Errr, you can’t compare absolute levels like that. Use one of the aforementioned ways to adjust levels to reference of the woofer.
a possible way of doing the combined driver/reflex measurement following klippel document, just in case it may be useful:
if your bass reflex tuning is very low you may not be able to find the nulling point, because the pink noise PM output can only be set to a min of 10 Hz. in this case you need to do several low frequency measurement sweeps (5-50 Hz or so) to find the correct microphone position.
- connect the speaker to an amp and the amp to the computer soundcard. don't set the volume too high!
- start the generator in REW and select "pink PM" choosing custom frequency range from 10 Hz to about 50 Hz
- start the real time analyzer with the same frequency range.
- now watch the graph on the screen while SLOWLY moving the microphone between driver and port(s), making sure to keep the mic near the box enclosure surface.
- once you find the deepest notch (cancellation of port and driver output far below bass reflex loading at frequency far below BR tuning) that is the correct spot where to put the microphone for the combined driver/port nearfield measurement.
if your bass reflex tuning is very low you may not be able to find the nulling point, because the pink noise PM output can only be set to a min of 10 Hz. in this case you need to do several low frequency measurement sweeps (5-50 Hz or so) to find the correct microphone position.
Maybe I'm misunderstanding, but what I'm comparing in the second shot is the summed port+woofer (port compensated -6.8db), to the measurement taken midway between the two, trying to approximate acoustic center. it was just a "why not?" quick look at what's going on. I'm going to try stv's pink noise method and see what that brings, and also play with decreasing the Vbox in steps closer to the Vas of the driver and see if that makes things prettier. My neighbors are looking at me like I'm a crazy person running these sweeps at 100db in the driveway, and my next door neighbor has a big bird that mimics the sweep, it's hysterical!
Going back to the original question, is the Radian coax really that great for flea power? I see that a lot of high efficiency pro-audio type speakers may be interesting for home hi-fi because the distortion drops to lowrr levels when used at low power.
But if we're really looking at peaks of 1-2mm of cone displacement and 4V, I think some of the design choices like a 750W rated voice coil start to become a liability. (Even if it is with the usual wink and a nod: ultra-super-max music program power, calculated with nominal impedance and enough overhead to never clip, band limited, etc... )
Full range drivers come to mind like the Visaton B200, or Supravox as low power alternatives. Something with a super-fine voice coil, low over-hang or under-hung. Light-weight cones that can resolve high frequencies...
But the Radian is a coax. How important is that? DIY co-axial speakers have been done before... But right now most of the design effort is trying to get the 10" woofer to produce deeper bass. The obvious alternative here would be a 12 or 15" woofer. A small subset of those allows a compression driver to be funneled through the pole piece. But again, this seems like a solution where high power and similar dispersion to a full-range driver are both required.
But if we're really looking at peaks of 1-2mm of cone displacement and 4V, I think some of the design choices like a 750W rated voice coil start to become a liability. (Even if it is with the usual wink and a nod: ultra-super-max music program power, calculated with nominal impedance and enough overhead to never clip, band limited, etc... )
Full range drivers come to mind like the Visaton B200, or Supravox as low power alternatives. Something with a super-fine voice coil, low over-hang or under-hung. Light-weight cones that can resolve high frequencies...
But the Radian is a coax. How important is that? DIY co-axial speakers have been done before... But right now most of the design effort is trying to get the 10" woofer to produce deeper bass. The obvious alternative here would be a 12 or 15" woofer. A small subset of those allows a compression driver to be funneled through the pole piece. But again, this seems like a solution where high power and similar dispersion to a full-range driver are both required.
^ it appears that many proaudio drivers are very well made. Yes, overhung motors and lots of power handling capability, but usually there is shorting rings for example. Obviously set of compromises is different for different application, but still the end goal is similar as in hifi, enabling good sound quality for the whole system within a context. It also looks like there is very little price added for the brand, and the cost is mostly due to high technology, durability. Proaudio application usually assumes digital signal processors, while hifi drivers need to cater for passive crossovers so there is also compromise in there if you will.
