Jump to content
The Classic Speaker Pages Discussion Forums
Sign in to follow this  
r_laski

My AR-9 Capacitor Recap/Upgrade Project

Recommended Posts

It’s been a long time since I posted on this forum. After several years without an audio system my better half urged me to replace whatever needed replacing and get back to enjoying my music collection again.

I have two pairs of AR-9 speakers. I am using one pair of these in my “new” audio system. I knew part of my system rebuild would be recapping a pair of AR-9s. I spent a several days on this forum reading every thread I could find on what you all have done with your AR-9s. Thank you all for the wealth of information. I found it helped me develop a plan for recapping my AR-9s. Years ago I downloaded and saved all the AR-9 engineering drawings that were posted at the time. Glad I did as I couldn’t find them again on this web site.

I also spent considerable time researching various types (electrolytic/film), brands, and sources of capacitors. I also found some valuable information on a few crossover component sellers’ websites / links. Below is a rundown of my project.

Scope/Considerations:

Some things I considered for the AR-9s:

1. Internal or External crossover. I decided to keep the crossover internal. There is only a minor change to how it is configured. Each Crossover section remains as originally laid out but is built on a separate pegboard.

2. What crossover parts to replace / retain. I decided to reuse all inductors and left the switch board as is. I replaced all capacitors, 6 Ohm resistor in the upper mid-range (UMR) crossover, binding posts, and all internal wiring.

3. Capacitor selection. Due to size and cost, woofer crossover capacitors are electrolytic with film bypass capacitor. All other capacitors are film capacitors (details below). I replaced the original wires with color coded 16 gauge tinned copper wire.

4. Cabinet changes. The cabinets are in such great shape I left them as is. The only “modification” was removal of plastic Velcro dots on tweeters and upper mid-range drivers and cabinet. I did not remove the original crossover backboard.

5. Drivers. Years ago I re-foamed all the 12” and 8”drivers in both pairs of AR-9s. The foam surrounds are still in new condition. Thanks to the speakers being boxed up or covered up while stored during a long period of non-use . I attached adhesive backed gasket tape to all drivers prior to reinstalling them in the cabinets during final reassembly.

 

Crossover tear down:

I removed nuts securing woofer/Lower mid-range (LMR) crossover boards removed them intact.

image.png.97831d408767c360354ca0ab83a88eee.png

This one had the small axial 470uF capacitor. The other AR9 had the large 470uF radial can capacitor.

Before removal of any component from a board I took many pictures and labeled what each wire inductor lead and capacitor lead was connected to. I removed everything on the UMR/High Range crossover board without removing the board from the cabinet. Then the worst task of the entire project – removing hot glue from all inductors.

Original capacitors measurements:

I used the same basic capacitance meter I used when I recapped my AR-11s years ago. It tests at a single frequency, but for me that is enough to indicate whether a capacitor is in or out of specs. Below is a chart of capacitor measurements from both AR-9s in this project.

Capacitor

AR9K013212

Dif

%

AR9K013494

Dif

%

2500

2250

-250

-10

2330

-170

-6.8

470

576

+106

22.6

510

+40

8.5

80

106.4

+26.4

33

151

+71

88.75

40

48.1

+8.1

20.25

40.7

+.7

1.75

30

50.5

+20.5

68.3

65.4

+35.4

118.0

24

33.0*

+9.0

37.5

27.1

+3.1

12.9

8

9.31

+1.31

16.375

8.79

+.79

9.875

6

5.94

-.6

10

5.57

-.43

7.17

4

4.07

+.07

1.75

9.25

+5.25

131.25

* Capacitor reading was unstable, continuous changing.

No doubt in my mind new capacitors would improve these AR-9s. Only one capacitor in each speaker measured within my threshold of within +/- 5% or rated capacitance.

I spend a considerable amount of time deciding what capacitors to use for this rebuild. I considered capacitors I used for AR-LST, AR-11 and AR-90 recaps in the past, what others in this forum used, and what capacitors are being used and recommended by professional crossover builders/re-builders.

I made several cost comparison spreadsheets of capacitor brands from various sources. I benefited from a 25% off seasonal sale on all the film capacitors I chose. It took purchases from 4 different sources to get all these capacitors. I decided on the following combination of capacitors:

2500uF - M.D.L. (Jantzen) (2) 1000uF and (1) 500uF electrolytic bypassed with 0.1 uF Clarity Cap CSA 630VDC film.

