Replacement AR-4x Tweeter: A Mod In Work

[ReliaBill Engineer]

I have no confirmation bias at present. Just comparison listening. I learned decades ago that just because I had a hand in the work, doesn’t make it better than another sample, or an original sample. I learned to try again until I got it right, or achieved what I set out to achieve. 
 

I worked doggedly for 6 months on a new stylus for the GE VR1000 cartridge. Originals were awful. Only when the stylus passed my tests and comparisons did I call it “good”. Then I sent one to a friend in Phoenix who had worked on the cartridge for 3 years trying to improve the stylus. He gave it a good listen over a week. He said it was the best sounding, best performing stylus for the VR1000 he’d ever heard. He already had a stylus business and began selling mine.

[Aadams]

And now we have page three.  This is a rare thread, drifting into a dendritic form that is reminiscent of something I have seen on TV.  I sense someone is wearing an onion on their belt.

[K6cdxkms]

I have not seen that term "dendritic" used in a many a fortnight, so long in fact that I had to look it up and confirm the usage.

Now back to the real subject at hand..............trying to improve the frequency response/tone... from a replacement "mid/tweeter" for the 4x.

Personally I am willing to wait on Bills video results and go from there as to what the listening experience results. 

Bill , thanks you for your work,

Ken

 

[ReliaBill Engineer]

10 hours ago, Aadams said:

And now we have page three.  This is a rare thread, drifting into a dendritic form that is reminiscent of something I have seen on TV.  I sense someone is wearing an onion on their belt.

No idea what that means. Veiled cynicism? I’m not sure.

Next 3 days will be overcast and raining here. So I should be able to spend time on this project, huddled away in the garage.

I’m anxious to hear the original 4X configuration. The only thing I’ll be changing is the pot and driver seals; original cap, original rock wool damping, original drivers. 
 

So I’ll continue on. It was never my intent to publish a white paper on this with complete Hirsch-Houck lab results.

Neither is this a preamble to a patent application. 

[ReliaBill Engineer]

Scouring through many posts and pages on AR speakers. Even the AR3/3a document. I see very few comprehensive measurements of output from the drivers and speakers, before and after refurbished. Or contemporary (when new) charts comparing to refurbished woofers, mids and tweets. 
 

Same for those refurbishing original AR drivers, particularly dome mids and tweeters. I find that a bit curious. Especially by those calling for exhaustive tests on a budget replacement tweeter. Maybe I looked in the wrong places.

Just lining up what to test and how.

[ReliaBill Engineer]

Dentritic. Broadly means branching out. But doesn’t apply here. 
 

What we have here is pretty straight forward:

1: Discussion of a replacement 4X tweeter. PRT was mentioned, with its faults.

2: Decision to get a pair of PRTs, look at it.

3: Decide to change the PRT and listen to it. Compare to unmodified. Use 2ax as a “test bed”.

4: Decision to purchase a pair of 4X speakers. To listen to. To compare PRT to original.  Fix speakers first.

So there you have it. All the dentrites.

[genek]

You might not have noticed, but the site is a bit stale because its owner, who hosts it as a personal project rather than a business, hasn't had time to update it for a long while. And several of our long-time members who were the primary source of deeper technical content like measurements have sadly passed away. So you have to dig long and deep to find some of that stuff.

In the name branded forums, the focus is on preservation and restoration, and discussions about changing drivers are focused on finding or creating "holy grails," new drivers that can be made to reproduce original sound. This forum is anything goes as far as changes go. But pretty much everywhere on the site, people will almost always want to know how something sounds and how it measures, because we're a bunch of audio nerds. Otherwise, we'd all just have home theater systems from BestBuy.

[ReliaBill Engineer]

Thank you Gene for the context.

This is a hobby for me. A great break from a highly technical world in which I work. I would love, at some point, to invest in measurement tools. It has to meet with the WAF; she doesn’t understand much of my work, nor my hobby. In a WebEx meeting now, where NASA has changed the timelines (again) for delivery of the 2nd and 3rd SLS vehicles for Artemis. 
 

It would be great if there was a consolidated repository for various test results on work done on AR legacy and heritage speakers over the years.  
 

I promise I will make every effort to provide some measurements, warts and all. Good and bad. Sink or swim. I can’t make promises on a time frame right now.

[genek]

Sadly, the forums on this site are the closest thing there is to a repository for classic "East Coast sound" speakers. If we can ever get to update the libraries, there's a good 8-10 years worth of content that needs to be extracted from old forum posts.

[ReliaBill Engineer]

So here is one of the 4x speakers. The only part not original is the pot. I had to change that out. But original 20 uF cap, which measured 22.3 uF. Untouched tweeter and woofer. Set at the “dot” setting. (I marked the original pot position and wiper setting at the “dot” on the XO board prior to removing, 11.3 ohms on mine.)

Again, my amp is set on “mono”, so both stereo channels are being played through the single speaker. 
 

I think my modified PRT is going to sound very close to this original tweeter.

 

[ReliaBill Engineer]

So…..

I mounted the modified PRT in place of the original AR tweeter. Gave it a listen.

I had to add a parallel resistor to the PRT to match the DCR of the AR tweeter. 4.6 ohms AR vs 7.4 ohms PRT. End result is the PRT measures 4.75 ohms. 
 

