advertisement


Active crossover design musings for E-IIIR

If I understand correct PD you are suggesting reducing components by using the input of the amp for 6db and the feedback for another 6db?

Sounds good to me, especially if it reduces out of frequency range work in the amp (being that the treble amp will amplify anything, including power noise)?
 
Yes.

You can also get HF cut by increasing the size of the input low-pass filter capacitor.
 
hacker said:
In the second Word doc, it states "Without baffle step compensation – standard textbook LR4 @ 265Hz and 3kHz", but has very good FR and phase alignment.
Click to expand...
I've got this document, but I can't for the life of me figure out why I chose such a low LF/MF crossover. The M15CH001 prefers to operate above 300Hz. I'll have to crank up LspCAD to have another go at replicating the passive transfer functions with the active filters.
 
I just sent you a copy of both docs. Thanks very much for taking the time and effort to work on this :)

PD, thanks for the input. I think I'm going to avoid messing with my power amp gain because the only way I'd know to do it would be to mess with the feedback resistors, and I believe the NAP series are prone to instability when messing with the feedback network; best to avoid that can of worms!

Another approch is to lower the feedback capacitor value. Currently I use 68uF films, which give a nice low corner frequency. Using 22uF should raise the corner enough to ease the load for treble/midrange power amps.

I'm glad to hear that unity gain buffer opamps won't futz with the phase - it's nice to avoid opamps where possible.

Thanks,
Carl
 
Carl,

Plan on a big cap protecting the tweeter, such that if it were the only crossover (ie big fault elsewhere / broken amp or XO) it would save the tweeter, while also making its crossover below the active crossover itself - ie no effect on the audio sent to the tweeter.

Lets put it another way ... wish I had!
:D
 
trancera said:
Carl,

Plan on a big cap protecting the tweeter, such that if it were the only crossover (ie big fault elsewhere / broken amp or XO) it would save the tweeter, while also making its crossover below the active crossover itself - ie no effect on the audio sent to the tweeter.

Lets put it another way ... wish I had!
:D
Click to expand...
I'd be more concerned about the midrange. The Seas M15CH001 is virtually unobtainium. With an x-max of 1mm, it won't take much DC to destroy what I consider one of the finest midrange drivers ever made. The Hiquphon OW-1, OTOH, can be bought in bulk, at a price, from Oskar Wroending.
 
hacker said:
I just sent you a copy of both docs. Thanks very much for taking the time and effort to work on this :)
Click to expand...
This may have to wait until the weekend. LspCAD is on the lappy I gave my daughter. I haven't installed bootcamp on my Mac yet.

hacker said:
I'm glad to hear that unity gain buffer opamps won't futz with the phase - it's nice to avoid opamps where possible.
Click to expand...
Have you considered designing an all discrete circuit that does away with opamps? I don't know enough about opamps/discrete transistors to know what to do, but someone else will.
 
No worries, there's no particular hurry - this thread is one of gestation, not immediate action.

With regard to discrete opamps, it really would increase complexity (and size and cost) quite considerably. I think the OPA2134 is a good compromise between cost and performance and keeps the PCB small and tight... and I've already got 12 of them in my parts box ;-)
 
James said:
trancera said:
Plan on a big cap protecting the tweeter, such that if it were the only crossover (ie big fault elsewhere / broken amp or XO) it would save the tweeter, while also making its crossover below the active crossover itself - ie no effect on the audio sent to the tweeter.
Click to expand...
I'd be more concerned about the midrange. The Seas M15CH001 is virtually unobtainium. With an x-max of 1mm, it won't take much DC to destroy what I consider one of the finest midrange drivers ever made. The Hiquphon OW-1, OTOH, can be bought in bulk, at a price, from Oskar Wroending.
Click to expand...

I'm using 4-pole Neutrik Speakon connectors, one per driver, and plan to wire pins 1+ / 1- direct to the tweeter, and 2+ / 2- via a large film cap so that I can choose which configuration I want by simply rewiring the speaker connector.

I don't think I can get a cap big enough to protect the midrange, which, as James says, would be nigh-on impossible to replace should the worst happen. That's why all my amps use Velleman speaker protection!

trancera said:
If going discrete then some LR version of the usual suspects maybe?

http://www.neilmcbride.co.uk/naxocircuits.html

ok the 2134 works for me too ;)

If I was designing this I would incorporate the required VBE/Regs, a little like Neils motherboard for 72 style boards. Individual boards got a little messy.
Click to expand...

I've got 4 spare ALWSRs that I plan to use for the XOs, so no on-board regs necessary. I'll trick out the upstream PSU with QSpeed diodes, 100R/100nF snubbers and LCLCLC for a nice fast, quiet supply. The XO PCB itself will have a ground plane and careful attention to local decoupling of the opamps to keep things nice and stable.
 
You must log in or register to reply here.


advertisement


Back
Top