I recently spoke with Paul Eizik regarding the criticality of the open-air mounting of that driver instead of baffle-mounting. I decided to go extreme and to get rid of that plate that I had under the driver no mater now small it was, also I thought to lift the driver up slightly. As the result I came up with new mounting harness for my T350.

In the picture the old harness on the left and the not painted prototype of new on one the right. After time-aligning of the “new” driver I played today mono recording for each channel trying to determine which positioning would be better.

Certainly the “new” positioning has less reflection from the mid horn and it produces more auditable “space”. The most fascinating that the “new” positioning yields higher output. The “old” positioning used .2uF cap but at the new positioning I got the same subjective amplitude of HF using .047uF capacitor. Here is where the Gintaras Sakenas’s Teflon caps were quite handy….

It was quite intriguing and very positive result. The concussion of that story is that if you use the Electro-Voice T-350, T35 then stay as further form baffler as possible. Actually, as far as I concern any tweeter and a panel baffle are not big friends but I would not generalize it too widely…

Some useful documents about the EV T350 Tweeters: Page1, Page2

Rgs,
Romy the Cat

Rich,

if you have T350 (not T35) then it would be shame do not use them. Make sure that they are not ST350 and not 350A 350B or whatever later modification they were. If you did not use the truly wide dispersion driver before then it might take some time to use to how they work with room and with the …. lower octave. 

I finished my new T350 harnesses and made a picture. It is not time-aligned yet. Also to have an idea how the new harness works with the entire speaker I updated the image in the "My Playback" section of my site.

The Cat

A lot of people ask me about those drivers and I’ll try to compile my vision about them. I did try most of the possible tweeters: dome tweeters with cloth cones, varies type of hard metal domes, compression drivers with aluminum, titanium, phenolic and beryllium diaphragms, ribbon tweeters, electrostats, Hales tweeters, piezo tweeters, early version of plasma tweeters…

First, to understand what I mean it would necessary to “get” my requirements to the tweeters. I do not really care about the foolishness that many audio Morons spread about the SACD-compatible tweeters that reportedly go to 60K or 100Khz. Those tweeters are something that completely out of my interest me. What I concerns me is only HOW the tweeter dose 10K-14K. Certainly the sensitively, directivity and the rest are important but this would a different subject that I would not touch not. Second, it is important to acknowledge I use tweeters over 10KHz and I religiously despise any tweeters operating from 3-5K (most of the 2-3 ways box loudspeakers). So, all my future “analyses” is applicable to the tweeters crossed above 10K.

My experience indicates that the best result I was able to get from silk-dome tweeters and compression drivers with phenolic diaphragms. To me all searches for better tweeter concentrated around a proper reproduction of string group. When the entire string group of a full orchestra kicks-in then practically all tweeters do gown despite that those violin not do not even reach the frequencies where tweeters operate. What I mean “tweeters go gown” is following: a half-hundred sting instilments produce a fat lover frequency tone and the tweeters add to this pitch a specific amalgam of glitzy “schmultz”. As the result a violin section sound too sharp and too hi-fi-ish. This artificially-effective pseudo-sound converts the violin tone of Vienna Philharmonic into Leningrad Philharmonic and produces an impressiveness of reproduction instead of the impressiveness of original musicality

Pretty much 3 second of listening of one weepy violin tone of 34 instruments playing together instantaneously allow to discard or accept a tweeter. A violin groups in real live (if they are not Boston Pops) never sound metallic or bright but rather soft and syntactic. The never sound impressive of the do not play imprecise. This “not impressive sound” I was able to get only from the better silk dome (the best) or phenolic tweeters. Unfortunately the silk dome have no sensitively and therefore I stick to the EV 350T with plastic cones.

