Mark K's Speaker Pages
...when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of Science...Lord Kelvin
Personal thoughts on NC DIY AUDIO Spring 2010 1/2010 (What the heck am I talking about? Go to the NC DIY AUDIO page.) |
Posted 5/09/10-NCDIY Postmortem
It's been a couple weeks, and attending a DIY always makes me think about my design philosophy, goals, as well as compare and see what others are working on. Below are a couple of offbeat musings.
SPL baby! There was a lot of grumbling about the music being too dynamic and having too much low end. I admit, I tricked folks with my introductory cut from Wynton Marsalis' Standard Time, vol 6. It starts out very low and then bam! with the horns. Everyone naturally cranks up the volume too much during the initial quite passage, then quickly scrambles back to the volume control when the horns kick in.
Well, not everyone. Curiously, folks with 3 ways, like the Statements, or the Orion-did not sound "too loud." Loudness is a funny thing. Think about this quote from SL from his Pluto spec page-
With small speakers that produce a very clean sound there can be a tendency to drive them to their maximum capability. We judge a speaker's loudness primarily by how distorted it sounds, and not by how loud we have to talk in order to be understood by the person next to us.
This is key. Many of the two ways were being driven to their limit, so to speak. Before obviously audible distortion sets in, a kind of strain is perceived. It's "too loud," but it's not the absolute SPL level. The Statements were playing at 80-100dB, but they were not perceived as "too loud." When the Orions were playing I overheard SL murmur that the volume was not high enough! I believe a choral piece was playing, the same one that one of the posters afterwards thought was not a very good piece. I went home and listened to that piece again at a higher SPL level (90-100+dB) and guess what? The piece sounded much more alive. I had to agree with the poster that, at the DIY, the choral piece did sound lifeless. Realism, especially on live pieces does require realistically loud SPL's. This requires significant displacement below 80 Hz. There is really no way that any two way is going to reproduce a dynamic piece at realistic levels in a large room, like the ones that the typical DIY are held in. I suppose in fairness, we should turn down the spl level for two ways, but that begs the question, "Is a two way good enough?" That's another debate however.
To this end, I'm working on an active circuit that would basically make it very easy to make a biamped 3 way out of any current two way and a woofer module. See the picture below.
Here we have the NC DIY audio budget unit. A nice unit, but very limited spl. Add the following circuit below along with a big fat woofer. I know, a bit tiny. Well, squint!
The active circuit would look something like the circuit below. Not exactly as it needs a buffer, bypass caps, a level circuit, but you get the idea. Really, just build the circuit from the parts you get from digikey, scrounge up a case and power supply and you have a really nice three way.
You need a four channel amp, but the flip side is that you are protected from bad juju amp distortion into the midrange, unlike a passive configuration, and you can have a low end level control which would markedly increase placement flexibility.
You could build something like a Zaph midwoofer/SB29 in as tiny of a box as you canand use the above circuit to add a Peerless 12" woofer, say in two of the cabinets below. You'd instantly have a really nice three way.
I'm hoping I can get PC boards designed for a reasonable fee. Let me know if you are interested.
Posted 4/11/10
Wow! These cabinets from DeepSurplus are very nice. Charlie and I bought a pallet. (Because that's how they sold it!) Neither of us needs 17 cabinets, but that's how it came. We're going to try to sell subwoofer cabinets to anyone that wants them. I'll take $25 a cabinet. These are nice, lighter cherry-not the dark cherry- 15" cubes with grills. Very nice. Check them out when you come. You can get four for $100 bucks, two per side for a set of dual woofer bases. 4 of these is cheaper than one new one at PE. In fact, since I only want ~6-8 and I have 17, I'll sell 6 for $125. You could make a stack of 3 a side, or an entire three way. It's so cheap it's worth stripping the finish if you don't link it and lightening/darkening to suit your taste. Might even give one away as a door prize.
