Just an update after I've had a chance to listen to them for a few weeks ( it seems that car speaker drivers take a while to break in ), I'm quite impressed with them. I had heard that omnis didn't really suit rock music, and I was beginning to agree, however after a change from an Audiolab Q DAC and Onkyo amplifier to a class d and Sanskrit 10th DAC there was a dramatic improvement, now they work well with any music. The change from the Onkyo to class d was interesting, with the Onkyo ( presumably class ab ) there was a lack of deep bass and treble, using the tone controls and loudness was really necessary, however with the class d, the bass and treble are still slightly lacking, however not so much as is particularly detracts from the music, perhaps more break in will yield further improvement; I had thought that XT25 speaker cable might boost the frequency extremes, although I am dubious what difference 4 foot of cable will do. The class d and Sanskrit 10th also give a much more dynamic sound.
I've recently played some classical, RUSH, early Opeth, Public service broadcasting, Rodrigo and Gabriela and Yazoo, and the result is I just want to play more and more music. Because of the 4.5 KG cast iron lump bolted to each magnet,( taking most of the reaction forces from the voice coil and cone ) and the light weight upper chamber, partially damped by the foam gasket there's very little left over vibration, this results in a very clear, detailed natural sound.
I've recently played some classical, RUSH, early Opeth, Public service broadcasting, Rodrigo and Gabriela and Yazoo, and the result is I just want to play more and more music. Because of the 4.5 KG cast iron lump bolted to each magnet,( taking most of the reaction forces from the voice coil and cone ) and the light weight upper chamber, partially damped by the foam gasket there's very little left over vibration, this results in a very clear, detailed natural sound.
Last edited:
A further update, I've tried them with rear facing, out of phase tweeters but then decided as I was using home made attenuator plugs between the DAC and amp, I could use a bypass capacitor to boost the treble, keeping the speaker ( reasonably ) omnidirectional throughout the frequency range, and have all frequencies coming from the same driver. This is a bodged proof of concept vero strip board made from components I had to hand, I sacrificed two old interconnects, splitting them half way and soldering them to the board
I initially used the top chamber empty, relying on the internal shape to control internal reflections, however they sound better stuffed with cotton wool.
They're not perfect; a teardrop shaped cast iron weight and rounded edges on the top chamber would reduce diffraction, as would a more open driver chassis, but the ( almost ) total lack of leftover vibration ( from the weight on magnet and lightweight top chamber ) and all frequencies coming smoothly from one driver makes them exceptional at revealing subtleties and very rapid percussion. The sound stage is still #### though.
I initially used the top chamber empty, relying on the internal shape to control internal reflections, however they sound better stuffed with cotton wool.
They're not perfect; a teardrop shaped cast iron weight and rounded edges on the top chamber would reduce diffraction, as would a more open driver chassis, but the ( almost ) total lack of leftover vibration ( from the weight on magnet and lightweight top chamber ) and all frequencies coming smoothly from one driver makes them exceptional at revealing subtleties and very rapid percussion. The sound stage is still #### though.
Now that I've had more time to listen to them, I 've made a few more tweaks. I've modified the amp ( increased the power supply capacitance ) increased the treble boost on the interconnect, improved the speaker cable from thin bell wire with push on spade connecters to medium bell wire soldered to the terminals, and re-positioned them farther from the wall, all of which have reaped improvements. When I first listened to them I was disappointed that the cast iron weights didn't seem to do much, but now I realise that the dull sound was masking the effect of the weight, with the treble boost percussion is now exemplary, I'm now genuinely amazed at what a large ( ish ) flexible speaker cone can do, including the subtle details that were one hidden and I've never heard before are all ( presumably ) there. There's one word to describe the sound, it's clean. With the cast iron sucking all the vibration from the magnet, and the enclosure separated into two parts, further reducing vibration, there is very little box colouration. I presume that there is also advantages from all the frequencies coming from one cone, with an even dispersion. Although the bass is very good, I think they could still reach a little lower, but I guess you can't expect much chest thump whilst listening at 60 Db. I'm certainly glad I made them, and now I'm working my way through CDs.
While the application of cast iron on the magnet makes sense in the design, it makes me wonder about damping magnet vibration in more traditional enclosures. Most people are using plywood braces it seems, and I'd guess whatever radiates from a magnet inside is insignificant compared to everything else that's radiating sound, but still interesting.
I wonder if cast iron damping the magnet would indeed sonehow sound like different coloration than with other materials. Always something to chase.
I wonder if cast iron damping the magnet would indeed sonehow sound like different coloration than with other materials. Always something to chase.
That idea behind the cast iron wasn't really to stop sound radiating directly from the magnet, it was to stop vibration being transmitted to the rest of the speaker and to stop the magnet vibrating ( the speaker chassis can act like a spring ), so it would give the voice coil a more stable " datum " to act against. I'm unsure about bracing the magnet in more conventional designs, as it could just transmit the vibration directly to the cabinet, that would act like a sound box, but if there's something compliant inbetween it could act like a dampener.
Yes, the brace would certainly need to be damped itself, to stop the magnet vibrations from reaching the walls. I'm more or less wondering about the audible impact of doing so. The BBC paper suggests bracing the driver is one of the more important aspects, damping the most problematic mode. But now I'm wondering about the material coupling the magnet to the brace.
I have read a little on this site about it, but never anyone using metal. Just curious about it. I do understand your application is achieving a different goal, but still, in yours, I wonder about the material used. It seems your main goal is weight/mass added, but would different materials coupling the weight/mass to the magnet produce different coloration? Or once enough weight is added, is it the absence of coloration that you are enjoying?
Sort of just spitballing here, sorry it's not directly related to your build.
There is an MDF disc between the magnet and cast iron, partly to stop the iron effecting the magnetic flux ( I have NO idea if this would be a problem ) and to add an element of damping, cast iron is slightly self damping. As I'm using a bolt to hold it all together, I could experiment with several discs of metal, sandwiched between rubber, or something similar. I do like the idea of the BBC style " lossy " cabinet, I have made a few speakers that have the front and back held on ( against soft gaskets ) by screws or tension springs, inspired by the BBC ls3/5a, it's just I thought that I'd have a go at stopping the problem at source ( with the weight ) rather than further down the line.
I've taken the opportunity to try them out in a better room (concrete floor, and less of a cube), I've also removed the (cotton wool) stuffing from the top chambers and tightly stuffed the bottom chambers, this with the new treble boost circuit seems to be as far as I can go with them. They certainly sound very clear, no muddiness and a very good reproduction of timbre. The sound-staging is still poor for some reason, but all the frequencies coming from one driver helps them reveal the music that they're playing, good for classical and jazz, but the Prodigy lacked punch.
Because the driver cone is (just) visible from the listening position, it should radiate the sound itself, much like the German Physiks speakers.
It is always funny to see that people presume that a loudspeaker is directing its sound from the front of the driver in forward direction and in sidewards direction from the back of the driver. Thats is physically not possible. Any sidewards radiation is only caused by difraction of the magnet and frame, not caused by the shape of the cone. Purely by shape the directivity pattern from the front and the back is the same and you are better of flipping it over and use a proper shaped reflector instead of relying on, not designed for omni, diffraction of the frame/magnet structure. Oh, and don’t forget the output of the spider which causes doppler and phase distortion.
The german physics is a bending transducer, it is not a stiff cone it is made to flex. It is a totally different designed cone shape than your flip over driver.
The german physics is a bending transducer, it is not a stiff cone it is made to flex. It is a totally different designed cone shape than your flip over driver.