EV T350 in a way reminds a SET amp that “clips softly”. It does not sound glitzy, aggressive and for many out there who are addicted to the brutal force of JBL-like compression driver the EV T350 might appear “not impressive”. The “audio impressiveness” usually comes from positioning a microphone 3 inches from a violin. This positioning is very nice for recording the Patricia Barber Trios but it is absolutely not appropriate way to listen a serious music. A symphonic orchestra’s sound form a normal listening position, when the orchestra is property integrated and balanced should not sound striking of extraordinary. The impressiveness of orchestra comes from the performing quality not form the sonic grandeur.

The Furtwangler’s Orchestra with Bruno Kittel Chorus in 1942 sounded impressive, probably the more impressive then any other orchestra in 20 century. That impressiveness is true impressiveness based upon a performing mastership not upon the reproductive means. Reproducing a good transfer of this for instance performance of Beethoven IX and adding 1/2dB to properly balance tweeter will severely destroy the impressiveness of Furtwangler’s Orchestra. It will make it slightly faster and slightly sharper and slightly hysteric. Do not forget that we are taking about the recording the hardly when over 10kHz.

I can testify that the majority of audio installations that I heard sounded more “instantaneously impressive” then sound of live orchestra. The poor quality of tweeter is one of the major contributors the HF misery of those installations.

It is hard to explain how a tweeter should be tuned, as there are very-very many variables involved in the tuning. ¼ db matters and if a tweeter ¼ louder then it should be then it will add  “audio impressiveness”. Forget about the “air” and “space”, this is not the job for the tweeters. My inexperience indicted that it is very good to hold tweeter on a “frustration diet”. It means that better result you can get if you underused the tweeter then overuse it.  Run your system without any tweeter and get good sound. Get familiar and comfortable with this sound: it should not frustrate you and it should be musical and self-contained. If the sound is not good enough then change the MF driver. Then, in a couple weeks add a tweeter. Peace your head horizontally at ~ .5M from the tweeter at the same distance from MF driver and begin to add the tweeter output (or raise the value of the cap moving the tweeter dope on the transition slop). Play some slow weepy violin section (Tchaikovsky, Smetana, and etc…) and turning you tweeter up. As soon you hear that Sound become slightly “whiter”, “softer”, “bright”, more or less “busy”, “neurotic”, of more “messy” (you tweeter might not to time aliened) then stop. Form the discovered setting remove approximately 2/3dB and let yourself to listen your playback system like this for a t list 2-3 weeks. Then add or most likely  to remove 1/4-1/2 db and you would be all set.

The rule with the tweeters is very similar like with LF module – if you hear it then it set up too loud. Be advised the practically all capacities the you will be using in your tweeter will change in 3-5 days quilt dramatically. BTW, the T350 would also care a LOT of abuse and the excessive 1db form T350 would be equally musically-damaging as 1/3 dB of JBL-077 or Fostex T90.

The recent events in my playback made me to think about he new prospective duty of tweeters. Since the Super Melquiades was enlisted for duty the Vitavox S2 got it’s own dedicated amplifier channel and stopped to “see” the coil of upper bass driver. As the result the S2 go up very clean and without it’s own typical idiosyncratic harshness.

For a number of years I use Electro-Voice T350 tweeter that hardly works as tweeter but rather as anti-Vitavox HF modulator. It sits deep in transition slope (somewhere at 60kHz) and it juts slightly soften what S2 driver dose at 12kHz. I like what EV-T350 dose and how it sounds and practically in context of my MF channel. It might be a bit soft but my tweeter has some tweaks, making them slightly punchier, and that eventually made T350 a perfect match for the goals they try to accomplish in my playback.

Now my midrange channel need no upper range diffusion and now I feel that the result I’m getting from T350 is too soft and too transiently slow for S2. So, I kind of very mildly begin to look out there and to try top fugue out what tweeter I might try. Unfortunately I do not see a lot of interesting tweeter of high sensitively and high dispersion pattern, did I mention the quality of sound?

