mtm to tmm 2way

[top gnome]

Hi (short version) I would like to build new speakers and use the components from a mtm and make a smaller shorter tmm keeping it a 2 way speaker system. I would like to use the same crossover parts. is this possible without changing the crossover too much. x sim does not seem to care if the speaker is tmm or mtm. there is no BSC in the current design.

(long Version) I build a small thor cabinets with the seas millenium and magnesium woofers. The woofers corroded and I replaced them with the Seas Nextel drivers. the Small thor cabinets are approx 59 liters and an mltl design.(scottmoose) I have spent the last couple of months trying to get great sound like the thors and altho the sound is not bad by any stretch it is not great. if I play with eq to flatten the response it improves the sound somewhat. there is a hump from 600 to 1000 and a dip from 60 to 300. my thought is the cabinet is just too big. Playing with the port length improved it some but I am quite sure that the port is close to optimized now as is the crossover. I would like to build smaller cabinets and for a few reasons would like to try a tmm design. At least to start with I would like to keep it a 2 way design. and not add a circuit to make it a 2 1/2 way. The Thors were also used in a much larger room. Our current house does not have a large room so a smaller speaker may work better.

[top gnome]

[Chris Roemer]

go right ahead . . .

W/out specific driver #s & XO component values and topologies, anything is possible.
A smaller box won't go any deeper. No BSC sounds thin to me. Do you have a sub?

[top gnome]

Hi no sub, really did not need it with the Thors not for music anyway.. w18nx001 - e0042 and t25cf002 -E0011 the w18nx001 replaces the w18e001. does the bsc show up in the measurement?. the measurements do not show a 6db drop in fact there is a hump in the midrange that is part of the w18nx001 MO. it shows up in many of the designs using that driver. the small thor is 9 inchs wide so I think that means 500hz would be the baffle step frequency.

the crossover is the thor crossover with r2 c4 and r3 removed and a 1uf added to c1 to make it a 9.2 uf

I realize i will not get the same bass level and I think trying to use the bigger box is mucking up the mid bass some but I could be wrong. I was thinking the soundstage and imaging may improve with a tmm and a smaller box not sure.
I did not know much when I started this and know a little more now and wondering if the nextels are just not a great replacement. They have a great sounding mid and do voices great. but the detail found in the mag drivers is not quite there.

[Chris Roemer]

Try this new XO please (keep the box, keep it MTM for now). (I tried to re-use some of your parts) It looks like I helped you w/this 3 mos ago...

2nd order HP w/L-pad: 6uF series cap, 0.80mH shunt coil (to gnd), "L": SR=2n(ohms) / PR=12n (use your old resistors here, inc. the 12.5n)
1st order "tanked" LP w/Zobel: use a low DCR (iron core) 2.5mH series coil.
Across that coil (in parallel with IT) use a 4n resistor in series w/a TINY 0.33uF cap. Run a Zobel across the woofers' leads using a 5n resistor (20w maybe?) and an 8uF cap.

This crosses a bit lower (2.0kHz - instead of near 2.4k)
It pulls your hump (600-1000) down by -5dB (as well as the range above it), but offers nearly full BSC (where you had none before).
If you feel the highs are too subdued, we can adjust the L-pad later.

[top gnome]

thank you Chris -- I should have listened to you in October and traded the nextels back for the mags. but I keep going back to the fact that the mags only lasted 4 years here. I spent quite a bit of time and realized the spl mic was quite far out of calibration so in december I bought a unik mic and started using holms so feel better about the measurements. I was working trying to use the parts I had. and Allen B was helping me with the measurements and the box understanding how to use the gating function. and the impact of the edge defraction. He was very helpful.I also have been doing quite a bit of reading and at least now I understand what you were saying in October about the drivers not matching up that closely. . Here are the components and the crossover layout as I understand it. If you would please take a look at it and see if I understood the topography correctly. I used xsim so some of the values may be off by whatever the standard available component is like 8.2 cap for an 8. for some reason the tweeter always measures quite a bit differently than x sim so not to worried about the rise in the treble. it may not show up that way once measured. just wanted to say that the green line is the nextel drivers in the above fr chart. [post #2) below I will post a cleaner fr chart showing the thor vs the nextel thor with both smoothed at 6.

[top gnome]

here is a better fr chart that shows the thor and nexel thor both have been smoothed to 1/6 octave

[rpb]

My .02 on the measurements, and comparisons.

What I like to do is get the speaker 4' away from any boundaries, and measure from about 1.5m. I use gating, and can usually take it down below 300 hz. No smoothing involved this way, and room reflections don't affect the measurements.

Under options, you can tweak things to show 5dB scales. To me, it's easier to interpret the measurements this way,

I don't even look at anything below 100hz. You can crop the measurement there, or slightly below. All the x-over effects are above 100hz, and the room affects the bass below 400hz. You will see the baffle step.

If you compare woofers without any x-over, you can see peaks that might need attention.

If you measure the woofer with its filter, you can see if your filter is getting things right. You can work on the woofer filter independently from the tweeter.

[rpb]

Here's an example. This is a speaker I'm currently working on. Top trace is the woofer without a filter. The lower measurement is with the filter. Note the baffle step can be seen between 300hz, and 1k.. Also note that I notched the peak above 10k, and the bump around 5k.

[top gnome]

thank you - the room is very small 12x14 and when I set holms to gate in auto it usually gates at over 1k. Here are the chart. the raw woofer pair is at 32 inches and the tweet is the same the gated with cross is at 1 meter. I do not have the woofer pair with the cross but can measure that.I have moved the gating to 300hz The second chart is at 2 meters. with and without felt (thank you Allen B) I can certainly do other measurements as needed. the scale has been reset to 5db,

[rpb]

My room is no larger, but has minimal furnishings to get in the way. If I measure with the speaker on it's stand, (roughly 33" from side wall, and 42" from the back wall), I get reflections, and the measurements look similar to yours. If I move the speaker to a position near half way (side to side), and 48" from the back wall, and 48" off the floor, then all those reflections will be gated out, and I can really see what's going on. It appears that you have a lot of early reflections that are combining with your speaker, and distorting the measurements. These reflections can come from anything within four feet of the speaker, or within four feet of the mic.

[top gnome]

I tried moving the speakers into the middle of the room but I am going to have to move some furniture out. I just cannot get good gated measurements. here are the latest. but I will work on more tomorrow the 7 k dip is the edge of the speaker I can turn that one on and off by adding felt to the baffle.

[rpb]

If you could take a few pictures of your speaker, and measurement setup, there may be something obvious that is causing issues. I'm not trying to make work for you, I'm just trying to help you get good results.

[top gnome]

thank you here is a photo of the set up I do my woodworking glue ups in the same room so keep cardboard on the floor. That is also the reason I am thinking about TMM because some of my listening is done standing up

[top gnome]

Hi Chris I was wondering if the topography is correct on the crossover you recommended? I am not sure of where to order the iron core in canada but will keep looking i sure do like the price compared to air core. thank you