You could forget about the markets and segments and traditions and design a speaker system with ideal drivers, have open mind at the system design phase. At the end, select best fit drivers for the set of compromises you came up with, what ever those are.
While there might be a difference in sound between a high and low power handling drivers I'm not sure how big that is. I think you must try out and listen yourself because if different driver allows to design a system where there is now completely different set of compromises but higher distortion, than some other system with lower distortion, I'm not sure how you would evaluate which one is better. You could make high or low distortion systems with both. Very likely the set of compromises you chose determines the sound more than the distortion spec of a particular driver. If you just swapped a proaudio driver to a system that was designed with constraints/realm of hifi drivers, it is likely the sound isn't any better, perhaps its worse, but not because the driver is bad but because the system was designed for something else. Same would apply the other way around, try put hifi drivers to a pa box and go for a sound reinforcement gig and it's not going to end up well. Common thing is, use suitable drivers for application.
I'm just a hobbyist and have been working with small set of drivers I could say the drivers are important but in the end just one part of the playback system. The driver needs to fit the system you've designed and that's about it, design a system without driver constraint and then buy best drivers with the budget constraint. Common thing for pa and hifi is the observer, they want good sound at the distance and environment they are at, so it's mostly about the system design and choosing appropriate drivers for it. Doing other way around, buying drivers and designing a system around them is likely not going to give as good sound, it's backwards and falls into marketing department logic.
You could forget about the markets and segments and traditions and design a speaker system with ideal drivers, have open mind at the system design phase. At the end, select best fit drivers for the set of compromises you came up with, what ever those are.
While there might be a difference in sound between a high and low power handling drivers I'm not sure how big that is. I think you must try out and listen yourself because if different driver allows to design a system where there is now completely different set of compromises but higher distortion, than some other system with lower distortion, I'm not sure how you would evaluate which one is better. You could make high or low distortion systems with both. Very likely the set of compromises you chose determines the sound more than the distortion spec of a particular driver. If you just swapped a proaudio driver to a system that was designed with constraints/realm of hifi drivers, it is likely the sound isn't any better, perhaps its worse, but not because the driver is bad but because the system was designed for something else. Same would apply the other way around, try put hifi drivers to a pa box and go for a sound reinforcement gig and it's not going to end up well. Common thing is, use suitable drivers for application.
I'm just a hobbyist and have been working with small set of drivers I could say the drivers are important but in the end just one part of the playback system. The driver needs to fit the system you've designed and that's about it, design a system without driver constraint and then buy best drivers with the budget constraint. Common thing for pa and hifi is the observer, they want good sound at the distance and environment they are at, so it's mostly about the system design and choosing appropriate drivers for it. Doing other way around, buying drivers and designing a system around them is likely not going to give as good sound, it's backwards and falls into marketing department logic.
Last edited:
If the drivers are locked-in, what about bi-amping? The tubes could play high frequencies, and active crossovers could split the frequencies. A small class-d amplifier could play bass (I guess 100W still counts as small.)
That way, a Linkwitz Transform (or similar style of bass boost) could be applied without modulating or overdriving the tubes. Meanwhile, the woofer will have plenty in reserve for EQ.
To potentially get a smoother response in the woofer above say, 200Hz and up, I would include a passive coil (e.g: 1st order passive + 2nd order active, tuned for a combined 3rd order filter). That way, as frequencies increase, the coil impedance goes up and modulation effects in the speaker have a smaller effect on the overall current.
That way, a Linkwitz Transform (or similar style of bass boost) could be applied without modulating or overdriving the tubes. Meanwhile, the woofer will have plenty in reserve for EQ.
To potentially get a smoother response in the woofer above say, 200Hz and up, I would include a passive coil (e.g: 1st order passive + 2nd order active, tuned for a combined 3rd order filter). That way, as frequencies increase, the coil impedance goes up and modulation effects in the speaker have a smaller effect on the overall current.
Yeah, absolutely. Or, if the fleawatt amplifier is a must, then one just needs more sensitivity for lows which means bigger box and/or bigger drivers and so on, third way to the system if necessary. If it must be two way system for some reason then use 15" instead of 10" if it wasn't enough, big horn, more amplifier power, what ever seems most reasonable way to go to reach end goal, what ever that was.
The thing is, a system must be designed to work fine and to do that goal must be known and then make a system that successfully meets that. Wishing some performance because it feels nice emotionally, but doesn't actually work, doesn't actually work 😀 Don't be let down, there is no reason to be miserable due to how physics work, quite the opposite, it's all learning. I could say mistakes are mandatory for learning experience, sure way to success is to do lots of mistakes to learn a lot.
The thing is, a system must be designed to work fine and to do that goal must be known and then make a system that successfully meets that. Wishing some performance because it feels nice emotionally, but doesn't actually work, doesn't actually work 😀 Don't be let down, there is no reason to be miserable due to how physics work, quite the opposite, it's all learning. I could say mistakes are mandatory for learning experience, sure way to success is to do lots of mistakes to learn a lot.
Last edited:
That Radian doesn’t have a real impressive Vd. Also, there are subs in the equation. The OP doesn’t require high levels as per fleawatt amp. So a passive design, closed box is more than feasible.
Yeah the Vd is one of the big issues here. I'm reading with interest and appreciation what is being said here, but I'm not giving up yet since the journey and the struggle are all teaching me a lot, so even if I ditch this project at the last minute, it will not have been a waste.
Having said that, I want to reiterate my original goals, maybe I didn't do so very well at the beginning.
First, I chose the Radian 5210 because I'd seen it "successfully" used by a few people who seemed to be getting high praise. Then I actually spoke with one of those designers, who encouraged me to go for it, and said I should have no problem in an 80L enclosure. The retailer also said it was feasible. Not knowing what I do now, I jumped in. Also, the Radian compression drivers are well regarded in the pro audio world.
My goal was to make a pair of speakers with sensitivity of 94db min. specifically to work with a SET 45 amp, and being a can-do type without the budget or space to buy a set of LaScalas or big horn loaded refrigerators, I figured I could make something that would fit my needs. I love coaxials, having owned coax speakers in the past. For midfield, I've found they present in a really coherent way, and imaging can be spooky good. I wanted high sensitivity not only because of the flea watts I'm dealing with, but also because of their tendency to play well at lower levels. I can't always be listening to sleepy ECM recordings at night, sometimes I want to hear more dynamic stuff and my current system really doesn't come alive below 75db or so.
So I basically wanted something of the BLH sound on a budget. I have indelible imprints of having listened to big Klipsch and Altecs, and I loved the sound and was really taken at the times I heard these setups. Many of those old designs don't play very low, since excursion is the enemy of sensitivity, as I've become painfully aware! So if I can get a solid response to 80-100hz, it's a compromise that may be worth it if I can have the responsiveness of a sensitive and low mass driver. I listen to a lot of jazz and smaller ensemble stuff, so realism or the impression of it is a top priority.
I considered and had pre-ordered some full range drivers, but the compromises weren't to my liking, and the Vas of the ones I could afford was just too large to be practical for me. So I decided this was the way to go. I still think it's possible, and if I can just get to the point where the response looks like it is, I'll get a crossover together and see how it sounds in some crap enclosures before I commit to building nicer boxes. If it's a fail, I can sell the drivers off to some DJ guys and I'll start over.
So I guess I have a question for anyone who is doubting this is the right way to go, and I'm really interested in what people have to say (wish I'd asked this question 5 months ago): If you were given the design goals I've mentioned, how would you proceed, within a budget of $1500-$2000 USD?
Having said that, I want to reiterate my original goals, maybe I didn't do so very well at the beginning.
First, I chose the Radian 5210 because I'd seen it "successfully" used by a few people who seemed to be getting high praise. Then I actually spoke with one of those designers, who encouraged me to go for it, and said I should have no problem in an 80L enclosure. The retailer also said it was feasible. Not knowing what I do now, I jumped in. Also, the Radian compression drivers are well regarded in the pro audio world.
