augerpro said:
Give this a try, run both drivers really hard equally and measure the rise in DCR of both. I have never had a chance to play with these new neo pro woofers, but power dissipation has to be an issue.
Hi Guys,
Well, I have the baffle set up and I'm about to run the near/far field responses. I measured the impedance of the TD10M, and it looks pretty good. I have attached it below.
Brandon:
If you take a look at the impedance graph for the 10NW64, you'll see two traces, each for a different driver. I always test two samples fully to make sure I don't post something that might be a fault with a single driver. They're quite consistent with one another, so the bump is inherent to the driver. I measured these in the test baffle, with nothing anywhere close to the driver. I also did a second test with the driver held in free air, and the bump was still there.
As for the motor linearity test, I'm not sure how I could reliably hold the driver at the extremes of it's excursion and still get a reliable measurement. How would I go about doing that? I suppose a DC offset could work, but I'm not sure how happy the driver would be with that.
ShinOBIWAN:
Thanks for the comments Shin, your projects are legendary in my books, and in a way they're an inspiration for this project. Your build threads have provided me with a great deal of fun reading!
Both you and Brandon are probably correct about the chaffing noise being a non-issue with the 10NW64. I wouldn't hold it against the driver in any way, but it is something that kinda surprised me when I heard it. It's so far down in level to the actual driver output that you'd never hear it under normal use. These certainly aren't subwoofers, so to get that kind of excursion you'd have to be playing at ear bleeding levels.
I'm hoping to get my hands on the latest version of SoundEasy (a Christmas present I hope) so I'll post some distortion measurements for all these drivers when I get the chance. I'm not sure how to measure distortion with Speaker Workshop, if it's even capable of doing it. Once I measure the ribbon distortion I'll be in a better position to make the call on the crossover point.
Nick:
I've never really agreed with the whole "Ceramic has more thermal mass and therefor less power compression" train of thought. I would think that better air cooling around the coil would be a greater advantage. Both motors under heavy use in the same cabinet would eventually reach a thermal equilibrium, at which point I think the better air cooling would have an inherent advantage.
I live in a condo, so I don't think I could get away with playing 120dB pink noise for three hours without finding myself homeless.
I consider this a non issue for myself anyhow since I'll probably seldom ever play these at that high a level. I'll be driving these with an F5 amp, which is only capable of 25 watts/channel. Life will be a walk in the park for whichever driver I choose.
If you can suggest a way to thermally stress these without making to much noise I'll certainly give it a try.
Back to the measuring!
Cheers,
Owen
Well, I have the baffle set up and I'm about to run the near/far field responses. I measured the impedance of the TD10M, and it looks pretty good. I have attached it below.
Brandon:
If you take a look at the impedance graph for the 10NW64, you'll see two traces, each for a different driver. I always test two samples fully to make sure I don't post something that might be a fault with a single driver. They're quite consistent with one another, so the bump is inherent to the driver. I measured these in the test baffle, with nothing anywhere close to the driver. I also did a second test with the driver held in free air, and the bump was still there.
As for the motor linearity test, I'm not sure how I could reliably hold the driver at the extremes of it's excursion and still get a reliable measurement. How would I go about doing that? I suppose a DC offset could work, but I'm not sure how happy the driver would be with that.
ShinOBIWAN:
Thanks for the comments Shin, your projects are legendary in my books, and in a way they're an inspiration for this project. Your build threads have provided me with a great deal of fun reading!
Both you and Brandon are probably correct about the chaffing noise being a non-issue with the 10NW64. I wouldn't hold it against the driver in any way, but it is something that kinda surprised me when I heard it. It's so far down in level to the actual driver output that you'd never hear it under normal use. These certainly aren't subwoofers, so to get that kind of excursion you'd have to be playing at ear bleeding levels.
I'm hoping to get my hands on the latest version of SoundEasy (a Christmas present I hope) so I'll post some distortion measurements for all these drivers when I get the chance. I'm not sure how to measure distortion with Speaker Workshop, if it's even capable of doing it. Once I measure the ribbon distortion I'll be in a better position to make the call on the crossover point.
