I use BMS's 18" drivers (modified 18S430v2, 18N850 and 18N860... Home). I have had them since 2010. They are very good even as stock in an OB, the venting of the VC is silent. Motor of these things is high tech and employs 3 shorting rings. The inductance is very low and Le(x) and Bl(x) are class leading linear (BMS 18N862 was tested on Voice Coil magazine 2/2012 with top scores).
They are waiting for walnut phase plugs in the picture
An externally hosted image should be here but it was not working when we last tested it.
B&C also makes nice woofers, they have also been tested on the Voice coil (propably 3 or 4 different models, all performing pretty equally, BMS is propably better). One can get free student internet subscription to the Voice Coil...
I see you have a 1" and a 2" compression driver with horn in there also. (Hopefully not the coaxial 2" BMS, as their CD has some pretty bad reviews from the DIY community.)
This makes you a perfect person to ask what you think about line array planar/ribbons crossed pretty low, compared to horns!
Last edited:
Just curious, Stig Erik, is there a substantial improvement with 8-21" drivers per side compared with 4 per side at moderate listening levels, or is it subtle?
I wish you were not on the other side of the world - I would really like to hear your set up!
Russell
I have Fostex TL16H horn supertweeters (~0,65") and 2" JBL 2446J in tempopary P. Audio horns. At some point I'll get new horns and Truextent Be-diaphragms... I also have these and some 8"-12" fullrange drivers to reproduce the upper band alternatively. I usually "current drive" the compression drivers and the fullrangers.
I would propably point out the usual and stereotypical sound signatures of horns and ribbons. The bigger ribbons I have heard limit to Magnepan 3.6 and 3.7. They are too gentle sounding to my liking. Maggies lack midrange percussive energy a little and the sound was somewhat uninvolving (as opposed to lively) as a whole. But I'm always with open mind and would like to hear them in different room and also other bigger magnetostatic drivers/speakers. I usually cross over low (and low order, usually 1st) and try to get as wide bandwidth for the "main driver" as possible. I might get some ribbons also one day as they are capable of taking low xo points (but not propably able to explode one's ear drums with low xo).
Massive compression driver in big horn is unsurpassable in the dynamics sector, they shoots you with uncompressed and firm sound pulses like a rail gun
. Horns require livelier room than dipoles imo to be at their best and the chosen horn and the diaphragm are also very critical. I have not yet tested the compression drivers without the back chamber (as a "dipole"), I hope it would give some characteristics of dipole to the sound as my room is quite dead (RT60 0,2-0,3s).
Last edited:
I have Fostex TL16H horn supertweeters (~0,65") and 2" JBL 2446J in tempopary P. Audio horns. At some point I'll get new horns and Truextent Be-diaphragms... I also have these and some 8"-12" fullrange drivers to reproduce the upper band alternatively. I usually "current drive" the compression drivers and the fullrangers.
I would propably point out the usual and stereotypical sound signatures of horns and ribbons. The bigger ribbons I have heard limit to Magnepan 3.6 and 3.7. They are too gentle sounding to my liking. Maggies lack midrange percussive energy a little and the sound was somewhat uninvolving (as opposed to lively) as a whole. But I'm always with open mind and would like to hear them in different room and also other bigger magnetostatic drivers/speakers. I usually cross over low (and low order, usually 1st) and try to get as wide bandwidth for the "main driver" as possible. I might get some ribbons also one day as they are capable of taking low xo points (but not propably able to explode one's ear drums with low xo).
Massive compression driver in big horn is unsurpassable in the dynamics sector, they shoots you with uncompressed and firm sound pulses like a rail gun
Great answer. This is also what I have experienced when comparing the TPL-150 and compression drivers, but my comparisons are limited to above 800Hz really. AMT's and probably also planars are very gentle. I doubt I will be pleased with a AMT lower than 800Hz, but I have been wrong before. 2" compression drivers are great, but they have a tendency to make the sound cut in your ears at certain parts of a song. So I am looking for equally as dynamic substitutes. I am having a hard time finding one.
