Bi amping component set with (2X) 2 channel amps

[Jonny Hotnuts]

I have a Dynaudio 240 set and are very power demanding drivers.

I acquired an A/D/S/ PQ20 to run the set with (not sure of the impedance of the set but would think it was somewhere between 2.9-4). The amp should make a clean 125-150 rms per side.

I have a chance to get another PQ20 and was wondering about bi-amping the set by using the supplied passive XO for the woofer and then use a like value cap to filter the tweeters.
I would loose the XOs 3 tweeter attenuation points and have do some wiring but think it’s very doable.

This way I could use one PQ for the woofers and one for the tweeters.

I have never done much with passives but if I remember correctly I think if impedance load as seen by the OX changes, the points of filtration change also, meaning that if I don’t use the tweeter on the Dyna XO it will change the XO point….this would be bad if true.



So am I crazy for considering such a project?????

I still think that (2X) PQ’s to the Dyans would be killer!

I only wish the PQ was a 2 channel amp….would make life soooo much easier!!!


Thanks for any advice
-JH

 

[tsmith1315]

I have never done much with passives but if I remember correctly I think if impedance load as seen by the OX changes, the points of filtration change also

That is true of 1st order filters, but only per XO section. Since the inputs to the HP and LP sections are paralleled, if you leave off one set of drivers the other section's -3dB point and slope should be unaffected.

I don't know how complicated those filters are, but Dynaudio has been known to be a proponent of 1st order crossovers with proper impedance compensation and phase correction. At least when they sold raw drivers.

I wouldn't be surprised to see a very complete filter inside your boxes, certainly more than a cap in the tweeter circuit.

If you want to continue to use the passives, it would be worthwhile to try to isolate the HP and LP sections and either use them in their entirety or make a duplicate of one section for the other amp.

Does Dynaudio disclose any specific information about their crossovers?

A PQ20 bridged on a pair of tweeters... Is this a great place or what?

Tim

 

[Jonny Hotnuts]

I have found another set of "X-250" passive XOs on fleabay (same that comes with the 240 Dynas).

You are correct that they come with more than one cap for the tweeter....

Would I be OK using one set of the XOs for the Mid and the other set of OX's for the tweeter????

Seems like this would be toooo easy!

 

[ppia600]

I would let the tweeters use the crossovers as is, and replicate the lowpass sections for the woofers. You could also do like I am and run one amp for the right and the other for the left channels to help seperation. You'll still be able to independently control level with the deck either way.


(oops, didn't read your last post entirely.. that seems like a winner, you were lucky to just pick up the crossovers, good deal 🙂 )

 

[ocool_15]

You could passively bi-amp them. Assuming the crossovers are parallel type there should not be an issue.

Buying a second set of the same crossover would be a waste. Without a driver pair connected on to each crossover both amps would see a essentially a short at some frequencies with most crossover setups.

Passive Bi-amping may not be worth the expense.


Rod Elliot information on Bi-amping

 

[tsmith1315]

Without a driver pair connected on to each crossover both amps would see a essentially a short at some frequencies with most crossover setups.

Looking at the network from the speakers' perspective, yes. But, looking from the amp's perspective:

Any component that would be a short or low impedance at some frequency will have another component in series that is open or high impedance at those same frequencies. An exception could be a zobel on a 1st order tweeter network, but that would still have a resistor in series.

The end result is there will be a set of reactive components still in circuit demanding some amount of power and consideration from the amp in the frequency band of the set of speakers still being used on that crossover.

The easiest way to remove this concern is to disconnect the first series component (cap) from the unused tweeter network, and the first series component (inductor) from the unused woofer network. At that point, everything following will be out of circuit.


Tim

 

[Jonny Hotnuts]

(pic of XO)

The easiest way to remove this concern is to disconnect the first series component (cap) from the unused tweeter network, and the first series component (inductor) from the unused woofer network. At that point, everything following will be out of circuit.

I measured the impedance for both drivers (3 ohms each).
Any idea if removing the unneeded components from the respective XOs will change the load as seen by the amp?

 

[ocool_15]

Looking at the network from the speakers' perspective, yes. But, looking from the amp's perspective:

Any component that would be a short or low impedance at some frequency will have another component in series that is open or high impedance at those same frequencies. An exception could be a zobel on a 1st order tweeter network, but that would still have a resistor in series.

The end result is there will be a set of reactive components still in circuit demanding some amount of power and consideration from the amp in the frequency band of the set of speakers still being used on that crossover.

The easiest way to remove this concern is to disconnect the first series component (cap) from the unused tweeter network, and the first series component (inductor) from the unused woofer network. At that point, everything following will be out of circuit.


Tim

I am talking from that amp side.
As an example what is a 12dB /octave filter with no driver connected? At a certain frequency the reactance will be equal between the C and L. It would not be a complete short primarily due to the DCR of the inductor but I would not do that to my equipment. I said most crossover setups as some will not have this risk like a 6dB oct filter.

 

[ocool_15]

(pic of XO)





I measured the impedance for both drivers (3 ohms each).
Any idea if removing the unneeded components from the respective XOs will change the load as seen by the amp?

Assuming you want to keep the same sound you would not remove any inductor or capactior from there. If you are going to passively bi-amp you could likely remove some resistors if they are in a L-pad configuration and use the amplifiers gain to adjust.

You should show the bottom of the crossover pcb and probably draw it out similar to the link below if you want a better understanding of what parts of the crossover are doing what.

Parts express layout

 

[tsmith1315]

As an example what is a 12dB /octave filter with no driver connected?

Ocool, you are quite correct. There would be resonance with a diving impedance around the particular frequency where reactances are equal.

Assuming you want to keep the same sound you would not remove any inductor or capactior from there.

Removing or disconnecting unused components would open the unwanted tweeter or woofer circuit. That would leave only the filter for the speaker being used in effect.
In *absolutely* no way would that have a negative effect on the remaining speaker's sound. The only effect it will have is to remove the load of the unneeded components from the amp.

For some reason, I can't get the crossover picture link to work. I can't even get Fiat Forum to load...

Tim

 

[Jonny Hotnuts]

This may be the biggest idiot question of the year but what would happen if I were to put a 3 ohm resistor (impedance of the drivers are 3 each) on the crossover output for the respective driver as a "dummy" load on the XO and leave everything there?



BTW:
Dont know why youre having probs with the pic....works fine for me.

 

[Jonny Hotnuts]

I’m committed now, here’s No 2, bit more than I wanted to spend but you dont see these everyday!

http://cgi.ebay.com/ws/eBayISAPI.dl...id=m37&satitle=+180201240891&category0=&fvi=1

 

[tsmith1315]

what would happen if I were to put a 3 ohm resistor (impedance of the drivers are 3 each) on the crossover output

It would really just be a waste of power. Effectively the amp would still be delivering full power output in the unwanted frequency band, only to be dissipated by the resistor instead of converted to audio by a speaker. It would also take a resistor capable of dissapating that amount of power. Large, expensive, and hot at 70W.

The biggest advantage IMHO of using two amps over one along with passive XO's would be in dynamic range and clarity at higher volumes. That's simply because of having separate power supplies for each set of drivers.

I got the picture to load, and it looks to be a 12dB on HP and LP with Lpad for tweeter and no impedance compensation. Pretty simple. If you can post a shot of the underside and the terminal labels, we can easily figure it out exactly.

The big Solen polypropylene cap is in the tweeter circuit, and the yellow electrolytic cap is in the woofer circuit. Chances are you can remove the cap from the unwanted circuit and the remaining components would be out-of-circuit if nothing was connected to the corresonding output terminals. Again, a couple more pictures would reveal all.

An electronic crossover would be more effective because the amplifiers' work would be entirely dedicated to the desired frequency band. Given what you already have invested, it shouldn't be a major expense. If you consider the option, be sure to look for something with resistor/chip programmable filter frequency instead of pots for more precise control.
With that kind of power on those tweeters, you want to be dang sure your HP frequency isn't 1kHz lower than what you think you've dialed in.

Tim

 

[Jonny Hotnuts]

It looks like an active XO would be a better way to go, I didnt even consider going active (I need an active XO anyway).
I have never used active filtration without using the supplied XOs, always felt that it was somehow bad for tweeters.
It has been a million years sense I was looking for an outboard XO….is there any you would recommend????

BTW: Good suggestion with the chip and not pots….but I always loose them and live to tweek pots.

 

[tsmith1315]

I have never used active filtration without using the supplied XOs, always felt that it was somehow bad for tweeters.

Kinda feels like there's no safety net?

The most immediate concern is that there is no series capacitor to block DC. And while that could kill any driver, tweeters are more sensitive with their little voice coil wires.

You can always add a cap inline with the tweeter to remedy this. Choosing the value isn't critical, but some thought should go into it.

Otherwise, the concern is having a variable crossover frequency. With premade passives, you can't run them out of their safe range.

As far as playing with the pots go, I'm right there with you. The fun really begins after everything is installed and you get to tune it. However, I don't trust that the silkscreened writing is position-accurate enough on the crossover cases to risk expensive tweeters.

More on particular units later...

Tim

 

[Jonny Hotnuts]

As the mids are much easier to dial in requiring the Hz and are more sturdy than the tweeters couldn’t I use the supplied passives in conjunction with the actives and use the active filtration for the mid without having to purchase an additional set of XOs?

This way I could take advantage of the Dynas tweeter attenuation option.

