Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by Drewan77

  1. Peaks are easier to deal with than nulls unless you are very flexible with chair and speaker positioning. A peak can be brought down very precisely & quite easily with DSP (& acoustic treatment which can be more difficult with specific frequencies & widths - Qs) after accurate measurements whereas nulls don't really respond to DSP eq - by pouring more volume into a null, it often just makes the issue worse & the null deeper! The best way to deal with a null is to firstly identify it (using the methods described in prior posts) & then alter physical positions, aided by some acoustic treatment if practical.
  2. Please follow the advice Tony gives above... If you recall Andrew, when we first measured these speakers, I set an 8th order (48dB) crossover at 2khz to avoid potential damage to the tweeters. As far as setting MiniDSP filters is concerned, I suggest that Tony would be best placed to advise as I am not familiar with that setup.
  3. Glad it's worked out Andrew - well implemented DSP has improved all the speakers I have tried. Any cynic who has never heard what it can do should visit your setup at Kegworth.
  4. Andrew, if you recall when we measured these speakers in my garden, I checked time alignment between the individual drivers and they were fine so as you are now working on in-room eq, it should be for the speakers as a whole including crossovers, not individual drivers.
  5. As far as a 'flat' response goes I am in complete agreement, most music sounds too lifeless and thin without some sort of house curve. You may recall that on my system I use the four DEQX remote presets with varying levels of bass boost, position 1 being neutral & unadjusted (sounds good with bass heavy albums) & progressive uplifts through to position 4 (for the thinnest sounding albums). I can achieve very similar sounding frequency response whatever the source, to my listening preference (deep but clean, tight & very fast bass with a lot of transient slam) At the end of the day it's about enjoying what you hear rather than trying to recreate some sort of anechoic chamber 'perfection'.
  6. Andrew, the way I read your last post suggests that after initial confusion with measurements you have reached a satisfactory result from tuning by ear (your edited post). if you do need my input I am happy to come over & help, just let me know. Andrew
  7. Turntable occasionally used in a bedroom system, complete with Ortofon OM10 & speedbox speed controller (+ a Vinyl Engine printed Pro-Ject arc protractor). Fully working. £185 collect only or meet-up (NW UK location)
  8. Purchased new by me from Audio Counsel Oldham in 2003. Colour - cherry Fully working with a room filling, wide soundstage, typical of Omnis. Condition: Some small marks if viewed close up but in generally very good condition. One of the top grill meshes also has a slight top dint visible from the rear. Cabling from below, either banana or spade. Description: 4 top mounted super tweeters & 2 midrange tweeters , front mounted 8″ Seas woofer & rear mounted 10″ passive radiator. I believe the original boxes are in the loft. £1,850 ono – collection or meet up (location NW UK).
  9. These might help you get your head around parallel & series wiring Andrew:
  10. You could hook up & measure each amp in turn at something like 25% of its volume on the same channel (I would use the bass channel). This should enable you to view their relative outputs. Once you have a general idea, then adjust the settings in MiniDSP & connect both amps/channels. If there is no easy way for you to measure again and view then you will need to adjust by ear. As mentioned previously - I would expect Tony to have a more accurate & repeatable method. Andrew
  11. Andrew - Something we didn't discuss while you were visiting.... You may find it useful when viewing measurements to change the 'smoothing' level which makes the frequency response graph easier to interpret, especially at higher frequencies. To do this, go to the title bar at the very top of your screen and select 'graph', then choose something like 1/12 smoothing. (I don't normally use REW because the Earthworks M23 mic, even with its calibration file doesn't produce data that I'm happy with compared to FuzzMeasure Pro or DEQX software. Having said that, your Umik mic measurements via REW were very close to M23/DEQX which is as it should be.... & reassuring!) Tony J is clearly more familiar with REW & Minidsp so he will offer the best advice with those of course. Good luck
  12. Although I don't use Minidsp, that looks correct for one amp/speaker provided you connect CH1 to the woofers & CH2 the tweeter! (repeat with CH3 & 4 for the other amp/speaker) Andrew
  13. I'm the 'other' Andrew mentioned above!..... Here are the outdoor measurements we took when using DEQX to set up the speakers via my own system so we could then move on to REW as in Andrews' posts above (only one speaker measured for this purpose): 1. Anechoic measurement from 1m showing only a slight mic reflection @10.2ms & then its decay refections 22ms, 33ms etc ('clean' data @10.2ms was therefore used for analysis) 2. Raw measurement of woofers/tweeter (as this is outdoors & in 'free air' & because of the 10.2ms boundary, bass response below 200hz drops away in the plot even though the speakers actually produce bass well below this): 3. Phase response of speakers - green = tweeter, blue = woofers. 4. Crossover applied at 2khz because above this, the woofers display a significant null & the tweeter is smoother. I 'guessed' that Mini DSP would only allow LR up to 8-way so I set at 48dB rather than the brick wall Xovers DEQX allows (up to 50-way 300dB) : 5. Just for fun - DEQX speaker correction applied within the blue box 'windowed' area (red line is the result - flat response for phase, time alignment & group delay) NOTE: THIS WAS JUST TO ILLUSTRATE TO ANDREW WHAT FULL DEQX CORRECTION CAN ACHIEVE & WAS NOT USED FOR ANY OF THE SUBSEQUENT MEASUREMENTS TAKEN WITH REW - we played back some music outdoors using the uncorrected speaker & then using this correction filter with a razor flat speaker response 100hz - 22khz. The difference was significant, as I expected. Andrew should have an opportunity to produce some pretty nice sounding music using these as 1st off active speakers & will learn a lot during the process.....