Nick:
I've never really agreed with the whole "Ceramic has more thermal mass and therefor less power compression" train of thought. I would think that better air cooling around the coil would be a greater advantage. Both motors under heavy use in the same cabinet would eventually reach a thermal equilibrium, at which point I think the better air cooling would have an inherent advantage.
I live in a condo, so I don't think I could get away with playing 120dB pink noise for three hours without finding myself homeless.
I consider this a non issue for myself anyhow since I'll probably seldom ever play these at that high a level. I'll be driving these with an F5 amp, which is only capable of 25 watts/channel. Life will be a walk in the park for whichever driver I choose.
If you can suggest a way to thermally stress these without making to much noise I'll certainly give it a try.
Back to the measuring!
Cheers,
Owen
Hi Blue934,
Those are a pair of Lynn Olsen's Ariels. The veneer is Brazilian Rosewood, which was old-stock from a place called Oak Wood Veneers in Michigan. I'm guessing (hoping) it was imported sometime before Brazilian Rosewood became endangered.
Its bloody expensive but worth every penny.
Cheers,
Owen
Those are a pair of Lynn Olsen's Ariels. The veneer is Brazilian Rosewood, which was old-stock from a place called Oak Wood Veneers in Michigan. I'm guessing (hoping) it was imported sometime before Brazilian Rosewood became endangered.
Its bloody expensive but worth every penny.
Cheers,
Owen
opc said:
"Ceramic has more thermal mass and therefor less power compression" - First I have heard of this and I agree that it is 100% bogus for a woofer. A motor will never get to equilibrium as to do that is about 200-400F across the entire structure. By that time speaker cones either melt or catch on fire. Rather the pole piece and top plate are always localized hot spots.
If you pound those TD drivers hard you will see that the center phase plug is a serious heat exchanger. The entire center pole/voice coil design of the TD is meant to pull all heat as quickly as possible to the air in front of the woofer.
You could try the power test with the drivers face to face and out of phase to lower some of the audible output.
opc said:
Thanks Owen but its now your turn 🙂
Speaker workshop can measure distortion but stick with you plan of Soundeasy, its a better program all round.
In my haste I forgot to attach this in my first reply. Its a scan showing the Neopro 5i distortion as tested by German magazine Hobby Hifi (March 07 issue):
An externally hosted image should be here but it was not working when we last tested it.
Look forward to seeing how your design develops. Keep us updated.
Hi Guys,
Well, I only got around to two nearfield responses before it was bedtime for my wife, so I'll have to put off the 1m and off axis measurements until tomorrow.
As for the nearfield, here are the test conditions and limitations:
Actual SPL level means nothing. The drivers were measured at different distances (due to the TD10M phase plug) so don't compare one to the other. I don't have absolute levels yet because my SPL meter is at work. I'll have those tomorrow.
The mic was 0.5 cm away from the center of the dustcap with the 10NW64 woofer, and 0.5cm away from the center of the phase plug with the TD10M.
Both drivers were mounted in the same location on a solid 2' by 4' baffle. Microphone is a Behringer ECM8000, sound card is M-audio. No gating, both graphs smoothed to 1/6th octave.
TD10M driver is in BLUE
10NW64 driver is in RED
Notice the really odd dip in the 10NW64 response right at 200Hz. Both these woofers were measured in the same baffle, so that actually is the driver's response, which I thought was a little odd. I measured both drivers I have here, and they're identical in that region. The TD10M looks very smooth, and as of right now it looks like it's a shoe-in for this project.
ShinOBIWAN:
Thanks for the distortion plots! I've been looking around for something like that. It's a little worse than I was hoping for, but I'll wait and measure my pair when I get SoundEasy before jumping to any conclusions.
If they are like that, then I may very well have to cross closer to 2500Hz, or at least use a steeper roll-off at 2000Hz to avoid that scary rise at 1.6K.
It's odd... I didn't see anything at all in the frequency response or the impedance of the NeoPro that would suggest a resonance or problem at 1.6K. I know it's not always obvious, but with a peak that large, I would expect to see something. We'll just have to wait and see.
Nick:
I certainly agree with the phase plug being a good heatsink, and I've witnessed it first hand while breaking in a pair of Usher drivers a while back. I got a little ambitious with the level, and when I came back after a few hours the phase plug was almost too hot to touch. The magnet on the other hand wasn't anywhere neat as hot. Volt does that radial chassis design which takes advantage of that phenomenon, and I think Eminence makes a Kilomax driver that appears to have some sort of lame heatsink thing as a phase plug.
Like I said before, it will never be a real issue for me as I'm not really a super loud music kinda guy. It's good to know I could play these for hours at 115dB if I ever throw some sort of high end stereo party!
Cheers,
Owen
Well, I only got around to two nearfield responses before it was bedtime for my wife, so I'll have to put off the 1m and off axis measurements until tomorrow.
As for the nearfield, here are the test conditions and limitations:
Actual SPL level means nothing. The drivers were measured at different distances (due to the TD10M phase plug) so don't compare one to the other. I don't have absolute levels yet because my SPL meter is at work. I'll have those tomorrow.
The mic was 0.5 cm away from the center of the dustcap with the 10NW64 woofer, and 0.5cm away from the center of the phase plug with the TD10M.
Both drivers were mounted in the same location on a solid 2' by 4' baffle. Microphone is a Behringer ECM8000, sound card is M-audio. No gating, both graphs smoothed to 1/6th octave.
TD10M driver is in BLUE
10NW64 driver is in RED
Notice the really odd dip in the 10NW64 response right at 200Hz. Both these woofers were measured in the same baffle, so that actually is the driver's response, which I thought was a little odd. I measured both drivers I have here, and they're identical in that region. The TD10M looks very smooth, and as of right now it looks like it's a shoe-in for this project.
ShinOBIWAN:
Thanks for the distortion plots! I've been looking around for something like that. It's a little worse than I was hoping for, but I'll wait and measure my pair when I get SoundEasy before jumping to any conclusions.
If they are like that, then I may very well have to cross closer to 2500Hz, or at least use a steeper roll-off at 2000Hz to avoid that scary rise at 1.6K.
It's odd... I didn't see anything at all in the frequency response or the impedance of the NeoPro that would suggest a resonance or problem at 1.6K. I know it's not always obvious, but with a peak that large, I would expect to see something. We'll just have to wait and see.
Nick:
I certainly agree with the phase plug being a good heatsink, and I've witnessed it first hand while breaking in a pair of Usher drivers a while back. I got a little ambitious with the level, and when I came back after a few hours the phase plug was almost too hot to touch. The magnet on the other hand wasn't anywhere neat as hot. Volt does that radial chassis design which takes advantage of that phenomenon, and I think Eminence makes a Kilomax driver that appears to have some sort of lame heatsink thing as a phase plug.
Like I said before, it will never be a real issue for me as I'm not really a super loud music kinda guy. It's good to know I could play these for hours at 115dB if I ever throw some sort of high end stereo party!
Cheers,
Owen
opc said:
There is some good news. If you cross at 2Khz it'll already be down some 6dB with a filter such as LR etc. It looks to me like a resonance of some kind since it affects 2nd and 3rd order in a very similar fashion. You'll have to see just how objectionable this is.
From your own measurements you may even find that the anomaly was a batch issue or that an unannounced revision has fixed/lessened the problem.
As always the only data that's reliable is that taken of your own drivers.
Email me and I'll send PDF with impedance and FR data from the same magazine. You can compare to what you already and see if some truth can be uncovered.
Email sent.
If you want to see what I measured, take a look back to the first few posts for the on/off axis FR and impedance of the ribbon.
Cheers,
Owen
If you want to see what I measured, take a look back to the first few posts for the on/off axis FR and impedance of the ribbon.
Cheers,
Owen
PDF sent.
The data doesn't match but try taking more data using software from the good suggestion by sqlkev.
Its free and accurate:
http://www.fesb.hr/~mateljan/arta/
The data doesn't match but try taking more data using software from the good suggestion by sqlkev.
Its free and accurate:
http://www.fesb.hr/~mateljan/arta/
Something weird is going on with that B&C driver, I wonder if you removed the dustcap if that 200Hz problem would go away. You would lose the forced cooling I imagine though.
