OB compact 3way nearfield monitor

[jeshi]

X that is looking pretty good, but too large for my application. I need it more compact.
Since you are getting away from the compact aspect of the design (or at least my compact definition) you could go for dual TG9s to better match the levels.
Do you think you could put a pair of these 1.2meters (48") apart from each other (tweeter-to-tweeter center) and listen from 1.2meters?

 

[xrk971]

If I lay the Kazba on its side and put the trapezoid on top it sounds pretty good. Better integration. 1.2m is a comfortable distance for these and it sounds good and coherent at 1.2m. I admit it's a bit large. Maybe the 8in woofer and 3in mid and tweeter in a rectangular box like all of these pro monitors is the way to go.

I would just go sealed and LT like you suggest. OB really needs clearance in back or else doesn't sound good. For a small studio may not be the best thing.

 

[jeshi]

If I lay the Kazba on its side and put the trapezoid on top it sounds pretty good. Better integration. 1.2m is a comfortable distance for these and it sounds good and coherent at 1.2m. I admit it's a bit large. Maybe the 8in woofer and 3in mid and tweeter in a rectangular box like all of these pro monitors is the way to go.

I would just go sealed and LT like you suggest. OB really needs clearance in back or else doesn't sound good. For a small studio may not be the best thing.

Good to hear that it is well integrated at 1.2meters.
My home studio setup/room is a bit odd, it really is not small, but narrow.
I have 3.3meters behind my speakers, 2.5m in front of my speakers (where I sit), but the room is narrow, only about 3meters wide. Also I only have 550mm between my LCD monitor and the nearest side wall on the right side, so maybe 510mm (20") is the absolute max width I can do and would prefer to be under 400mm wide. Also to get the tweeter/mid at my ear level it and above my piano keyboard, I only have about 390mm (15") max height, otherwise the woofer is below the top of the keyboard or the tweeter/mid would be too high above my ears. My speakers are on stands behind my piano keyboard.

This is why I set up my design criteria the way I did and why I thought my room/setup could benefit, from an open baffle / dipole design
Sorry maybe my design is really custom tailored to my setup. But I do think if I can pull it off with a dual 8" SLOB + TG9 + DQ25 in a 280x390x280mm HWD folded OB box (only back is open, with folded sides top and bottom), it would be very cool.
Sorry that I am a much slower builder than you X. All I can do right now is say my plan and give experience on my DQ25/TG9. I am still loving and using my little DQ25/TG9 OB for causal listening a lot, especially at about 1m listening distance. It actually sounds better at 0.8-1.0m then at 1.2-1.5m distance (maybe something in the dipole pattern or room reflections). Maybe I can have my SLOB sometime later this week or next week.

 

[xrk971]

I'm listening to this speaker for past couple of hours - it sounds quite good. Very dynamic and detailed without being harsh or fatiguing. SPL levels seems adequate for near field.

Here is a sound clip (mono) - change .asc to .mp3 to listen:

http://www.diyaudio.com/forums/atta...-tg9fd-mylar-ob-3-way-mg-so-we-meet-again.asc

 

[xrk971]

Update on TG9FD OB with ND25FA tweeter

I received the Dayton ND25FA-4 tweeter and installed it. With a 4kHz 24dB/oct HPF it measures exceedingly flat and well behaved in this baffle despite not having a rebated bezel. The ripples you see are the baffle edge and non-flush mount diffraction that remain.

Here is the measured response - no EQ of anykind. Sensitivity at 2.83v and 1m measures 91dB (a bit higher than spec):

Here is the measured HD - fairly low at 91dB as can be seen:

Here is the driver mounted in the trapezoid baffle:

A tighter CTC spacing can still be achieved, and removing part of the bezel of the ND25FA will allow yet even closer spacing. A very nice value at $20 for a silk dome Nd magnet tweeter that measures this flat.

How does it sound? Wonderful - smooth, detailed, clear. Here are some sound clips with an RS225-8 in a sealed cabinet providing bass below 350Hz XO of the TG9FD. Change .asc to .mp3 extension in order to listen. This is running a symmetric 1st order XO at 4kHz between the TG9FD and the ND25FA.

 

[jcarr]

xrk971, great-sounding results for the budget!

jonathan

 

[xrk971]

xrk971, great-sounding results for the budget!

jonathan

Thanks! This combo does sound exceptionally good, period - price not withstanding. The TG9FD is a superlative example of what a mid driver should sound like. Blends very nicely with ND25FA. I suppose I should simulate polars of both to see where XO should technically be set for the smoothest transition. Walking around the room though I don't notice any anomalies with XO at 4k.

One thing to note: OB doesn't have to be big to sound good. This one is 9.5in wide x 9in tall.

 

[jeshi]

Finally got back to this project. Decided to buy another pair of TG9 so I could try a compact MTM arrangement. Decided to also try the 4ohm TG9FD10-04. I will do more testing between the 4ohm and 8ohm TG9 in the future, but so far the 4ohm are working really well.

Not sure if I prefer the TM or MTM versions. The MTM can play louder with lower distortion, but it suffers from the line-array vertical lobbing. And for a nearfield application this can be more obvious. I can easily adjust the height and angle to avoid the lobbing in my sitting position, but still something to consider. For some reason this MTM required much less EQ steps to get it flat compared to me first TM, but this could be more related to the wings or the baffle arrangement.

Crossover is still LR2 at 4khz. I imported the target LR2 curves into REW and then adjusted the EQ/XO until it matched the acoustic target well.
EQ and XO for this DQ25 TG9 MTM OB v1

TG9x2 XO LR12db 4000hz
LS 150hz 7db q0.7
PK 488hz -5.7db q1.89
PK 642hz -3.7db q3.3
PK 1562hz -5.3db q1

dq25 XO BW6db 7000hz, inverse phase 0.06ms delay
peak 1900hz -8db q0.5


Overall the MTM version measures as well if not a little better HD than the TM version. I really like the looks of the MTM though and I like the extra headroom. It is still really small.

Here is the HD of the MTM vs TM. It is most different in the 200hz-600hz range and below 60hz.

vs the original TM

Not certain yet if the lower distortion of the MTM is due to the two drivers or to the 4ohm version of the TG9. I will do more 4ohm vs 8ohm measurements in the future. Also once I decide if I prefer the TM or MTM I will make a matching pair.

 

[xrk971]

Thanks for the update - looking really good. Your XO, even though its a textbook LR2, seems to have a good SR, as if the tweeter main pulse if positive polarity to match the TG9 (maybe because electrically, it is still only BW1?). So the transient response must sound very good. There is a slight dip up high with the response and I wonder if that is because the tweeter so close to the edge of the baffle now that is a diffraction cancellation. I like how compact it looks though.

 

[jeshi]

Yeah the >10khz dip also appeared in my other one after time. Maybe it is a burn-in effect of the DQ25 or maybe how I am measuring slightly changed. It doesn't sound like there is a dip up there since it has beautiful crisp accurate treble. But I also know that I am sensitive to excessive treble so I definitely appreciate a small dip over 10khz if it is there. I am not worried about it.

 

[jeshi]

Oh I forgot to include the phase graph. I am not sure if this is right, the impulse is aligned with the major rising peak (same if I estimate IR delay), but the phase shows a slight rise in the treble as if it was not aligned correctly.

 

[xrk971]

Jeshi,
I noticed you are using rectangular baffles now. The trapezoids work so well at controlling diffraction. Is there a reason you aren't using them here?

 

[Zonneschimmel]

I did some sims in Edge before deciding on a final baffle for mid-tweet for my 3-way and found that the smallest width worked best for the mid driver (even compared to off-center in trapezium). The tweeter always had huge peaks and dips regardless of placement and various shapes between full trapezium and smallest straight. Did I miss something?









I went with huge (4cm radius) roundovers just to be safe.

 

[jeshi]

mainly because I am planning to mount the SLOB woofer to the side of the DQ25/TG9 in a design similar to the KH310 and SCM25a. I did some baffle simulations and with the long side to make room for the pair of 8" woofers in a SLOB it didn't really make much difference. And then to put a small triangle on just one side seemed like a lop-sided idea and also didn't really make the baffle simulation that much better. Seemed to be just as good to get the driver really close to the edge. Basically I couldn't get a baffle diffraction simulation to look good with a KH310 style arrangement
So far my measurements are not showing any serious baffle diffraction issues.

Here are some examples. Overall no matter what I tried it didn't really smooth out like when it was a single driver in a truncated pyramid.

Putting a chamfer on it seems to do more good. Still not sure if I will do the chamfer in the final build since right now the DQ25/TG9 is measuring nice and flat.

 

[jeshi]

zonneshimmel, nice speaker! I also agree that it seems that just a narrow baffle and getting the drivers close to the edge, seems to be one of the most reasonable compromises. Was really hard to find a more complex geometry that yielded a significant improvement over the "close to the edge" approach when dealing with so many drivers.

 

[Zonneschimmel]

Thanks, it's a SS22w discovery, 10F 8414 and a nd25fa-4. All LR2 crossover at 700 and 3K at the moment but I also intend to try Harsch and B&O filler type. SS26W discovery will be added later (I can divide the internal volume of the basscabs in different ways).

What is the program you used for the simulations? It seems a lot more powerfull and user friendly then edge.

 

[jeshi]

Very nice drivers. I will be building my SLOB first with quite cheap GRS 8PF 8" drivers, but I am thinking about alternates already and the SS22w are on my short list.

I am using this excel spreadsheet "Baffle Diffraction Simulator"
Baffle Diffraction Simulator
For the pictures I posted, I composited the speaker layout picture over the frequency response curve. It is really nice program/spreadsheet with several models for simulating chamfer, regular, or rounded edges with reasonably flexible geometry, but maybe not as flexible in the geometry as Edge.

 

[forart.eu]

Hi everyone, I actually use quite-customized Canton GLE 70 in my home studio as main monitors, but I've always wanted to mix trough a KH310 clones (originals are too expansive for me).

Since i'm absolutely "noob" in DIY realizations, I've started - some years ago - to collect resources from the web about DIY sealed monitors (this 3ad included) at this page.

Anyway AFAIK best performace/price material for cabinets *should* be concrete:

Refined by a variety of additional ingredients, concrete today has evolved into a high-tech product with highly varied applications. Resilient, durable, and pourable into almost any form and especially, largely resistant to vibration.

Concrete forms, by its very nature, the perfect base for the housing of our loudspeakers.

The other approach is the (software-driven) multiamplification both to obtain a more flexible speakers appliances and to reduce costs.

Hope to inspire someone...