Passive Crossover bypass – Quest / reQuest

jedi35 said:

…

Could you tell us more about this DSP-based speaker processor(brand, cost, functions, etc.)? Do you think that ML would share with me all the info about EQ, phase, and whatever else the passive Xover is doing in my Quest Zs? Since the high voltage traces are mixed into my crossover board, I was thinking that maybe I should ask ML if they can swap my internals/back panels for the older Quests cabinet backs, if they have any still sitting around. What do you think of that? Atleast that would give me the ability to bypass the internal crossover. Perhaps they would send them to me already setup for active bi-amping.

Click to expand...

Adrian, I used the DBX DriveRack 260 for speaker processing duties. I’ll refer you to my center channel thread for further details on using it to set all the parameters.
You really do not need any info from the factory other than crossover points as everything else will be different with the active crossover, but you must have a good quality measuring rig. Here is a nice long thread on that listing what gear one can use:
http://support.supermegaultragroovy.com/wiki/index.php/Hardware_Suggestions

Rather than swapping panels, I’d ask the factory if you could send in yours and have them mod them for you (for a fee I’m sure), that might be best and quickest.

jedi35 said:

…
In your email to me, you mentioned that it doesn't hurt anything to run from my external Xover into the internal passive one, as long as I stay well enough above the range of the passive so that I don't double up on things and cancel frequencies. 1) Isn't it true that I'm running through 2 Xovers that feature "a bunch of power robbing, frequency and phase skewing passive components"? This means that I'm losing a lot, right? 2) My Quest Zs are crossed over at 150 Hz. You suggested that I should set the external LP at something like 315 Hz. I did try that, and the panels didn't sound too good to me. Bass was weak, but maybe that was because I left the woofers set too low. After several tests, I went the other way and set it below the passive, at about 100 Hz. My woofers, which are directly connected to the Xover, are set at 200 Hz. My panels sound more life-like with these settings, and the bass is quite good. I figure that if my Krell monoblocks don't see anything below 315, then I'm robbing the panels of any chance of seeing signals in the 315 to 150 range, which they should be getting. Setting the woofers at 200 should cover any dips at the Xover point of 150.

Click to expand...

It won’t hurt to have the passive there, but it doesn’t help much either. As long as you are a good bit away from the original crossover point, the overlap is not horrible.

The reason it sounded weak to you is you did not move up the low-pass to the woofer to a similar point. So you had a huge mid-bas gap.
Moving the other way, you now had an overlap between 100 and 200hz, which should have sounded a bit better in the mid-bass, but again, without control of Phase and delay, who knows how well it’s really working. As I mentioned above, you really need a measurement setup to accurately gauge what needs to be done.

jedi35 said:

…
As I'm typing this, it occurs to me that what you probably wanted me to do was set both the panels and the woofers to 315, so the woofer would be covering whatever frequencies that were meant for the panels. Having the woofers playing higher than normal would give a warmer mid-bass. I haven't tried that yet, but I will. This setup would have to be pretty darn good, because I love what I'm hearing right now. My Quests sound better than they ever have.

Adrian

Click to expand...

Great to hear there is some improvement. More improvements can had for sure, just keep persevering.
A call to factory next week about them modding your x-over panels should be high on the ‘to-do’ list.

HK-Steve said:

I have modified the Quest z crossover schematic for those who are interested...

Cheers
Steve

Click to expand...

Steve, great job!

Thanks for the schematic and the pics, most helpful

Jonathan,

Thanks very much. I greatly enjoy your detailed answers. I can tell that you really love Martin Logans, and helping us learn. Of course, it seems that a little more knowledge always brings about more questions. You say that I'll need an excellent measuring rig to get the results I need for my system. What about the mike that is used for measurements when plugged into the DBX processor? Combined with the computer, won't that tell me what I need to know about room and speaker response, phase, EQ etc.?

I called Jim Power about bi-amping my Quests about a week and a half ago. He seemed interested in helping me. First of all, ML doesn't have anymore boards for the Quests. They are willing to effect whatever repairs need to be done on them, but that's all they can do. He had some concerns about arcing when driving the transformer directly, as shown in the Quest Z mod schematic posted by Steve. He said that the Quest Z is already borderline, and felt that there is a danger of the high voltage jumping traces if there aren't any inline resistors put in place. He sent me a different mod schematic from the head ML tech, which I'll email to you, along with a pic. I'm hoping that you'll post it here for me, as I don't know how to post pics. Here is the email that he sent me:


Hello Adrian,
Our chief engineer has the following thoughts. See below. Also see
attached.