  14. I'm still struggling with this height thing - it only happens if I artificially change time alignment between mid-highs within the OBs, whereas leaving them as calibrated, images are very precise at a point between midrange driver & ribbon positions. Image clarity is outstanding, as is soundstage depth (less so width beyond the speakers) but never height. Mono recordings always play at a vertical point in space between speakers & chair you can almost touch. I can only surmise that this is as a result of the corrections made from my original speaker measurements (quoting DEQX themselves..."DEQX-Cal breaks the measured response of the speaker into thousands of separate frequency groups, minutely adjusting their timing so that all groups arrive at the listener’s ear at the same time."). I guess that this level of correction may account for what I (don't) hear. In some ways I wish I did!
  15. Ross, I completely agree with this. My Open Baffle speakers are about 5' tall with ribbon tweeters on the top baffle, angled to the listening chair and all drivers phase & time aligned. The soundstage has great depth & reasonable width, sometimes beyond the speakers (depending on the original recording of course). Image clarity is pin-sharp vertically between the tweeters & midrange drive units, about 4' up from the floor. Here's the interesting bit - it's quite easy for those of us with an active system to change the time alignment of any set of drivers & this will blur the imaging of instruments/voice but it can also affect the impression of height. I just did it! Delaying all drivers except the tweeters by 8ms, then instruments with a lot of high frequency information seem to come from a higher point, confused by the blend of other frequencies produced. Imaging is less precise & fluctuates. Delaying by 15ms, the system sounds uncomfortable. Therefore it seems logical to me that with less than perfect time/phase alignment, some of the effect we 'hear' is caused by this. The rest may be expectation bias which we 'audiophiles' all seem to possess in abundance!
  16. From my own experience, the M&K subs I have used are 3rd order (18dB/octave) to 125hz, then up to a maximum 6th order (36db/octave) above that. My B&W PV1D can use either 2nd order (12dB/octave) or 4th order (24dB/octave). 80-85hz is often regarded as an optimum crossover point although I have found anything up to 100hz is basically 'below the (frequency) range of orientation perception'. I use a crossover at 50hz 4th order with my OB speakers & a pair of subs - internal crossovers are switched/modded out, measured and DEQX corrected on both subs so they don't have both passive + active crossovers competing with each other. As for dovetailing with main speakers, before I went active many years ago, I used an M&K HP80-2, 2nd order (12dB/octave) passive high pass filter box which sounded really great with a pair of Royd Doublets and a small M&K VX7 Mk2 sub. I haven't listened to any of these 3 items for many years & they sit unused in a spare room - I really should sell them as they annoy my wife, especially the Doublets! Here are a couple of pics of the HP-80 and I can thoroughly recommend one (long discontinued but see if it can be found on ebay etc) for anyone wanting a simple but effective passive subwoofer crossover. It takes a full range output from a pre-amp & then splits the signal after applying the crossover/slope and feeds out to the sub and power amplifier for the mains. It adds 80hz/12dB crossover to your main speakers, not as good as the best active setup but pretty impressive nonetheless & I was happy for many years (until I heard someone elses' system using good DSP which can be in another league altogether). I must stress (as others have observed in this thread) that crossover & simple phase matching are only part of the equation here - absolute phase & time alignment are necessary to achieve seamless integration. It takes some understanding, effort & cost - but it can be done!
