New speaker project - Goodmans Magisters

13»

Comments

  • edited July 6
    Latest tweaking has been in the digital domain.

    Despite painstaking research into what was sensitivity of the original Magister tweeters (Goodmans DT3), it turns out that they were less than the 91dB that I reckoned on. Hence the replacement 91dB Monacor DT-19SU were a little too sensitive. In fact a lot too sensitive!
    Thankfully, JRiver DSP Studio to the rescue again. As luck would have it, in JRiver's DSP Studio feathure, there is a parametric equaliser facility that includes a "high-shelf" cut. This cuts all frequencies from a selected point to a selected degree. Based on listening this turned out to be -4.9dB from 4910Hz (how's that for accuracy...?! ;-D  (the Magister's crossover frequency to the tweeter is (I think) 5000Hz, so I used that as a guide).

    Ideally of course, I would have got the sensitivity of the tweeters themselves exactly right. ;-)
    Although, tweaking the speaker crossovers themselves to the settings I achieved digitally through JRiver would have been impossible for me.
  • edited August 17
    Latest update.
    The ribbon tweeters are great.
    Also helpful is the results of the time I took to integrate them using the high-shelf cut.
    I'm as happy with the sound of the system as I've ever been. Really natural, revealing, and out-of-the-way. I think the lack of graininess of Col's Class A is really shining through. I'm chuffed to bits.

    image

    ...Not so good is the fact that the speakers are reliant on the digital high-shelf to sound listenable. The 98dB sensitivity of the ribbon tweeters is way louder than the speakers overall. The shelf is now a -12.34dB cut, at 5107Hz using a 0.71Q slope (2nd order, 12dB/octave). Without the shelf, the speakers are 12.34dB too loud in the treble!. X_X

    I'm thinking of putting in an L-pad to make them more portable into other systems, but am not confident the degree of accuracy achieved by the digital high-shelf could be replicated using resistors. Maybe stick the L-pad in to make the speakers generally flatter, then tweak within my own system using the high-shelf....

    As an aside, the experience has left me thinking about the value of digital crossovers generally (if such things exist). Tweaking settings by ear to very fine degrees is surely more effective (and quicker) than using crossover components with relatively wide tolerances.

    Anyway, I must be feeling satisfied with the sound of the Magisters as today I felt driven to start on my next project ... <:-P
  • There are digital (and analogue) outboard crossovers (I'll find links later if nobody else comes up with them in the meantime). The crossovers should come before the power amp, so you're looking at bi-amping or tri-amping - a use for another of your rapidly breeding stable of amps?

    Funnily enough, I was talking about running my Royds actively to Colin earlier this week (I've been wanting to try a proper active bi-amped setup for ages). He came up with the idea of putting an analogue crossover circuit in the amplifier. Basically, I'll have a new quad-mono 25-amp SECA kit, with the crossover in the power amp case, fed from the pre-amp. Fewer boxes and generally neater solution. We'll see if it comes to pass.


  • uglymusic said:

    There are digital (and analogue) outboard crossovers (I'll find links later if nobody else comes up with them in the meantime). The crossovers should come before the power amp, so you're looking at bi-amping or tri-amping - a use for another of your rapidly breeding stable of amps?





    Interesting...
    Gives me permission not to bother selling them! :-D
  • Here we go:

    Digital crossovers - miniDSP

    Happy reading!
  • MiniDSP is very well thought of, there's a lot of users on pfm. It also has an inbuilt DAC iirc which is supposed to be good.

    I heard the cheap Beringer in Jim Kempston system once, it was great and quite plain to use.
  • edited August 24
    Thanks for that fellas.
    I wasn't at all aware of these products.
    From a quick scan the Xkitz one looks most appropriate to me right now as it is 3 way (and possibly easier to tweak). The miniDSP ones are in a way more similar to what I'm doing now, but appear optimised for home cinema use, running 2-way main speakers. There may be a way of programming them for 3 ways, which are what I'm interested in ATM. I'd have to have a closer look.
    I can certainly imagine myself trying either or both solutions on future builds, as both fit with my need and desire to tune things by ear.