-----Original Message-----
From: Joe Vojtko
Sent: Friday, July 13, 2007 9:31 AM
To: Jim Power
Subject: RE: Quest Z Xover

Hello
He could add a resistor as shown in diagram. This prevents 2 bad
items

1) DC short circuit through transformer primary
2) Stat panel capacitance - Transformer leakage inductance resonance
causing zero ohm input impedance around 25kHz. This can destabilize some
amplifiers.

My assistance should not be misconstrued of endorsement of these
activities. Modifying our product may result in severe injury or death
and is not recommended by MartinLogan.

Joe V
In addition, I would offer the shown assembly modification to the audio
transformer output to offer greater reliability by means of better high
voltage isolation. Under extreme levels, the 10K resistors have the
tendency to arc over to other components damaging trances and the
circuit board as a whole. This hard wire mod offers greater spacing to
the circuit board and other grounds. The photo shows an installation
using different terminal types between the stator panel connection and
the audio transformer output connection.
Good luck and be careful.
jp


OK, so how much danger am I really in? If I follow the advice from Jim and Joe, should I be ok? I'm figuring that for reasons of liability, certain things have to be said about modifications. I'm willing to take the risks as long as there is good reason to believe that I won't blow up my speakers or kill myself in the process. I'm not as concerned about the EQ issues now, because I will endeavor to get a speaker processor like yours. I'll retry your crossover point suggestions, and this time I'll make sure to raise both the high and low pass to compliment each other.

Regards,
Adrian

jedi35 said:

Jonathan,

Thanks very much. I greatly enjoy your detailed answers. I can tell that you really love Martin Logans, and helping us learn. Of course, it seems that a little more knowledge always brings about more questions. You say that I'll need an excellent measuring rig to get the results I need for my system. What about the mike that is used for measurements when plugged into the DBX processor? Combined with the computer, won't that tell me what I need to know about room and speaker response, phase, EQ etc.?
...
Regards,
Adrian

Click to expand...

Hi Adrian,

Yes you only need one set of measuring tools. The PC is used as the foundation of the measuring solution. I actually never plug in the mic to the DBX (even though it supports it) as the Software based measurement tools are much more flexible and easier to use.
This is will be clearer with the DBX setup doc I'm working on. I might send you a draft, as it covers some of this early stuff on what's needed.

I'd agree with Joe V that bit of resistance before the ESL transformer is warranted for both safety and for protection of the amp. He's not kidding when he mentions the impedance dips waaay low, and combined with the reactive load of the ESL, well, some amps do not like...
But also, there's a bit of CYA there as well, they have to say those things.

Boy, modding the QuestZ seems like a good bit more work than any of the other models. Using Circuit boards is what makes it so onerous.

Keep pushing, and soon, you will have amazing sound.

Cheers,

Here is the diagram Adrian got back from ML.

As I note in the original post, having some resistance in line with the ESL transformer is a good idea if you are unsure about the amp.

For a Quest, I'd recommend 60w or more of rated resistance. This can be accomplished by ganging in parallel three or four 20w 0.5 ohm resistors.

And here is the picture of the mod recommened to Adrian regarding the resistors near the ESL panel feed block.

MartinLogan said:

I would offer the shown assembly modification to the audio transformer output to offer greater reliability by means of better high voltage isolation. Under extreme levels, the 10K resistors have the tendency to arc over to other components damaging trances and the circuit board as a whole. This hard wire mod offers greater spacing to the circuit board and other grounds. The photo shows an installation using different terminal types between the stator panel connection and the audio transformer output connection.

Click to expand...

Thanks for posting the pics, Jonathan. I got my variac today. I'll play around with it and let you know what I come up with. Also, last night I experimented with the higher Xover points that you suggested. The midbass from the woofers increased so much that I had to back the level down. I think I'm now getting a much more open soundstage with more hall presence. The lower midrange and midbass are more prominent, with good low bass extension as well. I'm confident that things will be fantastic once I get the speakers full modded, and after I start using the speaker processor.

Adrian

jedi35 said:

Thanks for posting the pics, Jonathan. I got my variac today. I'll play around with it and let you know what I come up with. ...

Adrian

Click to expand...

Adrian, that's cool.
Please post your impressions and experience with the Variac in this new thread I started on the topic of "more volts = more / better sound?"

I also asked the Mods to move Greenipsum's post as well as mine on the topic of internal AC step-up mods to that thread.

Great to hear your test with the higher crossover points were satisfactory. It can only improve, as the match between panel and woofer really requires some delay tweaks as well as phase for best integration.