I have just used the Avantgarde Trio 2" CD called Community M200 and will soon continue with the JA6681B soon in a Goto horn or Avantgarde trio midhorn. These I will not cross higher than 1500Hz I think. The alu diaphragm is to heavy for faster movements. The AMT will take over after that, and possibly a 1" CD on low volume, but the 1" CD would just be for spice. To liven things up a bit when the AMT is too boring.
I was actually trying to make a diploe of my Beyma 1" CD yesterday, but it was attached to the plastic lid. I would have to destroy it, just for an experiment. And I probably will not experiment much with the JA6681B as these are one of a kind. Once the diaphragm blows they are useless.
Last edited:
I'm using this one:
http://profesional.beyma.com/ingles/pdf/21SW1600Nd_tec.pdf
30 mm p-p Xmax, or +/- 15 mm
http://profesional.beyma.com/ingles/pdf/21SW1600Nd_tec.pdf
30 mm p-p Xmax, or +/- 15 mm
You have some nice equipment. I'd propably like to get copies of TAD TH-4001 (Yuichi) with the Be-pharams in the JBL's, and maybe some big round "sinks" that are more modern like Autotech's low diffraction designs.
I have found that current driving reduces the "screaming" with many drivers and full rangers. It depends how prone the driver is to the electric motor force induced distortion types (the shorting ring topology, motor's strenght, voice coil carrier material etc.) that how much it benefits from the current drive.
I believe that in some cases (personal preference) it is better not to try to achieve everything in one system. Instead I try to expand the system into new areas and make it more flexible, ie. by changing the driver that reproduces the 200Hz-10khz one can change the the sound signature of the whole system (that is usually the band where the music lives, of course organs ect. are different
). This way one can enjoy the perspectives of the AMT, ESL, ribbon, horns, fullrangers etc. have to offer. Not at the same time though, and there will be hell of a lot to do to "get there" (if getting there is ever possible even with one system). (starting to get philosophical vibes)
Last edited:
Beryllium is toxic, - think of the beryllium dust created by that vibrating membrane! - and some say it is useless, unless you are trying to EQ the 2" up to 20kHz, which is a bad idea since you will only get good sound in a very small area. Or, so I have read.
On the other hand, I sense the aluminum diaphragm I have in my Community M200 is not optimum. I would like to try a lighter diaphragm. Magnesium is lighter and not toxic, but it can easily take "fire."
Do you mean field coil? Is this possible with any driver? How does it work?
Well, you may be right. In hifi one has to make choices. I will pop things in when I feel like it. The minidsp makes this possible in a heartbeat.
But, the true ribbon supertweeter stays, no matter what I use below it.
On the other hand, I sense the aluminum diaphragm I have in my Community M200 is not optimum. I would like to try a lighter diaphragm. Magnesium is lighter and not toxic, but it can easily take "fire."
Do you mean field coil? Is this possible with any driver? How does it work?
Well, you may be right. In hifi one has to make choices. I will pop things in when I feel like it. The minidsp makes this possible in a heartbeat.
But, the true ribbon supertweeter stays, no matter what I use below it.
Prima facie I'd presume there is not health danger. The membrane will be here almost forever even after we die (of natural causes
), so the "vaporiging/dusting" of the Be into the air is propably pretty small but I could be wrong. One can also coat the membrane with small layer of shellac/dammar etc. to totally prevent it (but would be quite expensive experiment).No field coil. Current drive means basically that you drive the transducer from high impedance rather than low impedance source like we are used to (= voltage drive). I do the current drive by adding mucho series resistance and step-up transformer. With 8 ohm speaker I add around 30-80 ohms of series R, depending how much current drive is wanted, how much sensitivity I'm wiling to sarcifice etc., how hard load should be in the end etc. Step up transformer matches the impedance and rises voltage sensitivity, pretty close to voltage drive situation but impedance becomes harder, but naturally the most of the power is dissipated in the series R. This way one can use "normal" amp and still realize the current drive.
A transconductance amplifier is also possible, like the First Watt F1, they do the current drive au naturell. I'd like to test one of those and compare to the step-up trafo system.
