AR-9 Refurbishment

[SHOKDU]

Well, I'm going to skip responses to the spiked feet issue, as it is a very mixed bag of opinions. All I can say, the granite slabs are functioning just as intended.

My system fully powered and driven by Emotiva is up and running. These AR-9's sound absolutely fantastic. I love listening to them, my neighbor does too. He came over and looked at my "Maxell Guy" photo hanging on the wall, he said...."that's how I feel right now, blown away...."

Enough said. Time to listen again.

[Joel]

Shokdu- does your neighbor like listening to them from his house, or yours?

I recently got a Emotiva USP-1 preamp. I like it quite a bit. Compared to my Nikko preamp, it seems much cleaner especially in the upper range. The only thing missing is bass and treble controls, but I never really used them on my old preamp either.

-Joel

[SHOKDU]

Still loving the Emo and AR9's!!

The XSP-1 has the controls you are looking for...and most likely a price tag you are not looking for. Haha.

Time to listen again.

[SHOKDU]

While the results of the granite plinths and rubber isolators was very satisfying, I may just have to get some spiked feet to test out. I will still use the granite as a plinth, as I very firmly believe that coupling a speaker to a wooden floor is a poor choice, you're sending all that bass directly into the frame of the house!

Another change I will be making is removing the switches from the circuit. The switches, or possibly the resistors, are noisy and I should have replaced or eliminated them in the first place. Now I know, and am wiser for my second pair of AR9 rebuild.

[SHOKDU]

Btw, the neighbor doesn't mind listening from his house, but he prefers to be enjoying the concert from my sofa inside my home!

My grandma was less than thrilled when I cranked em up to about 500 wpc at our thanksgiving meal though!!

[SHOKDU]

Some more photos to share:

I couldnt resist hearing the first 9 when it was finished, so here it is sitting next to my JBL L36 and Polk Monitor 7b's, and playing off of a Yamaha M-70.

WIth grilles on

After assembling my flexy rack

Another view

And another

A couple night shots, and after adding the billet pieces to the Emotiva's again...

I had to go to work for 2 weeks and leave the setup, but upon my return home I am going to stain and poly one more shelf to add to the rack. I'll leave the AudioControl C101, blu-ray player, and the ERC-2 (cd player) on a rack. I'll move the USP-1, DVR box, and the Sony XDR-F1HD to the proposed 3rd rack, and may add an XDA-1 or XDA-2 (DAC) as well. I think having a fourth shelf in there will fill space a bit better and make things look more symmetrical.

Hope you guys enjoy the yet again, terrible LQ pictures. I'll have to bust out my fancy camera sometime and take more pictures when the system is "fully complete".

[SHOKDU]

I see you left your attenuator pack in place. I took mine out. Didn't want those switches in the circuit. Not sure that this makes any sonic difference.

When you removed your switches, what kind of work did that detail? I can not find a schematic for the switches, but photos I've found seem to point to having only 4 wires that go to the switches. Is that right? Did you just disconnect them, or what did you do to remove them from the circuit?

Thanks!!

[Mar]

Very good work. Nice set up. Kudos to you!

[rrcrain]

My opinion on removing the switches, based on my own experience, is don't do it. If I hadn't thrown them away, I'd have reinstalled them by now. Not that my speakers sound bad without them, it's that I can no longer adjust the speakers without an EQ to match room acoustics.

[Robert_S]

I can not find a schematic for the switches, but photos I've found seem to point to having only 4 wires that go to the switches. Is that right?

If you do decide to take the switches out of the circuit, it looks like it would be a very simple job to bypass them, with the option of leaving them in place should you later wish to reconnect.

For reference, the AR-9 Crossover Assembly Drawing:

Robert_S

[SHOKDU]

I will certainly be removing them, potentially replacing them at some time. I can hear audible static through the tweeter and the UMR, when i move the switches to different positions the static/noise becomes more/less audible. I believe the problem lies with the old resistors, or maybe it is just old switches that need replacement.

