m0tion said:
There was some talk about materials earlier on in this thread.
The general concencus was that concrete with lead shot was the most economically sound, easiest to implement and above all, offered the most material resonance damping outside of complex composites and tailored polymer constructions.
I found that link BTW:
http://www.zelfbouwaudio.nl/index.php?option=com_content&task=view&id=18&Itemid=2
Its in dutch so you'll have to babelfish for a rough translation. Concrete with lead shot isn't listed unfortunately but the best tested was a composite of fibreglass, lead, MDF and glasswool with bracing - very very time consuming and expensive to build. Certainly impressive though:
An externally hosted image should be here but it was not working when we last tested it.
Compared to 18mm MDF:
An externally hosted image should be here but it was not working when we last tested it.
Wouldn't the double or trible layer sound proof window structure prevent panel vibration better?
m0tion said:
Both alu and steel have a main narrow Q resonance that shifts frequency in relation to mass and dimensions. They also have relatively poor damping characteristics with long resonance decay times at this main resonance. When dealing with thinner panels this can be lessened greatly by applying pressure at the edges(think tuning fork) or simply using a rather massive piece bolted to a lossy material.
Might be worth investigating if you have access to the relevant CNC machinery required. Big money though for something like the baffles I'll be doing in concrete and lead shot.
Greggo said:
Greggo,
The Zu design is an ELF system - the "upside" here is that they are using multiple drivers, the downside is that they are using considerably less cab. volume and so the very steep 4th order roll-off happens even higher in freq.. All this means is that as the freq.s extend lower the drivers become increasingly non-linear. ..AND this is at fairly low spl's. So despite what it looks like - its still a significantly compromised design.
The Escalante design is nothing more than a "push-push" system that halves Vas AND truncates driver decay character by using a simple filter to delay the visible driver in-relation to the non-visible driver. Beyond halving Vas, suppressing decay in this manner is a bit misguided IMO. It tends to clean up the sound (and imaging to an extent) at the expense of "scrubbing away" some of the soundstage character. It does this better in the drivers linear operation than other methods like a heavily damped cab. volume via stuffing. But not as well near the driver's in-box resonance (i.e. its non-linear region) as a heavily damped TL. It will however (because of the added mass in relation to the overall Bl) have a more "punchy" character than most designs. Note that the offset design WILL effect non-linear behaviour in a negative fashion to at least some extent.
Neither of the methods should ever have been allowed a patent IMO. The only thing novel about either design is the Escalante driver "offset" for the compound configuration - which again IMO does not represent something new and is more akin to a flavor on a theme.
As to Escalante's compound load and the B&C drivers:
This would not be a good idea.
The design was created to operate in a fairly linear fashion over a small passband. I.E. below the driver's in-box resonance but above the port tunning.
IF you operate the driver above its in-box resonance - your are then necesarily operating it AT the in-box resonance. At and near the driver resonance it doesn't behave in a linear fashion (..though it isn't as bad here in this design). This is something I wanted to avoid.
Next, there is a rather marked difference in output from the 50 Hz up where the B&C driver is MUCH more efficent. (i.e. 80 db at 20 Hz vs. 94 db at 200 Hz.) This would of course require a LOT of attenuation and I don't think it would be worth the effort.
Finally, the phase relationship between the linear operation near the port tuning freq. (say 24-25 Hz on a 19 Hz tunning) doesn't rotate a great deal. I think it was around 50 degrees *total*. That however is NOT the case if you extend the bandwidth up higher in freq.. THEN the phase will rotate significantly more over the now extended passband.
Sorry. Don't me to bum you out here, they ARE interesting designs.
m0tion said:
Its not just about the material but also about the shape.
From a DIY perspective I think its more to do with ease of construction and cost. Cement is just plain cheaper and easier to mold. As Shin' has noted, its also a material you can "load" with other material to disrupt resonances.
IMO there is one area where certain metals might have an advantage - rigidity and driver coupling.