My goal was to make a pair of speakers with sensitivity of 94db min. specifically to work with a SET 45 amp, and being a can-do type without the budget or space to buy a set of LaScalas or big horn loaded refrigerators, I figured I could make something that would fit my needs. I love coaxials, having owned coax speakers in the past. For midfield, I've found they present in a really coherent way, and imaging can be spooky good. I wanted high sensitivity not only because of the flea watts I'm dealing with, but also because of their tendency to play well at lower levels. I can't always be listening to sleepy ECM recordings at night, sometimes I want to hear more dynamic stuff and my current system really doesn't come alive below 75db or so.
So I basically wanted something of the BLH sound on a budget. I have indelible imprints of having listened to big Klipsch and Altecs, and I loved the sound and was really taken at the times I heard these setups. Many of those old designs don't play very low, since excursion is the enemy of sensitivity, as I've become painfully aware! So if I can get a solid response to 80-100hz, it's a compromise that may be worth it if I can have the responsiveness of a sensitive and low mass driver. I listen to a lot of jazz and smaller ensemble stuff, so realism or the impression of it is a top priority.
I considered and had pre-ordered some full range drivers, but the compromises weren't to my liking, and the Vas of the ones I could afford was just too large to be practical for me. So I decided this was the way to go. I still think it's possible, and if I can just get to the point where the response looks like it is, I'll get a crossover together and see how it sounds in some crap enclosures before I commit to building nicer boxes. If it's a fail, I can sell the drivers off to some DJ guys and I'll start over.
So I guess I have a question for anyone who is doubting this is the right way to go, and I'm really interested in what people have to say (wish I'd asked this question 5 months ago): If you were given the design goals I've mentioned, how would you proceed, within a budget of $1500-$2000 USD?
TangBand have anything that suits your needs? https://www.parts-express.com/Tang-...tity=1&irclickid=SQu38OwsCxyPWNd0ClTlUzXFUkFR
General question, something I'm seeing with my measurements as well, your post #44, the port peak and the driver dip don't line up. Anyone know the reason behind this?
The enclosure obviously isn’t working as a normal vented enclosure, more like some kind of a MLTL. The port output also could be influenced by the first order mode inside the tall cabinet.
These cabinet shapes make for bad vented enclosures in the way that their behavior needs modeling with HornResp rather than with normal ‘box’ apps like WinISD.
Having said this, I haven’t got a clue what the internal shape has become after the volume reduction…
Didn't realize his enclosure was misshapen. My recent build was a bass cab with a JBL, pretty standard. It also exhibited the misalignment. Was discussing it with others, but didn't get anywhere with it. Then I saw this thread and his measurements and it reminded me of it. My example:
All I did was partition the existing enclosure about half way, and now the interior dimensions are 11.5"w x 12.5"d x 19"h, with very moderate damping behind the driver, and at the top and bottom. I'm going to play around with further reduction tomorrow and leave the ductless 3" ports as they are, which should raise the port resonance a bit. I think this is going to make the driver more happy.
I was very fond of a speaker that had 101 db 1w1m that I usually listened to 1 Vav or 1/8 watt. Soft orchestral passages. They were huge, sat on poles near the ceiling, had a 14" long Peavey logo on them, and attracted the attention of a burglar that carried them to the flea market for me. Paid $600 for the pair.
Low excursion, wide dispersion, low distortion were the signatures of the SP2-XT. I've since replace them with SP2(2004) which has an actual distortion chart in the spec sheet (at 5 w). Paid $400 for the pair, + $100 to drive 150 miles to pick them up.