I've got the AudioControl C101 for an EQ, if I ever need one.

[SHOKDU]

So, after studying the wiring diagrams and the photo that was provided, thanks Robert, I was able to draw the connections that I believe should be made. I am human, and do error quite often, so could somebody confirm this for me?

PS - be warned of wire color discrepancies between drawings.

[xmas111]

So, after studying the wiring diagrams and the photo that was provided, thanks Robert, I was able to draw the connections that I believe should be made. I am human, and do error quite often, so could somebody confirm this for me?

Looks good to me.

I did the same thing with my original 9's.

The set I rebuilt a few months ago I decided to let the switches intact. I took the switch pc board out and Dexoited the switches. They worked fine after putting everything back together.

The worst part was putting the damn switches back in.

John

[Mach3]

I took the attenuator switch panel out of my 9's for two reasons; 1) the switches (or the pots for other AR speakers) are always a point of noise, particularly as they have aged and 2) I rewired my entire cabinet for I felt that the original wire (from circa 1980) was barely adequate for the job.

Since 1979 the hobby has "learned" (and this has been accompanied by a great deal of bogus advertising and absurd claims) that wire, whether it be speaker wire or IC, does make a difference in the sound. Sometimes a signficant difference though usually a small difference but still a clearly discernible difference. The factory wire looms with all their spade clips and small gauge were, at least in my mind, simply not up to the job - particularly on the large drivers (woofers and LMR). The factory wire was chosen for production reasons (cost and ease of use), not for sonic performance. Why should I default to what a production manager in 1979 decided? In that era all wire was the same - as were all amplifiers, et cetera.

As I was into the cabinet I decided to remove all the factory wiring and replace it with runs of 14 gauge "flat" wire. For the woofers I used double runs of the same wire. I also soldered all my connections - no more noisy, failure prone spade clips.

Once I made this decision it was more or less a given that I would not be using the resistor pack. Mainly because I couldn't determine how to remove the resistor pack and its switches and soldering at that location was impossible - at least for me.

How did this turn out? Well it is hard to determine how much difference the heavier wire and soldered connections actually made. The replacement of all the original electrolytic caps with film devices (Solen and Jantzen from Parts Excess) radically changed the sound of the speaker for the better - in fact it was a totally different speaker in terms of sound character. Therefore it is hard to actually determine what contribution the new wire made.

The downside of all this was that within a day of bringing the speaker online I was back inside adding some attenuation to the UMR and the Tweeter. It came out a little too bright for my taste and room (my default start point was to run the thing with NO ATTENUATION on any driver). Man it was smoking - bright. BUT please note that this sound character is driven largely by my room and my taste. YMMV greatly.

Right now I have it dialed in at about 5 dB of attenuation on the UMR and 3 dB or so on the tweeter. Mine are still somewhat "there" in the upper registers - I have been toying with the idea of upping the resistor on the UMR to its full limit. But please remember this is taste and room driven.

What I wish I had done - and I toyed with the idea - was to move the entire crossover outboard. However at the time I rebuilt the crossovers and rewired the speaker I was doing so with the idea that I was going to sell the things and I didn't see an outboard crossover as a strong selling point. The sound became so much better after the rebuild (it became a completely different speaker) I sold my Von Schweikerts (VR-4SR) and kept the 9s.

I will eventually move the crossover outboard. Though going into the thing to modify the attenuation resistors (which now sit on the board with the rest of the crossover) does provide me with the opportunity to "rotate" my woofers (keep spider sag from happening).

My history with these speakers was interesting. When I first purchased them and fired them up I was underwhelmed to say the least. The sounded dark, at times tinny, very "boxy" and were just very poor sounding speakers. They were quickly relegated to the garage where they gathered sawdust for a couple of years. I rebuilt the crossovers to help me move them out to some other person.

