DSP and not DSP

[IslandPink]

Has anyone tried using DSP to create a shallow crossover slope ?
By which I mean getting a first-order slope at the crossover point, which then steepens further up, once the bass amplitude has dropped more ?
Only, this is how I have done some of the more successful crossovers with passive components.
What I hear from DSP that I don't like also shares some characteristics with steep passive slopes, which I don't like - I hear what I would call a 'tone hole' in the area around the crossover.

[rowuk]

Has anyone tried using DSP to create a shallow crossover slope ?
By which I mean getting a first-order slope at the crossover point, which then steepens further up, once the bass amplitude has dropped more ?
Only, this is how I have done some of the more successful crossovers with passive components.
What I hear from DSP that I don't like also shares some characteristics with steep passive slopes, which I don't like - I hear what I would call a 'tone hole' in the area around the crossover.

Yes, but with the MiniDSP, the default is 12dB/8va because it uses biquads (2 poles). To get 6dB we need to program the function in the Advanced mode.

Here is a link that helped me a long time ago. It is not hard but we do have to brush up our maths.

[Nick]

I think we should first credit the cello recording to Janos Starker and Robert Eberenz
Secondly, although I would say it's one of the best DSP bass integrations I have heard, I could hear the effect of the crossover when the cello went down to the lower registers and it wasn't entirely natural.

Yep, that was the one. Two things I should add to that. When we got Colins speakers into the show room, we had the DSP set for Colins room based on measurement with a peak to correct a suck out and a hole to counteract a mode. We removed both of those corners but didn't have measuring kit to do it correctly. We also added a low cut filter to prevent cone wobble which was a result of the arm and cartridge combination not be ideal. I wonder if it was the effect of that almost randomly added low cut that Mark was hearing (there is no need for that cut in Colins system).

[simon]

Thanks rowuk, interesting stuff to mull over. Particularly that you couldn't find a valve buffer to your liking. The second half of an Aikido drives most stuff, did you ever try one? Just curious.

[brig001]

Has anyone tried using DSP to create a shallow crossover slope ?

My Behringer will do 6, 12, 24 and 48dB per octave depending on what feature you are adjusting. I am using a 6dB per octave low shelf to reduce the bass slightly below 200Hz

Brian

[rowuk]

Thanks rowuk, interesting stuff to mull over. Particularly that you couldn't find a valve buffer to your liking. The second half of an Aikido drives most stuff, did you ever try one? Just curious.

I said that I could not find an exceptionally "neutral" valve buffer (including the Aikido). We must always keep in mind how many stages of amplification we have and that a simple single buffer will flip phase.

[vinylnvalves]

The challenge I see with just using DSP on the bass, is the time lag that convolutions add. Chris can probably elaborate in the magnitudes we are talking about, and how to get around it. It’s not an issue I have as the DSP feeds all channels. In the past I used PLLXO’s with limited success.

[Nick]

Yep I guess it depends on sample rate and if ffr then number of taps.

[IslandPink]

OK, so does a typical simple bass DSP implementation have a significant time-lag vs. the midrange ?

[brig001]

According to the mini DSP manual, in normal mode (not DIRAC) the delay is 1.5 to 2 ms
If I limit the phase lag to 90° that puts me at 125Hz
Should be ok as I only really need DSP below that

I could put everything through the DSP but it can’t do what I need. Add this to wanting an improvement sonically, I came up with this idea of keeping DSP out of the mid range and treble

I think I need a mini DSP alternative that has the functionality I need then I can compare everything through DSP vs DSP for bass only - any ideas? Must be analogue in and out. Thanks

Brian

[IslandPink]

Sounds OK, that gives you some options.

[Wolfgang]

I don't understand how mixing DSP with passive x-over could ever work. The delay of the DSP would need to be corrected (added) on the passive x-over side. With multiple drivers one could adjust the distance of the drivers but with one driver this is not possible. How could one tune the phase (delay) in such small increments with passive x-over components so that it would match the DSP delay?

I have experimented with passive, analog active, mini DSP x-overs and nothing was really satisfying. The best sounding results with all necessary controls for fine tuning I could get with a heavily modded Behringer Ultradrive.

https://www.audiosciencereview.com/foru ... iew.40968/
https://linearaudio.nl/linear-power-sup ... er-dcx2496

[Nick]

and that a simple single buffer will flip phase.

Not true, a cathode|emitter|source follower will not alter phase.

[Ray P]

I'm planning to explore this general topic with my SLOBs when I can finally get to finishing them and the amplifiers I'll use with them. My own thinking is to use the Nelson Pass LXMini crossover network, for the bass section (I recommend having a read of the article and thread on DIY Audio). For the mid/top I plan to use bandwidth limiting in the amp to roll-off the bottom end at the crossover frequency but also use the other half of the LXMini network with a conventional bandwidth amplifier. Initial set-up will be with DSP.

My assumption is that because the LXMini network isn't doing any processing any latency will be so trivial as to be irrelevant.

[brig001]

I don't understand how mixing DSP with passive x-over could ever work. The delay of the DSP would need to be corrected (added) on the passive x-over side

You could well be right, but at 80Hz it’s only 57° out, and it gets better lower down

I’m not saying this will work, but it’s an interesting concept that could give the best of both worlds

Thanks for the links too, I’ll have a look to see if any of that is applicable to my Behringer amp (NX1000D) as that is another option

Brian