The more rigid the material, usually the better the apparent definition (up to a point). I'm not sure if the difference between both very hard materials (hard metals vs cement) would become audible. I have however noticed a difference between mdf, plywood, hard maple, and cement. Each time there is a subtle shift in apparent "resolution" - and it isn't that subtle from standard mdf to cement.
re front baffle
...have you thought about rear driver mounting (i.e. holding the driver by its magnet structure and not by the basket, and sealing it to the front baffle with some weatherstrip which the driver only lightly touches).
Linkwitz has a page on his site where he discusses the driver mounting issue together with front baffle rigidity etc. . His conclusion was that the rear mounting technique improves the situation to a much greater extent than optimizing the front baffle can achieve.
Mounting a driver to a baffle
...have you thought about rear driver mounting (i.e. holding the driver by its magnet structure and not by the basket, and sealing it to the front baffle with some weatherstrip which the driver only lightly touches).
Linkwitz has a page on his site where he discusses the driver mounting issue together with front baffle rigidity etc. . His conclusion was that the rear mounting technique improves the situation to a much greater extent than optimizing the front baffle can achieve.
Mounting a driver to a baffle
Re: re front baffle
Hi MBK
I think its more beneficial when dealing with cone drivers. The W22 used in the Orion has an Mms of around 40g compared to 4g for the ATC mid and 0.3g for the treble. The ATC is rear mount so it would be fairly easy to adapt the design with a metal bracket that attaches to the existing mounting holes.
MBK said:
Hi MBK
I think its more beneficial when dealing with cone drivers. The W22 used in the Orion has an Mms of around 40g compared to 4g for the ATC mid and 0.3g for the treble. The ATC is rear mount so it would be fairly easy to adapt the design with a metal bracket that attaches to the existing mounting holes.
Aaaah... now I get it - I thought you referred to the sub mounting & baffles (where vibration control probably matters most). If the ATC is rear mounted - and given the Mms involved - there are no real issues here...
Scott,
Thanks for the reply on the various designs I was reading about. I am not bummed out at all, and in fact really appreciate your perspective and good info regarding the issues at play.
Not to hi-jack this thread, but would the Delta 12LFA be a good "poor mans" starting point for using a 2-4 drivers in a sub woofer array and/or using a pair tuned below resonance and and a pair actively crossed above resonance and run up to a decent tweeter?
If you had a bunch of the same drivers of any type and wanted to play around with the concepts you have talked about here, would you set an active crossover just below and just above cab resonance (by the way, how do you go from the cones free air resonance to the calculating or measuring the resonance in cab?), thus leaving a dip where the cone would have the greatest distortion? If I had a cab resonance of 55hz, would you low pass around 50 and high pass around 60? Just wondering.
Thanks for the reply on the various designs I was reading about. I am not bummed out at all, and in fact really appreciate your perspective and good info regarding the issues at play.
Not to hi-jack this thread, but would the Delta 12LFA be a good "poor mans" starting point for using a 2-4 drivers in a sub woofer array and/or using a pair tuned below resonance and and a pair actively crossed above resonance and run up to a decent tweeter?
If you had a bunch of the same drivers of any type and wanted to play around with the concepts you have talked about here, would you set an active crossover just below and just above cab resonance (by the way, how do you go from the cones free air resonance to the calculating or measuring the resonance in cab?), thus leaving a dip where the cone would have the greatest distortion? If I had a cab resonance of 55hz, would you low pass around 50 and high pass around 60? Just wondering.
Greggo said:
Thats just a matter of modeling the driver to see if you get what you want with respect to low freq. extension vs. spl for a given freq.. Try it out - might be good, but remember to factor in things like room gain and boundry gain.
Member Tenson has more experience (success actually) with running a 12 inch up beyond 1 kHz. In general my advice is don't do it - particularly when I look at its polar radiation (which almost always suffers a lot of combing and is more directive higher in freq.). It could also have considerably worse non-linear behaviour at higher freq.s. In general a 12 inch should have a low pass crossover at 800 Hz or less.
...
