Biamping Magnepan Typani IVa's with the XO-1
I have heard my Tympani IVa's with ICEPOWER ASP1000 monoblocks I built and also a Krell KSA-50 Clone I also built. I love what the Icepower does on the Bass, but prefer the Krell's mid and top end. So after looking all over gods creation, I found my Magnepan XO-1 Biamp crossover I had and will try biamp the Tympanis with both amps.
Now, looking inside the XO-1 there is no capacitor for the high pass (probably because Magnepan doesn't know the input impedance of your amp). The pots are there to balance the tweeter gain to bass gain and I know I will need to dial this in as the gains are quite a bit different.
For the cap value, I need to do a little math (C = 1 / (2*pi*f*R) with R being the amp impedance and f the crossover frequency for the mid/high pass), so Input impedance on the KSA-50 is 22K and the crossover is 400hz, yielding a cap sized at .0181uf.
But question is this... when the internal crossover inside the Mid and ribbon tweeter section already have the high pass cap, why would I need one on the XO-1? Am I missing something?
Attached is the Crossover Schematic for the Tympani IVa. You can see the External Crossover, and the Internal Midrange and Ribbon tweeter crossover.
I would think I could send preamp out to bass amp to External Crossover and connect bass panel to Bass output of crossover, and for tweeter, send preamp out to tweeter amp to tweeter connection on speaker...the caps should knock out all bass frequency lower than 400hz... and using the XO-1 (without HP caps) connected to bass amp , to dial in gain differential on the tweeter and bass amp.
After looking closer, it appears that the midrange has no Hi-pass cap, just the ribbon does. So I would need the XO-1 to "trim off" all sub 400hz frequency before getting to the Mid/Tweeter amp, then go to the speaker panel.
Now I have also read thet the 75uf in the main external crossover needs to be cut out of circuit...don't understand why. I will put a switch on it. Maybe thats what that Biamp Switch someone wrote in on the attached crossover schematic means....
Hey, I just noticed I am the only one posting here, and I answered all my own questions....I feel weird. Kinda like talking to yourself and everyone snickering behind your back.
I had Karen at Magnepan send me the XO-1 Crossover Instructions (I love Magnepan Customer support as much as their products!)...I had them 99% correct...Anyway, I am attaching a link to the PDF file for anyone who wants it - for the MGIIIa and the Tympany IVa from my web server...sorry for the duplicate pages...
Um, Thanks for all the help everyone!!
Given that the instruction manual you posted says "<i>for IIIa <b>and </b>T-IVa</i>" I would assume these two models must be configured similarly. By this I mean:
* my IIIas had an external XO box which had the bass LP & mid HP filters, and an "internal" XO which was the mid LP & ribbon HP filters.
* so the T-IVa must be the same?
Now the external box has 1 pair of input connectors and 2 pairs of output connectors. So I can't see why you wouldn't configure it as follows:
* wires from bass amp go into the external XO
* just connect the bass output wires to the bass driver input connectors on the back-plate on the bass panel
* don't connect any speaker wire to the mid/ribbon output connectors on the external XO.
Thus you have the bass amp connected to the bass driver.
Now for the XO-1. AIUI, this consists of a series cap for the input to the mid/ribbon amp, plus an attenuator. (So it assumes the gain on the amp used for mids & ribbons is higher than the gain of the bass amp! Which may not be the case! :-)) )
You calculate the appropriate cap value (according to the table given in the manual), solder in the cap and connect the XO-1 between preamp and mid/ribbon power amp. Then connect the output of the mid/ribbon amp to the mid/ribbon input terminals on the back-plate.
IE. no unsoldering of that cap inside the external XO box is required?
BTW, if you have both amps, why not get a 2-way active XO instead of using the XO-1? You will deliver more power to your IV-as this way, for a start ... and doing away with the passive XOs will make them sound better IMO!! :-)) And active XOs come with gain adjustment.
Thanks for your reply. If you look at the schematic for the XO-1, it is the bass amp that gets attenuated, not the mid/treble amp, which is exactly what I needed as teh ICEPpower has more gain than the Krell.
According to the instructions, that External 75uf cap needs to be removed from the system or it will cause issues with the bass amp...
I hooked everything up as detailed out and it sound OK. I wasn't blown away as I thought I was going to be. I even tried my Pass Aleph 5 Clone (without changing the tweeter cap, so highpass filter was a bit higher) and actually sounded a little better (the Aleph 5 has a 50k input impedance - raised from original 10k, and the Krell has 22k input impedance)
Next is my Behringer DCX2496 Electronic crossover. I haven't played with it in a while, so don't know what to expect, but I will report back with my findings...
Hey Andr - your probably the same Andr from MUG? Just a wild guess...
What would you recommend as a starting point on the DCX as far as crossover points? I want to keep the tweeter caps in tact so leave it as a 2 way cross for now...
Please tell me what the DCX is capable of - is it limited to symmetric slopes for LP & HP ... or can it supply, say 18dB LP for the bass and 6dB HP for the mid? :confused:
Andy, it is capable of all the following
Butterworth, Bessel and Linkwitz-Riley filters with slopes from 6 to 48 dB/octave. The LP can be a different slope and/or filter from the HP...neato huh?
I have the cross on a 18db LR at 250hz now.
You have quite a few options to try - the only thing that counts is how you think it sounds in your room.
But what you have now - 18dB LP & HP at the same frequency - is not recommended as this delivers a 3dB hump centred on 250Hz. This is because while even-order filters (eg. 12dB & 24dB) are normally described by their -6dB ("1/2-voltage") frequency - and 2 halves make a whole, odd order filters are described by their -3dB ("70% voltage") frequencies.
You need to read the DCX user manual to confirm this is the practice it uses (as it can get confusing when you are combining odd & even-order slopes).
So at 250Hz, you have 2 x 70%-signals adding - which gives you more than 1 ... ie. you will have a hump at that frequency.
NB: The suggestions I make below refer to -3dB frequencies for odd order filters and -6dB frequencies for even order.
According to lspCAD, the stock passive XO has:
* bass LP 18dB @ 250Hz (-3dB point), and
* mid HP 12dB @ 330Hz (-6dB point).
These are basically flattened Butterworth slopes.
So you might like to try this first - with the drivers connected the way they are, stock.
Then reduce the mid HP slightly, to 300Hz (still 12dB) - lspCAD gives this combination a slightly smoother overall frequency response.
If you want to try 18dB then try:
* bass LP 18dB @ 250Hz (-3dB point), and
* mid HP 18dB @ 400Hz (-3dB point).
You would keep the same driver connections with this change of the mid HP slope to 3rd order.
Again, you'd need to experiment with Butterworth vs. Bessel slopes.
Good luck. :) Tell me how it goes!
|All times are GMT. The time now is 12:31 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio