Assorted driver tests

[chrismercurio]

Well,

I think it is a deal. Perhaps not quite like a $20-30 Seas or Vifa tweeter is a deal, but a fabulous measuring tweeter for the price. For my application it is also shielded which is hard to find these days.

AugerPro will send mine out next week, and I have a pair of TI-100's arriving friday. Does anyone know how to get in touch with Zaph? I'm wondering if he would like to measure them?

[Jed]

I think it is a deal. Perhaps not quite like a $20-30 Seas or Vifa tweeter is a deal, but a fabulous measuring tweeter for the price. For my application it is also shielded which is hard to find these days.

AugerPro will send mine out next week, and I have a pair of TI-100's arriving friday. Does anyone know how to get in touch with Zaph? I'm wondering if he would like to measure them?

Chris, you saw the scan I had up of the Ti100 at audiocircle, right? Distortion is very low for that driver too.

I did end up picking up a pair of KE25s for a project. :B

As for contacting Zaph, just post a message on one of the forums and he'll respond when he gets a chance.

[chrismercurio]

Yes

This is one of the things that attracted me to the TI-100. That, and hearing it as part of the Sonics Passion speaker. The reports from the DIY Iowa event were also encouraging.

[JonMarsh]

Well, this week I ordered a pair of the KE25 to test, may be an upgrade for my M8ta system, if I convert it to a three way.

[ColoradoTom]

Well, this week I ordered a pair of the KE25 to test, may be an upgrade for my M8ta system, if I convert it to a three way.

Jon,

Does this speaker have possibilities for the three-way Isis like design as well?? From what I'm reading it seems to be a possible Accuton replacement.

Also - enjoyed meeting you last week at ThomasW's!

Tom

[JonMarsh]

It could, Tom, a lot depends on the distortion performance measured, what it does a above 20 kHz, etc. I don't think it will match up to my C30N's below 3 kHz, but it looks pretty promising for a 3 kHz and above three way crossover, at a very moderate price. But with what mid is 3 kHz practical?

Basically, I just want to confirm that it will do what other folks have measured and reported.

I'd have to see what kind of filter performance I can come up with for the C89, and where I think the optimum crossover point it.

Looking at the distortion spectra for the C90, where you get a rising characteristic, this defines your upper crossover point more than just the simple on axis frequency response, though that has it's own clues, too.



From this, I'd want to be rolling off the C90 by 2 kHz. The wavelength issues for a driver that large in diameter also dictate 2 kHz or less for the nominal crossover, I think.

Here's HD4 at 105 dB output... same pattern vs frequency. I believe getting the best sound out of these drivers means attenuating the region where distortion starts to rise sharply.



In the Isiris, these will be used up to a crossover point of ~1.6 kHz.

The C89 may be a skosh smoother in frequency response, but it shows a similar rising HD characteristic.



Curiously, the HD3 doesn't rise as much, but has a definite plateau....



HD4 also suggests staying away from the area above 2 kHz.



The 30mm Accuton tweeters don't go very high, but distortion down to 1 kHz is quite low. That gives more flexibility for crossover choices.





Looking at the C23, it goes lower than the KE25, and higher, and seems to have lower distortion- but then it costs 2X or more.

I'm considering the KE25 for the three way M8ta with RS52 because of the highish crossover point possible, and the fit in the existing cabinet.

~jon

[Jed]

Well, this week I ordered a pair of the KE25 to test, may be an upgrade for my M8ta system, if I convert it to a three way.

:T Looking forward to seeing your results, Jon. I have a hunch this might be "it" for a match with the C79s.

I also picked up a pair of RS52 and the New SEAS ER18s with the money I saved instead of getting the C13s.

[chrismercurio]

Jon,

I'm going to send the Ti-100's to Zaph for testing. Do you want to test them also as you are local to me? I'm in Santa Clara and work in the East Bay.

C

[augerpro]

Jon,

I'm going to send the Ti-100's to Zaph for testing. Do you want to test them also as you are local to me? I'm in Santa Clara and work in the East Bay.

C

Maybe Zaph could measure both of your KE25s also? I'd like to see how the one I sold you is compared to another.

[TacoD]

My experience with Visaton is that they offer good series consistence. Hopefully the Chinese production is still as good as the German production.

[augerpro]

Seas H1461 and TangBand W6-1139SI to be tested soon.

http://www.madisound.com/catalog/product_info.php?manufacturers_id=149&products_id=8178

http://www.partsexpress.com/pe/pshowdetl.cfm?&Partnumber=264-832

What a night :M Just got back home from family reunion stuff and started packing stuff I've sold lately to be shipped. First I discovered that getting my XO out of my Bose Buster project has proven quite a task. Gorilla Glue is tough...PoorBoyMike give me a day or two to get them out and shipped. Everything else is already boxed though.