Here's some theory and measurements: Current-Drive - The Natural Way of Loudspeaker Operation
edit. The amount that current drive can reduce the distortion at maximum is given by (ZO+ZL)/ZL, where ZO is the output impedance (output R + series R) and ZL is drivers impedance at given frequency. If a 50 ohm series resistor is used with 10 ohm speaker (with normal voltage amplifier), we get (50+10)/10 = 6,( "voltage ratio" 6:1) which means that distortion component reduction can be 15,6 dB (=20 lg(6)) at maximum. this means that the voltage to current conversion is mostly done in the series resistor, which does it linearly. Driver's voice coil is not as linear voltage to current converter, because of the electric motor forces that distort the conversion (and thus create distortion). I have mostly quoted Esa Meriläinen's texts here, I hope I got it right
Last edited:
You should try to make your own graphite resistors then. 30ohm is what you get from a normal HB pencil.
I use them to get rid of the transformer for the Ribbon tweeter. Not sure another transformer will be better for sound. Is it like those Autotransformers I see in certain altec crossovers?
+++++Crossover+ Frequenzweiche+für+Altec+604++605+Special+Edition++++ | eBay
I wonder now if I am making a mistake by just using a graphite resistor for my true ribbons, without the stepup transformer. (I always do the mistake of having the transformer to close to the magnets and then I wonder why nothing is working, when the iron in the transformer is magnetized, hence my aversion against transformers. LOL)
I found that a 0.5 ohm graphite resistor is good. Then I just hook it up to a solid state amp and everything is fine, and at 40W I have slightly too high volume for a movie. I crossed the ribbon at 7000Hz yesterday, which was good and does not stress the amp at all. When crossed down at 500Hz the amp get warm fast.
I will have to read more about this. Do you have any links to a good step up transformer for your JBL compression driver so that I could do some experiments on my own?
I use them to get rid of the transformer for the Ribbon tweeter. Not sure another transformer will be better for sound. Is it like those Autotransformers I see in certain altec crossovers?
+++++Crossover+ Frequenzweiche+für+Altec+604++605+Special+Edition++++ | eBay
I wonder now if I am making a mistake by just using a graphite resistor for my true ribbons, without the stepup transformer. (I always do the mistake of having the transformer to close to the magnets and then I wonder why nothing is working, when the iron in the transformer is magnetized, hence my aversion against transformers. LOL)
I found that a 0.5 ohm graphite resistor is good. Then I just hook it up to a solid state amp and everything is fine, and at 40W I have slightly too high volume for a movie. I crossed the ribbon at 7000Hz yesterday, which was good and does not stress the amp at all. When crossed down at 500Hz the amp get warm fast.
I will have to read more about this. Do you have any links to a good step up transformer for your JBL compression driver so that I could do some experiments on my own?
Last edited:
Going from 4 to 8 per side was not subtle, but not primarily because of more cone area. The main reason I wanted to do it was to get the woofer towers the same height as the RD75 ribbons, and I think that was a good idea.
And yes, the bass quality below 40 Hz improved.
Here it was discussed a little which kind of transformer would be good: http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design-28.html#post3449594
I took them with dual primaries, dual secondaries and bifilarly wound. Right now I'm using transformer at 1:4 step-up ratio, primaries and secondaries in parallel. They can also do 1:8 and 1:2 step-up if one changes the primary/secondary connections.
One has to calculate the resulting impedance before connecting anything, the load the amp sees can become very hard if you put a step-up trafo and drive an improper load (too little series resistance).
My trafos (at 1:4) don't have any attenuation at 20khz and do not increase the THD (in the band I'm using them) at least what I can measure acoustically. I will try to measure them electrically at some point. The benefits of current drive seem to outweight the possible negative sides of using transformer but more testing is still needed. I think usually this way in similar questions, that for example transformers might never be better from "high fidelity standpoint" compared to no-transformer, but subjectively the transformer can be "good" or "better" if one likes the change they make to the sound (if one can hear it).
I don't know the function of the autoformer/transformer in the Altec's crossovers exactly. They might also use them to just match the levels instead of using resistors. Autoformer has wider bandwidth but cannot make electrical isolation or balanced lines like a transformer. I have never compared an autoformer and a transformer. Transformer or autoformer is irrelevant regarding the current drive, what matters is that the transducer is driven from high impedance source, and the resulting "voltage ratio" like I wrote in the previous post after the "edit". Current drive is realised without using a transformer/autoformer, it's made by the series resistor.