The plasma tweeters sounds like interesting but the problems with radiated gases, low sensitivity and very limited life-span do not make them unattractive. Some exotic Japanese tweeters, like Onken OS-5000T, might be interesting to try… but to pay $2000 juts for a try (who know how they will sound) is not practically attracts me. TAD ET-703s are syntactic garbage. The JBL-077 are too mechanical and granular. A good silk dome if to cross them above 10.000Hz might be interesting but they have no sensitively. The good ribbon tweeters (for instance Pioneer PT-R9) might be an interesting direction but never tried them + my fear of the ribbons... The armies of the titanium cone tweeters are worst then a nuclear war. The AMT tweeters have too genetic HF. The Fostex tweeters are too Hi-Fi….

I really do not know what I would try. Any suggestion?

The CaT

Romy,

Put them in a box and send them to me. Did you notice the replacement diaphrams available on eBay? Gone now, but they did not sell so will probably return.
Bud

Bud,

the replacement diaphragms are available all time. On Ebay and everywhere else they are being sold by Bob Crites. He holds the original manufacturing rig and he can do them in unlimited quandaries. I own the original diaphragms but I do not use them as I find that the contemporary 16Ohm production is better, it is what I been using. Look somewhere in “My Playback” section I mentioned his email….

Yes, the T350 is fine driver but a driver could not be fine on it’s own. It should be a tonal, color, contrast and transient match with the associated channels. I feel that since I put to S2 a deeded amplifier the T350 can’t do it anymore.

Well, perhaps after screwing with number of tweeters I will return back to the T350, which had happened in past with me. For a time being I’ will be trying to use a reserved, open-back another S2 as a direct radiator. It might be useful…

Rgs,
The caT

Just some scattered points:

1) I gather the Acapella plasma tweeters have a catalyst so you don't get the nasty oxides/ozone etc. Got some old ionofanes on the way so will know soon how well they freshen up the room... They are self-amplified so efficiency is not an issue.

2) The PTR9 might be discontinued so if you want a pair, act soon. Actually, I've never tried mine with the S2 as it would mean padding the horns down 12dB.

3) I gather the Onken is long since out of production. I know JMLC loves them and found the ET703 "somewhat digital" sounding...

4) Seems to me like the following tweeter might be worth a try, so long as you can bypass the internal x/o:
http://www.e-speakers.com/pdf/LCY-100K-specsheet.pdf

except you'd need another amp channel.... which goes for all ribbons I guess.

5) Have you tried boosting the HF to the S2? Given that you have a potential divider at the input of the amp, you can just shunt the upper R with a cap (or RC for shelving filter)  Actually, forget that, it will just exacerbate the S2s breakup.

6) What about that little JBL Be compression driver as used in the K2? No experience of them but they might fit the bill sensitivity and transient wise?

cheers
cv

cv wrote:

I gather the Acapella plasma tweeters have a catalyst so you don't get the nasty oxides/ozone etc. Got some old ionofanes on the way so will know soon how well they freshen up the room... They are self-amplified so efficiency is not an issue.

cv wrote:

I know JMLC loves them and found the ET703 "somewhat digital" sounding...

cv wrote:

Seems to me like the following tweeter might be worth a try, so long as you can bypass the internal x/o: http://www.e-speakers.com/pdf/LCY-100K-specsheet.pdf except you'd need another amp channel.... which goes for all ribbons I guess.

cv wrote:

Have you tried boosting the HF to the S2? Given that you have a potential divider at the input of the amp, you can just shunt the upper R with a cap (or RC for shelving filter)  Actually, forget that, it will just exacerbate the S2s breakup.

The caT

Bud wrote:

I have a really barbaric suggestion... of course. I can set you up with 8 lLinaeum, soft, live horn, dipole tweeters. Stacking 4 per side should see you at 103 dB sensitivity and 8 ohms of impedence. You can stack them without interference diffraction effects and I would even treat them for you. Vertical dispersion is about 80 deg, horizontal is 140 deg and then there is the backside output, true dipole.

Bud wrote:

Meanwhile I was serious about the boxing of the EV's. If you want them faster, with higher dynamic excursion capability, no loss lf musicality and a bit more intensity I can perform that miracle. And, since there are good diaphrams available and the pattern cleans off with alchohol, we could experiment....