  17. I believe I already mentioned in another thread that I am very familiar with the Linn setup (Exakt on two Linn only systems in different rooms though) and have helped setup these systems to the point where one sounds very close to my own. We both agree that the DEQX setup is marginally better but that is probably because I use a dedicated listening room with no domestic compromises, open baffle speakers & I have acoustic treatment as well. I'm pretty confident that if Exakt was used with my setup, in my room I could get an almost identical result so I have to agree. I still have a nagging doubt though that time of flight and room simulations for correction are going to be slightly less accurate than actual real world measurements (having seen variances in pairs of several manufacturer brand speakers I have measured and corrected - Royds, JBLs, Shahinians, B&Ws). Give that the Linn Beta is designed to work with other manufacturers speakers, I don't understand how a simulation could factor out those variations or that of the actual listening space which is rarely a perfect shape with windows, doors, fireplaces, furniture etc all affecting the sound at the listening position. We probably have to agree to disagree on that one! For me, the best of both worlds would be the Linn system with a measurement & correction facility or DEQX with the slightly better in room eq facilities post correction from Linn Exakt.
  18. You don't have to do this & many DEQX users seem happy after just an in-room measurement. Both methods are valid, it's just that an outside measurement can produce exceptionally clean data to work with. DEQX does not work with simulations, the more accurate the data it collects, the better the result. It corrects everything downstream of the processor (which can also be the preamp) including the impact of cabling and power amps. The 'clever' part is calibration after measurement so that the response of a given set of amps/cables/speaker drivers with each channel become totally phase and timing coherent whereas natural driver responses mean that higher frequencies with shorter wavelengths will arrive progressively sooner than lower in any standard passive speaker* and most actives*. This results in removing 'smearing' to give exceptionally precise imaging and instrument separation within the soundstage (dependent on original source of course). *so 'time alignment' in this case is actually a compromise because you are only really aligning impulse peaks of the highest (fastest) frequencies in each driver set, whereas mic based DEQX / Trinnov & the calculations provided by the Linn system will correct all frequencies properly.
  19. I may be the only person here using DEQX (?) but this does solve the issue of phase integration pretty much once & for all: 1. Measure the main speakers (pref outdoors) & allow the software to calibrate & correct all frequencies for timing, group delay, phase etc. 2. Measure subs (pref outdoors) & allow the software to calibrate these including all frequencies for timing, group delay, phase etc. 3. Re-measure the system at the listening position & align timing by delaying the main speakers so the impulse peaks exactly correlate with any or all subs at their initial peak. Because DEQX has already corrected every frequency to be in phase and time coherent, the subs and main speakers are then in perfect alignment. Stefan's (Orangeart) observation of the best sub alignment he had heard when listening to my system bears this out. It's as if the subs are part of the main speakers and any crossover is undetectable with all frequencies sounding remarkably coherent.
  20. Yes, I use both a DEQX HDP-5 and HDP-3 in a 4-way system (the HDP-3 controls two subs) & vinyl is my preferred medium. DEQX processing is transparent to my ears and seems to be to anyone else that listens. Although Rodney is correct in saying that distributed subs could solve Steve's bass issues, surely it would be better to try and get the best from the Wilson Sophia 3 s first (after all these are rated as very high end speakers).
  21. Steve (Buffalo) I can imagine how frustrating this must be when you are using such an expensive amp & speakers. If domestic harmony dictates the seating & speaker positions & this correlates to room nodes giving bass suck out then there's relatively little you can do without either changing positions or moving to a different room. Sorry if that seems harsh but if it was me and I had your gear, I would take Purite's advice, then either reposition or relocate somewhere else in the house. Better still, measure the room to take out the guesswork & calculations as others of us have done (it doesn't give the solution but at least you will understand the problem better). The most common solutions for a difficult room would normally involve either acoustic treatment, some form of digital processing or a combination of both. Properly aligned subwoofers can also resolve the issue you describe but that should be last resort for speakers like your Sophias. In my experience, acoustic treatment is useful to reduce but not add when dealing with room nodes, likewise DSP. My system uses both very effectively - dealing with too much bass at a couple of frequencies rather than too little (it's little use using eq to boost a bass frequency to fill a void as this usually just pours more energy into the node). ....is 'domestic compromise' an option?
  22. Up - finally getting a system I am totally happy with after more than 40 years of trying & a hifi graveyard in the loft/spare room. Another up is OTT releasing one of my favourite albums, Fairchildren on vinyl (inc best album cover ever!) & Skylon coming soon. Biggest up was retirement at 60 in May & loads more time to enjoy the system but bit of a down when the wife joined me in September (turn that down...). .... Keith, glad I was able to help & again anytime you may need in the future.