Hi Guys,
Well, I spent the night reading up on ARTA and re-calibrating my setup to re-measure the B&C driver.
I have tested several passive components as well as re-tested the B&C and I'm very confident the impedance/response graphs are accurate. This applies to all the impedance measurements.
On that note, I took a less drastic approach to Nick's suggestion of cutting off the dust cap, and instead just plugged the vent holes on the rear of the magnet. The resulting graphs are posted in .pdf below.
The cooling system is indeed responsible for the strange behavior seen at 200Hz, and to be honest, it's pretty disappointing. With the vents sealed, the Fs obviously goes up, but the resonance at 200Hz is gone. The effect on the FR graph is even more pronounced. The driver now looks well behaved with a normal roll-off.
I put the response with the holes plugged in BLUE and with the exact same setup, I removed the tape and took a second measurement which is in RED with the vents open. The exact same technique was used to measure the impedance. Same colors apply.
I'll be setting up for the farfield responses tomorrow, as tonight was just a little too rushed. I'll also hopefully get Arta up and running for some measurements.
Cheers,
Owen
Well, I spent the night reading up on ARTA and re-calibrating my setup to re-measure the B&C driver.
I have tested several passive components as well as re-tested the B&C and I'm very confident the impedance/response graphs are accurate. This applies to all the impedance measurements.
On that note, I took a less drastic approach to Nick's suggestion of cutting off the dust cap, and instead just plugged the vent holes on the rear of the magnet. The resulting graphs are posted in .pdf below.
The cooling system is indeed responsible for the strange behavior seen at 200Hz, and to be honest, it's pretty disappointing. With the vents sealed, the Fs obviously goes up, but the resonance at 200Hz is gone. The effect on the FR graph is even more pronounced. The driver now looks well behaved with a normal roll-off.
I put the response with the holes plugged in BLUE and with the exact same setup, I removed the tape and took a second measurement which is in RED with the vents open. The exact same technique was used to measure the impedance. Same colors apply.
I'll be setting up for the farfield responses tomorrow, as tonight was just a little too rushed. I'll also hopefully get Arta up and running for some measurements.
Cheers,
Owen
opc said:
WTF?
Why would they even include those vent holes considering the damage they do. Did they even bother to try a driver with and without them.
Maybe they think the VC cooling is worth the downside, afterall they are PA drivers that often get driven hard so cooling helps here. But for us with less demanding tasks, it doesn't look good.
Stick with the Lambda I say.
Hey ShinOBIWAN,
It's pretty bad eh? I was taking a better look at the magnet structure, and unfortunately it looks like the back cover with the vent holes is an integral part of the magnetic circuit, so I can't just break it off and run the driver without them.
I suppose they're just not the right driver for the application. This is one of those pro drivers that just doesn't belong in hi-fi applications.
I would still like to try one or two other 10" drivers, especially the Faital 10HP1020 or the Volt BM2500.5. The cost is starting to get prohibitive though, so it might just be better to stick with what I've got now. The TD10M's would likely still come out on top, but I do like to have options.
With the Fs being lower than expected on the AE drivers, I might just go back to the original dual TL design and build a box. I guess I have some big decisions to make over the next few days.
I wish companies had a return policy on drivers. It would help to keep them honest about measurements, and it would save me having to spend $3000 testing random 10" drivers until I find one I actually like.
Cheers,
Owen
It's pretty bad eh? I was taking a better look at the magnet structure, and unfortunately it looks like the back cover with the vent holes is an integral part of the magnetic circuit, so I can't just break it off and run the driver without them.
I suppose they're just not the right driver for the application. This is one of those pro drivers that just doesn't belong in hi-fi applications.
I would still like to try one or two other 10" drivers, especially the Faital 10HP1020 or the Volt BM2500.5. The cost is starting to get prohibitive though, so it might just be better to stick with what I've got now. The TD10M's would likely still come out on top, but I do like to have options.
With the Fs being lower than expected on the AE drivers, I might just go back to the original dual TL design and build a box. I guess I have some big decisions to make over the next few days.
I wish companies had a return policy on drivers. It would help to keep them honest about measurements, and it would save me having to spend $3000 testing random 10" drivers until I find one I actually like.
Cheers,
Owen
Owen,
I know what you mean about a cooling off period with drivers. If your slightly dishonest then in the UK there's the "distance selling regulation act". What this means is you order something online, over the phone or by any means other than in person and you get 14 days to return for a full refund(minus P&P) if the product wasn't what you expected. Its kinda useless though because it also specifies that the item must not have been used and the packaging as new.
I've bought stuff tried it out and sent it back but I was careful about handling and packed up exactly as I received it. Always got a full refund but haven't tried it with drivers.
Besides this you shouldn't have to enter a legally grey area to try something for suitability and ascertain legitimacy of claimed specification/performance. I suppose the opportunity is there for abuse though.
I've had the Volt B2500.1 but this is going back to when I started DIY in 2004. I was impressed back then but can't help feel that its been surpassed as I've heard more and more drivers. On pure conjecture I'd say if I heard them today I'd likely rate them as above average to good. Volt are overpriced for performance on offer, they don't measure as class leaders either. I wouldn't rush out to get a pair considering you have what is regarded as a very good 10" in the Lambda.
I know what you mean about a cooling off period with drivers. If your slightly dishonest then in the UK there's the "distance selling regulation act". What this means is you order something online, over the phone or by any means other than in person and you get 14 days to return for a full refund(minus P&P) if the product wasn't what you expected. Its kinda useless though because it also specifies that the item must not have been used and the packaging as new.
I've bought stuff tried it out and sent it back but I was careful about handling and packed up exactly as I received it. Always got a full refund but haven't tried it with drivers.
Besides this you shouldn't have to enter a legally grey area to try something for suitability and ascertain legitimacy of claimed specification/performance. I suppose the opportunity is there for abuse though.
I've had the Volt B2500.1 but this is going back to when I started DIY in 2004. I was impressed back then but can't help feel that its been surpassed as I've heard more and more drivers. On pure conjecture I'd say if I heard them today I'd likely rate them as above average to good. Volt are overpriced for performance on offer, they don't measure as class leaders either. I wouldn't rush out to get a pair considering you have what is regarded as a very good 10" in the Lambda.
opc said:
This is one of the biggest problems, actual T/S testing of speakers.
Make the room warmer, Fs goes down.
More humidity, Fs goes down.
Work the driver hard for a few minutes then test, Fs is down temporarily do to suspension hysteresis. Wait a few more minutes, Fs is creeping back up.
The only thing actually changing in all 3 scenarios is the Cms (and Rms slightly), but it affects all the other parameters. Cms can move that much in a driver from day to day and even minute to minute.
Rather you should compare Mms, Bl, and Re as they are the only real parameters you can QC so closely unless you are in a controlled environment that matches that the manufacturer used to make their tests.
The beauty of a speaker cabinet is the box volume air spring goes in series with the driver Cms (suspension spring) and usually swamps its effects creating a new Fs for the driver that is not so temperature/humidity/hysteresis sensitive. The smaller the Vb is in relation to the Vas the more this happens.
Hi Nick,
Sorry, that comment wasn't intended for the TD10M drivers, which so far I'm very happy with.
It was more of a blanket statement to cover little issues that seem to crop up far too often. I know my testing will be different than your testing, and different from the testing of others, but a problem like the one with the 10NW64 should have been apparent from the FR and IMP graphs on the datasheet, but they don't show up there.
That means they either fudged those graphs, or measured something quite different from the driver I got. I suppose the use of a small chamber during testing could have had that effect, but that's not really industry standard testing. Humidity and room temp didn't cause what was going on at 200Hz.
Cheers,
Owen
Sorry, that comment wasn't intended for the TD10M drivers, which so far I'm very happy with.
It was more of a blanket statement to cover little issues that seem to crop up far too often. I know my testing will be different than your testing, and different from the testing of others, but a problem like the one with the 10NW64 should have been apparent from the FR and IMP graphs on the datasheet, but they don't show up there.
That means they either fudged those graphs, or measured something quite different from the driver I got. I suppose the use of a small chamber during testing could have had that effect, but that's not really industry standard testing. Humidity and room temp didn't cause what was going on at 200Hz.
Cheers,
Owen
- Status
- Not open for further replies.