:-)
The Cat

Hi Romy,

Give the ribbons a try http://www.e-speakers.com/products/beengineering.html.  These are too expensive I think but believe me the sensitivity is real.  The best part is also that the impedance is very flat and an easy amplifier load.  I wouldn't worry about blowing them but if you are worried you can try something a bit cheaper than these.  I do not think the planar magnetic drivers are performing quite to the same standards but they can also be quite good with high sensitivity (93-95 db).

A second opinion needed.

Interesting, I recently found a very-very well performing setting for my EV T350 tweeter. I changed the source for HF channel form amplifier’s output to the MF channel and now I source the HF channel AFTER the MF high-pass filter. Well, let see. If the MF is crossed at 1000Hz (metal diaphragms) and the HF are crossed somewhere in ultrasound and roll to the auditable spectra at the transition slope… then what kind order at HF I have?

It looks like I use two sequentially located first orders but they are divided by 8-9 octaves. The tweeter dose sound more like second order and lost a lot of its upper-midrange softness and “noise”. In fact it sound now quite nice along with S2 drivers. So, is it the real first order or not? Interesting that I did not change the location of the driver (it sits still at it’s first order spot) and when I went to this knew crossover setting I did not feel sonically that it slided to anywhere time wise….

Rgs,
Romy the caT

Thanks, Greg

Greg B wrote:

Cap and L-pad resistor quality is critical even in the cheaper models.

I do not use any L-pad resistors with tweeters. If you need to reduce the tweeter output I drop the caps and move the tweeter up at the transition slope. Do not forget that the EV tweeters are dead over 16kHz and all that I need form then is juts to add ½ octave at the top.

Greg B wrote:

The main problem with ribbons is that they need a steep crossover, which is inherently problematic. The other problem is that they seem to have more non linear distortion than good conventional types. Multitone burst tests show this. It is audible as a sort of shimmery silver sound - especially at high SPL, but for low efficiency systems they work OK. Details and image are good to excellent, but integration is a bitch. Best integration I got was with a simple quasi second order series type, which seemed to lock the drivers together. I was shooting for something that started shallow but increased cutoff below the crossover point, sort of like what was just mentioned.

You nailed down the ribbons perfectly. The only thing that I would add that when ribbons sound I feel that that ribbons have diarrhea nd afraid to make any fast movements. Although I still would like to hear those new HE ribbon drivers.

BTW, if you would like to be really entertained then get this :-)

http://www.angelsoundaudio.co.uk/html/golden_sounds.html#UtraTweeter

Rgs,
Romy

Romy the Cat wrote:

Now my midrange channel need no upper range diffusion and now I feel that the result I’m getting from T350 is too soft and too transiently slow for S2. So, I kind of very mildly begin to look out there and to try top fugue out what tweeter I might try. …

I decided to re-appoint again the T350 as a “Phenomenal Driver” ....as for the last 3-4 months while I doubt it the T350's price on Ebay went down. Now the justness is reinstates and the dues are paid.

Hate me... :-)
Romy the Cat

Listening my Macondo with new “Fundamental channel” I noted that my tweeter is not as effective as it use to be. The effect of “not-effectiveness” is very minor but at some recordings it is notable. My T350 still sorts the Vitavox S2’s top but it also begin to pick a tiny bit off fussiness. I presume that it comes from reflections. The EV T50 notoriously want to sit aside very far from everything, including any baffles. Also it wide radiation pattern does not really comply with it’s current location. I will do further experiments, trying to put my T350 horizontally and some other action that “might” be successful (as I do not know the real reasons why the fussiness came to the picture) but I also ask myself if any other tweeter might be utilized in new Macondo.

T350 is wonderfully soft but it 120x60 degree radiation is very completed in context of my current setup.  My initial idea was to use the beautiful copper tweeter from Celestion  (I LOVE that tweeter) but it needs 100+ amplification and no any amplification but very good amplification. It would be too ridicules to have 100W PP SS to drive a tweeter.

