Replacement AR-4x Tweeter: A Mod In Work

[ReliaBill Engineer]

I decided it would be best to discuss this in here, rather than in the Acoustic Research section. This way it can be seen without potentially hijacking a thread, nor competing with genuine original AR tweeter discussions. 
 

In essence, when the original 4x tweeters blow, or are otherwise damaged and non-functional, many replace them with the Phenolic Ring Tweeter (PRT) sold in various outlets such as Parts Express, eBay vendors, Amazon, etc. However, while the PRT fits nicely, and is cost effective (around $20), it doesn’t perform the same as the original AR tweeter. 
 

The small AR-4x speaker is a nice sounding legacy speaker, despite its small size. A common failure in these is the tweeter.

Out of pure curiosity I decided to buy a pair of these PRTs and have a look at them. They are reasonably well made. But a bit “harsh” in my opinion. 

Can these be made to be better sounding? Can they more closely mimic the performance of a genuine AR-4x tweeter? 

That’s what I want to find out. I may fail or succeed. 
 

Original 4x tweeter:

 

PRT, stock:

 

Typical listing:

[ReliaBill Engineer]

In rough shape, but intact:

These will be my baselines for comparative listening. After some refurb, of course. 
 

 

[ReliaBill Engineer]

The new PRT:

 

After removing the factory surround. I decided to add some acoustic damping under the cone, to attenuate the back wave. Black wool felt and a little wool fiber:

 

After the felt is installed:

 

 

 

[ReliaBill Engineer]

I started with a butyl rubber roll edge. But it turned out to be too stiff:

 

So I removed it. Now I’m working on a different surround. It’s curing as we speak. It should be more flexible, yet damp well. 

[ReliaBill Engineer]

Instead of a butyl “roll” edge, I’m trying a treated edge.

High quality, high thread count silk. Treated with butyl rubber. High flexibility but sealed and damped. I used parchment paper so the treated silk just pulls off after fully curing. 
 

Later I’ll cut what I need for the tweeter. Silk is bad about fraying; treating it first will allow me to cut it without fraying.

The 2 rods were used to keep the parchment paper flat. It wants to curl.

 

 

 

 

 

 

[ReliaBill Engineer]

Letting the sealant cure.

 

[ReliaBill Engineer]

The thin butyl sealant is fully cured. Pushing in on the dust cap of the cone, the surround is flexible and moves evenly all around the cone edge. 
 

Now to mount these in a pair of 4x speakers. But I’ll need to get the original 4x speakers working fully. The pots are corroded. Fortunately, I have spare parts from my 2ax refurb; 2 new pots and some metal film polypropylene capacitors. 
 

I will be able to listen to the original tweeters, then compare original to the stock PRTs, original tweeters to my modified PRT, and stock PRT to modified PRT.

If all goes well, I’ll take these over to a friend for some measurements of impedance curves and FR plots. 

[ReliaBill Engineer]

Decided to give the tweeter a test drive. Used my 2ax speakers. Disabled the super-tweeter using the pot, already had the AR mid-tweets disabled, wires disconnected and running out behind the mid-tweets. 
 

It should be a relatively good test run. The 2ax is a 2-way at its heart. The mid-tweet is a nominal 8 ohm driver (6.6 ohms DCR). I used a 60 ohm resistor across the PRT terminals reducing its DCR to 6.5 ohms from 7.3 ohms. Tweets mounted on the top of the cabinets in acoustical (leftover from room treatment panels) foam blocks.

Crossover should be around 3000 Hz using the 6 uF capacitor. 

At the 70% Mid-pot position, this modified PRT sounds quite good!

I did the same for the other 2ax speaker, but used the stock PRT.

Comparison using a mono signal source, same to both speakers, switching between L and R, reveals a difference between how the 2 PRTs sound. 
 

The stock PRT is brighter, with a seeming larger suck-out in the midrange. My modified PRT has a much more present midrange and less bright, smoother treble. 



 

 

 

 

[genek]

Fyi, there's a cap on uploads. You'll fill up fast with crossposts. Better to use the share function to let people know there's a new post in another forum.

[ReliaBill Engineer]

Ok. Good to know. Thanks!

But I didn’t load photos on the AR section of my last post.

I noticed no comments, but quite a few views in here.

Guess I’m being stalked….

[genek]

The upload cap is cumulative across all forums you post to.

I expect that there are some who will follow your project's progress in Mods and Tweaks because they're interested in trying it themselves and others who will just wait to see what results you post in the AR forum and whether you'll start selling them because they're only interested in drop-in replacements they can buy.

[ReliaBill Engineer]

After listening to this modified tweeter, is it worth the effort? 
 

I’ve played 8 of my favorite LPs, and changed out cartridges 5 times. I can clearly hear the differences between the cartridges, which is as it should be.  I put on the ADC 10E MKIV, ADC XLM MKIII, Shure V15-III and V15V-MR,  and Pickering V-15 with the nude elliptical stylus. Each makes its own fingerprint to the music; each was clearly differentiated through this combination of AR woofer and this tweeter. None of them sounded harsh or strident. Nice amount of “”air” and a more musically effortless sound. The cartridges sounded good, the music sounded good.

The stock PRT sounds “thin”, strained. Relatively “tinny” in comparison. On cymbals, blurred, not really clear and focused. 
 

Without sophisticated test equipment, all I can give are my subjective impressions at the moment. But we all know that graphs don’t tell us how something “sounds”, either.
 

So “yes!”, I like the sound of this tweeter. It no longer has the sound of a “cheap” $17 paper cone tweeter. So yes, I think it was worth the effort on this. 
 

Now I need to hear the original AR tweeter. Hear for myself how different they are, or possibly alike they are. 
 

