Jump to content

Le Cleach crossover


Horatio

Recommended Posts

I have been messing around with a miniDSP active crossover that I am using on my Khorn bottom and DH1012/EVHR6040 top.

I initially set this up with Linkwitz-Reilly 48 db/octave filters, at 400 Hz, with some eq dialed in for the HR6040 horn, which is a CD horn. I also dialed in 6.5 ms of delay for the top, based on the Heyser article values, just as a starting point. Set levels by ear. This all sounded quite nice, even though it was rather quickly set up.

I have not yet verified the goodness of the 6.5 ms value, and will get to that during later testing.

I implemented the Le Cleac'h crossover in about 5 minutes with the miniDSP, and this is basically done by moving to 3rd order Bessel filters, setting the cutoff frequencies shifted up and down from the original Fc (down for the bass, up for the midrange/tweeter) by about 0.87 (bass cutoff), and 1.14 for the mid/upper. Then, you dial in an additional .22Fc(wavelength basis) of delay for the mid. One other detail: you need to invert the polarity on the midrange, too, in this crossover topology. This is for a two-way configuration, mind you.

The advantages are purported to be a more uniform phase response, manifesting in a broader sound stage.

I did this, but I did one other thing, which is why I am posting about this: instead of using 400 Hz as the crossover, and shifting up from there for the mid-range, and down for the bass, I MOVED THE CROSSOVER frequency DOWN, so that the mid was still being crossed at 400 Hz, -3db. This means that the bass, in the Le Cleac'h setup, is crossed out at something like 305 Hz(!) This bypasses the ragged response of the bass horn that is present from 300 Hz up. The kick to the delay was not much, compared to the 6.5 already dialed in, but, it's there.

The result is stunning. I have not heard a full-on horn system image like this. Moreover, I have never heard the kick drum thwack with a Klipschorn (not, say, like you get from an Altec VOT) until now.

So, doing this crossover resulted in two advantages!

Link to comment
Share on other sites

Thanks for posting what you have done. Sounds very exciting and it would be great if you would post some pictures. How would yu discribe working with the mini dsp set up?Thanks again it great to se members here step out and have fun with their speakers and have them share their results. Best regards Moray James.

Link to comment
Share on other sites

The miniDSP is very easy to set up and work with. I got the miniDSP and the digital input board, and the digital amp board. These can get stacked, and used in a variety of ways, and getting it configured can be a bit daunting. This is why I began with just the miniDSP first. That part was really easy.

The thing that convinced me to look into this approach was this:

http://www.hifizine.com/2011/09/digital-direct-minidsp-stack/

As for pictures; I will try to take a couple and post them when I return from travel. That may take a couple of weeks, unfortunately.

Link to comment
Share on other sites

How is the sound quality of the minidsp? Are you using a digital input? How does it compare to other cdps or dacs you have used in the past? I am considering using a minidsp 2x8 as an active crossover setup with a digital input, and I am curious as to the sound quality potential. The flexibility and tunability seems to be amazing with this setup.

Link to comment
Share on other sites

Thus far, I'd have to rate the sound quality as outstanding, although I would also say that I have no comparative experience with another digital crossover. I am using the analog in/out facility, so, I rely on the quality of the A/D that is on the board. I have purchased the digital input module also, as well as the digital amp module, but have not yet tried these out.

I front end this with a Hafler preamp, and the sources are all analog (tuner, CD, etc.) thus far, although I am intending to eventually set this up as pure digital from the CD forward, and simply switch in analog content as I change sources.

The output section is very quiet. Presently, I get a very faint noise signature from the upper horn, but I suspect this has more to do with getting the gains set well (e.g., I have the gain on the amp for the upper horns set rather high, before the miniDSP hits it) than anything else. I can say this, however: having used a Rane AC23 and a McClelland CX series through the same amps, the miniDSP is substantially quieter, regardless how one sets gains. I could never get those units to quiet down, no matter what I did with the gains. On the miniDSP, I would suspect that with the gain issue sorted a bit further, the noise that I very faintly hear between tracks (and I have to stand in front to hear it) will likely disappear into the background of my listening room.

Don't be put off by the low price (I know I had concerns about audio quality at this price point before I got it); this thing hits much harder than the price would indicate.

Link to comment
Share on other sites

Have a look here:

http://www.diyaudio.com/forums/multi-way/121175-lecleach-paper-crossovers.html

There is a reasonably good discussion of this crossover in this thread, and at least as of this writing, the links still work. The main feature of the crossover is the relatively minimal phase distortion. This, to my ear, results in very audible consequences. I was leary of setting the crossover point at what becomes a 5 db down point, but, the proof, as they say, is in the tasting.

One thing to watch out for: nearly everyone, it seems, messes up the delay. The delay specified by the spreadsheets is to be added to any delay already dialed in to acheive acoustical alignment. So, if all drivers are physically aligned, there is no delay other than that from the Le Clea'ch calculations. If you have delay(s) dialed in to 'time-align' already, the spreadsheet values are IN ADDITION to these. Read it through a few times, and try it.

The other thing that tends to throw people is the need to invert the polarity on the mid range. It all comes out in the wash, due to the delay that's added, and the bessel response, set up at 5 db down.

Take a look at page 53 of the powerpoint (in pdf) presentation, and compare to p. 50's.

-M

I have not *measured* to see how this performs, but my *ear* tells me it's pretty decent.

Link to comment
Share on other sites

Horatio,

I read the papers... why did you choose Bessel over the butterworth config? The spreadsheets allows for BUTT, BESS, LR but LeCleach suggested using a 3rd BW alignment.