Looking (again!!!) at the army of available tweeter with narrowing deposition I discover a new opportunity. Fostex has their T-line compression tweeters, they are aluminum in alnico: T-90A, T-500A, T925A. I had them. They do well with slow and sleepy Altec’s MF (although they even with Altecs a bit too hard to me) but with Vitavox they very brutal. So, how big my surprise was when I discover that Fostex did acknowledge the problem and introduced a new version of thier T-tweeters: T90A-EX and T500AmkII. Those new drivers are the same old drives but with bigger magnets and the diaphragms made not from aluminum anymore but with magnesium.

http://www.goodsoundclub.com/pdf/Fostex_magnesium_foil.pdf

Wow, this "might" be very interesting and practically in context of the Fostex public revelation: “ Pure magnesium diaphragm having superior internal loss, having less unpleasant overtones as heard in other types of metal domes.” Those “unpleasant overtones” was exactly what bothered me in the old Foxtex tweeters, but again, that revelation very much might be a marketing BS and the new drivers might have own “magnesium problems” in sound. I would not speculate at this point as I never heard them the new Fostex tweeters and I fine them as a good candidate.

Did anyone herd them? Did anyone come across any new interesting tweeters some my last tweeters frustrations?

Rgs,
Romy the Cat

It is incredible how stupid sometimes our audio steps are.  A few days ago I was complaining that sound of my T350 tweeter suddenly started to bother me but it was no the problem with the tweeter but rather the problem with my own moronity.

A few months ago I implemented a transaction slope attenuator. It was basically a pile of capacitors sitting in a good quality rotary switch. That allowed me very precisely to dial-in the attenuation of my tweeter. I tuned in my tweeter (it was a step #6 on the rotary switch) and then sine I dissemble my entire tweeter assembly I time-realigned the tweeter.  I was so pleased with the result that I made a post: “Something that a loudspeaker must have.” in my “Audio For Dummies ™” section:

http://www.goodsoundclub.com/TreeItem.aspx?PostID=1487#1487

Since then I have been using the depicted configuration, playing the tweeter all time in the in step #6 and turning it down when electricity is bad.

Well…

although I do not argue with the concussions of that post and do find that an ability to precisely set tweeters amplitude is very important and powerful tool but when I look at THE WAY HOW I did it is really question who is the Dummy!!! I juts realized that THE WAY in which I did it was so much bogus that it is scare. Running the tweeter across the transition slope I unavoidably introduce in each new “click” a new phase offset. It means the correct position of the tweeter relative to MF driver (or angleing of the tweeter) would be different at the different “clicks”.  I spent a few hours confirming that my tweeter was misaligned but I failed to acknowledge that my attenuator was not in the #6 step. This perfectly explains the “mud” and “fussiness” that I experienced. At this point the attenuator was in position #6 and after I moved the teeter a few mm forward then the problem is gone and the tweeter sounded as perfect as before. How much was my surprise when later on during the same evening I moved the attenuator to the default step #6 and begin to hear the same “fussy dirt”.  Of coursed I immediately re-measured the time aliment and off course I was deep in cancellation mode…. What else I was expected ands what the hell I was thinking!!!

I think I have too much time on my hands or use my time in highly stupid fashion but I decided to do today the experiment about witch I was thinking for a while. The experiment was “methodology clean” and lead me to quite interesting discoveries. Since you follow my blog and have as much time on your hands I think it is very much worth to share my observations with public. Also, leaving all foolishness aside, if the subject of this blog and your interests is hard pushing an envelop of advanced audio methods and evolved music reproduction techniques then read on.

I hate tweeters and I hate HOW Hi-Fi reproduce high frequency. There was literary a hand full moments within my audio experiences when I heard produced HF that I liked.