Frequency response SP2 is +-3 db 54 hz to 17 khz, and that is enough bass for me. I do listen to piano & organ LP's that go down to 26 hz. Box tuning is 55 hz, they are ported, and I don't hear boominess or group delay. (?) I have them against a hard plaster wall which boosts bass 3 db below 54 hz. Xmax of the 15" woofer is 0.9 mm, Vas is 143 L, F0 is 51 hz. So low excursion, as you say, promotes low distortion. They are not boomy and I don't hear variable delay as people talk about. I do have some full rangers, based on Visaton bg17-8, in a 1 cf vented box - in the HDTV room on a 120 w/ch amp and I find the big SP2's much more realistic.
I've played 90 db dynamic range music in high school band, and I really enjoy a system that will do the whole 72 db dynamic range of a classical CD. So from 1/8 to 50 watts is about that: I have a 70 w/ch stereo amp. The full watts is for the cannon shot in 1812 overture. If I decide to play crowds of 300 outdoors I have a CS800s amp that will to go 260 w/ch for 8 ohms. (the cabs will play pink noise to 500 w). If I get popular, not likely since I am playing crowds of 7 people indoors these days.
I'm trying to duplicate SP2(2004) in ugly MDF with US assembled Eminence Deltapro-15A woofer and N314-T8 CD, so my ugly home built cabs will have zero pawn shop value. I don't have any room limitations that would encourage a bookshelf cabinet, except I would find something as big as 2 Danley SH-50 overwhelming my 14'x11' front wall, leaving no room for the front door. The SP2's straddle and tower over a hammond organ at one end of a 33' long room. Currently I have $320 in the woofers, $470 in the CD's, about $50 in crossover parts, $50 in sandel plywood +mdf, and $50 in a circular saw & sawblade. No shop table yet, no flat end CD horn yet. I plan to test outdoors against my cinder block garage wall, with a line level graphic equalizer to set the driver curves, (1 channel tweeter, 1 woofer) before I buy passive crossover parts to try to duplicate that response. No BSC required for cabs backed into a hard concrete wall.
So press on with what you have purchased, Getting the best out of what you have bought is the sign of an astute designer. You do have a sub to do the deep bass, I think you said. I'm interested to reading how you are doing.
Last edited:
That's one of the best reads I've had in months, your writing is very entertaining and a great, sober reality check.
Your post reminded me that the heritage of many or most of these high sensitivity horn speakers is soundly from the pro PA realm. I think even the original Klipschorn was designed for PA systems but I could be wrong.
You remind me of the phrase "perfect is the enemy of good", thanks for helping me re-center! It's easy to get lost in all these parameters, endless measurements, graphs, etc, and still end up with something that sounds very mediocre with a perfect response curve, while I'm quite confident I've heard systems that blow me away that would look pretty ugly on the page but just have that facsimile of realism that tickles my hi hat. Not saying I shouldn't be trying to optimize what I have like you say, just that I might need to relax a bit and live with some imperfection. Thank you.
Your post reminded me that the heritage of many or most of these high sensitivity horn speakers is soundly from the pro PA realm. I think even the original Klipschorn was designed for PA systems but I could be wrong.
You remind me of the phrase "perfect is the enemy of good", thanks for helping me re-center! It's easy to get lost in all these parameters, endless measurements, graphs, etc, and still end up with something that sounds very mediocre with a perfect response curve, while I'm quite confident I've heard systems that blow me away that would look pretty ugly on the page but just have that facsimile of realism that tickles my hi hat. Not saying I shouldn't be trying to optimize what I have like you say, just that I might need to relax a bit and live with some imperfection. Thank you.
OK so maybe this thread is getting a bit ragged but I'm still at it! Today what I did was to fill the existing 42L enclosure with bricks, and measuring at approx 35L and 31L, to see the direction things would go in a smaller enclosure (Vas is only 25L or so). What really surprised me was how subtle the changes in nearfield measurements were, and for the port too more or less. I expected the port resonance to be a bit higher as the box got smaller, but the main thing I see is how the 42L alignment seems better in both regards. So below I'm posting shots of NF sweeps at 42, 35, and 31, and port sweeps at 42 and 31. I left 35L out because it's pretty much right between the two. The summed plot at 42 looks really good to me, I don't understand why I shouldn't just proceed at this volume?
