Very nice looking baffles! Thanks for the photo.
I have run the Manzanita with digital crossovers, but I could never get it to sound quite right. There are different approaches, imitate the passive, do active EQ, try both. The passive version always sounded better to me. However, you might be able to do a better job than I!
Hi Pano! And thanks for compliment and your experiment info! I think ill have to trial & error. Starting points noted!
Passive is always there if things go wrong.
I noticed when tweaking passive xo that it really is sensitive to different caps and resistors. Last summer at holiday i exprerimented almost two weeks in countryhouse in asperger like mania
(ok, i did other things also) I had xo board which was equipped with easy connections to change different components fast. Got most detailed and crispy/sharp string details with Dale RN60 metalfoil resistors, also my preferred cap was Audyncap plus 5.6uF. These components worked best with polite / soft sounding Denon AVC-A1XVA. I think different amps may also move your choices bit.
Uwe, Thanks for comment!
Can you enlighten me more. Do you mean that when i do highpass for Vifa and then lowpass for Peerless i use 6db filters for both and vary xo points to get sound there. Or, i think you meant something else which i didnt get..
OK I'm excited by all the comments on here as to how good OB speakers can be. I'm seriously tempted to build the MJK version (just 'cos size fits best), I'm currently running IMF TLS80mk2's with crossovers rebuilt to 1% tolerance. But although the speakers 'disappear' I am still aware that the sound comes from a box. Would these MJK OB's possibly sort my problem? or would I be looking at a reduction in sound quality?
Many thanks
Andrew
Many thanks
Andrew
Hi Andrew,
I've only heard the MJK open baffles with highly modified crossovers so I can't really comment on the sound differences between the stock MJK speakers and the Manzanita/Ultras.
But I recall reading that the MJK speakers seem to be optimized for solid state amplification and aren't quite as good driven by tubes. That was why the owner of the speakers I heard had reworked the crossovers.
John's open baffles will do just fine with tube amplification provided you have 30 - 60 watts. If you're into tubes it's worth considering as it might save you some time and effort.
If you run solid state amplification you'll be in good shape either way.
Barry
I've only heard the MJK open baffles with highly modified crossovers so I can't really comment on the sound differences between the stock MJK speakers and the Manzanita/Ultras.
But I recall reading that the MJK speakers seem to be optimized for solid state amplification and aren't quite as good driven by tubes. That was why the owner of the speakers I heard had reworked the crossovers.
John's open baffles will do just fine with tube amplification provided you have 30 - 60 watts. If you're into tubes it's worth considering as it might save you some time and effort.
If you run solid state amplification you'll be in good shape either way.
Barry
I thought the high-pass was more like 2k and the low-pass/notch at 1k, otherwise you get a very forward presentation in the upper-mid (700-2000 Hz). I wrestled with this for a week before realising that the high DCR on my inductors and therefore higher pad resistor values had shifted the crossover points. Recalculated and used an Lpad based around John's original design in post 455 and was rewarded with the laid-back mids many posters had mentioned. This overlapping is the genius of the design imho.
We almost need two x-over designs - one with the premium low DCR components that these were designed with, and one for those who can only get a 20mH inductor with a DCR around 1 ohm. This can increase your pad values on the mid unit by 2 ohms, enough to move your x-over points and mess things up.
We almost need two x-over designs - one with the premium low DCR components that these were designed with, and one for those who can only get a 20mH inductor with a DCR around 1 ohm. This can increase your pad values on the mid unit by 2 ohms, enough to move your x-over points and mess things up.
Hi, I posted this about a month ago...
I believe this is the most current parts list right?
L1 inductor. 18-22 MH Iron core. 15-16 ga. PE P/N 266-960 (16 ga. 20 MH)
L2 H.F. trap inductor 2.75 MH air core PE P/N 266-836
C1 H.F. 1st order electrical high pass capacitor. 6.0 MFD. PE P/N 027-236
R1 HF trap bypass resistor... 15.0 ohm, 5-10 watt P/N 015-15
R2 H.F Pad...8.0 ohm, 5-10 watt PE P/N# 016-8.2
Woofer. GRS-15PF-8 15" woofer. PE P/N 292-415.
Mid-tweeter. Vifa TC9FD18-08. PE 264-1062
Pano replied that it looks mostly right but maybe John can weigh in on values. I have not heard back yet.
Also, any idea how these would compare to DLK/2's from the mid to late 70's? I'm using these for my vinyl listening area and think they sound great but am inspired by this thread.
I believe this is the most current parts list right?
L1 inductor. 18-22 MH Iron core. 15-16 ga. PE P/N 266-960 (16 ga. 20 MH)
L2 H.F. trap inductor 2.75 MH air core PE P/N 266-836
C1 H.F. 1st order electrical high pass capacitor. 6.0 MFD. PE P/N 027-236
R1 HF trap bypass resistor... 15.0 ohm, 5-10 watt P/N 015-15
R2 H.F Pad...8.0 ohm, 5-10 watt PE P/N# 016-8.2
Woofer. GRS-15PF-8 15" woofer. PE P/N 292-415.
Mid-tweeter. Vifa TC9FD18-08. PE 264-1062
Pano replied that it looks mostly right but maybe John can weigh in on values. I have not heard back yet.
Also, any idea how these would compare to DLK/2's from the mid to late 70's? I'm using these for my vinyl listening area and think they sound great but am inspired by this thread.
I may have missed it in this long thread, but has anyone tried an active version of this speaker? A miniDSP and a pair of low cost class D amps can be had for about $130. I just priced the above parts in post #1207 for the XO and it is $60 for one speaker or $120/pair and that doesn't include the amp. That is awfully close the price of a miniDSP and a pair of TPA33116D2 amps.
The active is hard to get right apparently, because the slopes in this crossover do not follow your typical 6/12/18dB curves. There are component-speaker interactions that are part of the design and transcribing these accurately for a DSP would not be a trivial task. They can be approximated for sure, it's just nobody has done it yet.
Yes, as mentioned in the thread I've tried several active versions but was never as happy with them as the passive version. The active versions I've built were OK and had some strong points, but overall were just not a satisfying to listen to as the passive crossover. I don't really know why.
Can someone kindly show me the main post describing the ultra with circuit schematics?
@Pano:
Perhaps the passive version includes some impedance flattening effects on the TC9FD and this is the difference?
One could add a complementary impedance flattening circuit to the active version and see if it helps. They should sound the same if the XO's and phase are the same.
@Pano:
Perhaps the passive version includes some impedance flattening effects on the TC9FD and this is the difference?
One could add a complementary impedance flattening circuit to the active version and see if it helps. They should sound the same if the XO's and phase are the same.
Last edited:
Ultra crossover
Hello all... sorry not back sooner on this.... busy time just now. Always fun to see continuing interest in the Manzanita & Ultra DIY builds!
The Ultra schematic and material list shown on page 86 (posts 851 & 855) are correct. The only thing you may want to add is a 0.10 UF bypass cap across the 6.0 UF HF high pass cap. I have found the $1.00 each PE P/N 027-452 to work well.
As for doing an active crossover... not easy. A passive description would be something like 2nd order, reverse phase, spread frequency asymmetrical Butterworth with HF trap and pad. Lots of non-standard shaping using as few components as possible. Acoustic crossover is around 750 Hz. IIRC.
Once can also add a broad woofer trap in series with the GRS centered around 1,900 Hz and clean it up a bit. Don't remember the values, will have to go look at my home pair!
John
Hello all... sorry not back sooner on this.... busy time just now. Always fun to see continuing interest in the Manzanita & Ultra DIY builds!
The Ultra schematic and material list shown on page 86 (posts 851 & 855) are correct. The only thing you may want to add is a 0.10 UF bypass cap across the 6.0 UF HF high pass cap. I have found the $1.00 each PE P/N 027-452 to work well.
As for doing an active crossover... not easy. A passive description would be something like 2nd order, reverse phase, spread frequency asymmetrical Butterworth with HF trap and pad. Lots of non-standard shaping using as few components as possible. Acoustic crossover is around 750 Hz. IIRC.
Once can also add a broad woofer trap in series with the GRS centered around 1,900 Hz and clean it up a bit. Don't remember the values, will have to go look at my home pair!
John
Thank you! Can you explain what this bypass caps function is? Just trying to learn as we go. It is put in parallel with C1, correct?
Bypass cap
A bypass capacitor of a small value with a relatively high voltage rating (>300 Volts) will usually improve the overall sound quality / clarity of midrange or high frequency driver(s). Lots of conjecture as to why this works. Do not have the time at the moment to get into this topic. Might be a nice thread topic for someone else to initiate.
A 0.10 UF value is small enough to have minimal effect on the crossover frequency but still yield a nice sonic improvement. A typical bypass is around 5-10% of the main capacitor. However, with this design the small 0.10 UF value works well. A 0.33 or 0.47 can also be tried. And yes, place the bypass in parallel with C1. I think you will like the results!
A bypass capacitor of a small value with a relatively high voltage rating (>300 Volts) will usually improve the overall sound quality / clarity of midrange or high frequency driver(s). Lots of conjecture as to why this works. Do not have the time at the moment to get into this topic. Might be a nice thread topic for someone else to initiate.
A 0.10 UF value is small enough to have minimal effect on the crossover frequency but still yield a nice sonic improvement. A typical bypass is around 5-10% of the main capacitor. However, with this design the small 0.10 UF value works well. A 0.33 or 0.47 can also be tried. And yes, place the bypass in parallel with C1. I think you will like the results!
John - I think you posted this many pages back:
"There is the option of adding a woofer trap in series with the woofer (coil - cap 15 uf & 0.40 mh in parallel with each other). I have never posted this as it is part of the commercial crossover and it does add to the build cost. This trap does clean the upper end of the woofer band pass."
Thanks Barry! I never did that version.
I did build a version with 2X15" GRS woofers. Nice, but the poor little Vifa couldn't keep up! Ended up using a vintage 8" fullrange up top and reworking the crossover. They made some good party speakers. Gave them to a friend when I moved back to Hawaii - he loves them.
I did build a version with 2X15" GRS woofers. Nice, but the poor little Vifa couldn't keep up!
Ended up using a vintage 8" fullrange up top and reworking the crossover. They made some good party speakers. Gave them to a friend when I moved back to Hawaii - he loves them.