1) A few years ago I heard Grand Utopia (old one, not today’s Be crappy version) driven by some huge German-made PP tube amp. The sound was typical Grand Utopians… with obnoxious laser-like HF. However when I got myself in the very precise distance from the speaker then it was it. Grand Utopia is freaky speaker and knowing how to deal with it I was literally moving inch by inch from 25-20 feet and somewhere when I hit the correct distance then… the tweeter suddenly despair and the HF where execrably how they should be: not auditable and non-existing but hypnotizably non-existing

2) Last year I was listening Wilson Alexandria with Spectral 360 SS PP. I would not comment on the general Sound but HF were incorrect. However, when I was walking out of the room, somewhere near the door it suddenly hit me. In this, many feet away from listing position spot the Alexandria’s HF suddenly collapsed and the effect of the phenomenal HF calmness come to the existence.

3) A few ears ago in a used LF shop there was some very crappy no-name 30 years old dirty bookshelf speaker that was sitting on the top shelf of the store. The speaker was driven by a consumer receiver. The HF were so spectacular (from any place of the store) that it was not even funny. I asked a clerk to play an opera peace that I picked from the shelf and it was literally like nothing else.

4) In June-July 2000, in my own listening room, my speaker (I do not remember which tweeters I used at that time as I changed them monthly) were driven by Lamms ML2 and during the typical movement experiments my speakers hit the  DPoLS. During those few weeks that HF were absolutely incomparable with anything I ever hear in audio. It is absolutely imposable to convey to anybody who is not familiar with the effect.

5) The Celestion SL-600 with copper tweeters and driven by Lamm M1.1 amplifiers. The Celestion copper tweeters are class of their own and there is nothing like they if the amplification is correct. I know, Lamm quite suffers from quite harsh and quite deserved criticism on my site but let pay dues when it is needed: what Lamm M1.1 does compare to other amps when it drives Celestion SL-600 is nothing less then deserving heights respect.

It was it. Any other listening of a playback I ever did annoyed me with HF tremendously. Still, there is a tweeter that always annoyed me lees then any other: it was Electro-Voice T350. It’s phenolic diaphragm sound VERY different from ANY other tweeter I heard. I use this tweeter for 4-5 years and I always suspected that I do not use it properly. I do not let T350 “to sound” but rather inject T350 at the strategic minus 8dB at ~13kHz -15kHz to cure some issues of my midrange Vitavox S2 driver. It did the necessary touch and it was acceptable, at least not too compliable (when electivity is right). However, today I took a single Celestion copper tweeters, placed it atop of Macondo, connected it via Dominus to Lamm M1.1, EQed the volumes and listen the Macondo powered by the Celestion’s copper. What can I say? It gave me sense of new direction. Nope. The Celestion’s copper was not better, in fact it was harder then T350 but …

You see the Electro-Voice T350 is a MF driver. I meant to be used from 3500Kz and it has very flat and very fat MF. It is not good quality of MF but it is there. The first order was not juts able to kill and made me use the “injection method”. I set the T350 with sharper filter and it did quite deferent result. Somewhere at 2rd -3rd order the T350 enters the performing domain of Celestion’s copper but it has very soft stress and very “yellow” HF tone. Mixing with Vitavox S2 it has different result then it use to be but still it has very elegant, very sophisticated and very-very-very-very none-twittery sound.  I do not know in witch setting I end up but the facts remain the very same: there was no other known to me driver that has comparable to T350 ability for sound reproduction. The Celestion’s copper might be very much close but with 82dB sensitively and heavy impedance EQing it might be a problem. You see, to drive the Celestion’s copper you need a >100W PP but the push-pull have inherent issues at HF (die to the transition line asymmetry of push and pull, not to mention if they use feedback).

So, Electro-Voice T350 is potential “the only correct none-metallic HF”, 110dB sensitively and 120 degree deception. Thanks to my public blabbering the prices of the used T350 jumped from $300 to $1000. Good for you. I think it is still much cheaper then many other tweeter and much more interesting sounding. Dose it has anything to do with advanced audio methods and evolved music reproduction techniques?