For your 0.22FC did you use the median value of do you use the shifted value for the mid?

Link to comment
Share on other sites

I used this method on my three way Khorn a few years back and really liked the effect. I have since been busy with my two way system and have not tried it again. I played around with the spreadsheet in a two way setting and could not get it to work correctly. I need to revisit the process when I get some free time.

Link to comment
Share on other sites

Re-read the LeCleach slides... the midrange delay is 0.22 x Fr. So I effectively have answered my own question.

Speakerfritz, I am implementing this with a DX38. LeCleach has the settings for a Behringer 2496.

It will be pretty expensive to build third order filters...even if I choose the passive route, I don't know how to implement the recommended delays using passive devces alone.

Link to comment
Share on other sites

Sorry to be late on this response, I am traveling on business this week.

I think I misspoke on the filter type: I am using Butterworth, not Bessel, as I originally wrote.

With regard to the crossover frequency, yes, down .022F I believe is correct. When you do this with the original Klipschorns' crossover frequency of 400 Hz, you are effectivey raising the Fc of the midrange and lowering the Fc of the bass such that at 400 Hz, you are 5 db down. What I did was to KEEP the midrange's -3db down point at 400 Hz...this means I moved the Fc of the bass horn down even more...I don't have the notes in front of me, but I think the -5db point on my system is 350 Hz, which lets the midrange run with a -3db 400 Hz point, and makes the bass horn run with a -3db point of 305(?- again, don't have my notes with me) Hz. Bottom line, is, it's the Le Clea'ch topology, but I have effectively chosen a different (lower) crossover frequency than 400 Hz, but I am still running the midrange such that it retains its -3db point at 400 Hz.

I had questions, too, about how to use the basic Le Clea'ch scheme with a two-way, as this is what I am running. The spreadsheets and notes do not provide clear instruction about this. A bit of googling returned a thread on this that I eventually used to guide my setup:

http://www.diyaudio.com/forums/multi-way/131660-dxc2496-lecleach-setup-2.html

Jean-Michel addresses this 2-way question directly.

For those who've asked about circuit diagrams: I don't use analog filters for this...all done digitally.

Link to comment
Share on other sites

I played around with this crossover scheme yesterday and had different results than when I used it with my Khorns years ago.

This time, I applied it to my mains which are set up as follows. RSW-15, 20-80Hz, JCs DBB 80-900Hz, Jabo horns 900Hz-20kHz. I normally run them with L-R 4th order filters on my Ashly crossover. I took system delay measurements and set the system up according to the directions.

Switching back and forth between both setups, could not discern any audible difference. When I used the setup on my Khorns, there was a noticeable improvement in soundstage and depth. However, with this non-conventional setup I could not make out a difference either way.

The filter simulator spreadsheet shows similar results for both setups and in both cases, the on axis frequency response remains within 1dB.

I ended up going back to the L-R setup since I have been very satisfied with those results for the last year or so.

Link to comment
Share on other sites

I can understand this, to a degree, as my experience with miniDSP led to a similar result.

That is, I initially set up the miniDSP with L-R filters, at 400 Hz, and I set the delay for the HR6040 horn. This resulted in a remarkably good sounding setup, relative to the K400+T35 going though a Type A or AA filter network. I think there are several reasons for this improvement: the HR6040 is very stable in terms of its polar response, the DH1012 is head and shoulders above the K400, and getting an accounting of the driver spacing via delays made a very noticeable difference when A-B'ing the setup.

I then moved to the Le Clea'ch topology, and this seemed to improve the sound stage a bit more- wider, deeper. Not as dramatic an effect as moving from passive to active, but noticeable: I had pretty much given up on the notion that an all-horn setup could image; this one is quite good, as presently set up.

The benefits I see in the LeClea'ch approach are: 1) this noticeable improvement in the sound stage (which I would hope could be captured via measurements of the impulse response) and 2) the approach enables one to move the cross point of the bass horn down to 305 Hz, while keeping a 400 Hz -3db point for the mid/upper end. This part, I think, is most valuable, as it enables to sidestep the more troublesome response region of the basshorn; it's working pretty hard for anything above 300 Hz, and what it does yield is pretty raspy. Being able to sidestep that is a decided advantage, in my opinion. After all, the whole point of the 'Jub was to be able to solidly reach through and beyond 400 Hz, to enable a two-way configuration. If we can get this with a stock K horn and some digital magic, I am quite happy to have that result.

Now to move to the next step: replacing the T-amp driving the top with a single-ended triode (SET) tube amp; I have my eye on that little 2 watt wonder, the Zkit1 version of Decware's SE84C. With a good hefty mosfet solid state amp on the bass horn, this SET 2W amp ought to be more than enough for handling the top end alone.

The biggest problem I have run into in any of the active approaches I have taken has to do with more prosaic problems: depending on how complex the equipment, one needs a 'take-off and landing checklist' for how to power the system on and off. This is fine for the enthusiast, but my wife complains loudly. Gotta find a simpler way.

-M

Link to comment
Share on other sites

I have a fairly complex system setup and have solved the 'checklist' problem with power sequencers. I use three power sequencers power all my amps and crossovers in a given sequence. I use my room for both two channel and 7.1 movie listening. I have it set so that a single programmed remote has a one button program to start the sequence, depending on what you are going to listen to.

If just two channel, it only turns on the appropriate amps and a single crossover controlling my mains. If 7.1 movie is in use, it sequences all the amps and both crossovers.

I also use a HT, triggered power strip and a furman trigger to run all the power sequencers.

The bottom line is that anyone in the house can run the room without any in depth knowledge on how to run the various components.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...