 

 

After listening, I got the results I was expecting. The PRT is considerably more efficient, so I turned down the level control, from 80% to 40%. Played with it some, going up and down from 0 to about 65%. Too much volume from the tweeter above that. 
 

What doesn’t come across well in the video is a more “full” and wider sound from the PRT.

 

[ReliaBill Engineer]

Listening to recordings with vocals. Very smooth and natural. Bass is quite good for this small speaker, but that’s all AR’s doing. (But I will be treating the surround later. Ultra thin butyl rubber. But not a sticky formula; it will cure dry to the touch. I don’t like a forever tacky compound that attracts and holds dust and fuzz/hair.)

Still playing with the level control. About have it dialed in now. 

This tweeter has more “air” to the sound than the AR tweeter. While it reveals high treble extremely well, the midrange is smoother and more balanced. Not at all bright. Not irritating at all. Smooth and pleasant.

 

Emmylou Harris:

 

 

[ReliaBill Engineer]

After extensive listening, I would say this modified PRT is a major success story. It maintains the character of the original AR tweeter; It sounds very much like it. But the AR tweeter has a certain “graininess” to it that I find taxing over time. It lacks depth in its presentation. I would add that the unmodified PRT tweeter mirrors that graininess in its presentation. Neither the AR nor stock PRT are “pleasing” over time.
 

Turning down the tweeter level helps with reducing listener fatigue, but then (for me) frustration sets in because detail is muffled and the overall presentation is too warm and “chesty”.  
 

Doing the same with the stock PRT ends with the same frustration. 
 

I can see why vintage_ar parts added dope to the surround of the PRT, in an attempt to “tame” it. Not sure how well that works, since the source of the “graininess” is not eliminated; it’s coated over with dope. 
 

Other than aesthetics for both my modified PRT and the 4x speaker, my work here is done. I accomplished my goal and proved my theory and point. Start with a smooth driver, and the crossover becomes easy. It’s very difficult to design a crossover to “fix” a problematic driver. This PRT can be made to be smooth and detailed while keeping its overall sound character; a sound character very similar to the AR tweeter without its flaws. 
 

From vintage_ar parts, regarding their replacement PRT tweeter for the AR-4x: 

 

 

 

[ReliaBill Engineer]

Forgot to add that I will be changing out the AR paper-wax cap and using metal film p-p caps. Also I will add a “dressing” to the modified PRT, so it looks more like the AR tweeter. Not paint, in case anyone is wondering.

And unrelated to the tweeter, the AR woofer surrounds will be sealed, as well as cabinet refinished and new grills installed. 

[ReliaBill Engineer]

Listening to big band jazz (Glenn Miller) and Bobby Hackett on the Epic (Columbia) label. These can be bright and brash. Trumpets sound “bloomed” and a bit harsh on both the AR tweeter and the stock PRT. With the modified PRT brass is smooth, detailed with “bite” but not harsh. Even Al Hirt sounds nice! And his horn can sound irritating at times. Brushed cymbals and cymbal strikes come through clearly and with nice metallic shimmer. 

[ReliaBill Engineer]

Maybe tonight I’ll get the caps changed to metal film p-p. I expect this to exacerbate the problem with bright tweeters on the 4x, both AR and stock PRT. The old AR wax-paper caps tend to “mellow” the tweeters. 
 

After that, I’ll apply butyl rubber to the woofer surrounds. As they are, they leak air. I used a rubber tube against the surround and was able to pass air through the rolled surround. I’ll snap a few pics as I apply it. Aesthetically, there won’t be much change in appearance from the untreated surround. But I prefer that. I don’t want a shiny appearance nor a sticky substance on it. Nor a change in the color. I don’t use toluene. I use a more gentle, slower evaporating solvent. It allows better “flow” and soaking of the fabric. 

[ReliaBill Engineer]

This second 4x in the pair is going to need a lot of work!

Everything measures ok. Just aesthetically not too pleasing.

[ReliaBill Engineer]

Today I disconnected the original 20 uF capacitor and added the metal film polypropylene cap. I took a pic of the pot position and measured the resistance at optimum setting, just for my own record keeping. Then I replaced the back cover on the pot. 
 

I like these 15 ohm 25 watt pots. Beefy, robust and smooth. The position shown is for the PRT tweeter. 

The wires from the original cap have been cut:
 

 

Listening after the cap was replaced gave the expected, and somewhat unexpected results. More crisp clarity from the modified PRT with the new cap. But yet still smooth and effortless. No fatigue at all. 

[ReliaBill Engineer]

One very interesting thing occurred to me as I was listening to this modified PRT, (post cap replacement) vs the original tweeter and the stock PRT. It also occurred to me when I was using this modified PRT on the 2ax, in place of both the original cone midrange and original phenolic dome tweeter pair. In fact, I stated it when I was listening to the 2ax using this modified PRT.
 

This modified PRT sounds so close to the 2ax mid+tweeter it’s just uncanny! It possesses that articulate and smooth nature of the AR upper tier speakers. Conversely, the original AR-4x tweeter does *not* have the characteristic sound of those upper tier AR speakers! The original AR 4x tweeter is somewhat brash and has (to me) an unwelcome “in your face” quality to its sound. Also, it beams, having less dispersion off-axis. Using the pot to lessen this brash sound results in a more numb, warm and chesty sound that I don’t care for either. There’s no middle ground. I’m pretty sure this is the sound SpeakerDave was trying to tame using crossover mods for the original 4x tweeter.