470uF - Mundorf E-Cap AC Raw electrolytic (matched pair) bypassed with 0.1 uF Clarity Cap CSA 630VDC film.

80uF - (2) 40uF Clarity Cap PX 250VDC.

40uF - 40uF Clarity Cap PX 250VDC

30uF - 30uF Clarity Cap PX 250VDC

24uF - (2) 12uF Clarity Cap CSA 250VDC

8uF - 4.7uF + 3.3uF Clarity Cap PX 250VDC

6uF - (2) 3.0uF Clarity Cap CSA 250 VDC

4uF - (2) 2.0uF Clarity Cap CSA 250VDC

Note: Woofer capacitors are electrolytic bypassed with 0.1uF film. All other parallel capacitors are Clarity Cap PX film to save cost. All other series capacitors are Clarity Cap CSA except for 80uF LMR series capacitor. PX was selected to save cost. At the sale prices, the Clarity Cap film caps ended up costing about the same as a regularly priced set of Solen film capacitors would have.

Prototype:

As I stated above, I built each driver’s crossover on a separate pegboard.

image.png.8c25fb2cddbdea55d61e2a7faf990d04.png

Preliminary layout

image.png.795897ca2ad489c9a4676ffe1fa7bcd4.png

Final layout - Holes in 4 corners of woofer pegboard match up perfectly with mounting studs in bottom of cabinet. Homemade bus bar of 14 AWG tinned copper. On the second AR-9 crossover I stacked the 500uF cap on top of the two 1000uF caps to reduce space and moved the bus bar over. This enabled me to move the entire layout down one row of holes and still have the corner holes open for the mounting stud in the cabinet. This also gave me more space between this board and the LMR board mounted behind it on the back of the cabinet.

 image.png.cf4dc6d9bdf88fbf9af08eaebaad02d2.png

Crossover from second AR-9 (big can 470uF) and new woofer crossover - I tried to orient inductors in different planes as much as possible on each board and across boards.

image.png.190cfe9f9e1509d52e5e9502557b5a0f.png

Original AR-9 LMR section (lower left quadrant) and new LMR crossover. The new board is mounted to the back of the cabinet below the UMR (original backboard).

image.png.429fcec3c955b15dcf4f8b2f460e3766.png

Woofer and LMR boards mounted in cabinet. Best effort to orient inductors in different planes between the two boards.

 image.png.7ec59c8a4a641ef3c1cd0bc245c9c4c9.png

UMR and High Range (tweeter) capacitor pegboards. UMR is built around original binding post holes. One side of pegboard holes had to be enlarged for all binding post holes to line up.

image.png.ac926681069ed62625c3e9344f15926d.png

UMR and tweeter boards in cabinet.

Close-up of tweeter and top of UMR pegboards. One blue wire on tweeter board goes to tweeter. Second blue wire goes to black (-) binding post. Originally inductor and blue wire were connected directly to black (-) binding post.

image.png.144a9f80cefed92360661406c1b024b5.png

Wires: With few exceptions I kept all new internal wires the same length as the original wire lengths. However I changed some of the wire colors for convenience/ standardization. All woofer section wires were red (+) / black (-). I changed the wire from the upper red binding post to the switch board from white to gray. I kept the purple wire color from the switch board to the 80uF cap on LMR crossover. All other LMR wires were white (+) green (-). All UMR wires Orange (+) brown (-) All tweeter wires yellow (+) blue (-).

Final assemble and testing - While I was working on the project AR-9s, I hooked up my other pair of AR-9s to my new audio system to “break-in” the new components and establish a subjective baseline of what an original pair of AR-9s sound like. I have to say, not knowing the actual condition of the capacitors in this pair the sound was very good. However, “s”, “t”, “st” harshness was very evident.  I also had to switch the UMRs to -3db to tame its overpowering the other drivers.

After reassembly I replace the left original AR-9 with recap project AR-9. All switches on the recapped AR-9 set to 0db. No burn-in. Just put some music on and subjectively listen.