Ribbons need step-down, not step-up. Step-down rises the impedance the amp sees, step-up drops it. With step down one can make the 0,1 ohm ribbon to seem like a 4-8 ohm load to the amp. With step-up it's the opposite, one can make a high res load to seem like a 4-8 ohm load for the amp, so no current capability of the amplifier is lost.
I will definitely try those carbon resistors from pencil at some point (poor man's Duelunds?), anyone ever tested how much power they can dissipate?
The hardness of the carbon might also affect it's parametres and sound, there are the hard ones used for writing and the soft ones used for art etc...I hope Stig is not offended from the little OT
.edit. By the way, ribbons do not benefit greatly from current drive as their electric motor is not as nonlinear as the typical voice coil. For example ribbon's voice coil (= the ribbon itself) does not have much inductance that can vary when the ribbon moves in the air gap or when the current in the voice coil alternates.
Last edited:
Prima facie I'd presume there is not health danger. The membrane will be here almost forever even after we die (of natural causes
No field coil. Current drive means basically that you drive the transducer from high impedance rather than low impedance source like we are used to (= voltage drive). I do the current drive by adding mucho series resistance and step-up transformer. With 8 ohm speaker I add around 30-80 ohms of series R, depending how much current drive is wanted, how much sensitivity I'm wiling to sarcifice etc., how hard load should be in the end etc. Step up transformer matches the impedance and rises voltage sensitivity, pretty close to voltage drive situation but impedance becomes harder, but naturally the most of the power is dissipated in the series R. This way one can use "normal" amp and still realize the current drive.
A transconductance amplifier is also possible, like the First Watt F1, they do the current drive au naturell. I'd like to test one of those and compare to the step-up trafo system.
Here's some theory and measurements: Current-Drive - The Natural Way of Loudspeaker Operation
edit. The amount that current drive can reduce the distortion at maximum is given by (ZO+ZL)/ZL, where ZO is the output impedance (output R + series R) and ZL is drivers impedance at given frequency. If a 50 ohm series resistor is used with 10 ohm speaker (with normal voltage amplifier), we get (50+10)/10 = 6,( "voltage ratio" 6:1) which means that distortion component reduction can be 15,6 dB (=20 lg(6)) at maximum. this means that the voltage to current conversion is mostly done in the series resistor, which does it linearly. Driver's voice coil is not as linear voltage to current converter, because of the electric motor forces that distort the conversion (and thus create distortion). I have mostly quoted Esa Meriläinen's texts here, I hope I got it right
That looks interesting! Like here: http://www.firstwatt.com/pdf/art_cs_amps.pdf
Do you use a ~47 ohm resistor? How much does that attenuate the signal?
I tested my poor mans Duelund up to 40W for many a long evenings, IF my amp is correctly labelled. They could probably take more. I use a big chunk of solder as a homemade cooler. Posted a pic a few posts ago. I use a thicker naked and very soft 6B graphite that I found in a bookstore. Ended up with 0,5ohm. 2cm of HB pen gave 4 ohm.
Sry for my stepdown/stepup mixup. Transformers confuse me.
Sry for my stepdown/stepup mixup. Transformers confuse me.
Right now I'm using 2x33ohm = 66ohms (balanced config because the amp is also balanced). This has worked well with 16ohm compression drivers and 8ohm fullrangers.
This results in 5.125:1 and 9.25:1 voltage ratios respectively, meaning 14.19dB and 19.32dB maximum reduction of harmonic components caused by the voltage drive. Like I said, some drivers show massive reduction in THD/IMD, while others show smaller reduction, it depends on the motor and how good "voltage to current converter" it is.
