Jump to content

AR2ax rebuild in Finland


iso

Recommended Posts

I bought pair of decent condition AR2ax slightly over year ago. Seller in the UK was very friendly and he told me that even his wife thought that these were probably the nicest sounding pair of speakers they have ever heard, considering that the woofer surrounds were rotten off. When the speakers arrived here, it was clear that none of the pots worked but everything else seems to be OK and in original condition. I got no answer from seller, when I told him that these were absolutely worst sounding speakers I have ever heard, but they were in perfect condition for rebuild. I suspect that he misunderstood what I did mean...

There are no before pictures now. Cabs had some scars, dented corners and looked quite dull. I repaired dents and sanded them lightly and wiped them with BLO. This was not clearly enough as they looked dull and faded still. I sanded lightly again and wiped with BLO, finish got better... but not good enough, Then I sanded again, but now I sanded too much. As you can see from pictures, one cab top veneer is sanded almost trough. Grain of chipboard is clearly visible trough veneer. I learned also the hard way that curing time estimates of BLO seems to be quite optimistic, as it took something like 9 months to cure cabs completely. Cabs are not finished yet, but they look quite good to me. One further note from cabs... crossover board has been installed upside down.

Cabs are clearly not from same production run. Different glues were used for front panel mounting, wood grain in veneer is slightly different and molding colors do differ considerably. Drive units do not have similar production dates but production dates are from late 1973 to early 1974, so most likely no replacements have been needed. Woofer do have 3903-2 part number, tweeters are front wired ones and mid´s do have this 4uF+ 2 x 3ohm factory modification installed. Electrolytic has been replaced in the pictures.

I cleaned the pots first with ultrasonic cleaner. These cleaners do not cost too much and are quite handy, but do not replace old fashioned elbow grease. I had to soak two pots in salt&vinegar solution to remove some oxidation. This worked, but corroded tags so badly that some more elbow grease was needed. Tooth paste and toothbrush were also handy for cleaning pots. Finally I used ultrasonic cleaner to clean pots before final assembly.

More pictures and information to follow...

Best Regards

Kimmo

post-126436-0-46656300-1409741730_thumb.

post-126436-0-75544300-1409741746_thumb.

post-126436-0-72584000-1409741758_thumb.

post-126436-0-69897700-1409741775_thumb.

post-126436-0-67566500-1409741791_thumb.

Link to comment
Share on other sites

Hi Kimmo,

I repaired a pair of AR2ax earlier this year.

The pots were slightly corroded and cleaned up with minimal efforts.

The woofers on mine have an adapter ring where they fir to the cabinets.

I used many thin layers of BLO on the cabs but I understand that they are not meant to be glossy.

Please see attached a few pictures.

Best regards,

David.

post-125249-0-19913700-1409745338_thumb.

post-125249-0-58848500-1409745341_thumb.

post-125249-0-34404200-1409745344_thumb.

Link to comment
Share on other sites

Looking good guys. Those pots look pretty clean Kimmo. Do they now work?

David, you need another a badge. JKent knows where you can get one. We just had a discussion about the ax having just the a badge because apparently later models in the ax line do not. Mine also has the a and one is missing!

Link to comment
Share on other sites

Hi David,

I am still thinking of whether I should get the missing 'a' badge or not.

Being in the UK makes things a lot more expensive and difficult to source from the USA.

Some do charge a lot on postage for a small badge.

I have attached the serial number of the speakers.

Not too sure they are the early or later production but the numbers suggest that they are the early version.

I have also attached a picture of the two worst pots of the four.

Best regards,

David.

post-125249-0-61427800-1409757393_thumb.

post-125249-0-44582100-1409757396_thumb.

post-125249-0-42416700-1409757399_thumb.

Link to comment
Share on other sites

I have learned that after about 10-15 min curing time, I should wipe all remaining BLO from the surface. I used to mix 30-50% turpentine, made from pine oil, to BLO. Previous layer of BLO should be perfectly dry before new layer... or it will take 20-100 times more time to make next layer dry. How you do apply the last layer of BLO, can change gloss of finish.

My pots were also in quite good condition. One pot had badly corroded center track. I tried to cover cleaned pits with silver solder without too much success. Maybe it is better now... as track was pitted near end of travel, I decided to use it as mid control. This should be not too bad problem, as wiper is situated closer to mid point of track anyway. Remaining 3 pots did have close to perfect tracks that needed only mild cleaning. One stud of center post had more severe greenish corrosion around it, but I was able to make this one clean enough.