    Edit: Also, the idea of having all 3 of my Wonfor class As (ID25, Toca 20 and One4 20) all in use in the system at the same time exciting (and terrifying). Into the cost of any build, I would have to factor in an air conditioning unit...
  • Docfoster said:
    Thanks for that fellas. I wasn't at all aware of these products. From a quick scan the Xkitz one looks most appropriate to me right now as it is 3 way (and possibly easier to tweak). The miniDSP ones are in a way more similar to what I'm doing now, but appear optimised for home cinema use, running 2-way main speakers. There may be a way of programming them for 3 ways, which are what I'm interested in ATM. I'd have to have a closer look. I can certainly imagine myself trying either or both solutions on future builds, as both fit with my need and desire to tune things by ear. Edit: Also, the idea of having all 3 of my Wonfor class As (ID25, Toca 20 and One4 20) all in use in the system at the same time exciting (and terrifying). Into the cost of any build, I would have to factor in an air conditioning unit...
    Come off it! It never gets that warm in Glawster :-)
  • edited October 25
    Wanting to touch base with this thread (just for my own satisfaction).

    I tried to convert the -12.3dB digital cut I was using into an L-pad. I used various on-line calculators to work out the resistors required (2.5R in parallel, 6R in series), but this sounded way too much attenuation. The upper frequencies were far too low. As proved in my recent personal bake-off (clicky).

    So I tried a -10dB pad (3.5R in parallel, 5.5R in series)...
    image

    Still excessive attenuation. :-/ 
    So then a -8dB (simpler, as a 5R for both, which I think actually gives about a -8.4dB attenuation)...
    image

    ...And that's about right.

    It's curious why the 12.4dB attenuation that was spot-on digitally was way off using an actual pad. Maybe I'm misunderstanding / neglecting something important...

    But anyway, in terms of stand-alone speakers I think I've got the Magisters about as good as I can get them without any behind the scenes digital jiggery-pokery.

    (Except that I might possibly put some spikes on them. :-D )
  • Now that you've lived with them for wee while, would you/do you add "any behind the scenes digital jiggery-pokery" or have the physical teakeries proved to be fully sufficient?
  • "teakeries" should be tweakeries ;)
  • cj66 said:
    "teakeries" should be tweakeries ;)
    Yes. B&Q’s pricing for teak compares unfavourably with MDF. ;-)
  • edited November 1
    cj66 said:
    Now that you've lived with them for wee while, would you/do you add "any behind the scenes digital jiggery-pokery" or have the physical teakeries proved to be fully sufficient?
    Great question. :-) Three answers to this. My strictly honest sonic answer is that I got easier and better results using digital stuff. Things sound smoother, more balanced and clearer. There was no audible downside. Tho I spent a lot longer tweaking digital things to the Nth degree, than the morning I spent with the L-pads. Second answer is that long term I could not tolerate the speakers being unusable in their own right. And +8dB tweeters makes them unusable! So I always knew that I would have to pad the speakers as best as I could. With the L-pad in place the speakers are very listenable on their own. Final answer is that if I were to use the Magisters long term I would now be working on the digital processing as finishing touches to the L-pad attenuation. I’ve committed myself to enjoying my big DIY speakers however, which I feel are just better, so the Magisters are back in storage in the summer house. Will get them out at some point to fix on some spikes.
  • edited November 6
    Well, that last exchange prompted me to order and fit some spikes.

    And, not the for the first time, I'm reminded that a bit of coupling to the floor and / or overall stability makes a real difference.
    The "Magisters" do sound tighter sitting on their spikes.
    In fact, I was sufficiently impressed to feel motivated to spend some time positioning them properly (not just sticking them in front of the usual speakers), and they sound pretty decent. I think I'm going to re-run my little bake-off to give the Magisters a chance at a rematch. The bass is still a bit of a broad brush stroke, but it's a nice warm, deep colour. And although the midrange is a bit uninspiring (lacks nuance), the ribbon tweeter brings something pleasing and smooth to the party.

    One interesting footnote...
    Earlier in this thread I posted about the speakers being at fault in having something of an unstable soundstage; higher frequencies pulling to the right (or bass and mid pulling to the left). I re-arranged the drivers and went to other efforts with room re-arrangement attempting to address this. Recent listening with other speakers is confirming that the sole cause of this slightly unstable soundstage is my room's sonic signature. 
Sign In or Register to comment.