Step up ratio being 1:4, this results in ~1,79dB loss in voltage sensitivity with 16 ohm compression driver and ~6,9dB voltage sensitivity loss with the 8 ohm fullrangers. The impedance the amp sees is (66ohm+16ohm)/(4*4) ) = 5.125ohms and (66ohm+8ohm)/(4*4) = 4.625 ohms. I have not yet fine tuned the values or even tried anything else. Works so good that there is no real hurrry
.Beware that this rises the Qts greatly as the electrical damping is basically lost. With woofers current drive cannot be used greatly without eq'ing the hump around the Fs away(if one wants to get bass notes that aren't all over the place and boomy). When used mainly above the driver's Fs this is not needed but the frequency response is altered regarding the driver's inductance and the resulting rising impedance towards 20khz. Also some smaller parasitic resonances that show in the impedance plot, will be somewhat boosted in the freq response when current driven. Short version: the flatter the impedance plot of the driver is, the less current driving changes the freq response.
Imo current drive usually works best for midranges and tweeters. Voltage drive artifacts are not that objectionable in the bass region so there is no "real" need to reduce them (at least not as great?). But I have not really tested much current driving woofers (sole bass transducers) yet. The higher amount of high order harmonics that results in voltage drive might actually be good to some extent at bass freqs, giving the notes better definition (one can trail/follow the notes easier) and snappier nature, just like when driving woofers with an amp that has crossover distortion. These objective vs. subjective questions interest me, I don't usually care for better performance if it does not sound better.
The mentioned question about "High fidelity" or "objectively accurate" which can usually be seen from measurements, opposed to "subjectively pleasing", which can be anything and does not relate to measured performance necessary.
Last edited:
Hi StigErik
Have you seen the following post:
Dipole wool carpet baffle experiment
I was looking at that idea and the subsequent measurements OllBoll made, and something suddenly occured to me: why don't you make a carpet underfelt\wool 'frame' to fit around your RD75s, all the way around from the edge of the ribbon to the wooden frame. Just the wool felt, nothing else, like a soft absorbtive baffle. It should not show the nasty response deviations that a solid baffle showed. It would be interesting to see how this will affect the measured response. Worth a try maybe? I really hope you think so, as I would love to see what the effect will be.
Enjoy,
Deon
Have you seen the following post:
Dipole wool carpet baffle experiment
I was looking at that idea and the subsequent measurements OllBoll made, and something suddenly occured to me: why don't you make a carpet underfelt\wool 'frame' to fit around your RD75s, all the way around from the edge of the ribbon to the wooden frame. Just the wool felt, nothing else, like a soft absorbtive baffle. It should not show the nasty response deviations that a solid baffle showed. It would be interesting to see how this will affect the measured response. Worth a try maybe? I really hope you think so, as I would love to see what the effect will be.
Enjoy,
Deon
I found the dynamics of the neo 10's really opened up by 2nd order high passing them at 500hz. This being done at line level, tried a simple first order passive x over between them and a Neo 3(backless). Agree with your assessment of optimum frquency range. Neo 10 did need to be in quasi enclosure transmission line loaded (@500hz) and helmholtz tuned with 19mm long port 10.16 mm wide to maintain some dipole benefits. This led to a pretty odd looking "uncabinet" with a ported enclosure @ 6.5"x11.5", sandwiched between two 12"x19.375"x3/4" baffles, front with flush mounted Neo3, 4" hole behind in rear baffle to avoid reflections. I do not find this set up warm or cold, just fairly lifelike, paired with a pretty good sub.I do want a " bigger" sound, however, and and am in the process of trying out 2nd kick at the polar opposite of these- no sub, dipole 15" woofers sharing an open baffle with compression driven 12" waveguides. These were easy to get good , but seem likely to be hard to get very good. Advice is invited on the open baffle horn hybrids.
Two RD75 per side
Hi
How would two RD75 per side compare to one per side.
Will there be any benefit and in such case which crossover frequency will be recommended between the RD75 in an Active setup with a dedicated power amplifier to each.
Could one RD75 between 180-800hz and one from 800-20000hz be recommended.
Hi
How would two RD75 per side compare to one per side.
Will there be any benefit and in such case which crossover frequency will be recommended between the RD75 in an Active setup with a dedicated power amplifier to each.
Could one RD75 between 180-800hz and one from 800-20000hz be recommended.
In my head one would screw up things with a setup like that. Main reason for me to choose RD-75 is it's ability to be full range with satisfying volume without crossover in the critical 800-3000Hz range.