Best Regards

Kimmo

Link to comment
Share on other sites

Hi Kimmo,

I applied the BLO very thinly almost wiping all off after each layer and let it dry over 3-4 days between layers.

I didn't want to make the finish too glossy so just buffed it out using a soft cotton towel.

By the way, I was the other bidder on your AR3a Improved but obviously lost the bid!

Best regards,

David.

Link to comment
Share on other sites

Dear David,

Sad to hear you lost the bid... good thing was that price did not sky rocket to unreasonable level.

I will report how 3aImproved´s will proceed...

Best Regards

Kimmo

Link to comment
Share on other sites

Last week I sealed all internal seams with PVA glue to make sure that there are no air leaks in the cabinets. I assembled the pots using hi temperature grease on shafts to avoid air leaks from there. Ultrasonic cleaner was handy for cleaning all mounting hardware. I discovered that only terminal nuts seems to be made of brass, so they will not rust. Other crossover board nuts, screws and washers did leave brownish mark on white paper where I left them for drying. This is why I did solder wires directly on mounting screws on crossover board.

After this I wired new caps and clean old parts on crossover board. Coils measured 2,00mH and 1,6mH when I received them. Coil weights were 325g and 292g so... I unwound 17 g of coil wire from 2,00mH coil and soldered this to 1,6mH coil and now I have two 1,74mH coils. Slightly less than target value, but within 7%... so I used 5,6mF mid cap to avoid bump in crossover region. I think this is close enough due 1st order crossover slope. You may note heat shrink tube used over solder joint in coil former.

I added tag strip and screw for capacitor anchoring points and glued new caps with acrylic putty to crossover board. After double checking everything inside, I stuffed original stuffing material in the cabs. Stuffing material was this nicer purplish one that is less dusty than regular glass wool. I did insert layer of cardboard behind tweeter and mid openings to make sure that stuffing material will not seize mounting screw as it does so often... even there is slight possibility that this arrangement will introduce some rattles from wiring. I used also glue to secure woofer mounting nut inserts to cabs.

My first re foam was not 100% success, as I placed one foam on the cone slightly off center. Fortunately my good friend Jukka was able to attach foam to the frame, even fit was not good. So after all drive units were assembled, cabs were ready for rock´n´roll.

More tho follow...

Best Regards

Kimmo

post-126436-0-84705600-1410161467_thumb.

post-126436-0-98397900-1410161490_thumb.

post-126436-0-22605000-1410161506_thumb.

post-126436-0-02229300-1410161516_thumb.

Link to comment
Share on other sites

Last weekend we made first listening session with rebuilt AR2ax´s. Results were not too good, as there were quite many issues to repair in addition to most obvious ones. I will report this later, but I have now one question to experts on this site. Is there any other way than cracked or detached input terminal to develop leak to back side of CTS mid range driver? I can not prove that there is leak as cone can not be seen, but when playing something like pipe organ, sound is like organ was played through Leslie.

Best Regards

Kimmo

Link to comment
Share on other sites

Roy

I have tried to find where I made mistake... but I have not found the bug. Picture is not too good, but can you tell where you see faulty connection?

Best Regards

Kimmo

Link to comment
Share on other sites

Roy

I have tried to find where I made mistake... but I have not found the bug. Picture is not too good, but can you tell where you see faulty connection?

Best Regards

Kimmo

Kimmo,

It appears both pots are not properly wired.

Attached is an old photo taken by a forum member during a similar discussion showing the proper wiring of the 2ax with new caps. Note that the #2 pot terminal is on one side of the circuit, and #1 is on the other.

If it does not make sense to you, send me a PM.

Roy

post-101150-0-96660000-1411052784_thumb.

Link to comment
Share on other sites

I just went back to your forum thread, and realized your 2ax's have an

unusual midrange modification. Originally there was no mod connected to the

mid, which is the most 2ax's are. When AR began using the ceramic magnet

woofer, the resistance of the mid was raised from around 6 ohms dcr to 9

ohms. Some of the earlier mids were used with *one* 3 ohm *series* resistor

here in the US, and the later 9 ohm mids went back to nothing at all.

