Jump to content

Some additional AR4x stuffing info. and off axis response testing


Carlspeak

Recommended Posts

I've just completed the restoration of two AR4x speakers Ser.#FX75842 & FX785854. The tweeters were dead and thanks to Roy C. I was able to replace them with used but good units. The remainder of the restoration consisted of the usual 20 uF cap replacement and cabinet exterior renewal.

The pots didn't clean up well so I replaced them with new Ohmite units.

These were original, unopened speakers so I decided to do some additional tests on the stuffing and replacement thereof. I first did a WT test on one of the speakers and got a Qtc of 1.09 and followed up with a close mic response test. Below is a ATB-PC response chart of the as yet, unopened 4x speaker. The results show a hump in the 80-150 hz range with a slow tapering off of the response up to 1 khz and then a rapid drop beyond that due to the series inductor. I have seen this same response shape previously in a different set of 4x's with similar Qtc's.

Both speakers were opened and the rock wool removed and weighed in at 9.6 and 7.8 oz. This was suprisingly low based on my previous work and also what John O'Hanlon has published in the past. http://www.classicspeakerpages.net/IP.Boar...cabinet+damping

I found the rock wool had been installed mostly around the inside edges with very little located behind the woofer.

Both speakers had inductors marked with the number '200'. This indicated 200 turns or, a #4 coil which is expected for ser. no's as low as these.

Following John's recommendation on stuffing 4x's with #4 coils with 18 oz. of FG, I weight out 18 oz of OCFG removed from a roll of st'd R-13 paper backed insulation. Of course, I removed the backing paper prior to weighing. When I stuffed the cabinet with the OCFG I found I could only put in 12 oz without jamming more in with excessive force. So, I decided to install the cloth surround woofer which I had resealed with some new material Roy is developing. The Qtc measured at 0.95, Fs=60 hz. A close mic response test was run which resulted in a response shape with no hump and the bass extending lower with an F3 of about 50 hz (see resp. chart marked AR4x ser FX75842 close mic'd woofer 12 oz. OCFG).

I ran one final set of response tests on the completed restoration which included off axis measurements which, I believe, have not been shown here before.

post-100237-1240237348.jpg

post-100237-1240237367.jpg

post-100237-1240237392.jpg

Link to comment
Share on other sites

Thank you for posting more excellent measurement data on vintage ARs, Carl!

Murphy's results are here for comparison:

http://murphyblaster.com/content.php?f=AR4x.html

My own findings, and some suggestions as to mods and tweaks, are disclosed in this thread on AudioKarma:

http://www.audiokarma.org/forums/showthread.php?t=123113

Clearly, with the stock tweeters, AR4x must be toed-in for direct-field response to have audible content in the highest octave, as -10 dB effectively equates to "Nobody home."

Higher density fiberglass is available if you want to get more damping accomplished (or the same, probably, in less space,) but remember that nearfield woofer response measurements reflect 2-pi, one-boundary response; those using them in freespace, as in on stands out from the wall, may prefer less damping....

Link to comment
Share on other sites

Nice work, Carl! I'm glad the tweeters worked out. Make sure there are no small rodents stuck to that sticky sealant I sent you. They have a tendency to alter woofer response! :angry: Thanks for posting your data!

Zilch, The AR fiberglass was of a high density. Can you share a source for the higher density fiberglass you mentioned? Thanks!

11 or 12 oz of fiberglass was typical of the majority of 4xs. The earlier ones were packed tightly with 18oz (mostly with rockwool)... so Carl should be in the ballpark.

Roy

Link to comment
Share on other sites

The AR fiberglass was of a high density. Can you share a source for the higher density fiberglass you mentioned? Thanks!

Owens-Corning sells theirs under the name "ProPink." For other brands, look for 9" batts rated R-30 or 12" batts rated R-38.

Link to comment
Share on other sites

Let's not confuse thickness with density. Your local lumber yard will have R-15 home insulation, or be able to get it, but the real poop is here:

http://www.bobgolds.com/AbsorptionCoefficients.htm

Higher R-values for the same thickness = higher density. When trying to talk to a building supply stock person, a 9" batt rated R-15 won't be considered "high density;" it's in the lower range of "medium density." That's assuming you're able to find someone who even knows what "density" is.

