DIY 3 way main speaker with 15" drivers?? - Page 14

Actually I'll take some measurements for you. The off axis response is not nearly as bad as you would think. Keep in mind that the TD15M is only -3dB at 2KHz at 30 degrees off axis. Now I wouldn't suggest taking the pair of TD12S's up to 3KHz for most practical purposes. The fact is though that the off axis of the TD10's in the MTM at 2KHz is as good as the off axis of the 6.5" drivers at 3Khz. While I do agree that off axis power response is important, I'd put more emphasis on what happens with in +/- 30 degrees than what happens outside that range.

John

Thy should share and recommend penny the super isodynamic planar tweeter. (customised RAAL would be too expensive). Run active (good for future builds/upgrades). Problem solved.

Still saving hard....

T/S parameters are not useful at that frequency. You can not accurately model high-frequency response with them.

I'm currently listening to Dire Straits.... just now on Money for Nothing, also the La Bamba CD by O-zone percussion group. I peaked at nearly 120. I wanked the TD15M-A up to 3k. No breakup. No fancy boxes or baffles, I have the 2x TD15M-A on the floor on the LDPE packing to reduce vibrations to the floor.

Is the low Le a major contributing factor to the sound quality? I am sure the freq response and off-axis specs are nice. But to me as an audio enthusiast, my ears are the final arbiter. I can't get such sounds from 5-digit $$ hifi boxes (6-digit boxes can though).

Even if WinISD could measure that accurately, it's not telling you what you want to know; It's off axis response. WinISD is purely 0deg off axis.

Not sure, I just maxed the volume on the integrated amp and hoped the DCX/CDP is supplying enough line level voltage. Dire Straits is quite high in recorded level. I did cut at 100Hz and handed it to the TH-112 though.

You want me to use the 1850HD tomorrow? That's 300+ into 8.

Jim Salk brought their prototype unit down from Michigan for us to listen to. Actually, it was also for Jeff to hear it since he had not heard the build was done. Jeff had a different looking prototype that he sent to Jim and further tweaks in the crossover were done via email, etc., after Jim created the new box design.

We had a shootout approach at that DIY meet. I believe the levels were at 90db on a test tone, IIRC. The Bagby/Salk design did not compete in the competition, so it was cranked up quite a bit higher during the demo, though no spl levels were measured. I know Jeff and Jim said they would play cleanly much louder than we had them. I thought they sounded as clean and clear as any speaker I have heard. On one Doc Severinson piece, I remember thinking Doc could be in the room playing. I can't tell you how impressed I was with them.

Hi John,

I have a check in my wallet for you that I keep forgeting to send out. It is for the $12 I still owe on the TD12H's.

I can't attest to what you are hearing, but I can tell you that the midrange of the Bagby/Salk speaker was very special. I know they did an all passive crossover and Jeff did some very unique things with it, which I can't divulge. I can't imagine how it can get much better than that combination of woofer and midrange. Perhaps the tweeter could be improved, but I had no complaints. I heard not a hint of strain, distortion, etc. They sounded as live as anything I've heard.

Personally, I would not do an MTM with two 10's or two 12's. It would simply have to be to big and set to high in my room to get the tweeter near ear level. My wife would not be happy with that either. This is one reason why I like the 3way approach with the tweeter on top and a smaller, 6"-8" mid, which allows the mid/tweeter baffle to be narrower or tapered to narrower. But, that is all personal choice and based partly on how spectacular I thought the Bagby/Salk speaker was. I think the open baffle mid is part of the reason it sounded so clean and live. No sound reflecting back through the driver and attenuated reflections from the side walls.

Well in his defense the PHL crapped out 2 minutes into testing I am sure he did not have time to listen to the driver. I am working on getting another driver ordered but looks to be out of stock till late Sept. The MTM design is great and still considering it but me looking for serious midbass kick in the chest I wanted an extreme 3way setup...Ultimately I want the best sound possible for my system. I have extremely great speakers now just looking for some more output without breakup. That and a new design for behind my screen.

Oh and how are they building those speakers? That tweeter has been discontinued for some time it was one of my original plans to use that tweeter and 2 ScanSpeak 6 1/2" mids and a bass driver.

Nate

Penn, if you have the money, I say go for it. Plenty of power and passive cooling = no noise.

Another quick measurement of the modified 6ND410 response:



Yellow curve is the modified version with damping and no dustcap. Blue is the old response. Efficiency is knocked down a little where we need it, and the huge breakup at 4KHz is now gone. It sounds much better now. I'll do some more experimenting with it in the 3way system.

John

I'm not sure about the dome tweeter, but I know they are planning on offering a ribbon tweeter version, so perhaps that is the long term answer.

...... nuke

Too bad I can't find any other PHL driver available besides the 1120.

Even if sensitivities don't match up nicely for a passive XO, the DCX has a built in eq right? The 500-2k area could be eq'd up a couple db to compensate.

It has EQ and you can adjust levels for each output separately.

RIght, but what I am getting at, is this 3 way design is geared towards being bi-amped, with the MF and HF crossed over with a passive XO. That would allow someone to build 3 LCR's and use a single DCX and 6 amp channels. So adjusting the output level will not only effect the mid, but also tweet since they're on the same XO network. The eq, however will not effect the tweeter...well, not enough to worry about since it's below it's XO point.

