Very good question and few months ago I would not hesitate to go with LR2 and do not bother with LR4. But my view changed after Ellips-A active crossover and LR4 vs LR2 experiment for woofer-midrange part.
I started with LR4 and after few itterations I was satisfied with the sound, but out of my curiosity I tried LR2 and was really surprised with the change I could hear. Whole soundstage become larger but also less defined, bass and lower mid structure was worse and midrange overall lost precision and certain sparkle.
Not sure exactly to explain this. On axis FRs were the same for both version, or as close as possible. 32W distortion is very good and low up to midrange. The obvious difference was vertical off axis response in crossover region.
This experience also led to use LR4 for Shamal project (22W Rev, MW16TX, T25B in WG) and yes, crossover is really .... complex and large, but I am glad I did so as overall sound is close to Ellips-A.
I also realized that ellips-a LR2 experience might be the reason I had problems to get good sound with my previous projects. I just did not want to go LR4 and accept added complexity and parts count.
LR4 for woofer-midrange when done properly (not easy as the sound can easily be dry and too analytical), based on Ellips-A and Shamal, is closer to my ideal of a good sound.
^Your description of sound or LR4 vs LR2 active xo matches my experience well! But I prefer LR2! Could be because my no1 music is classical and acoustic jazz and I don't listen very loudly. To my ears, LR4 also gives a bit more precise stero image.
LR2 has more overlap of drivers, so vertical response is different, perhaps also IMD is higher. But excess group delay is better.
LR2 has more overlap of drivers, so vertical response is different, perhaps also IMD is higher. But excess group delay is better.
Juhazi, I agree, the difference was clear but I would not say LR2 was worse, that was rather about the matter of taste. I sldo liked some of my previous projects a lot, they used LR2 woofer-mid crossover, but in that case the proper driver combination was needed, there were combos I did not manage to get them sounding well.
And passive version with mid-tweeter crossover lowered to 1600Hz. Note change of DI in that region and mild power response dip....
And passive version with mid-tweeter crossover lowered to 1600Hz. Note change of DI in that region and mild power response dip....
Hello Pida, this project and the shamal have been very inspiring. Its time for me to start another project. It will be a 3way active: 12sealed+6+1. I am settle with bliesma t25b in augerpro waveguide (first tests and measurements in april). For the woofer and midrange, there are several candidates and my budget will be the limitation. For the woofer, i am thinking of 32w4778t00 or Dayton rss315hf (from 45hz to 200hz). And for the midrange, I am thinking of the new purifi midrange ptt6.5m04 (hificompass measurements are superb from 180hz to 2300hz), the textreme mw16tx or sb15nbac. I know that you have not test all of these drivers but for a given budget will you favor the woofer or the midrange ? At the end I know that drivers implementation and crossovers are more important than specs but these beautiful drivers are so tempting. The reasonable choice should be rss315 + sb15nbac but... Thanks again for all the work you share.
Hello,
I am using an old fashioned but interesting Audax tw34 in a monacor waveguide...
I would like to jump in the élitiste beryllium world...
I am using digital filters, BC 6md 38 and sealed 15" with an LT. The convolution file is generated with rephase. It is quite simple if you follow the tutorial as you juste 'compensate' the 2nd order roloff from the sealed box...
What kind of roll-off should I expect in the upper highs ? First order, 2nd order ?
Regards Greg
I am using an old fashioned but interesting Audax tw34 in a monacor waveguide...
I would like to jump in the élitiste beryllium world...
I am using digital filters, BC 6md 38 and sealed 15" with an LT. The convolution file is generated with rephase. It is quite simple if you follow the tutorial as you juste 'compensate' the 2nd order roloff from the sealed box...
What kind of roll-off should I expect in the upper highs ? First order, 2nd order ?
Regards Greg
usa_satriani:
Hi, I assume you mean upper rolloff of woofer driver. You do not mention woofer specs, but assuming 200-300Hz crossover we can consider the woofer FR flat up to ~1kHz so upper rollof 12dB/oct could work well as initial setting.
Yes sorry I was referring to the upper rolloff of the tweeter in the waveguide. Low pass behaviour...
I have a sealed 15" with Linkwith transform.
Open baffle 6md38 400-2500hz
Audax tw34 with monacor waveguide.
I think going with the bleysma beryllium tweeter would be a good upgrade! Wonder how it would interact with monacor waveguide and what would be correction needed...
Regards Greg
I have a sealed 15" with Linkwith transform.
Open baffle 6md38 400-2500hz
Audax tw34 with monacor waveguide.
I think going with the bleysma beryllium tweeter would be a good upgrade! Wonder how it would interact with monacor waveguide and what would be correction needed...
Regards Greg
Ok I understand now. For this project I used ~2uF Jantzen Alumen capacitor before the tweeter. I did crossover modelling in Vituix, it allows to model active crossover together with passive elements. So I ended up with ~2uF capacitor in front of tweeter and it basically compensated for WG boost. Further filters and EQs were realized in DSP. Such a small capacitor also serves as tweeter protection.
I do not like "DSP is almighty and can correct everything" approach, so I used it to make just necessary minimum. For example, there were no delays needed to be set for the drivers. Baffle tilt and WG makes driver acoustic centers aligned good enough.
Regarding T34B and suitable WG, maybe Augerpro works on this combo.....
WG300 is not actually good, I have seen some measurements, well...it was usable but top end could have been improved. T34B is top notch driver and deserves optimized WG.
I do not like "DSP is almighty and can correct everything" approach, so I used it to make just necessary minimum. For example, there were no delays needed to be set for the drivers. Baffle tilt and WG makes driver acoustic centers aligned good enough.
Regarding T34B and suitable WG, maybe Augerpro works on this combo.....
WG300 is not actually good, I have seen some measurements, well...it was usable but top end could have been improved. T34B is top notch driver and deserves optimized WG.
Thanks for that precise reply.
I remember using a 1 to 2 uF capacitor with fostex super tweeter...
Yes I am sure I can find a better waveguide than wg300, I bought that to make quick and cheap tests...so far so good...
I also have a stock wavecor tw30a12 to try lol...if you have some idea, please keep us updated :,-)
Thanks again !
I remember using a 1 to 2 uF capacitor with fostex super tweeter...
Yes I am sure I can find a better waveguide than wg300, I bought that to make quick and cheap tests...so far so good...
I also have a stock wavecor tw30a12 to try lol...if you have some idea, please keep us updated :,-)
Thanks again !
Last edited:
I agree. 5inch WG for T25 and 7inch for T34, for the beginning. Depths ~25mm for 5inch and ~30mm for 7inch, this would cover the most usecases.
Protection grid removal and dome installation to WG is quite risky as Be dome is fragile, so ready made solution plus protection grill would be part of WG, I think DIY crowd would welcome that, though Bliesma's business case and ROI may disagree..
Protection grid removal and dome installation to WG is quite risky as Be dome is fragile, so ready made solution plus protection grill would be part of WG, I think DIY crowd would welcome that, though Bliesma's business case and ROI may disagree..
Last edited:
T34 looks more advanced, I did not compare hificompass measurements to satori, just looking at mms, narrow surround, variable dome thickness. And yes the thing around 20k, but for ears this would be just cosmetics.
Actually satori be could measure well in jantzen wg. I measured tw29dn and it looked good.
Actually satori be could measure well in jantzen wg. I measured tw29dn and it looked good.
