I could quite happily listen all day to the P3ESR at background levels. In fact, we listened at very low levels for two hours this afternoon and I enjoyed the experience.

    Fourth. The fit and finish is examplary. You can't help touching and fondling them and admiring the build. They are beautiful in a traditional, British built but thoroughly modern way.

    This is a truly great speaker.

  7. #87
    Join Date
    Jan 2009
    Location
    USA
    Posts
    31

    Default Re: New Harbeth P3ESR

    Fifth. They are fast ! They remind me of my Quads. Attack is immediate. No ringing, no artifacts, no excuses. Wow, I am blown away, and have no urge to set up my SHL5's or any of my other dozen or so speakers.

  8. #88
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Re: P3ESR - measurements @ the Village Hall

    There are two fundamentally different approaches to 'accurately' measuring the frequency response of a loudspeaker. I say accurately with considerable caution. It's actually impossible to move the loudspeaker so far from reflective surfaces that the measurement is only of the speaker alone, with absolutely no contribution from the measuring space.

    Still, the early loudspeaker designers (probably in the 1930s) realised that something had to be done to minimise the room's ability to corrupt the frequency response measurement at the microphone (which couldn't separate the speaker from the room). One way to minimise the room's contribution is to soak up the reflections by lining the room with soft, absorbent cushion-like material, hence the anechoic chamber. There is one of these free-field rooms at the BBC Research centre, designed and built in the 1960s long before the digital measuring era and as far as I know, I am the only person who pays to use the BBC chamber these past few years. I always take my own digital measuring system and calibrated microphone which I've built into a blue portable box. I showed pictured taken there on the previous posting.

    However, the advent of high-precision digital measurement systems such as mine reveals that even the BBC chamber is far from reflection-free. In fact, there are many small but irritating reflections off electrical fittings and the steel mesh floor (needed for safety) and I always take a pile of cushions to position here and there to damp down the reflections. That then allows a measurement of the speaker with only a small (but still present) contribution from the chamber.

    There are two alternative measuring strategies: outside on a warm windless day, high off the ground and indoors as far from walls as possible. After some research touring village halls in the area, I found that the Scaynes Hill Millennium hall (built with support from the National Lottery recently) has a very high ceiling and an even decay curve. I've used that hall during the development of the P3ESR (see picture).

    How do the results compare between the traditional "analogue" BBC anechoic chamber and the "digital" village hall measurement? Well, under perfect conditions, they give the same result. Surprising considering one is very lively and one very dead? The secret is that the digital measuring system (unlike the old analogue pen and paper chart) can clearly differentiate between the direct sound reaching the microphone from the speaker first, and then the later arrival of that sound reflected off a hard surface. So the trick is to window the measurement so that all of the direct sound is allowed through the digital off-on gate and just before the first reflection hits the mic, the digital gate is snapped shut. After some experimentation I'd previously found that this room-asymmetric position of the mic and speakers gave the best result. Since sound travels very slowly, providing the walls are at least about 2-3m away, this gives long enough 'on' gate to get a really good measurement of overall frequency response, but as with all these sophisticated systems, there is a long learning curve. So, we have three methods which weather, cost and convenience permit virtually identical results when set-up and used carefully.

    Incidentally, many of the Compact 7ES3 measurements during development were taken at the hall so this is a proven technique. The picture taken in Dec08 just before I set off to launch the P3ESR in Las Vegas CES shows the P3ESR about 3m above the floor, securely strapped to the stand). On the stage is the blue-box all-in-one mobile test rig with B&K microphone amplifier and ruggedised Panasonic Toughbook laptop running the measurement software. To their right is a collection of other mini-monitors for comparison including our "green label" BBC-provided reference LS3/5a used to calibrate the measuring equipment, and a P3ESR.

    >
    Attached Images Attached Images
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  9. #89
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Re: Measurements at BBC v. local hall - the sound-energy curves ...

    I decided overnight that to complete this comparison of alternative measurement environments used for the P3ESR design, that it may be of interest to show the same reference speaker (not P3ESR) measured under near identical conditions at the BBC anechoic chamber and the village hall, shown in the last posts. The key point here is that it is not necessary for you, the end user, to be asked to pay for exotic million dollar facilities because accurate and repeatable measurements can be made in very modest environments providing that the constraints of those facilities are well understood. In other words, when measuring loudspeakers, using a little brain power gets the same results as throwing lots of money at the problem. This way minimises the cost of your Harbeth speakers - and is a satisfying intellectual challenge for me!

    At the BBC chamber, the speaker under test is on a short rigid stand, itself standing on the metal grid floor half way up the height of the chamber. At the hall, the speaker is on a telescopic pillar stand, and is raised high above the floor. My test equipment and B&K measuring microphone are the same for both. This was all part of the P3ESR design loop so to be sure that I was measuring the speaker, not the speaker + environment.

    There are some interesting things we can observe from the two graphs - BBC chamber results at the top, hall below. NOTE: these curves show the sound energy as it arrives at the measurement microphone after a sound pulse (like a hand clap) is applied to the speakers inputs. This pulse makes it easy to see reflections in the environment around the speaker. These graphs do not show the frequency response of the speaker! The big pulse on the left side of the graph is the bit which with suitable maths can yield the frequency response of the loudspeaker being measured; the rest is mainly the room's contribution. Note how in both charts, the left pulse - the direct sound - has very similar characteristics. This indicates that even six months apart under totally dissimilar set-ups, the speaker's latent characteristcs are unchanged.

    1. In both cases, the first major reflection (from a wall, floor or ceiling) occurs at about 20mS (one fiftieth of a second) after the direct sound from the speaker reaches the microphone. As the reflective surfaces are some way from the speaker, the speaker's signal has to travel to those surfaces and then bounce off them and then they must find their way to the microphone. This takes time, so the first signal to arrive at the microphone is always that directly from the speaker as that is the shortest route to the microphone.

    2. After the first reflection there follow many others as surfaces further and further from the speaker receive the sound and then bounce it to the microphone. The comparison between the two environments shows a marked difference in the long-term situation: the sound bounces around the hard, reflective walls of the hall each one causing a reflection picked up by the microphone. But in the lined, damped anechoic chamber, the longer term sound is heavily absorbed by the lining and lost into the background silence. It doesn't reach the microphone.

    3. The early reflection situation is different in these two environments. In fact, the village hall gives a better result than the BBC chamber. In the hall, the speaker and microphone are relatively far from the nearest reflective surface but at the BBC chamber there are many small hard surfaces in and around the speaker/microphone each of which give a little reflective sonic 'blip'. So the sound dies away faster and more evenly and completely in the hall before the first reflection compared with the BBC chamber. The reverse applies after the main reflection.

    Which gives the best overall measurement that minimises the sonic degradation due to room reflections? Not so easy to give a clear-cut answer. For a short measurement window (up to 20mS) this particular village hall is the best choice: for a long-window measurement, the BBC chamber is sonically more inert. But there are workarounds for both.

    Hope this is of interest.

    >
    Attached Images Attached Images
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  10. #90
    Join Date
    Apr 2008
    Location
    SG
    Posts
    225

    Default Re: New Harbeth P3ESR

    the P3ESR is with an enclosed design i.e. no port. anybody keen to elaborate in this particular design? and how does this design enhance the P3ESR (or P3ES2) performance? Tks

  11. #91
    Join Date
    Mar 2009
    Location
    Malaysia
    Posts
    243

    Default Re: New Harbeth P3ESR

    Quote Originally Posted by Lorpuris View Post
    Dear Mr. Shaw,

    As I write this note I am enjoying my SHL5. I have the 30th Anniversary Edition speakers. While I have owned many loudspeakers in my life, I have never been so utterly and completely impressed. So impressed in fact that very shortly after i received them I ordered a pair of the new P3ESR.
    May I ask how does the SHL5 compare with the new P3ESR in terms of sonic performance?


    Hi Alan,

    I noticed the P3ESR has a low sensitivity of 83.5dB. Does this figure imply the P3ESR would be slightly less efficient than the bigger speakers in having 86dB especially for folks who use low-powered amps? I am well aware that there wouldn't be any big issues to worry about but was just curious how the speaker would perform in a technical viewpoint with the lowish figure. I would also be interested to know the parameters that affect the sensitivity and impedance values of a particular speaker during the design process. Your enlightenment would be most appreciated.

  12. #92
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Harbeth P3ESR - efficiency, gearing and quality etc.

    Quote Originally Posted by ryder View Post
    ... I would also be interested to know the parameters that affect the sensitivity and impedance values of a particular speaker during the design process. Your enlightenment would be most appreciated.
    This is a good question which is at the heart of the design process for any quality speaker system - but especially for a mini-monitor which must sound like a much bigger speaker to satisfy the listener. I have actually covered this in quite some detail in the Tech Talk section on the Harbeth web site here. I work with these parameters every day so this is all second-nature to me, but does it make sense to you? If not tell me what I can better explain and I'll do my best. Please remember that efficiency* is an unavoidable consequence of the design process, not, in Harbeth's case, a design objective.