Anecdotally, my wife had a comment. I kept the modified PRT in the left speaker, switched back to my AR drivers in the right speaker. Went back to stereo listening. Her off-hand comment was, “Why all the work? I don’t hear a difference now.” 

[ReliaBill Engineer]

YouTube sound clip. AR tweeter turned off, AR mid-tweet disconnected. Mid-tweet pot at 75% going to the modified PRT from the mid-tweet crossover.

Mono signal from the Emmylou Harris record. Shure V15-III cartridge with original nude elliptical stylus; known for it’s excellent detail and balanced, flat response.
 

 

[ReliaBill Engineer]

The 4x has a lower crossover point than the 2ax. 1200-1400 Hz vs 2000 Hz for the early 2ax. So it should be interesting to see how this modified PRT responds. 
 

The 8” AR woofer of the 4x should be more capable of response up to 2000 Hz than the 10” of the 2ax. But that’s not what AR chose. 1200 Hz is a low crossover point for a paper cone tweeter. The paper cone tweeter has a free air resonance only slightly lower than 1200 Hz. SpeakerDave went to a 3rd order network to try to tame the tweeter hump close to its resonance. If I have the same issues, I’ll raise the crossover point to 2000-2500 Hz for both the woofer and tweeter. That way I can keep the first order network.

[genek]

Your efforts would benefit considerably if you could measure the FR of your modded tweeter. It's harder to design a crossover if you don't know where your driver response rolloffs begin.

[ReliaBill Engineer]

I have that scheduled. As I said earlier, I have a friend across town with the test equipment. He repairs amps, speakers, cassette decks and R2R tape decks. Not a tech shop, he works from his home. He’s known in the state for doing good work. He’s retired, so does it on the side. That’s also why I haven’t molested the 2nd PRT. 
 

I was his maintenance manager when we both worked for Goodyear. He was an excellent troubleshooter. I used to park him at a troublesome machine for half of a 12-hour shift, to document problems, mechanical and electrical. He would come up with a troubleshooting list that was nearly always correct.
 

As an added feature, once I know what’s going on with the pair of 4x speakers I bought, if the tweeters work, I can test those too. 
 

Currently, I’m not aware of anyone who has measured/plotted the response of the PRT, despite it being sold as a replacement for the 4x tweeter.

[genek]

It would definitely be interesting to see what the response curve of an old AR-4x tweeter looks like. The published curves for classic era ARs are usually made up of individual driver curves superimposed on each other, rather than a single curve of the assembled speaker, so direct comparison between original and aged can readily be done.

[ReliaBill Engineer]

Besides seeming like the “right thing to do”, I added the wool felt and wool fibers under the cone for this reason: Fiberglass batting under the original AR tweeter cone. (borrowed pic)

The original tweeter has AR’s “trademark” use of fiberglass damping pads. They used it a lot, and often, on their early mids and tweets. 
 

I chose wool. I’ve found over the years that it works well. (NASA uses melamine foam in large panels to absorbs sound vibrations.)

 

 

 

 

[ReliaBill Engineer]

The treated silk suspension:

RoyC said this in a thread over in the AR section:

” It should be noted that the replacement PRT type tweeter does not have the same construction as the original 4x tweeter...which has a more compliant suspension. The PRT replacement is not as capable in the mid frequencies primarily due to excursion differences.”

This is exactly why I gave the PRT a different suspension. The stock PRT cone outer edge is tightly secured to the phenolic ring. There is little chance for the cone to have “excursion”. To handle lower frequencies, 800 Hz-2000 Hz, the cone has to have much better excursion than above 2000 Hz. But that excursion has to be controlled, hence my use of butyl rubber treated silk. 
 

So now it has a damped, more compliant suspension. More like the original AR tweeter in that respect. In my opinion, it now has a “chance” to have a smooth roll off at the 4x crossover point of 1200 Hz

[ReliaBill Engineer]

Another sound check. I use this 1962 stereo album to check for resonance in cartridges. The xylophone tends to cause styli problems. Using a Shure V15-IV which is very well controlled. 
 

My phone’s ability to handle the resonance is challenged. In the room, there is no ringing or resonance.
 

 

[ReliaBill Engineer]

Should be here today.

 

 

[ReliaBill Engineer]

So, I got the 4x pair today. Neither has been molested. Both tweeters are good, and measure 4.7 ohms DCR. Pots are bad. But I have new ones. I have new caps. All the seals were in place, undisturbed. Woofers are intact, no glue failures that I can see/feel. I flexed the cones; spider looks good, surround looks good. Cone is good. Paper scrim intact behind the woofer, rock wool never disturbed. Cosmetically they need TLC. But I have a good foundation here. April 8 1968 on woofers and tweeters. 

[ReliaBill Engineer]

Found this on-line for the 8 ohm PRT:

 

 

 

 

 

 

 

 

 

[ReliaBill Engineer]

So my 1968 AR4x use a “#4” inductor on the woofer, measuring 1.02 mH. The capacitor is labeled 20 uF, measures 22.3 uF. According to charts I have, that equates to 1250 Hz XO frequency on the 8 ohm woofer, and around 1500 Hz on the tweeter (4.7 ohms DCR, 5-6 ohms impedance.) 

All drivers are in full working order in these 4x speakers. So I’ll replace the pots. Get them working with the original caps and have a listen. Later, when my parts come in, I’ll replace the caps. I want to hear the AR sound. Later, I can put my mod’d PRT in, but I have to adjust the impedance of the PRT so the 4x caps “see” the same impedance as the AR original tweeter. And vice-versa, so I can listen to the AR tweeter on my AR-2ax speakers. (That’s just for fun, and to familiarize myself with the sound of the AR paper cone tweeter.)

[ReliaBill Engineer]

AR-4x. April 11 1968 on the woofers: