MTM with Dayton RS 7" and Seas TBFC/G: will it work?

[tcpip]

Hi all,

I've been wondering about trying a simple MTM, either active or passive xo, with the Dayton RS 7" metal cone and the Seas TBFC/G, xo'd somewhere between 1.5K-2K, LR4. I am new to speaker design, as some of you know, but I intend to use Speaker Workshop or some other package, take SPL measurements, and build a software-optimised xo.

I'm not too worried about how to model the box ... well, not at this stage. What I was wondering about was whether this overall combo will work. What do you guys think?

The RS 7" metal cone has some resonance peaks. If I go by what the PE Website's SPL graph shows, then there are three or four peaks, all of which can be bunched together in one region between 6K and about 12K and pushed down by using a single, wide, low-Q notch filter. After that, if I put an LR4 slope starting at 2K or a little lower, then the cone breakup peaks should be really 50dB below base line or even lower... well below audibility. I don't know how easy it will be do do this wide notch in a passive xo, but I'm willing to go the (analog) active xo way if needed. If I do active, I'll probably do LM3886 chipamps, one amp per driver (three amps per box).

What do you guys think of this whole idea? Will it work? Will it be a poor man's SS Solist, do you think?

Driver costs are attractive: I'll spend about USD 200 for all six drivers.

Once you tell me that there's nothing crazy or stupid or hairy-difficult in what I'm thinking of doing, I'll go order the drivers.

 

[simon5]

With a LR4 at 1.5k or 2k I think you can discard the notch filter.
I mean, at 8k, everything will be 48 dB down.

If you use 1.5k, even better, you reach 48 dB down at 6k so that huge peak at 8 kHz is still over 40 dB down.

I think your project will work, but it will need clever crossover design to take into account that ugly dip at 2 kHz. Maybe use the LR4 at 2 kHz for the 7" woofer and put the LR4 for the tweeter at 1.5 kHz.

 

[Ed LaFontaine]

It will work

How well depends upon your crossover:

http://www.htguide.com/forum/showthread.php4?t=11321

is one place to start.

 

[DougL]

Modula MTM
Modula MT Natlie P

Three designs from Jon Marsh that use the drivers you are proposing. Also, Dr. K has an MTM on the Parts express site.

Good luck;

Doug

 

[tcpip]

Re: It will work

Wow, thanks guys. Lots of things for me to chew on.

I think your project will work, but it will need clever crossover design to take into account that ugly dip at 2 kHz. Maybe use the LR4 at 2 kHz for the 7" woofer and put the LR4 for the tweeter at 1.5 kHz.

I have the PDF file from PE giving me the SPL curve of the driver, and I didn't see any dip at 2K. Actually, there seems to be a very gentle, wide rise from about 600Hz to about 2.5K. Am I even looking at the right graphs?

http://www.htguide.com/forum/showthread.php4?t=11321
is one place to start.

I'm looking through it now. It'll probably take me a day to actually study that thread and get my own thoughts into place. Thanks for the link.

Actually, I'd visited that forum once, and seen a lot of discussions on an RS-based 3-way, with the RS 10", the RS125 and the RS tweeter. At that time I couldn't figure out why they are not using the Seas tweeter instead of the RS, specially after seeing the distortion measurements on Mark K's site. But I had not noticed any discussion on the two-way.

Modula MTM
Modula MT Natlie P

Three designs from Jon Marsh that use the drivers you are proposing. Also, Dr. K has an MTM on the Parts express site.

Thanks a lot. I'll study all of these.

Edit: Isn't the Natalie P using the RS tweeter? Of course, it's possible that with a very minor tweaking of the xo in the optimiser, I may be able to get the Seas to behave the same way as the RS tweeter, I don't know...

 

[Ed LaFontaine]

It'll probably take me a day to actually study that thread

Its a LONG thread, maybe 2 or 3 days

At that time I couldn't figure out why they are not using the Seas tweeter instead of the RS

Isn't the Natalie P using the RS tweeter?

The Modula thread begins with the RS tweeter, the SEAS tweeter is used as a substitution later, using the "final" x-over dated (late) March, 05. Note that it contains some options for speaker positioning. Jon Marsh even alludes to the SEAS H1212 tweeter for consideration this fall, as his time allows. It is one I want to try.

The Modula thread includes comments from users that I think are valuable to someone like yourself in trying to make a decision for which tweeter to select.

On the note about costs, the Modula x-over components called for amount to ~$400/pair. Careful substitutions may be in order.

 

[simon5]

Sorry tcpip, I was wrong, at least I believe Jon Marsh measurements, so the dip is at 4 kHz. Maybe I saw the wrong graphs hehe!