 

[ocool_15]

As the mids are much easier to dial in requiring the Hz and are more sturdy than the tweeters couldn’t I use the supplied passives in conjunction with the actives and use the active filtration for the mid without having to purchase an additional set of XOs?

This way I could take advantage of the Dynas tweeter attenuation option.


[/IMG]

I have a pair of ADS crossovers that that have a very similar layout to this at least for the woofer circuit.

You will be affecting the sound as active crossovers and your passive are shaping the response differently. I haven't followed car audio for a while but 24dB/oct linkwitz-riley and 12db/octave butterworths were most common, which probably wont match the passives electrical response. It may even sound better using the active filter but I would use an RTA and mic to measure the affect.

The question mark in your layout should be a poly switch for protection of the tweeter.

 

[jpruden]

Hi Jonny,

Damn. I was gonna buy that one... 😉

Dynaudio used to recommend using the high pass part of their crossovers for the tweeters to protect them from dc offset... basically confirming what tsmith1315 typed above. You can just hook the PQ-20 to the Dyn crossover and hook it up to the tweeters with no problems. If you'd prefer to change the xover point, you can just use a capacitor to block the DC offset rather than the whole crossover thing.

AudioControl makes very nice external preamp level crossovers, but you should remember that the crossover slope on them is 18 or 24 dB per octave. You can find a 24xs at a very attractive price on eBay. In fact, I was going to do *exactly* what you're doing right now when I originally started collecting equipment to do my ultimate system. I've long since gotten rid of both PQ-20s, but I still have the Dynaudios and the AudioControl crossovers... I ended up going with a McIntosh MC440 for the amp and not bi-amping the front stage.

These days, I've gone to JL Audio Slash series amps for mo' power... but I find it terribly amusing to see someone else coming up with the same solution for some serious sound.

 

[Jonny Hotnuts]

jpruden, make no mistake it was my original intent to get a Mtosh. and made a few bids in the 7 range and came close a few times but decided a PQ would be an acceptable substitute after prices were going in the 1K+. After getting a PQ the idea of 2 seemed like it would take things to the next level (and hopefully take some of the sting out of having to settle).

Still wish I had McIntosh amps though….but a very clean 300 watts per side of ADS power seems like it shouldn’t be too hard to live with.

BTW:

There is currently another PQ for sale on eBay now, and is being sold by a member of this forum.

http://cgi.ebay.com/ebaymotors/ADS-...ryZ38638QQssPageNameZWDVWQQrdZ1QQcmdZViewItem



I am looking very hard at the Audio Control 6XS.

 

[tsmith1315]

Jonny,

Here’s a redraw of the schematic in a more “standard” XO look, using the part numbers you designated. The sections are very recognizable.

The 27uF and 2R2 are there to flatten the woofer’s rising impedance, and L1 is the actual first order filter.

First order filters are Butterworth filters with a Q of 0.707. This can’t be varied.

“therm” is the protection device.

C2 (10uF) and L2 are the tweeter’s actual filter while R2, R3, and R4 form the attenuator for the tweeter.

Being a 12dB/2800 Hz filter, using C2 of 10uF, L2 will be about 0.3mH. The Q, or response shape of the filter in the crossover region will be determined by the impedance of the tweeter in that frequency range. Looks like the MD100 is pretty close to 3 ohms around 2800Hz. That puts the filter Q around 0.53 with these component values. A Linkwitz-Riley filter is represented by a Q value of 0.5. A Bessel filter is represented by a Q of .577. You’re within a few percent tolerance of either.

So within a reason given standard value components, you have a 2nd order filter with a conservative Q in the mid 0.5 range at 2.8kHz on the tweeter and a 1st order Butterworth at 2.8kHz on the woofer. Assuming the specified 2.8kHz crossover frequency is close, that is.

All that said, I wouldn’t beat my brains out over choice of crossover slopes or trying to duplicate Dynaudio’s exact filter response. Looking at the x250 info on their website, it’s a bit generic so as to work with any one of 4 woofers. In other words, it would be pointless for the passives to have been designed/tuned with an exact target response in mind, with the infinite possibilities between woofer choice and installation variables.

Yet Another option:
The 7” woofer rolls off pretty smooth above 3kHz on its own. Let it roll off naturally, and only use an active section on the tweeter.

IMHO, a good quality active crossover really should have a bipolar power supply, the control features that you want, and be quiet. That’s about it.

That’s not to say it must have a bipolar supply, the ADS 642CSi was a well-liked 12V unit. It was a 4-channel two way but IIRC, had independently selectable F/R HP points and Q factors by DIP resistor pack. If that’s the case, you could HP the 7's at the sub LP frequency and let them roll off naturally on the top end, and the HP the tweeters at 2.8kHz & Q of your choice with one of these.


Tim