Hi Derek,
Do you know if it is possible to feed digital out from a nanoDigi and use the channel DACs and amplifiers of the Yamaha.
Not sure how to use the DRC DSP inside the yamaha in this case.
Joji
I think it was just the degree of expectation and not maybe fully realising that any DIY kit (amplifier, speaker, robot, anything...) might not work straight out of the box and will require some amount of learning and elbow grease.
I agree. Rick has a huge amount of patience. I used to design speakers at one time (for myself) but that was years ago. This time, with time limitations and other responsibilities I had decided to get commercial models, partly also because the finish they provide is far better than I can ever hope to attain.
Rick's kit was an excellent half way point, more so when the carpenters who were doing much of the woodwork at our apartment took up the challenge of making the speakers and more importantly the finishing. Friends often ask how on earth did I find speakers with the same exact veneer as my walls. 😀
Yes I have used illuminators (again another Selah design) I think you have seen them (on the audiocircle thread).
Jojip I am not sure what you are trying to achieve here. Are you going active or passive?
I want to explore the options for active if the passive changes do not work out.
For active as i mentioned, i would prefer to use external multichannel DACs and miniDSP.
Can the crossover schematic help explain the low bass/lower mid energy?
Or do we need the values for that analysis?
I still havent understood the real cause of the large roll off on the lower end?
Or do we need the values for that analysis?
I still havent understood the real cause of the large roll off on the lower end?
Yes, it is the degree of expectation. A logical think to believe would be,
the more it costs, the less additional engineering work should there be
involved, or none at all. I mean what kind of a kit is it if it needed modifying
to properly work. This is only an assumption, not a qualification of WA kit.
Yes must be C5 it sits as a shunt.
If you interested dig deeper into XO and see what happens when C5 is shunt or serial to L1 here's great free tool http://www.diyaudio.com/forums/multi-way/259865-xsim-free-crossover-designer.html. Can be used with rough component values and build in textbook drivers to learn a lot, and if program is feeded components real values plus the drivers is updated with impedance and frequency response it can model exactly your speakers response.
Last edited:
What I was responding to in your post was mainly "and from what i read in regards to the crossover, he is absolutely spot on. a lot of the posts in this thread were misleading and pure guessing imho," suggesting that his input has influenced your view of the existing crossover design in a positive way.
The cause could have been the large C3,C4 values and small R1.
Regarding Rick's C5 comment, it may be there should have been
a series LC network in shunt to both W22's to get rid of the HF breakup,
just like at W15.
Is this filter version modified to some extent by previous owner
in respect to original one?
More comments please explaining the massive roll off from 500 Hz and down.
I repeated the measurements on Holmimpulse and see the same.
Can this really be explained by the room or wrong cabinet volume?
I am arranging to measure the individual XO components to see if there was some error in the values used.
I repeated the measurements on Holmimpulse and see the same.
Can this really be explained by the room or wrong cabinet volume?
I am arranging to measure the individual XO components to see if there was some error in the values used.
Thanks. Need to learn one of these tools to be able to do deeper investigations on my own
Ok - once the enclosure is modified then we can compare the new measurements to what I have. I'll also look at ARTA so we can be on the same page.
Dissi,
Noticed the neat piece of software you developed which you are using in your posts.
MZ SpeakerDesigner
Thanks
Joji
Joji,
What you have for the schematic layout is actually the same electrical transfer function as my schematic so both yield the same response. The important thing is that the board is wired correctly.
Rick
Yes the breakup of the W15 is well suppressed, that is the peak at ~8k has been tamed by the crossover. The problem is the huge whacking peak in the third order distortion at ~2.7k. This is why you cross the W15 at 2kHz with a 4th order slope, not to help keep the 8k breakup out of band, but to suppress the third order peak it creates at 1/3 the frequency of the breakup. This also applies to pretty much all drivers with very stiff cones, you generally have to cross unusually low for the driver diameter.
The neo3CD obviously cannot cope with that.
This might be the gin tonics talking, but if it was mine, I might be inclined to just notch it a little...maybe a slight dip is less of an insult to the ear?
*shrug*
No, don't use a l-pad because it won't work well for this design. I can give you specific values to try along with what amount of attenuation to expect. The midrange resistor is added in series; however, a second shunt resistor might be needed depending on the amount of attenuation you desire. I can also give you the tweeter resistor changes to go along with the midrange revisions. You could easily use some inexpensive sandcast parts and later replace them with the Mills or Mundorf Supreme resistors.
Please see my additional response on the topology. The filters don't appear to have been modified by the owner.
- Status
- Not open for further replies.