Immediately there was a noticeable difference between the left and right speakers. The left was a little louder, MUCH cleaner. The original AR-9s sounded pretty good by themselves. Compared to the recapped AR-9 however, my impression was the original AR-9 was now somewhat “fuzzy”. They sound between the two speakers was unbalanced. Putting the UMR switch on the original AR-9 to 0db balanced the output somewhat. But, it did not improve the difference in clarity (no pun intended).

I rebuilt the second AR-9 crossovers making minor changes in the layout. One significant change I made was to glue machine screws to the spacer boards.  I put spacers under the crossover pegboards to provide space under the pegboards for the cable ties that secured the capacitors and inductors. Now all crossover boards are secured with screws, washers, and nuts and can be removed. Only the spacers are glued to the cabinet in this AR-9.

Listening with both recapped AR-9s hooked up:

image.thumb.jpeg.d1319e817f10f22df5ae39ea5d5ea4e0.jpeg

 I’m not a GESR (golden-eared stereo reviewer). All I can say is I’ve never owned an audio system that sounded this amazing!

I remember when we were recapping AR-11s, AR-90s and LSTs. We would describe the improvement as being as if a veil had been removed from in front of the speakers. This is more like a heavy curtain has been removed. Female and male singers sound like they are standing in front of me in the room. Background vocals are distinct from the lead singer. There is air and space between instruments and around each instrument. As familiar as I thought I was of my music collection, and I can’t believe I’m saying this, I heard things I never heard before in the music.

 

Share this post


Link to post
Share on other sites

That's a lot of really good work.

Congratulations on gaining such an improvement in your speakers, and thanks for including your write-up & photos for future reference!

Share this post


Link to post
Share on other sites

Well done!  I'm personally happy to see that you used respected brands of capacitors, as well as using good, like capacitors for the series values.  In my experience, that really helps the speakers to continue to sound as a single voice.  Also, please update your thread, as the speakers break.  Just curious if you'll hear improvements in sound?

Thanks for sharing!

Share this post


Link to post
Share on other sites

Nice project r_laski.   I too will be curious as to what you think of your "new" AR9's after you've listened to them for a while.   I particularly had problems using poly caps in the 24uF series spot on the UMR and eventually went with Mundorf ECaps bypassed with 0.01uF F&F caps.

Share this post


Link to post
Share on other sites
1 hour ago, AR surround said:

Nice project r_laski.   I too will be curious as to what you think of your "new" AR9's after you've listened to them for a while.   I particularly had problems using poly caps in the 24uF series spot on the UMR and eventually went with Mundorf ECaps bypassed with 0.01uF F&F caps.

E-Caps.  Wimp...!  :D

Those ClarityCap CSA's will sound fine.  They should make for a nice match.  Warm and relaxed.  But, a long break in period.

Share this post


Link to post
Share on other sites
On 1/31/2020 at 10:58 PM, Stimpy said:

E-Caps.  Wimp...!  :D

Those ClarityCap CSA's will sound fine.  They should make for a nice match.  Warm and relaxed.  But, a long break in period.

Yes, as many of you recall, those Solens ? beat me into submission!! 

Share this post


Link to post
Share on other sites

I like the CSA Clarity caps, too. I didn't think 'break-in' took all that long. However, I played pink noise thru them for about 8 hours.

Also, I use a 3.9uF and not 2 x 2uF.

Share this post


Link to post
Share on other sites

Thanks to all of you on your compliments.

>>I particularly had problems using poly caps in the 24uF series spot on the UMR and eventually went with Mundorf ECaps bypassed with 0.01uF F&F caps.<<

>> Yes, as many of you recall, those Solens ? beat me into submission!! <<

I read your AR-9 posts while preparing for my project. I actually purchased Mundorf 22uF and 1uF ECaps (couldn't find a 2uF) for the 24uF UMR and was going to experiment with them. However, when I measured new 22uF ECaps they measured 23.8/23.9. That’s almost 9% over rated value. It should be no more than +/- 5%.  I never used them. I purchased two pairs of 470uF ECaps for woofer crossovers from a different source. I paid extra to have these precision matched. I received exactly matched pairs. The second pair is for when I rebuild the crossovers of my remaining pair of stock AR-9s.

 I had a similar experience of capacitor disaster with a pair of AR-90s. I reread a thread on a pair of AR-90s I worked on 16 years ago. I tried a combination of Dayton, Solen, and North Creek (no longer available) poly caps. It was very unpleasant, unlistenable.  I replaced the Daytons with Solens and all was well.