My buddy won't be able to get back over to complete the mod to my speakers until the end of August, probably. I also need to save my pennies and get the DBX DriveRack unit. I'm sure that all this will open a new world of sound for me. Bad news on the Threshold Model 4000 amp I put in for service. It appears that both channels are shot, and the repair costs are projected at $600-900. Wow, that is going to set my schedule back.

Adrian

best upgrade for all Quest´s

the best upgrade for all QUEST´s is to replace the original ( very poor) step-up transformer with a new one. You need a 1:50 Type . Best step-up transformer was developed from Menno van der Veen (http://www.mennovanderveen.nl/eng/index.html) He is a specialist for Tube amps and for transformers. He developed this step-up transformer two years ago. My REQUEST now sounds at least as good as the Summit, it´s an enormous improvement. You can buy it at http://www.plitron.com/shopping/shopexd.asp?id=376# or http://www.amplimo.nl/ .

Klaus

Hi Klaus,

Wow, you are in Heidelberg? I played violin for an opera festival at the beautiful castle ruins there, back in the summer of 1983. I was in music school in New York, and the orchestra went there every summer for 6 weeks. It was my first time out of the country. I've been back to Germany a few times since then. It was always a wonderful experience.

I've basically dropped my plans to bi-amp my Quests, since I'm in the process of buying a pair of Monolith IIIa speakers, which are already setup for active bi-amping. However, I'll probably keep my Quests and use them for a super rear channel in my surround setup(I know that's sick, right?). Still, after I've gotten the Monoliths up and running, it might be fun to upgrade the Quests like you suggested. I'd love to find these transformers in the states. Anybody know of a place?

Regards,
Adrian

Thanks for the tip.

I've been thinking about this a for a while. Good pointers to the Plitron site.

I'd like to point out to all a good article on the general topic of step-up transformers at that site: http://www.plitron.com/pdf/atcl_11.pdf

This is new area of decisions for me ! Now With my new rebuild of my Quests I may look into fully Bi Amping or new and better parts !

I may look into something like this ! I like the idea of even more tweaks to the newly rebuilt Quest's!

JonFo said:

This applies to most models of first-gen ESL speakers from ML.

Click to expand...

Has anyone tried this with Aeirus i? Are there are any special problems? I'm getting ready to try it and any help will be greatly appreciated.

I currently have them biamped using a Classe 70 for the panels and a B&K 2020 for the lows. A Rane AC 23B is crossed at about 600 Hz. I use Smaart 6 for adjustments and calibration.

They sound excellent as is but I would really like to check them out with the passive completely bypassed.

Thanks!
Scott
:bowdown:

ScottGardner said:

Has anyone tried this with Aeirus i? Are there are any special problems? I'm getting ready to try it and any help will be greatly appreciated.
...

Click to expand...

Scott,

The Aerius should be exactly like the other, the first diagram I posted (showing the added 0.5 to 1 Ohm resistor) should work as well on the Aerius.

Again, other than checking for polarity reversals between woofer and panel, the bypass is very straightforward.

To minimize risk to your speaker and allow you to go back to passive (for a sale or something), just buy some additional resistors and clip and tape up the old passive component crossover leads.

So the steps would be:

Follow the general steps in the Quest bypass listed in post #1

Once you have the crossover / HighVoltage stepup board out, see good pics of this that jjcarr took in this “Aerius Tweaks” thread, take your own pictures, study the wiring so you understand which wire is what (in terms of what is feeding the transformer). Then plot and document your mods on paper before snipping any wires.

If you want, post that before proceeding and we can review.

Once you clip out the passive, wire in the negative to the negative input of your transformer (wire colors change with model of transformer, so I can’t tell you which unless I have detailed pics of YOUR board).


The positive would go from the terminals to a bank of three paralleled 0.5 ohm 20W resistors, and from there to the hot lead of the transformer.

Woofers are easy, positive to positive, neg to negative. These might need to be reversed to match the panels polarity post-bypass.

Since you have a measuring rig, check the impulse response of both the panel and the woofer.

If running these with a center channel, also check that the Aerius panel polarity is equal to the polarity of the center (I.e. both signals show positive going spikes first).

If you need to adjust panel polarity, do so by swapping the stator leads (front / back). Note: only do this if there are NO resistors between transformer and stator. Models with resistors on the rear stator feed MUST keep them on the rear stator. I don’t believe the Aerius had resistors between the tranny and the stators.

Now, measure measure, measure.

Your Rane crossover will be fine for starters, but its delays are only on low freq. outputs, and typically one needs delay on the highFreq. for the panels to time-align to the woofers.