I have never seen the parallel component arrangement on the back of your

mids before! You obviously have the later 2ax with the ceramic magnet

woofer, but with a different mid mod.You can try disconnecting these

components and try the mids with no added components, *or* with just a 3

ohm resistor in series with the yellow wire, as it was done over here. Both

of these things will provide more extended midrange output....and perhaps

sound better to you.

Also, the old tweeters were never very strong, and many of them are very

compromised these days...mostly due to stiffening suspensions, and

decomposing foam under the dome (which gets into the voice coil gaps).

Roy

Now when crossover problems have been solved with Roy and above mentioned issue with midrange might be reason for midrange problems described earlier. Roy´s comment about tweeter is also true as some piano recordings exhibit clear distortion or ringing, when there is not too much treble energy present. I suppose that when there is plenty of overtone energy anyway, this will mask ringing of tweeter. I have used ARC D76A and Quad 405 power amps for testing, results are pretty pretty consistent.

I do not know if you did note that there was 4uf capacitor soldered between midrange terminals in addition to 2 pc 3 ohm resistors in series positions. There is no reason to believe that 4uf electrolytic cap has been added later,as glue used for mounting of original cap and resistors was same. DCR of this combination is 13 ohm so DCR of mid is about 7 ohm or early 6 ohm DCR unit. Production date in the picture seems to be from 1973. Resistors were used for padding purposes as Roy said... but was 4uf cap used to assist original low pass section made from yellow fiberglass?

4uf capacitor placed across mid terminals will cut 8 ohm mid output considerably (-6dB) at 5000 hz?... If 1975 advertising material is correct, midrange driver should be flat to 12-14K and tweeter output will have quite small effect on hf reproduction.

http://www.classicspeakerpages.net/library/acoustic_research/original_models_1954-1974/original_models_brochures/acoustic_research_loudspeak.pdf

Best Regards

Kimmo

post-126436-0-63798700-1411477114_thumb.

Link to comment
Share on other sites

When I removed 4uf capacitor between mid terminals sound quality of this pair improved considerably. There is now considerably more hf and ringing of of tweeter or mid can hardly be heard. If ringing was originating from tweeter, higher output of mid in 3K to 12K range might mask it now.

There is something I do not fully understand in resistor modification Roy mentioned and that was discussed in earlier tread http://www.classicspeakerpages.net/IP.Board/index.php?showtopic=1548&hl=%2Bar2ax+%2Bmidrange+%2Bresistors

I suppose my 3903-2 woofers are the late higher efficiency ones? Was this resistor modification made to late 9 ohm mid to reduce mid sensitivity to earlier 6,5 ohm mid level. I am asking this as my mids are most likely early ones with 6,5 ohm DCR and do have this resistor mod installed.

Even I tried to be careful when I used PVA glue to reinforce all internal seams, I did find one leak around masonite terminal for front wired tweeters. It is funny as I remember that this was the first place where I thought leaks could be. Anyway I poured some PVA glue in the seam, first layer slightly thinned so it could creep in the seam more efficiently.

Best Regards

Kimmo

post-126436-0-50323900-1411659419_thumb.

Link to comment
Share on other sites

When I removed 4uf capacitor between mid terminals sound quality of this pair improved considerably. There is now considerably more hf and ringing of of tweeter or mid can hardly be heard. If ringing was originating from tweeter, higher output of mid in 3K to 12K range might mask it now.

I suppose my 3903-2 woofers are the late higher efficiency ones? Was this resistor modification made to late 9 ohm mid to reduce mid sensitivity to earlier 6,5 ohm mid level. I am asking this as my mids are most likely early ones with 6,5 ohm DCR and do have this resistor mod installed.

Best Regards

Kimmo

I suspected removing the cap would improve things. This is a perfect example of a reason to take AR's published specs with a grain of salt. Models were always evolving, and actually differing internationally. Not all iterations of various AR models sound the same.

The ceramic magnet version of the 2ax woofer has different midrange characteristics. Later AR-2ax mids had higher dcr of 9 ohms, presumably to compensate for this....so, yes, earlier 6+ ohm versions of the tweeter in the same cabinet as the ceramic magnet woofer were fitted with a modification (probably simply to use them up). The most common mod to the 6 ohm tweeter I have seen was a simple 3 ohm resistor.

Obviously AR was playing around with other mods as well. Given that the crossover components in the cabinet did not change, there is no way all 2ax's were created equal. Of course, there was the earlier transition from the cloth surround woofer version to the foam surround woofer/different tweeter version. They do not sound the same either.

Roy

Link to comment
Share on other sites

iso, I hope you iron out the issues and get these up an running. The cabinets look very good!

Roy, as usual giving the benefit of his knowledge and expertise in the way of guidance. Roy, I am intrigued by your comment about the woofer change on the 2ax models. In fact I have two sets in the queue for refurbishing, one with cloth and one with foam surrounds. You say they do not sound the same, could you elaborate on which is considered the better sounding of the two types. My hope is to complete these two sets and do some critical listening to determine which set to keep in my main system.

Link to comment
Share on other sites

Roy, I am intrigued by your comment about the woofer change on the 2ax models. In fact I have two sets in the queue for refurbishing, one with cloth and one with foam surrounds. You say they do not sound the same, could you elaborate on which is considered the better sounding of the two types. My hope is to complete these two sets and do some critical listening to determine which set to keep in my main system.

The overall character of the various iterations of the 2ax is the same, but tonal balance varies somewhat. Much, however, depends on the condition of the original drivers of the specimens being listened to. The re-foam job and the variable (mostly reduced) output of many of the 3/4" tweeters are both possible issues relative to the later versions. The early cloth surround woofer version of the 2ax has the red/orange tweeter, which seems to have held up better than the later black 3/4" tweeter...and of course there is no foam to deal with. My opinion is that the alnico magnet/cloth surround 10 and 12 inch AR woofers were smoother into the midrange...great woofers...so I prefer AR models equipped with them.

Given the toll of time and wear, it is probably not possible to know which version (if any) was actually "better" than another today.

Roy

Link to comment
Share on other sites

Much, however, depends on the condition of the original drivers of the specimens being listened to.

Given the toll of time and wear, it is probably not possible to know which version (if any) was actually "better" than another today.

Roy

Rebuild is proceeding very well. Speakers are performing well. Only minor issues like slight woofer buzzing and how many padding resistors I should use has to be sorted.

One woofer is buzzing but not so much that it can be detected when listening music or sinusoidal waveform from signal generator... but when using low distortion 30-60 Hz signals from CD some buzzing can be heard from VC area. I have used all 4 mounting positions of woofer, tried different routing for woofer leads. VC to terminal leads seems to be far enough from cone and nothing special can be seen from back terminal opening in the VC and spider area. Surround seems to be properly attached and VC centering "feels" right. Maybe I should swap woofers from L to R and to R to L cabinet and try again.

Resistor modification bothers me still. Why these 3 ohm padding resistors were added to this lower efficiency midrange, if more efficient mid was needed due more efficient new woofer? These resistors will degrade efficiency of mid anyway. There must be some other reason for this resistor modification. Honkiness that was present with parallel 4 uf cap does not bother me anymore, even removing this cap should not effect too much to lower crossover area where honkiness should originate. Anyway I have used now 2 pc resistors and tried to get used to sound of the speakers. I will try other options later to see if there is some difference.

Tweeter degradation should degrade AR2ax performance considerably less than models like AR5 or AR3a. AR2ax mid level exceed tweeter level up to 14-15K whereas 3a or 5 tweeter exceed mid level at 6,5K... difference is something like 1 octave. Maybe 2ax tweeter should be considered as super tweeter that has been added mostly to improve dispersion of HF.

3a, 4 and 2ax resonse can be seen here

http://www.classicspeakerpages.net/library/acoustic_research/original_models_1954-1974/original_models_brochures/acoustic_research_loudspeak.pdf

Best Regards

Kimmo

Link to comment
Share on other sites

One woofer is buzzing but not so much that it can be detected when listening music or sinusoidal waveform from signal generator...

Resistor modification bothers me still. Why these 3 ohm padding resistors were added to this lower efficiency midrange, if more efficient mid was needed due more efficient new woofer? These resistors will degrade efficiency of mid anyway. There must be some other reason for this resistor modification. Honkiness that was present with parallel 4 uf cap does not bother me anymore, even removing this cap should not effect too much to lower crossover area where honkiness should originate. Anyway I have used now 2 pc resistors and tried to get used to sound of the speakers. I will try other options later to see if there is some difference.

Tweeter degradation should degrade AR2ax performance considerably less than models like AR5 or AR3a. AR2ax mid level exceed tweeter level up to 14-15K whereas 3a or 5 tweeter exceed mid level at 6,5K... difference is something like 1 octave. Maybe 2ax tweeter should be considered as super tweeter that has been added mostly to improve dispersion of HF.

Best Regards

Kimmo

Kimmo,

Woofer buzzing is almost always the result of re-foam issues. It is one of those variables I mentioned above.

<<Resistor modification bothers me still. Why these 3 ohm padding resistors were added to this lower efficiency midrange, if more efficient mid was needed due more efficient new woofer?>>

Becasue the 6 ohm mid is not the "lower efficiency" midrange. A less efficient mid was needed. Resistance was added to it to compensate for the increased midrange output of the ceramic magnet woofer. A 9 ohm mid was later used with no mods becasue the resistance was built into the voice coil.

Based on the variety of midrange mods we have seen, I'm guessing AR was wrestling with more than one issue related to integrating the later woofer. Who knows which mod works "best", but just one series (3 ohm) resistor with the 6 ohm mid is closer to the version using just the non-modded 9 ohm mid...which is where AR ultimately went with this model before production ended.

The tweeter output is equally important for ALL models if the goal is to hear them as AR intended. The air and dispersion characterisitics of this tweeter are part of the charm of the old beasts.

Stupidhead...All of the above are some reasons why I prefer the earliest version of the AR-2ax. :)

Roy

Link to comment
Share on other sites

Becasue the 6 ohm mid is not the "lower efficiency" midrange.

Roy

Thank you Roy... now I understand. In earlier discussion sensitivity issue was talked about, but not defined clearly enough for me.

3903-2 woofers (alnico ones?) were from time before this modification and they were supposed to be used without any padding resistor mods?

Best Regards

Kimmo

Link to comment
Share on other sites

Thank you Roy... now I understand. In earlier discussion sensitivity issue was talked about, but not defined clearly enough for me.

3903-2 woofers (alnico ones?) were from time before this modification and they were supposed to be used without any padding resistor mods?

Best Regards

Kimmo

Kimmo,

If you are referring to the woofer in your photo earlier in this thread, it is not the alnico version. It is the later version which would have been used with the 9 ohm mid, or the modified 6 ohm mid. If I had your speakers I would simply install one 3 ohm resistor in series with each of your 6 ohm mids.

Roy

Link to comment
Share on other sites

  • 2 weeks later...

Roy,

Mid with single 3 ohm padding resistor improved woofer to mid matching considerably. Performance is now quite good with ARC CD3 mkII, ARC SP6C, ARC D76A combo. HF level is down as expected and LF seems to be slightly under damped but otherwise balanced. Sound staging is nice but not well defined, as became apparent during discussion with Tom. This combination seems to be especially suitable for old America or CSN records.

When using Quad 34 and 405 as amplification, slight harshness in hf became back. It sounds similar than ringing when using mid with both 3 ohm resistors and 4uf cap between terminals, but only to considerably lesser extent. I will investigate more this issue...

Best Regards

Kimmo

Link to comment
Share on other sites

  • 2 weeks later...

I did add 560nF capacitor between mid driver terminals... in same place where there was 4 uF cap was placed originally. This made speakers perform quite much smoother, but removed some detail from performance. Overall performance is now better but some female voices like like Emmylou Harris were cutting too much in, due some harshness in HF even with ARC SP6C/D76A combo. Better now, but not perfect yet...

One tweeter flange does have crack close mounting bolt. Metal ring glued to the original mounting flange would solve this problem... but how to make good looking ring of 12 cm dia, without press and proper dies, is mystery to me. Any suggestions for smart looking repairs?

Best Regards

Kimmo

Link to comment
Share on other sites

  • 1 month later...

Some time ago I removed also the other 3 ohm resistor and 560nF cap from middrange driver... and used it straight. Mid to woofer integration was not good as earlier and it came more apparent that sustained piano notes sometimes distorted quite badly. This upper middrange hardness has bothered me more or less all time.

As it seems that middranges are reason for this hardness... are there any known issues that can be sorted with CTS middranges... or should I try to find better replacemets? Are 9 or 6 ohm models usually surviving better as asking if these are OK... I will receive reply like pulled from working speaker, DCR tests OK but no guarantee...

Best Regards

Kimmo

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...