 My point is different brands of capacitors do not always mix well together. So far same brand Clarity Cap CSAs and PX series work perfectly together in these AR-9s. The MDL and Ecaps combination in the woofer crossover also work extremely well together.  

My take on bypassing – I’ve read many views about bypassing. I bypassed the electrolytics in the woofer crossover with Clarity CSA poly caps. I chose this based on a similar practice by a company that rebuilds crossovers, mostly for classic Polk Audio speakers. They do not bypass Clarity Caps, so I also did not.

There is one thing I wish someone would research about bypassing. Is there a formula for an optimum ratio of bypass capacitor value to “primary” crossover capacitor value?  I’ve seen the recommendations for .01uF up to a certain capacitor value and 0.1uF above that. I’m not sure how much affect a 0.1uF capacitor has on a 2500uF or 470uf capacitor. Not going to remove them, just something to ponder.

>> I too will be curious as to what you think of your "new" AR9's after you've listened to them for a while. <<

I've got more than 30 hours of listening to these "new" AR-9s. So far I am VERY PLEASED. I'm running them with all switches at 0dB. I haven't heard any need to "fiddle" with them. I will post any changes to my listening impressions as I spend more time listening. At most, I expect them to “warm up” only slightly more over time.

There is one “improvement” in how these AR-9s sound that is worth mentioning. It’s related to speaker placement, listening position, and listening room acoustic characteristics. My listening room is a multipurpose room – separate audio and TV (no home theater). My stereo system is along one of the long (wide) walls, TV along a short (narrow) wall. I’ve placed the AR-9s as close to recommended in the owner’s manual as possible. My normal listening position is centered between the speakers seated against the opposite wall. The “sound stage” is very good in this position. However, if I move about 2 feet forward the sound stage really opens up. It appears to move a little forward toward me, but, more so it extends beyond the walls behind the speakers. Side to side it extends past the speakers to the corners and side walls of the room. Imaging, by that I mean separation and placement of individual vocalists and instruments, is much improved. This appears to be an acoustic sweet spot in the room.

>> I like the CSA Clarity caps, too. I didn't think 'break-in' took all that long. However, I played pink noise thru them for about 8 hours. Also, I use a 3.9uF and not 2 x 2uF. <<

I considered both options. A single 3.9uF cap is probably just as accurate to replace a 4uF cap and a lot less costly.  2x2uF is just my personal preference. I may go with single caps within 5% of desired value, where possible, in my second pair of AR-9s.

Share this post


Link to post
Share on other sites
3 hours ago, r_laski said:

> I like the CSA Clarity caps, too. I didn't think 'break-in' took all that long. However, I played pink noise thru them for about 8 hours. Also, I use a 3.9uF and not 2 x 2uF. <<

I considered both options. A single 3.9uF cap is probably just as accurate to replace a 4uF cap and a lot less costly.  2x2uF is just my personal preference. I may go with single caps within 5% of desired value, where possible, in my second pair of AR-9s.

I pointed that out as adding those 2 caps together cuts the ESR in half and THAT will cause more 'brightness'. I use Parts Connextion and for a VERY reasonable matching fee of $1 per pair You can ask for a matched pair that reads approx 4uF and they do their best to give you that.

On the ECaps reading high I had contacted Mundorf on that very issue. They are made for Mundorf and Jantzen by FT Cap in Germany,. I was told they had an issue with the winding machine and have since add equipment to correct that issue. Again for $1 per pair from Parts Connxtion you get what you ask for.

Congrats on your work !

Share this post


Link to post
Share on other sites
On 2/5/2020 at 3:46 PM, r_laski said:

I read your AR-9 posts while preparing for my project. I actually purchased Mundorf 22uF and 1uF ECaps (couldn't find a 2uF) for the 24uF UMR and was going to experiment with them. However, when I measured new 22uF ECaps they measured 23.8/23.9. That’s almost 9% over rated value. It should be no more than +/- 5%

As DavidR said, I had my my Mundorf E-Caps all matched.   (Remember that I am using AR9's as fronts and AR90's as surrounds, so I wanted all of the caps matched.)  Carlspeak once told me that we probably cannot hear differences in caps unless the values differ by 15% or more.   But some of us, including me, are more anal about it and prefer closer tolerances.

On 2/5/2020 at 3:46 PM, r_laski said:

I had a similar experience of capacitor disaster with a pair of AR-90s. I reread a thread on a pair of AR-90s I worked on 16 years ago. I tried a combination of Dayton, Solen, and North Creek (no longer available) poly caps. It was very unpleasant, unlistenable.  I replaced the Daytons with Solens and all was well.

Now that indeed is interesting because using Solen MKP on the UMR 24uF spot produced incredibly horrid results on my AR90's; and almost as bad on the 9s.

On 2/5/2020 at 3:46 PM, r_laski said:

I've got more than 30 hours of listening to these "new" AR-9s. So far I am VERY PLEASED. I'm running them with all switches at 0dB. I haven't heard any need to "fiddle" with them. I will post any changes to my listening impressions as I spend more time listening. At most, I expect them to “warm up” only slightly more over time.

I'm very happy to hear that you are very pleased with the recap.   I too would consider going to CSA Clarity, but the expense and time for redoing 5 speakers (2xAR9, 2xAR90, 1xAR915)...ugh!

On 2/5/2020 at 3:46 PM, r_laski said:

However, if I move about 2 feet forward the sound stage really opens up. It appears to move a little forward toward me, but, more so it extends beyond the walls behind the speakers. Side to side it extends past the speakers to the corners and side walls of the room. Imaging, by that I mean separation and placement of individual vocalists and instruments, is much improved. This appears to be an acoustic sweet spot in the room.

I have experienced that as well.  The stereo sound field is quite enveloping.   When sitting in the sweet spot, the AR9s can produce what seems to be surround sound fed the right material.

Share this post


Link to post
Share on other sites
On 2/5/2020 at 7:02 PM, DavidR said:

I pointed that out as adding those 2 caps together cuts the ESR in half and THAT will cause more 'brightness'. I use Parts Connextion and for a VERY reasonable matching fee of $1 per pair You can ask for a matched pair that reads approx 4uF and they do their best to give you that.

… for $1 per pair from Parts Connxtion you get what you ask for.

I purchased all the Clarity Caps (and the “defective” Ecaps) from Parts Connexion. For my other pair of AR-9s I am leaning toward purchasing matched pairs of single caps where possible vs DIY assembling matching pairs of 2 caps paralleled together. Just have to get over exactly matching original crossover cap values.

I purchased matched pairs of 470uF ECaps from Sonic Craft. I received exactly matching caps within 1% of rated value.

On 2/7/2020 at 9:11 AM, AR surround said:

Carlspeak once told me that we probably cannot hear differences in caps unless the values differ by 15% or more.   But some of us, including me, are more anal about it and prefer closer tolerances.

Yes we are. For both exactly matching original capacitor values and tolerances between speaker crossovers.

On 2/7/2020 at 9:11 AM, AR surround said:

Now that indeed is interesting because using Solen MKP on the UMR 24uF spot produced incredibly horrid results on my AR90's; and almost as bad on the 9s.

I think we solved the same problem differently with equally acceptable results. I replaced Dayton caps with Solens. You kept the Daytons and replaced the Solens with Ecaps. I suspect Daytons and Solens don’t mix well together. Like I said, I believe different brands of capacitors do not always mix well together.

On 2/7/2020 at 9:11 AM, AR surround said:

I'm very happy to hear that you are very pleased with the recap.   I too would consider going to CSA Clarity, but the expense and time for redoing 5 speakers (2xAR9, 2xAR90, 1xAR915)...ugh!

That would be a daunting task to do all at the same time. The research time, purchasing parts, prototyping the first crossover, initial testing vs stock AR-9, and building and installing crossover in second AR-9 took about 3 months. I still have a pair of AR-9s to do.

If you are happy with what you have already done to your AR speakers there is no need to redo them.
On 2/7/2020 at 9:11 AM, AR surround said:

I have experienced that as well.  The stereo sound field is quite enveloping.   When sitting in the sweet spot, the AR9s can produce what seems to be surround sound fed the right material.

Yes, exactly!

Share this post


Link to post
Share on other sites
2 hours ago, r_laski said:

I purchased matched pairs of 470uF ECaps

Let me know how they work out. The voltage rating on those ECaps (23VAC-63VDC) is much lower than what AR used (50VAC-100VDC) for Callins caps)

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...