Since the rebuild I have compared them directly to several so-called "high end" speakers. To my taste they put most such creatures way into the shade. They trounce in overall sound quality almost anything from Von Schweikert (a very poor speaker designer to my taste), run easily equal with the Revel Ultima Salons ($22k and a very good speaker), and have their way with anything from ProAc or Harbeth or Spendor. I prefer them to the Wilson W/P model 7 (though the Wilson Sasha is a different animal altogether).

The Nines are superb speakers - if you own a set you are indeed fortunate. But remember that their glory is only accessible with original drivers (that work) and a rebuilt crossover.

I still have plans for the Nines. I want to hear how they sound with the 2500 uF caps on the woofers removed. That capacitance was designed in for one purpose only - to keep the load presented to the amplifier from dipping too low (around 2 ohms without such caps). In 1979 amplifiers had a hard enough time with four ohm loads - two ohms would have popped most amps of that era (remember that the Nine was designed to be a mass market speaker). In this era solid state amps that can handle a two ohm load are easy to acquire.

The other thing I want to look into is the 6 uF cap in parallel with the UMR. Several people have claimed that this cap should be an 8 uF unit (though Shokdu reports that the original he pulled was in fact a 6 uf per the CSP schematic). I have run a pink noise spectral analysis of my Nines and they show a peak right at the 7 KHz crossover point - which may indicate that the UMR is running "past" its nominal 7 KHz and doubling with the tweeter. A higher shunt cap value would have the effect of steepening the crossover point and actually pulling it back in terms of frequency. This may in fact be the cause of the sense of the speaker still running a little hot around that frequency. Perhaps this would be better than adding even more attenuation.

As I implement these tweaks I will advise on this forum.

Best to all!!

[SHOKDU]

With respect for peoples opinions, I'm going to refrain from offering up my personal opinion on some of the "upgrades" mentioned, and simply suggest that people read Roger Russell's study on speaker wire.

http://www.roger-russell.com/wire/wire.htm

I will say that when I had the switches all set to 0dB, the speakers did seem quite bright. I can use my EQ to solve this problem though. The main issue I want to resolve and remove is the noisy static that currently hisses through the UMR and tweeter whilst no music is playing. I know that this static is not coming from my source or my amplifiers. The hiss/static is either more or less evident as I move the switches around, so to me, this indicates that the switches and/or reisistors are the source of the noise. Again, getting rid of this issue is priority. I can use the EQ to correct for brightness after the fact. I have a feeling the brightness may relax a bit too as the new capacitors break in.

[rrcrain]

With respect for peoples opinions, I'm going to refrain from offering up my personal opinion on some of the "upgrades" mentioned, and simply suggest that people read Roger Russell's study on speaker wire.

http://www.roger-russell.com/wire/wire.htm

I will say that when I had the switches all set to 0dB, the speakers did seem quite bright. I can use my EQ to solve this problem though. The main issue I want to resolve and remove is the noisy static that currently hisses through the UMR and tweeter whilst no music is playing. I know that this static is not coming from my source or my amplifiers. The hiss/static is either more or less evident as I move the switches around, so to me, this indicates that the switches and/or reisistors are the source of the noise. Again, getting rid of this issue is priority. I can use the EQ to correct for brightness after the fact. I have a feeling the brightness may relax a bit too as the new capacitors break in.

With regards to cap break in and speaker brightness. To this day, I'm not sure if the caps in my AR90's settled in or if the brightness broke in my ears. Some recordings, such as Sarah Brightman, can be almost painfully bright. I can affect this to a massive extent by rolling different tubes into my preamp.

[ar_pro]

How have you determined that the hiss isn't coming from your amplifiers, or another component?

[SHOKDU]

I agree, I think some of "break-in" of speakers is actually people just adjusting to what they are hearing.

I have tried different sources and different amps with the speakers, I get the same static hiss with every different combination I've tried. So, I'm pretty certain that the issue lies in the switches/resistors.