    A very crude analogy occurs to me: think of a big woofer as a Harley Davidson motorbike engine: lots of power (torque) at very low revs. Once you've heard the slow cylinder by cylinder throb of a Harley engine it's instantly recognisable. Think of a small woofer as a 50cc scooter engine: no torque at all at low revs but lots of action in the middle rev range. Somehow, we have to make our 50cc engine fool the driver into believing that he's driving a Harley. If we very carefully adjusted the gearing of the 50cc engine's gearbox (if it has one) we might just be able to reposition that engines power range more usefully - and that's really what we do in the circuit that drives the woofer. Nothing we can do to the woofer itself - that's fixed - but we can do something clever in the 'crossover' that drives the woofer.

    It's hugely important to appreciate that the division of signal between the woofer/midrange and tweeter is only one function of a crossover. Almost an incidental function. The real action is the ability to shape and contour the frequency response just as the designer wants, and that needs numbers of components. You can think of these crossover components as the gear cogs of an acoustic gearbox matching the woofer to the room.

    If you're only interested in the splitting of high from low frequencies, that can be done with just one or two electronic parts in the crossover; but to create an acoustic gearbox to match the speaker to the room you need a good handful of components. (See attached P3ESR circuit photo). There are three functions that can occur in a BBC-style crossover:

    1) frequency splitting
    2) bass/midrange 'gearing' to adjust for flattest response
    3) mid/top relative level adjustment

    To achieve all three simultaneously we will need several electrical components like coils, resistors and capacitors. These functions need components like a gearbox needs cogged wheels. So, if you take a peek inside a speaker (or a gearbox) and count crossover components it gives you a strong clue as to how well balanced the sound will be before you even listen. Because if there are only a few components then there is little the designer can do to 'gear' the woofer to the room. Just as with cars: a six-gear automatic gearbox will match the car to the road better and give a much smoother power performance than a three gear. There's no substitute for gears! And no substitute for crossover components!

    I believe that it was the Shorter and Harwood at BBC Research in the late 1950s (when developing the LS5/1speaker) who realised that once the frequency division function had been implemented, the crossover could then be further manipulated to provide this gearing action on a selective frequency band basis. But that duality needed lots of components which is why the 'BBC monitor' has a complex crossover. Furthermore, only a BBC inspired designer with a deep appreciation for the ingenuity of the BBCs cunning crossover design approach, the subtlety of their 'cogs and wheels' would really understand how to design such a circuit.

    That's why there are really two classes of hi-fi speakers .... the BBC-related designs, and all the others.

    * Loudspeaker efficiency: this means how loud for a (known) amount of electrical power at the terminals. Some manufacturers, those whose customers are ignorant and driven by specifications, would design for high efficiency and inevitability would sacrifice some other parameters including sonic quality.

    ================================================== =====================

    Now I realise that this analogy of gearing is not perfect. The point is to see that the 5" drive unit is like a small fast revolving cog, and we want it to behave in a more sedate manner, like a big, slow revolving cog. That way we can get a smooth response from even a small motor unit. And that means that by using some 'gearing' in the crossover even a small speaker like the P3ESR can be made to sound like a big Rolls Royce not like a city car buzzing around and making a lot of useless noise. We can give the P3ESR some weight and torque in the low frequencies.

    You can get the idea of cogs and wheels and driving forces here. Look at the video.

    Also here watch how we can get a nice slow revolution of the output shaft by selecting the optimum gear.

    >
    Attached Images Attached Images
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  13. #93
    Join Date
    Jan 2007
    Location
    Canada
    Posts
    80

    Default Re: New Harbeth P3ESR

    I had a brief audition of the new P3ESR during my lunch hour last week. I have to agree with the early reports of this speaker... it is something special.

    Very authoritative sounding small speaker.... I could not believe the dealer did not have the REL sub turned on. I have listened extensivey to the P3ES-2 and the the new P3ESR is a real step forward. That radial midrange sounded stunning on vocals. It joins the M40.1 and C7ES-3 in having an uncanny ability to reproduce spatial cues and has superb stereo imaging. I think it's no coincidence that these are the three most recent models designed by AS, as all 3 of these models seems to distinguish themselves in this area compared to older Harbeths.