I've got SE working with my Delta 410 soundcard very nicely. So much so that I'll probably just skip the Bryston and run the digital filter function of SE to biamp into the Parasound. What I'll do is generate a filter in SE on the fly, as a demonstration, a lowpass at 125hz into an LT circuit for the sealed sub cubes and then corresponding highpass active crossover circuit into any pair of two ways. See below. Note this is just a quickie that I drew out in SE in 3 minutes. The values/circuit hasn't been verified.
The idea though, is to have SE make a transfer function for the woofer at node 22 and MT unit at node 15 and emulate this. It's pretty cool if you haven't done/heard it.
Posted 4/7/10
It's looking dimmer and dimmer for the dipole. But, it's looking better and better for a set of dual 12" woofer cabs based on the older, but still very impressive Peerless XLS 830500. I've had a pair sitting in my closet for a couple years now. I'm just too lazy to build a cabinet, and and active setup. But I've recently obtained some very nice cabinets from www.deepsurplus.com They blowout overstock from M&K's bankruptcy. Their loss is our gain. For ~50 bucks a cabinet, a very nice set of cabinets, good maple finish, and already the perfect cutout for the Peerless. You can see the photo below.
The only issue is the cutout in the back. Thankfully, there holes with star nuts, so I can just cut out a piece of stock maple and screw it over the cutout. Viola, a solid coverplate.
I've got a pair of Walnut cabinets coming, and split a lot with Charlie for a bunch of 1 cu ft cabs in Cherry.
So what's coming? Well, a bit of a detour. The RS225 dipole panels go to 40-50, but could use a sub, for maximum full range effect. I just bolt on the Peerless, make a coverplate, and the pair of XLS woofers is done.
Now, you just need the electronics. The way to go, is of course active.
So the RS225 dipole panel will have the 4 RS225's along with a waveguide loaded RS28F. There is active dipole eq into a passive 4LR crossover between the RS225's and the RS28F and this will require one amp channel. The other amp channel will feed the Peerless XLS. See the block diagram below.
Posted 3/28/10
I plan to bring the following-
My measurement setup. I will measure for anyone who wants it-basic FR, impedance. If there is interest, I will review distortion measurements and maybe burn out some crummy drivers...
A Parasound 85watt 5ch amp along with a Bryston 10B crossover. Anyone who wants to try an electronic xover and use their MT as a satellite is welcome to try this. I do not have a pair of matched woofers/subs yet, though I have a set of cabinets on order. It may not come together so, unless someone brings a set of woofer cubes I can't guarantee this will work. (Hint, if you have a matched set of woofer boxes and can bring them, do so. They don't need amps since we'll be using the Parasound an Bryston.
I'm working on the music now.
The morning session will have set music tracks. I'm using a program called Audacity to slice up particular cuts. The usual drum solo/percussion, male vocal, female vocal, live, choral, classical, jazz, rock. The tracks are anywhere from 20 seconds to just over a minute long. They are strung together without any gaps.
There are pros and cons to this approach. Here is my thinking. Really, within 30 seconds you can appreciate the acoustic space of the recording, the dynamics, the detail. I know folks will argue this point. Still, I've been to a lot of diy's where someone feels the need to play 10 minutes of his personal favorite. I've pretty much gotten all the information I need out of the first 30-60 seconds. If I like the song, fine. If I don't now I have 9 minutes of my life gone. I'm all for deference to folks' music and designs but folks look worn out and beat up by 3 pm.
Think of this as the "compulsory program" in ice skating. It's a series of elements that every skater(speaker) has to do, so that we have a uniform comparison standard.
Yes, there's also a free skate program. In the afternoon, we'll have a freeform demo session. Jon Hancock is bringing a very nice setup. You sould be able to play either CD's or digital. You can use any cuts you like and there will be more latitude on time, positioning, amps and sources.
You'll be able to demo in the morning or afternoon-maybe both if there is time, but I'm not so sure.
Anyway this is what I envision.
What about my dipoles? Er...I'm going to do my best, but probably will only have a mono single prototype to show off, using SE to provide the eq and run it through the Parasound. Even that is iffy. What can I say, work and family life places a lot of demands. The detour into horn/waveguide loading has also slowed the process down.