You were right, the peaks are very high so you might need that notch filter.

Good luck!

 

[tcpip]

The Modula thread includes comments from users that I think are valuable to someone like yourself in trying to make a decision for which tweeter to select.

Good, I'll see what I can find (and understand). I've already taken a peek at the xo, and it's way over my head. I can blindly build it, but I can't figure out how it works. However, the discussions bring out some general concepts about what the xo must do to get good results, and that'll be very useful if I want to design my own xo for this beast.

On the note about costs, the Modula x-over components called for amount to ~$400/pair. Careful substitutions may be in order.

Yes, I saw! That's a truckload of money, as far as I'm concerned. In fact, I found the Natalie P much more attractive because of this reason. Or else I'll go active, and the cost of all components, including amps, will be less than that $400, I think. (Don't know about the big fat PSU transformer, though. )

You were right, the peaks are very high so you might need that notch filter.

Or I'll need some super-sharp xo slopes. I'm already dreaming of LR8.

 

[tcpip]

Hey guys, thanks to the pointer to that other thread. I sort of sauntered over across the street to that blue and white bulding across the street and look what I learned. It was worth a whole chapter in a textbook. Thanks for pointing me to that thread.

I have a suspicion I'll try an LR4 followed by a deep notch, active. Should be interesting, and will certainly be less expensive than the high-end all-passive xo they've done there, under the guidance of Gooroo Jon Marsh.

It's very rare to find one gang of veterans as patient and friendly as you guys here. Finding two such gangs is unbelievable. I guess we're blessed.

 

[tcpip]

Its a LONG thread, maybe 2 or 3 days

Even 2 or 3 days was an underestimate. But boy, was it worthwhile. I went into at least three other long threads, all handling different designs with the same set of RS drivers.

And about the parts costs, I think I can't afford the original Modula MTM. So I'll either build the Natalie P, or I'll try my own active xo design with a mix of LR4 followed by a deep and wide notch. I don't know what it'll do to the phase, but it'll certainly get the SPL graph looking sort-of right.

And I'll use the Seas TBFC/G. Xo not higher than 1500Hz.

 

[JoshK]

I have been reading and asking questions on that "other" forum for some time. I don't pretend to speak on Jon or Thomas's behalf, but I thought I'd point a couple of minor things out.

With the Modula MTM, actually the first tweeter used was a SEAS, the one marked Mar 05. (At least in this version. ) Jon then decided to help release a variation of his design for the Dayton RS28a. This is the one I built. However, the RS28a is a ~$50 tweeter while the H1212 is a $30 tweeter. Also some variation in production RS28a's have motivated some consideration for the H1212 to be adopted. I think the RS28a might have been the better tweeter if QC was a little better.

I think Jon's approach can be summed up as using very high quality (& preferrably lower cost) drivers within their best operating range. Then he goes balls to the wall on the xo to get the very best out of them. Having recently built the Modula MTM and having a lot of exposure to many expensive hi-end speakers, I am a believer in Jon's approach.

These aren't yo mama's MTMs! :lol:

 

[tcpip]

Originally posted by JoshK With the Modula MTM, actually the first tweeter used was a SEAS, the one marked Mar 05. (At least in this version. ) Jon then decided to help release a variation of his design for the Dayton RS28a. This is the one I built. However, the RS28a is a ~$50 tweeter while the H1212 is a $30 tweeter. Also some variation in production RS28a's have motivated some consideration for the H1212 to be adopted. I think the RS28a might have been the better tweeter if QC was a little better.

As my thread title says, I started out thinking of this approach with the Seas in mind. (Isn't the TBFC/G the same as the H1212?) I don't know why I should spend the $20 more for the RS tweeter... I'm not sure I can hear the differences more experienced listeners can.

I think Jon's approach can be summed up as using very high quality (& preferrably lower cost) drivers within their best operating range. Then he goes balls to the wall on the xo to get the very best out of them.

Yes, I can see that. He doesn't appear interested in using Audio Technology midranges or Skaaning woofers.

I saw his passive xo and decided I can do a reasonable facsimile (famous newbie comment) using active xo and chipamps at lower cost. Unlike Jon, I can't hear the sound of all those opamps in the path, and chipamps made using non-audiophile components are very inexpensive. Adding complexity to the xo in the active world doesn't add much to the cost, so I'll do that LR4 + notch approach I was talking about. Each channel should require two 2134 chips for the lowpass: one chip for the LR4 and one for the notch. I'll first pick up Jon's SPL curve from his thread, put it into some xo design software and model the resulting curve to see how low the 1800-12,000H region goes. That'll give me and others some idea of where one can head with this approach. What do you think?