    I'll be going back with my own CD's to listen more carefully. I am sorely tempted to dip into my M40.1 savings fund to buy this little "pocket rocket".

  14. #94
    Join Date
    May 2009
    Location
    England
    Posts
    847

    Default Re: New Harbeth P3ESR

    The P3ESR is an absolute stunner and by far the best 'mini monitor' I have heard in thirty six years of selling hi-fi. I have lost track of the number of times I have demonstrated it now but I have never had a negative comment except perhaps 'I can't afford it'. The usual first comment is about the prodigious bass and the scale of the soundstage.

    A sensational speaker by any yardstick.

  15. #95
    Join Date
    Jan 2006
    Location
    South of England, UK
    Posts
    4,276

    Default Re: New Harbeth P3ESR

    Quote Originally Posted by Will View Post
    I had a brief audition of the new P3ESR ...the new P3ESR is a real step forward. ... I think it's no coincidence that these are the three most recent models designed by AS
    That's an astute observation, and actually true.

    When the time came to upgrade the C7ES2 into the C7ES-3 (2006?) I was self-aware that I had reached a point of comfort in the operation of the test equipment, the (then new) crossover simulator software and my general awareness of the design objective and the path to it. The C7ES3, M40.1 and as you say, P3ESR were designed to the same specification, with common software and test equipment and of course, a common me! The only difference is that the C7ES3 was designed in a small spare bedroom/study at home before we established the R&D centre in the woods (The Cottage) at which the M40.1 and P3ESR were designed. So I know for sure that the C7ES3 is very small-room friendly. Unless something really radical results from my training course in New York next month, all future Harbeths will be designed in the same way.

    Just to remind you - I do not invite, encourage or permit anyone to hear a Harbeth design until I am ready to commit to production. I cannot and do not design by consensus - so it's always a real pleasure to hear from users after all the man months or years alone, that you can get real pleasure from my input. I wonder how common it is amongst speaker brands for one designer to have designed the smallest (P3ESR) and largest (Monitor 40.1) in the same listening environment. Littel wonder they share the same house sound.

    I's like to thank Derek Hughes for his invaluable contribution as my draughtsman, working remotely. His CAD skills are exceptional and he takes care of the drawing details leaving me free to get on with the design work.
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  16. #96
    Join Date
    Aug 2007
    Location
    Harrow, UK
    Posts
    479

    Default Re: New Harbeth P3ESR

    Quote Originally Posted by A.S. View Post
    ...my training course in New York next month...
    Do tell Alan! Sounds fascinating.

  17. #97
    Join Date
    Mar 2009
    Location
    Malaysia
    Posts
    243

    Default Re: New Harbeth P3ESR

    I listened to the P3ESR at a friend's place with the Rega Elicit. Superb sound from such a small box. Extremely musical. He is the owner of the 40.1 and so impressed with the performance of the P3ESR he has added this speaker into his collection which was set up in a 4th(or 5th) system in the house.

  18. #98
    DrewTurner Guest

    Default Re: New Harbeth P3ESR

    I'm curious when AS will offer up the new P3ESR for review with the Hi Fi magazines?I think I read that he wants fill the back log of orders from dealers before doing so.I'm crazy about my new P3ESR's,but ,it's always interesting to hear what certain reviewers have to say about Harbeth speakers.

  19. #99
    Join Date
    Jun 2009
    Location
    Canada
    Posts
    905

    Default Re: New Harbeth P3ESR

    Quote Originally Posted by DrewTurner View Post
    I'm curious when AS will offer up the new P3ESR for review with the Hi Fi magazines?I think I read that he wants fill the back log of orders from dealers before doing so.I'm crazy about my new P3ESR's,but ,it's always interesting to hear what certain reviewers have to say about Harbeth speakers.
    As a fellow P3ESR owner I admit to a bit of curiosity as well, but put yourself in Harbeth's shoes: if you're selling every single speaker you have the capacity to make, and are backordered for months, what's the point of sending speakers out for review? My sense is that we might see a review or two when - or if - sales start to tail off a little, and there's a need to rev up a bit of public interest. In a year or two perhaps? I doubt it will be sooner, and it might be later than that.

  20. #100
    Join Date
    Jan 2009
    Location
    USA
    Posts
    31

    Default Re: New Harbeth P3ESR

    Also a fellow P3ESR owner: Reviews are a double edge sword to say the least. If anyone really wants a review on the P3ESR, all they need to do is read this thread. They can then listen to the voices of the many instead of the few.

Page 5 of 15 FirstFirst 1234567891011121314 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •