Start a thread and list the drivers. We may help come up with something interesting for you.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Recommendations needed for a replacement for a miniDSP active crossover
- Thread starter Old Hi-Fi Guy
- Start date
I'd get the Flex 8 too, vs the Thomann.
JeremyFife
Addicted to Fun and Learning
True!That’s the problem: he doesn’t have a digital solution
This way though, options for a digital solution get simpler (or, rather, are not complicated by trying to deal with analogue)
Xilica Solaro QR1 and FR1 offer the same versatility as Q-Sys, and FIR capabilities and price are between MiniDSP and Q-Sys.
For comparison of FIR capabilities, see FAQ 'processors' in https://www.loudspeakers.audio/en/faq/
Some design decisions dictate the course and then it becomes a self fulfilling prophecy. When you make the crossover steep there is minimal interaction so minimal chance to trade one thing off against another. When you want that to be completely acoustically complementary, it forces a flat on axis phase match at the crossover point, there is only one delay value that will give you that.Not trying to change your mind either.....just communicating
This is a perfectly reasonable approach but it is not the only one that works. I try and make my decisions based on the total radiation of the speaker. It is possible with specific cabinet design and driver spacing to get a smooth DI and power response with a phase matched crossover, but often allowing a little asymmetry or phase offset helps.
Yep, I think designing for acoustically complementary response does become a bit of self fulfilling prophecy...in that it sort of defines that achieving no tradeoffs, one thing vs another, is a primary goal.
I've found it works great both on-axis and off, at least within a listening window as defined by the large conical horns I use.
I don't even know how to think about measuring the total radiation power for the big unity/synergys with subs, especially outside their designed pattern control.
Do you have a build with some measurements that show a total radiation design?
That's a nice FIR processor comparison you put together. And a very interesting website !
Yes, q-sys limits FIR to 16,384 taps per channel for whatever reason. But can run a lot of channels.
I haven't found any need for over 16k taps at 48kHz yet, even for subwoofer frequency work.
But i did find the Core110f insufficient after a while, but due to the number of channels I needed (8 minimum for stereo 4-ways), not so much the max per channel.
I've run 15 channels of 16k taps per channel on a Core 500i (three LCR 5-ways)
Right now, if there is any such thing as an incredible bargain in hardware processors, I'd call it a used Core 250i or 500i, when one with the right set of I/O cards pops up on ebay.
They typically go for less than Core 110f's.
Here's an example of a 250i https://www.ebay.com/itm/1759703526...BD1vTXHhBMN8rbfqpuFauiL2a7|tkp:Bk9SR5bM09frYg
Nothing anywhere remotely near $800 can touch its processing power imho.
Within a general listening window many approaches can be made to look the same, it's when the full 360 radiation is considered that it becomes harder.
The same as any other just with a lot more difficulty from the size and weight
I can show it to you better with some simulated data I am working on at the moment. This is a simulation of a Bliesma M74P and a SB29TX in a small waveguide. The directivity is likely to be quite accurate which is the important part.
This is a linear phase LR8, flat on axis as is the listening window, but there is a DI bump at the crossover frequency. There is a good argument that this is perfectly fine and dips in the power response like this are not a problem as per Lipshitz and Vanderkooy.
The bump can be smoothed out by making the slopes asymmetric and reducing the phase match with a small delay offset.
Thanks fluid, it's cool how far simulations have come, and that they are within the reach of us all ...given an investment in learning....
Bless Kimmo
Have you turned any of those sims into builds and measured/compared with the sims?
I see more folks are starting to do that on DIY, especially with Mabat's, yours, and others guidance ...awesome stuff.
I've been playing with VCad a little, looking at linear phase brick wall xovers, with varying Y axis separations......vs traditional IIRs with typical lower orders.
For the elementary designs I've tried, the lin phase brick walls appear to consistently win out (vertically), not matter how I juggle. Dunno what to think there yet.
Also don't know what to think about trying to create cardioid response, for any frequency range.
In measurement training classes for Smaart, the instructors set up sub arrays outdoors demonstrating the various cardioid techniques, and how that each of them has a purpose in attenuating sub response to a particular angle/area relative to stage; but also how each of them compromises the quality of normal on axis / designated listening area response. I'm not keen on the audible quality loss in the direct sound.
I have to believe that holds true for all frequencies, so being the simple-minded somewhat lazy dude I am, I've said why bother with cardioid.
And have kept it simpler, by only working on large horns for pattern control as low in freq as possible, given the need to stay halfway practical with size and weight.
Anyway, I can't really put my big builds on a spinorama, and I don't know what program can sim a big synergy well enough to make a design decision...maybe I need to see how far Hornresp has come? VCad doesn't do horns, does it?
Who knows, now that Baltic Birch has become unobtainable, I may turn into a sim-before-build dude, too
Bless Kimmo
Have you turned any of those sims into builds and measured/compared with the sims?
I see more folks are starting to do that on DIY, especially with Mabat's, yours, and others guidance ...awesome stuff.
I've been playing with VCad a little, looking at linear phase brick wall xovers, with varying Y axis separations......vs traditional IIRs with typical lower orders.
For the elementary designs I've tried, the lin phase brick walls appear to consistently win out (vertically), not matter how I juggle. Dunno what to think there yet.
Also don't know what to think about trying to create cardioid response, for any frequency range.
In measurement training classes for Smaart, the instructors set up sub arrays outdoors demonstrating the various cardioid techniques, and how that each of them has a purpose in attenuating sub response to a particular angle/area relative to stage; but also how each of them compromises the quality of normal on axis / designated listening area response. I'm not keen on the audible quality loss in the direct sound.
I have to believe that holds true for all frequencies, so being the simple-minded somewhat lazy dude I am, I've said why bother with cardioid.
And have kept it simpler, by only working on large horns for pattern control as low in freq as possible, given the need to stay halfway practical with size and weight.
Anyway, I can't really put my big builds on a spinorama, and I don't know what program can sim a big synergy well enough to make a design decision...maybe I need to see how far Hornresp has come? VCad doesn't do horns, does it?
Who knows, now that Baltic Birch has become unobtainable, I may turn into a sim-before-build dude, too
At the moment I don't have anything that I can share publicly, so you will just have to trust me that if the simulation and measurements are done correctly they are well within an acceptable tolerance
Cardioid sub bass requires quite a big effort regardless of how it is done and perhaps those efforts have downsides in the systems you have experienced.
I think you would be surprised how "cardioid" like your existing very large speakers are without trying to do anything more than you have already done.
For home use cardioid makes more sense in trying to control directivity in a smaller package.