  23. Update: I mentioned in an earlier post that I may replace the Alpair 7s with the Neo 3 ribbons. After listening to music over an extended period, the response of the Alpairs didn't quite match my memory of the Neo's (which I had used in the previous OBs for several years). With vinyl, the delicate percussion sparkle which I was used to was missing so I bit the bullet and decided to change things. Initially I planned to take outdoor calibration measurements on a dry day in November but by the time everything was set up in the garden, it had become too windy to get a clean response, even using 36 frequency sweeps per speaker (about 2 minutes) which averages out all rogue sounds like occasional birds/distant aircraft/exasperated wife etc. The wind became almost gale force by midday when I was ready so I abandoned that. I set it up again a couple of weeks ago on a day which was dry and wind free - with the added bonus that this followed several days of continuous rain, meaning the lawn was soft & wet, reducing possible floor reflections even though the speakers/mic were raised well above it. Twittering birds are scarce at this time of year too & the results were by far the cleanest I have taken. The measured treble response is much flatter than with the Alpairs & the only (tiny) reflection which appeared during this process was from the mic itself (which only appeared on the ribbon driver measurements) so I was able to window calibrations to 40ms & therefore allow the algorithms to correct bass almost all the way down. After letting DEQX do its thing 0hz - 20k with frequency, phase, crossovers, group delay and impulse, a second set of verification measurements gave the 'semi-anechoic' response below, (whilst each speaker was still on the raised platform - these are actual measurements of the speakers after correction. Phase plots are equally flat, step response is clean etc). The Neo's are crossed at 3024hz/300dB to the Apair 12s & because the speaker is measured as a 3 way, the interface is corrected automatically and seamlessly by the software - frequency, phase, timing, group delays. DEQX processors are very powerful in this respect & it's very difficult to find fault with the results, especially when room correction is finally added once in situ. It's not just about measurements of course so I have since listened to a lot of LPs, FLACs, CDs & SACDs. The system really does sound remarkable on everything regardless of source or volume. I have no 'real' need to play around with the system, purchase anything more or tweak further. I'm not sure the system or room could get any closer to the sound I have been after since my teens so since retirement I listen to more music but spend much less time behind a computer screen & that has to be a good thing! I also mentioned to Keith, MF1000 when he came over that I may arrange a small bake off next year if anyone is interested in hearing this system.
  24. Our house has a solar PV array & here's yesterdays measured grid voltage
  25. Thanks Stefan, it was very good of you to come over after such a busy two days & I'm gratified that you were so positive about what you heard. It's all very well having other visitors enthusing about what they hear but I appreciate critical and experienced ears like yours (even if still recovering from the excesses of Saturday!) Sophisticated DSP processing gives the opportunity to voice a system as you want and for me, this one is exactly right. I should just leave it at that but, as is the way with this hobby, I may feel the need to tweak further.... As an update to my earlier posts - below are the final in room measurements - Stefan listened mostly to one of the 4 DEQX presets with a slight bass boost house curve from what is shown below. When my immediate neighbours were on holiday, new calibration measurements were taken outdoors & in spite of a few uncooperative birds (and an aeroplane!), I was able to run multiple long frequency sweeps which averaged out everything except the speakers. The raw anechoic response is completely flat +/-0dB @ 30hz- 9.2k & apart from a very narrow, high Q null @ 11.8k, it's +/-3db from 18hz-25k (the limit of the M23 Earthworks mic). DEQX algorithms corrected phase, timing, amplitude & group delay across all drivers & frequencies. I never cease to be amazed by the resulting music from any speakers treated in this fashion. The four 15" drivers have taken an age to loosen up and run in properly & I also changed the high end crossover settings, created new correction filters, made a few room adjustments and did a new in-room measurement/bass eq adjustment. The subs are very precisely time aligned with completely undetectable crossovers, as Stefan observed. The system also portrays bass instruments with the most natural voicing I believe is possible with these speakers and, where present on the source material electric bass 'growls' in a very believable fashion. Individual notes are also clearly audible, even when there is a lot going on. These speakers are now transparent, fast & full range with height, depth and (some) width to the soundstage - as Stefan mentions. Voiced exactly as I want and no apparent room effects on any material that I have played or at any volume so far. Below are the listening chair plots (the room has naturally caused deviations from the outdoor measurements above, even allowing for acoustic & subtle bass eq treatment. Nevertheless the results are impressive, still almost 'flat' & they certainly sound great). 3-Way OB speakers: Two subs: Combined, final result (with time/phase aligned crossovers between both speakers/both subs):
  • Create New...