Then I'm breaking in the little TangBand sub using some bass heavy music and realize I left my brand new jig box setup for impedence measurements. So a lot of juice was running through a 10 Ohm resistor...for hours...the box is kinda soft where the resistor is and a couple switches are hot as hell. I have some new resistors coming but I wonder how the switches are? You could only touch them for about a second. I don't know how much current was going through that 12watt Mills resistor but it's blackish now instead of blue :E My big fear is I left my probe from soundcard tapped into the line and may have damaged it. I work the next 5 days so all these problems wont' get dealt with for a bit.

On a better note I reran some tests for the KE25 just to make sure the baffle/driver interface wasn't causing a buzz between 1k-2k. I made some modifications but the results are basically the same. In fact with 1.5k tone several higher harmonics are very audible. Unless this is an out of spec tweet you guys interested in using it must really cross it high. Chris the tweeter will ship out tomorrow.

[Jed]

Unless this is an out of spec tweet you guys interested in using it must really cross it high. Chris the tweeter will ship out tomorrow.

I plan on using it around 3K on up and that should be very good. Also, most speakers I've seen with it in a system use it at 2.5K and above, so I should be safe at LR4 @ 3k.

Jed

[augerpro]

I plan on using it around 3K on up and that should be very good. Also, most speakers I've seen with it in a system use it at 2.5K and above, so I should be safe at LR4 @ 3k.

Jed

Cool. Crossed to an Accuton? Are you going to use a waveguide? I'd like to see the results with one...Geddes posted some over at diyaudio concerning the equation for an OS waveguide and the proper waveguide/driver flange interface dimensions that I'd like to try on some tweeters.

[Jed]

Cool. Crossed to an Accuton? Are you going to use a waveguide? I'd like to see the results with one...Geddes posted some over at diyaudio concerning the equation for an OS waveguide and the proper waveguide/driver flange interface dimensions that I'd like to try on some tweeters.

I'm going to try it with my Accuton C79s. Should be a nice match- in theory at least.

[Jed]

Distortion of the SEAS ER18RNX with magnet shield installed. Pretty darn clean.

Image not available


Dayton RS52 Distortion performance.

Image not available

All of the above about 1 watt with mic 5" from the center of the driver.

[Dennis H]

Curious measurements, Jed. In all of the ones you've posted (I think), the 3rd harmonic is higher than the 2nd. I haven't seen that in other people's measurements. Is it just the drivers or something about the measurement method?

[augerpro]

I decided to check out the TangBand W4-1320SD. Since the W4-1337 is getting some notice as a fullrange driver I wonder if the bamboo cone version may not better in this regard. No cone breakup so need for shaping which is turnoff to new people wanting to build their first speaker. Besides even if the breakup is brought down in amplitude I wonder if the sound is more objectionable because it is afterall a metal cone breakup, compared to paper which is less pistonic perhaps but the breakup is relatively mild? Also the description and Z plot indicate to me that the motor is identical. And higher sensitivity, always a plus in single driver FR designs.

[augerpro]

Jed can you put those in dB scale? I really have a hard time understanding them at a glance since I have to look at the % and then calculate in my head how many dB down that is. That RE18RNX looks incredibly close to Zaph, especially considering the different software.

[Jed]

Curious measurements, Jed. In all of the ones you've posted (I think), the 3rd harmonic is higher than the 2nd. I haven't seen that in other people's measurements. Is it just the drivers or something about the measurement method?

I noticed the same thing, though it seems the drivers I have on hand measure that way. Note, the visaton MHT12 exhibits more 2nd order (attached), which is consistent with other measurements I've seen, and the ER18RNX looks very similar to Zaphs-maybe a bit less 2nd order than his. According to Accuton, the C79 has more odd order distortion, so my measurement is consistent with theirs.

That all said, I'm pretty new at this and I'll try to get some other graphs up and running. It appears most like the DB scale better than a percentage graph.

Edit: Another thing is my measurements are with the drivers mounted in a box, not IB like Zaphs, so that can change the results as well.

[Dennis H]

I like the % scale better. Keeps me from having to convert dB to % in my head. Each division is 20dB so it's really no big deal either way.

[Jed]

Took some more tests- note the distortion below 100HZ % has to be calculated from the declining SPL. Dayton Rs52 at about 1watt

Image not available

SEAS ER18RNX

Image not available

[augerpro]

Just found the SEAS tweeter on the front porch!! Hopefully the TB W4-1320 will be here Friday and get tests done over the weekend.

[cjd]

Geeze, that RS52 is, for lack of any other way of describing it, ruler flat! Nice.

[Jim Holtz]

If only it had a smaller faceplate, it'd be perfect.

Jim

[Jed]

Geeze, that RS52 is, for lack of any other way of describing it, ruler flat! Nice.