...... nuke

...... nuke

Don't forget that because you have another driver in the system, there is yet another issue with phase and time alignment, especially if the mid and tweeter are next to each other. As you move left to right, things will change at the xover point due to the path length difference. The current enclosures are in this alignment:



The current response looks like this, including the effects of the response off the floor and ceiling:



Red is with drivers in phase, yellow with the midrange polarity reversed to check for phase coherence. It sounds good, don't get me wrong, and I'm sure the PHL would be better than the 18sound, but we can't get it. I spoke with Elliot at Zalytron just a few minutes ago. They have 2 PHL1120's held for Nate now. They don't know if they are getting any more in though. They have something on order, but they don't know if it will be the 1120's of some other driver that they end up getting.


It's actually one less driver as the other is a WTMW as you have seen. Basically just saving the customer the money of the midrange. Nate's plan was for a pair of 12" drivers as he wanted lots of midbass kick. That's what I started working on here first. You could easily do a single woofer, mid and tweeter. But then again at that point, a single woofer with just the tweeter is a viable option as well. Greg from Phat Planet Studios is currently working on a 2way using the TD12M's and the neopro5i ribbon. Either way, getting rid of a midrange is always a driver savings.


Wisdom Audio has one now that people are raving about. www.wisdomaudio.com It uses a pair of 10" drivers and a ribbon tweeter in a freestanding or inwall mounting configuration. The cost is $2250 per cabinet and you have to buy their amplifier with processing starting at $3500 and going up from there.

You also have to realize that in an MTM, drivers act differently than in a line array as there are only 2. You don't have a 3rd, 4th, 5th etc driver to give you cancellations. When you are between the centers of the 2 drivers they actually sum quite well. This graph was at 2meters away from the large MTM's using the TD15M's along with the 18sound XT1464horn. This is the TD15M's only, measured at the height of the center of the tweeter.



Now if you had another woofer above and one below, you'd see all kinds of comb filtering. Or if you get above the height of the woofers you begin to see this as well.




Here's a nice test for you that has been done in the past by Paul Butterfield. You can read about it on the Yahoo Lambda Drivers Group. It's the reason the TD15 phase plugs were made longer. Play a 10Hz tone at high excursion and play a 1KHz tone at the same time. See how much level you need to drive it at before the 1KHz tone sounds audibly modulated. With a driver that has high inductance or non-linear inductance with Xmax, the results will be horrible. The TD drivers do this quite well. With the shorter phase plug, the acoustic loading on the cone was different on the inward and outward stroke. Adding the extra long phase plug kept the acoustic loading the same at higher excursion and eliminated this issue.

Most all studios end up with some large system with dual 15" woofers and a compression driver on a large horn that are crossed between 1KHz to 2KHz. This is what they are listening to during the recording and mastering process. It is not at all unrealistic to do the same type of setup when listening to the same music.

Again, I'm not trying to sway people in any given direction. DIY is about deciding what is best for you. If people want to wait for a PHL driver, they are fully welcome to do so. Whether they do a WMTW system and MTM, a TWM, or just a TW system is fully up to them. My only goal here is to present some measured data, give people a starting point to work from. To me, the simple MTM with the 10's sounds significantly better than this WTMW system. I think raw driver measurments is a good point to start to show why. You can eq drivers to flatten them but you're not changing resonances, etc. The impedance curve on the TD drivers extends much higher in frequency and has much less goofy stuff going on. That would be a big part as well.

John

Thylanter,

You seem to have experience with the PHL drivers. Are you familiar with the 8" PHL 2520? I've picked up a pair of them that I could use with the Lambda TD12H and the Peerless HDS tweeter described in my previous post. The mid will be open baffle, like the Bagby/Salk design. My other current option is the B&C 6md38, 6.5". I was planning the crossover points in the range of 450-500hz and 2500-3000hz. It will be an all passive crossover. Any thoughts about the PHL vs the B&C?

Best Regards,

Dan

I've attempted to contact PHL several times with no response. I'd be glad to look into purchasing them, but it is hard to even consider when there is nobody to even talk to at the company. I put in calls to Orca and E-speakers again today as well. I'll see their thoughts if they return the messages.

If we're looking at 6mos time frames, I could have a much better midrange driver available by that time at a similar price point. That is my overall goal anyway, as there just aren't a lot of good options out there when you start looking at over 95dB efficiency. Again, if I'm going to spend the time to really put a full kit together with passive and/or active crossover diagrams, cabinet plans, etc etc that someone can easily duplicate, the effort has to go into something that can be repeatable. The fact that the original people interested in this system wanted something capable of output in the mid 120's range really disqualifies any traditional midrange and tweeters from the mix.

FYI, current lead times on Lambda drivers are 3-5 business days as well.

John

...... nuke

John J:
I'd like to order a pair of TD15x's to play with for the time being (before the price increase). Can I still order the neopro5i's from you, along with mids if you're able to source/build some?

Just one other update. EQ'd the top end of the ribbon a little more and here is where we are at now:


Red is the tweeter before the final EQ and yellow is summed response of the system, time gated to get rid of floor and ceiling reflection. Response below 500hz drops off due to the time gating.

John