You could do that (or a notch filter), but it then becomes *to* complicated and phase really turns to cr@p. If I were doing something like this it would be a "rumble" filter for the port (imeadiatly below resonance), not for the in-box resonance.
If you want further replies consider making a new thread and perhaps request from a mod to transfer this stuff over to that new thread.
Shin' -
What was the final outcome on the gainclones (and differing power supplies)?
Also, how goes the UCD700?
What was the final outcome on the gainclones (and differing power supplies)?
Also, how goes the UCD700?
Just a quick note... the Eminence Magnum 12 (now called Diffinimax12) worked beautifully up to 1.3KHz with a 24dB crossover.
However, for about the same price you can get the Supravox 285GMF which, if you are going to have dedicated subs, you may be better off using as a midrange unit.
However, for about the same price you can get the Supravox 285GMF which, if you are going to have dedicated subs, you may be better off using as a midrange unit.
ScottG said:
I ended up quite enjoying the gainclones. Vikash lent me an LM3886 based GC with a beefy PS and I thought this was preferable to the Patek SE implementation that I'd tried previously - those seemed a but dull and flat in their presentation leading to something entirely less musical. I guess system synergy was also an important part of the equation because the ATC's like plenty of current and power so no doubt they benefitted from a substantial PS.
The SMPS I gave up on, horrid sound IMO. Sure its very controlled but the life is sucked out of the music. Bass seemed marginally more controlled than Vik's LM3886 but the rest was definitely worse, I preferred Peter's Patek PS to this one. After that I only tried it for a few days and threw the idea out.
The UcD700 and parts are still in the shipping boxes. I haven't really got an excuse to build them just yet as the system is in bits and the room is my first priority now - I've actually just stripped one of the wall this evening and found that I'm going to have to hire a steam stripper for the rest as the wallpaper just isn't comming off very well at all.
ShinOBIWAN said:
Ilckth..
I've done that several times before (both chemically and with steaming).
Learn from my mistakes here: Its far better just to remove the drywall IF you have drywall, and replace it. If its lath an plaster - then you are just better off sanding the whole thing down and re-applying a skim/finish coat of plaster (..here though the finisher (pro) can make a BIG difference with good marble/venetian plaster - makes a room look MUCH more expensive and it can be done on bathroom type drywall as well).
I'll bet it's neither of those and is simple plaster on brick 😀 I honestly can't see why you would sand the wallpaper off that. Loads of effort and loads of mess.
Ant, try soaking the paper with a splash of sugar soap in the warm water solution. Keep applying the solution with a car sponge or big paintbrush.
Ant, try soaking the paper with a splash of sugar soap in the warm water solution. Keep applying the solution with a car sponge or big paintbrush.
ScottG said:
The house has brick walls throughout, no cavity walls.
We never have to hire anyone in our family 😀 The old man is a trained joiner and painter/decorator, one uncle is a bricky/plasterer, the other uncle is an electrician and my brother is a trainee plumber. So we've got it all covered, quite often its a case of you scratch my back and I'll scratch yours, so it tends to cost much much less than someone that you got out the yellowpages.
I'm going for a light and simple feel with three walls in Magnolia(very light cream colour) and the wall with the equipment backing up against will be a light brown to provide a bit of contrast then its white ceiling, dark tan closed weave carpet and the treatments will be finished in light coloured wood such as pine/beech/birch type shades (I was going to go for silver with a high gloss but this would look tacky on the walls and ceiling I believe, so threw that out pretty quickly)
richie00boy said:
I spoke to the old man and he said I'd gone messed up painting over paper. It apparently makes removing the paper ten times harder. Only reliable way is a steam stripper, so I'll get on that soon as.
Cheers for the advice.
richie00boy said:
WOW. No insulation at all?
Btw, its actually very easy and fast sanding with a belt sander and a very coarse grade paper (..then use a shopvac to suck up the dust). It will gouge into the wall some but the finish coat cleans that up.
I've tried the soapy water approach as well - this depends a lot on the paper, wall material, and glue. Most of the time I've found that what I suggested was faster and easier.
- Status
- Not open for further replies.