While we're at it I use a mixture of Doepfer and Analogue Solutions modules in the Eurorack format and a Simmons SDSV brain.
No keyboard but sequencers, theremin, joystick and ribbon controllers.
Analogue Solutions is run by a really nice guy in Kingswinford which is only a few miles from where I live.
Analogue Solutions | real analogue synthesisers
No keyboard but sequencers, theremin, joystick and ribbon controllers.
Analogue Solutions is run by a really nice guy in Kingswinford which is only a few miles from where I live.
Analogue Solutions | real analogue synthesisers
A filter will always produce the same phase shift regardless if it is realized passively, with opamps or in DSP.
However my opamp based BSS allows for phase adjustment through 360deg in the xover region so phase shift can be minimized but I doubt it can be fully eliminated and in DSP you could use minimum phase FIR filters at the cost of pre-ringing.
As long as its nicely asked no reason anybody will answer you.
Active analog filters schematics are a whole subject by themself, but you should find easily info about them. A nice start is to find info about N.Pass Firstwatt's B5. In the Passlab you should find info about them. The main difference with other analog active filter is in the choice of using transistor for the active parts (versus more commonly seen op amp) but everything else remain on same principle.
http://www.linkwitzlab.com/filters.htm
For your Crown you re right. Analogue input then adc> sofware treatments>dac> amplifier section.
For a filter to be FIR (the 'linear phase' filter you are talking about) it needs to be digital and performed by dsp (or processor in case of pc). This is not something usually encountered except on particular applications because: 1) it induce latency of treatment (you need to perform a lot of calculation to perform them) and the more latency the lower the fc, 2) usually they re not needed in a home installation (typically the unit taking advantage of them are in PA because of very high slope filter* giving possibility to dump very large amount of power into driver safely and to approach limits of driver without to much compromise on quality (aproaching breakup freq of drivers)).
*In my unit up to 180db/oct filter availlable once 1khz up freq for fc.
That said, you could choose to use them with global approach of speaker design for home use a la Ulrich Horbach and D.B. (Don) Keele, Jr. and that's great because it is a spot on original topic design example
: http://www.xlrtechs.com/dbkeele.com...ear Phase Digital Crossover Flters Part 1.pdf
http://www.xlrtechs.com/dbkeele.com...ear Phase Digital Crossover Flters Part 2.pdf
For this you'll need 'acourate' software which is the only one i know implementing this particular crossover curves.
Used as regular filters results are great, but probably not worth the price to be paid except if you want 'the best' (you could probably have 90% the same at 1/10 the price to be more precise).
Last edited:
Oh. Apparently I misunderstood some article I read and came away thinking that DSP based crossovers don't shift phase. I had the idea that they just cause different time delays depending on the implementation but the delay was the same across the frequency range.
More reading to do...
Last edited:
Charles: Nice toys! I like AS, they sound great. I'm running Frankenstein, protoboard for the modules i use (filters, env,strange treatments,...
I use an AS Oberkorn and a bunch of modules they don't make anymore as sequencers.
But in the end it's the filter that makes the synth sound and by far my favourites are the AS SY02 (MS20 style) which rings/self-oscillates easily and then screams like a banshee and Doepfers Wasp filter which does not but it gets close and then makes wonderful gurgly, watery sounds.
May i ask why? If you managed to do it passively i don't see why it couldn't be done active at line level.
I suppose it's because of notch or eq related? It's relatively easy do do some clones of the 'original' parametrics eq and you could have it done without capacitors in signal path (up to five parraleled bands with high and low band switchable from shelf/parametric) if that's something which concern you.
And the unit i'm thinking of is one of the better sounding things ever done.
Perhaps I should have qualified
it would be difficult for me using the philosophy I used of no opamps in the "direct" signal path. I used RLC notches in series with the drivers and I'd have to simulate floating inductors (without a shunt cap) to do that, and that is quite a bit more complex, than the gyrators and FDNR's that I have used for the "simple" circuit. Doable but more complex than I'm ready for I think. I did make my MTM's fully active ready from the outset. However I wanted to do passive as a first try with active a possibility down the track. Pic shows that they are active ready
Tony.
Attachments
I start to become suspicious about you Charles! Do you know me, have we meet before? How is it possible? Here's schematic of one of the most used module in my frankenmodular:
http://www.schmitzbits.de/rs20.png
?!
Never ending! The more you learn, the more you know you need to learn... never ending!
True but it is also important not to miss the wood for the trees.
The process of completely or nearly completely eliminating tends to create problems elsewhere.
Like with the crossovers here: You can eliminate phase shift by using FIR filters but at the price of preringing and increased latency.
Latency is not really a problem for home HiFi because all that happens is that the sound appears a few milliseconds later but in live and studio situations it can make everybody's life a misery because it really matters if you hear your band mates late when you try playing in time with them.
I start to become suspicious about you Charles! Do you know me, have we meet before? How is it possible? Here's schematic of one of the most used module in my frankenmodular:
http://www.schmitzbits.de/rs20.png
?!
I'm watching you when you're not looking!
Charles: do you know Serge modular? If not listen to the track 'Blue Serge' on Skinny Puppy's "the process' and 'Consumed' from Plastikman. On CGS.net there is some clone schematics (approved by Serge!) of some modules. Worth the effort to built a sidecar for your main Eurorack.
Same for Buchla's... They have some vactrol filters which are NASTY! here's a variation on them:
Mutant Filter
Same for Buchla's... They have some vactrol filters which are NASTY! here's a variation on them:
Mutant Filter
Right. Got 2 preset for my monitors: one when i'm working 'live'(IIR) a second for 'high quality listening' (FIR). Anyway with the Lake once you use them past 500hz latency is not so high (i'll have to check but it's below 5ms and lessen the more you go up).
It's really problematic lower in fc and i don't feel the gain to be worth the latency in lower range. Better stay IIR using different strategy (JMLC filter topology work great in my case).
I know of them but never used one. Surely I must have heard one unknowingly.
I'm itching to try some Metasonix modules though.
They've had me when they called their VCO 'the worst, most evil, the most insane VCO ever offered in Euro format'.
Home
As long as its nicely asked no reason anybody will answer you.
Active analog filters schematics are a whole subject by themself, but you should find easily info about them. A nice start is to find info about N.Pass Firstwatt's B5. In the Passlab you should find info about them. The main difference with other analog active filter is in the choice of using transistor for the active parts (versus more commonly seen op amp) but everything else remain on same principle.
Active Filters
For your Crown you re right. Analogue input then adc> sofware treatments>dac> amplifier section.
For a filter to be FIR (the 'linear phase' filter you are talking about) it needs to be digital and performed by dsp (or processor in case of pc). This is not something usually encountered except on particular applications because: 1) it induce latency of treatment (you need to perform a lot of calculation to perform them) and the more latency the lower the fc, 2) usually they re not needed in a home installation (typically the unit taking advantage of them are in PA because of very high slope filter* giving possibility to dump very large amount of power into driver safely and to approach limits of driver without to much compromise on quality (aproaching breakup freq of drivers)).
*In my unit up to 180db/oct filter availlable once 1khz up freq for fc.
That said, you could choose to use them with global approach of speaker design for home use a la Ulrich Horbach and D.B. (Don) Keele, Jr. and that's great because it is a spot on original topic design example
http://www.xlrtechs.com/dbkeele.com...ear Phase Digital Crossover Flters Part 1.pdf
http://www.xlrtechs.com/dbkeele.com...ear Phase Digital Crossover Flters Part 2.pdf
For this you'll need 'acourate' software which is the only one i know implementing this particular crossover curves.
Used as regular filters results are great, but probably not worth the price to be paid except if you want 'the best' (you could probably have 90% the same at 1/10 the price to be more precise).
Thank you again. I suspect what I read over a year ago was specifically about FIR and I just associated that with all DSP filters. I need to find the manual for the Rockford Fosgate 3SIXTY.3 DSP I run in my car and see what it uses.
-Chris
Thanks. I was wondering about the latency thing and if it really matters in my case where all channels are going through the same DSP.
I'll search and read up on preringing.
-Chris
How is it possible tu put a chainsaw on a cute little yellow box?! Like them! Ts21... Hell on earth!
Their small modular S-1000 is unreal!This is synthesis!
Check this and the vaccum tube section especially:
CGS - Ken Stone's Modular Synthesizer
...nice article from Eric Barbour, owner and designer of metasonix... including schematics!
If you are listening to music I don't see a problem with a bit of latency. If you are watching movies however, it may be a bit more of an issue. I find even 10 ms of latency quite annoying (lip sync).
Tony.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.