[Mach3]

Shokdu,

Read Russel's report that you provided the link to. While there is nothing really new there just the same old argument rehashed - and remember like everybody he has his ax to grind as well - the bottom line was that small gauge wire does contribute excess resistance. Therefore putting in larger wire - as I did - can only help the situation. BTW: the wire I bought to rewire them with? It cost $10 for a fifty foot roll at Radio Shack. So we are not talking about expensive wire.

As to the larger issue of whether or not wire can be heard? That is a very religious topic that I am NOT going to address - believe what you will. However I do believe that installing 14 gauge wire inside the AR-9 is an excellent idea - the factory wire is a pretty thin puppy - and is way to thin for the woofer drive.

As for your noise - well as you pointed out it seems to be coming from the attenuation switch panel. No real surprise there. That switch panel is a bad idea and it was a bad idea the day Holl designed it. But the thing was a mass market speaker designed to work with as many different rooms and tastes as possible. That doesn't mean it has any purpose in this era.

My suggestion? Just wire past it. The switch panel is an anachronism and it harms the sound of the speaker.

BTW: Got a ANOTHER SET of AR-9s today. with these I am going to build my external crossover box so I can tweak to my hearts content WITHOUT going into the speaker. Maybe I can do "stacked Nines" ;-))). Wow that would be almost 10 feet tall.

[RoyC]

Mach 3,

What "ax" is Russel grinding? Is he trying to sell you thinner magic wire? He certainly has more credentials than most folks in this neck of the woods.

It could be argued that the resistive properties of the original crossover wire is part of the design and associated response measurements. Folks who find the need to re-wire everything with heavier gauge wire tend to think that all resistance is evil, which is not the case. The only resistance to be concerned about in a proven design like the AR-9 is that caused by corrosion or worn connections.

I believe nothing was gained by re-wiring your AR-9's with 14ga Radio Shack wire unless there was obviously something wrong with the original wiring. You could have saved yourself some work. In fact, since your original AR-9 wire was of the more durable and costly tinned variety, that aspect of your restoration may actually be a downgrade. The reason speaker manufacturers like AR used tinned wire in their crossovers was to minimize the corrosion typical of bare copper wire, which is of much greater concern than wire gauge. (Btw, Ever wonder why many speakers are equipped with inductors constructed of very small gauge wire? It has nothing to do with cheaping out.)

Hmmm, then there is this... http://www.engadget.com/2008/03/03/audiophiles-cant-tell-the-difference-between-monster-cable-and/

Roy

[ar_pro]

Since its introduction, I've listened to many AR-9 systems in many different environments, with all sorts of amplification & sources, and came away with the opinion that the level-control switches make an excellent system even better, by allowing that final degree of control.

The AR-9 envelope of performance takes the switches & resistors into account, and as Roy has mentioned, most likely the original wiring, as well.

[HarryM]

Both Acoustic Research and Hafler recommend 16 gauge wire from your amp to a 4 ohm speaker unless it's more then 40 or 50 feet. That tells me that both these companies had done testing on the wires. They were not guessing when they wired the speakers.

[RoyC]

Since its introduction, I've listened to many AR-9 systems in many different environments, with all sorts of amplification & sources, and came away with the opinion that the level-control switches make an excellent system even better, by allowing that final degree of control.

Agreed... Level controls were integral to the design. Crossover points are affected along with the attenuation provided by the resistance. Removing them is another downgrade.

Roy

[SHOKDU]

Wow, you guys are making me think about this even more. Thanks!! Finding resistor replacements is easy, but does anybody have any replacement switch solutions?

Very good info guys, thanks for helping me to think about this stuff even a bit more.

[SHOKDU]

Ok, so after a little bit more thought, honestly, removing the switches is not a downgrade. If you were to leave the switches at 0dB at all times the switch should not really add any more resistance, unless it's corroded (which all old switches are). The resistors are only in the circuit when the switches are selected at either -3 or -6dB. So, if you were to disconnect the wiring to the switches and utilize said wire to make the proper connections to the binding posts, little to none of the originally intended resistance would be lost (presuming the 0dB switch posistion).