Yeah, I'm getting worried it might outperform my Accuton C79s at 1/10th the cost. 8O I know, both are good, also the C79 can go much lower, but I AM looking forward to seeing how far I can go with each.

Jed

[mazurek]

I am exploring right now the variations within drivers and setups that are changing my performance level. I have had poorly fixed things around the room vibrate and contribute to HD, room nulls contribute to far-field HD, levels seem to make a big difference as well.

At comfortable low levels, nearly all distortion products on my SS18W4531G00 are very low, but when I get louder levels, the second order distortion elevates. Here is this test, I measured 93dB/meter, but the mic was at about 6inches (properly adjusted not to clip). The higher order distortion is good, but the second order is surprisingly high. Maybe this is just a paper cone phenomena. Jed, can you do some testing with ~95dB/m or higher with your drivers. Also, Jon when you say 105dB, is it per meter, that is really loud, at that level I would need to be more careful to set the bottom range of my sweep so it doesn't explode the driver, and bother the neighbors.

Thanks,
Lee

[JonMarsh]

Generally, i think you'll find the cone contributes to non-linear distortion (FR and energy storage) but it's the motor that determines HD harmonic emphasis. Certain manufacturers have a somewhat "signature" characteristic in this regard. Auccuton, for example, tends to have very low HD2, and HD3 is the dominant component (though still fairly low in most drivers). That just means the BL product is symmetrical with respect to displacement in the gap. (in and out motion of the cone/VC).

[Dennis H]

One problem with testing really loud is the mic can distort so you don't know if you're seeing driver distortion or mic distortion.

[Jed]

I am exploring right now the variations within drivers and setups that are changing my performance level. I have had poorly fixed things around the room vibrate and contribute to HD, room nulls contribute to far-field HD, levels seem to make a big difference as well.

Lee

Lee,

I'm working on a testing Lab in the basement to minimize these issues. I can take some tests at higher DB sometime, but I'll have to get ear protection, because even at the levels I had it set at, my ears were in pain even if I was in another room.

[capslock]

Generally, i think you'll find the cone contributes to non-linear distortion (FR and energy storage) but it's the motor that determines HD harmonic emphasis. Certain manufacturers have a somewhat "signature" characteristic in this regard. Auccuton, for example, tends to have very low HD2, and HD3 is the dominant component (though still fairly low in most drivers). That just means the BL product is symmetrical with respect to displacement in the gap. (in and out motion of the cone/VC).

The exception being second harmonic in paper cones (and to a lesser degree plastic / plastified fiber stuff). Just go to the Seas Excel vintage data sheet section. They had a couple of drivers that shared a common motor but were available with mag, glass fiber and paper cones.

[JonMarsh]

Interesting point Eric. I sit corrected or amended, though I think we both agree on the first order determinants. :W


~Jon

[capslock]

now what could those be? price? street cred?

[jkrutke]

I'm going to send the Ti-100's to Zaph for testing.

It must be titanium day in the house of Zaph. Chris' Ti-100's arrived today. Nice looking build quality. I hope they perform. Like the TB titanium, they have super fragile cones. Very thin and light. I should get to test them this weekend.

Seas H1461 and TangBand W6-1139SI to be tested soon.

Then, a pair of Seas 27TTFNC/GW (H1461) 1" Titanium domes showed up on my porch too. They are a little pricey and I held off picking up a pair fear that they aren't a notable improvement over the 27TBFCG. I guess we'll see. With your tests, we'll have a better look at consistency.


edit: I called the Seas a 27TDFNC/GW but that's the soft dome version. Madisound made the same mistake.

[TacoD]

Then, a pair of Seas 27TDFNC/GW (H1461) 1" Titanium domes showed up on my porch too. They are a little pricey and I held off picking up a pair fear that they aren't a notable improvement over the 27TBFCG. I guess we'll see. With your tests, we'll have a better look at consistency.

The new Seas neodym prestige tweeters sound different than the "old" ones. In my opinion these sound more natural and less harsh. The measurements in Voicecoil show little difference between the soft and hard dome version.

[Jed]

It must be titanium day in the house of Zaph. .

Looking forward to your results. The Ti100 did very well in K&T magazine.

[jkrutke]

The new Seas neodym prestige tweeters sound different than the "old" ones. In my opinion these sound more natural and less harsh. The measurements in Voicecoil show little difference between the soft and hard dome version.

Voice coil didn't post any non-linear distortion measurements for these tweeters. Also note that their response curves were done on a 17" x 8" baffle without noting the driver location (offset near top? dead center?) and without noting if it was a flat baffle or if it had depth like a normal enclosure. The measurements were at 1m, so the baffle has a huge effect. Those are major testing shortcomings making the results near useless.

Well, their impedance test is useful, showing us a little blip at 3kHz. With all my latest tests, I've been doing super-zoomed in impedance curves and just dumping them in my directory of measurements for personal use. I'll probably round them all up sometime and post what I have.

I do like Voice Coil's Klippel tests on their woofers, but other than that I'd rather see things done like Hobby HiFi or Klang & Ton. I also like Mark K's normal suite of tests. And of course Augerpro's are useful too.

[augerpro]

It must be titanium day in the house of Zaph. Chris' Ti-100's arrived today. Nice looking build quality. I hope they perform. Like the TB titanium, they have super fragile cones. Very thin and light. I should get to test them this weekend.



Then, a pair of Seas 27TTFNC/GW (H1461) 1" Titanium domes showed up on my porch too. They are a little pricey and I held off picking up a pair fear that they aren't a notable improvement over the 27TBFCG. I guess we'll see. With your tests, we'll have a better look at consistency.


edit: I called the Seas a 27TDFNC/GW but that's the soft dome version. Madisound made the same mistake.

Finally some tests on the Visiton Ti woofer! Too bad we both got the H1461, I would have liked to see the soft dome results too.

The TB W4-1320 and Aurum Cantus AC130MK2 both arrived on my porch today. I should have results for the Seas tweeter and the TB W4-1320 up soon since I have usable baffles already. The AC and the TB subwoofer may be next weekend.

[Jed]

Finally some tests on the Visiton Ti woofer! Too bad we both got the H1461, I would have liked to see the soft dome results too.

The TB W4-1320 and Aurum Cantus AC130MK2 both arrived on my porch today. I should have results for the Seas tweeter and the TB W4-1320 up soon since I have usable baffles already. The AC and the TB subwoofer may be next weekend.

Distortion testing is addicting.

[TacoD]

Voice coil didn't post any non-linear distortion measurements for these tweeters. Also note that their response curves were done on a 17" x 8" baffle without noting the driver location (offset near top? dead center?) and without noting if it was a flat baffle or if it had depth like a normal enclosure. The measurements were at 1m, so the baffle has a huge effect. Those are major testing shortcomings making the results near useless.

I know that Voice Coil doesn't post distortion plots, but this fact doesn't influence the other measurements. You can argue the completeness of the measurements but this has nothing to do with the quality of the other measurements. Also a large baffle minimizes diffraction so you actually see what the tweeter does. For you those measurements are useless, others (like me) can appreciate them. I also like the fact that Voice Coil like HobbyHifi present the facts and do not interpret them, let each for themselves decide if they like what they see.

[jkrutke]

You can argue the completeness of the measurements but this has nothing to do with the quality of the other measurements.

IMHO, complete documentation of measurement conditions is a very important part of the whole measurement process and determines the overall usefulness of the result. This applies to both response curves and non-linear distortion plots. I could probably understand that there's not enough room for extra documentation in a magazine though.

Interpretation of results benefits those who who lack the experience to correlate measurements with sound quality. For everyone else, they are easily ignored. In magazines, I suspect there is a lack of direct interpretation so advertisers are not angered, not to mention they don't want to appear too brutal. (direct interpretation = "these speakers are horrible, don't buy them")

[augerpro]

Finished tests for the TB W4-1320 and Seas 27TTFNC/GW. Take the Seas results as preliminary since I have to check out a different baffle tomorrow.

I've added Le(x) and some multitone HD testing to some of the 4" drivers. I'll add the multitone HD tests for some of the other high interest drivers I've tested in the future. I used the method of Linkwitz's site which is two sidebands at +/- 10% of F1 and -6dB down.

Also I reran the THD for the Visiton KE25SC after tweaking the baffle.

[augerpro]

Man what day...I got good news and bad news. The good news is the TB 6" sub and AC130MK2 will be done today. I may even be able to get some multitone HD testing of some of the other drivers I still have on hand.

The bad news is the absolute SPL calibration method I use. It works well enough as long as the mic gain is put back in exactly the same position. Which doesn't always happen. So I've put a fine mark on it and now I'm going to rerun all the responses of the drivers I have...lot of work but it should go fairly fast. Tolerance from driver to driver should be within 1dB or so now.

And I'm hungry...

[JonMarsh]

And I'm hungry...

Being hungry when working is not good- low blood sugar.

Oughta fix that....

The rest sounds good, though!

[Jed]

Being hungry when working is not good- low blood sugar.

Oughta fix that....

The rest sounds good, though!

Nope, the 2 soldergun burns on my leg from yesterday's non-stop crossover tweaking and development are proof of that. Note-to-self: don't wear shorts and solder near your legs next time!

[augerpro]

Jed why do you think they call it workbench? I've never heard of a worklap. Sheesh.

Ok, Seas 27TTFNC/GW, AC 130MK2, TB w4-1320 and W6-1139SI are all done and up.