A DBX DriveRack PA or 260 would be the next step in that component. The fact that they also have Parametric EQ is a huge advantage. The factory passive do perform some amount of EQ correction, so your external x-over needs to be able to supply that to get back to factory specs. The nice thing is one can improve on the factory spec performance with a DriveRack.

Cheers,

Jonathan,

Thanks I really appreciate it. For anyone interested Jim Power at ML has provided me with the Aerius I Crossover Schematic. He has requested that it not be published but I can answer questions if you PM me.

Scott

Wow

I upgraded my hardware so I'm now using a Sunfire TGA-5200 Amp and a modified Behringer DCX2496 Xover

I finally did the bypass on my Aerius I's. WOW the change is dramatic. The clarity is awesome, they really sparkle now. The soundstage is so deep I don't think I'll ever get out. The low end is soo much tighter. I just love this setup now.

The bypass was very easy two cuts and two solders. No resistors were used and the sunfire dosen't even get warm.

Thanks again JonFo!

Hello JonFo,

I am using the SL3 with a sub in a two-channel system and planning to go active. Based on your statements regarding the 260 I am considering it. Can you give any feedback regarding the DBX 260 or the Behringer DCX2496? If you have any calibrated microphone and USB preamp recommendations, it would also be appreciated.

Can you give any feedback based on your experience with Fuzzmeasure and REW? Especially if you have used either with OS X.

Thank you for your help.

JonFo said:

Scott,

The Aerius should be exactly like the other, the first diagram I posted (showing the added 0.5 to 1 Ohm resistor) should work as well on the Aerius.

Again, other than checking for polarity reversals between woofer and panel, the bypass is very straightforward.

To minimize risk to your speaker and allow you to go back to passive (for a sale or something), just buy some additional resistors and clip and tape up the old passive component crossover leads.

So the steps would be:

Follow the general steps in the Quest bypass listed in post #1

Once you have the crossover / HighVoltage stepup board out, see good pics of this that jjcarr took in this “Aerius Tweaks” thread, take your own pictures, study the wiring so you understand which wire is what (in terms of what is feeding the transformer). Then plot and document your mods on paper before snipping any wires.

If you want, post that before proceeding and we can review.

Once you clip out the passive, wire in the negative to the negative input of your transformer (wire colors change with model of transformer, so I can’t tell you which unless I have detailed pics of YOUR board).


The positive would go from the terminals to a bank of three paralleled 0.5 ohm 20W resistors, and from there to the hot lead of the transformer.

Woofers are easy, positive to positive, neg to negative. These might need to be reversed to match the panels polarity post-bypass.

Since you have a measuring rig, check the impulse response of both the panel and the woofer.

If running these with a center channel, also check that the Aerius panel polarity is equal to the polarity of the center (I.e. both signals show positive going spikes first).

If you need to adjust panel polarity, do so by swapping the stator leads (front / back). Note: only do this if there are NO resistors between transformer and stator. Models with resistors on the rear stator feed MUST keep them on the rear stator. I don’t believe the Aerius had resistors between the tranny and the stators.

Now, measure measure, measure.

Your Rane crossover will be fine for starters, but its delays are only on low freq. outputs, and typically one needs delay on the highFreq. for the panels to time-align to the woofers.

A DBX DriveRack PA or 260 would be the next step in that component. The fact that they also have Parametric EQ is a huge advantage. The factory passive do perform some amount of EQ correction, so your external x-over needs to be able to supply that to get back to factory specs. The nice thing is one can improve on the factory spec performance with a DriveRack.

Cheers,

Click to expand...

TYY said:

Hello JonFo,

I am using the SL3 with a sub in a two-channel system and planning to go active. Based on your statements regarding the 260 I am considering it. Can you give any feedback regarding the DBX 260 or the Behringer DCX2496? If you have any calibrated microphone and USB preamp recommendations, it would also be appreciated.

Can you give any feedback based on your experience with Fuzzmeasure and REW? Especially if you have used either with OS X.

Thank you for your help.

Click to expand...

I would definitely recommend the DBX 260 over the DCX (I've had both).

I've posted pretty complete instructions for setup of the 260 along with discussions with other members on the process of measuring and setup.

Do a search for posts I've made plus DBX or 260.

Here's the most recent thread:

How to properly set up the DBX 260 speaker processor

As for measuring tools, I've use REW and recommend it, and while I've not used Fuzzmeasure, I really like their UI and feature set. So either one would be good.

The Tascam 122 USB interface and a measuring Mic and stand is all you'll need.