If you can refer Carl to a supplier who's willing to sell him small quantities of FG spec'd by absorption coeffcient, that'd be great, but if he's stuck with getting the stuff in bags from his local retail building supply or Lowes, R-values and inches of thickness are pretty much all he's going to have to choose by.

Link to comment
Share on other sites

When trying to talk to a building supply stock person, a 9" batt rated R-15 won't be considered "high density;" it's in the lower range of "medium density." That's assuming you're able to find someone who even knows what "density" is.

I believe I said "higher" density. Not SURE, but I believe an R-15 3.5" batt would be higher density than an R-13 one, which is higher density than an R-11 one.

I believe Google will reveal sources for Owens-Corning 703 as well as many of the others shown there. I particularly like semi-rigid HVAC plenum liner for small cabs, and fiberglass of similar density may be sold as duct or water-heater wrap in home improvement centers.

In any case, there are online mail-order suppliers of fiberglass in varying thicknesses and densities; it's not terribly expensive to ship. I do recall Mr. Widget on the Lansing Heritage linking to the particular one he uses.

How 'bout YOU look some up for Carl and everyone else here? :angry:

Link to comment
Share on other sites

How 'bout YOU look some up for Carl and everyone else here? :angry:

I did, keeping in mind Carl's long history here of recommending O-C pink fiberglass as a replacement for original AR fiberglass and rock wool, and the fact that he had just described his results attempting to restuff an AR-4 with R-13 O-C pink batting. If he wants to add 50% more by weight to the same volume, going from R-13 to R-15 batts isn't going to be nearly enough of an increase in density. R-30 might need a little fluffing, but if Carl is installing one of his rheostat-protecting enclosures around the new ohmite, maybe not.

Carl, I plunged my hands into a box full of AR fiberglass for the first time in some 20 years just last fall, and my impression of the stuff was that it was much denser than the R-15 O-C pink from HD that I used to insulate my garage roof. Let us know what you think if you try the R-30.

Link to comment
Share on other sites

I particularly like semi-rigid HVAC plenum liner for small cabs, and fiberglass of similar density may be sold as duct or water-heater wrap in home improvement centers.

Without knowing the density, I've used the hot water heater wrap in the past with good results as well.

As a bonus, it is often yellow in color, like the old AR fiberglass...and it is common audiophile knowledge that the yellow fiberglass sounds better! :angry:

Roy

Link to comment
Share on other sites

I did, keeping in mind Carl's long history here of recommending O-C pink fiberglass as a replacement for original AR fiberglass and rock wool, and the fact that he had just described his results attempting to restuff an AR-4 with R-13 O-C pink batting. If he wants to add 50% more by weight to the same volume, going from R-13 to R-15 batts isn't going to be nearly enough of an increase in density. R-30 might need a little fluffing, but if Carl is installing one of his rheostat-protecting enclosures around the new ohmite, maybe not.

Carl, I plunged my hands into a box full of AR fiberglass for the first time in some 20 years just last fall, and my impression of the stuff was that it was much denser than the R-15 O-C pink from HD that I used to insulate my garage roof. Let us know what you think if you try the R-30.

And your findings were what?

http://www.owenscorningcommercial.com/

R-30 is more than twice as thick as R-15, and thus, one might reasonably suspect it is actually less dense. The ratings are for the assembly; we need to know the actual density to conclude anything. There's a toll-free number provided.

Home Depot used to carry the pink blown-in AttiCat, which is chunks, much like what came out of my AR3a's.

http://www.audioheritage.org/vbulletin/sho...amp;postcount=7

Link to comment
Share on other sites

Thanks, Zilch...

I pulled this from your link above.

http://www.wrap-on.com/products/pi.html. It also mentions the wrap being effective as a "sound-deadening material".

...and I just recalled getting the idea to use heater wrap from a Ken Kantor post here some years ago.

We may actually have a consensus of opinion forming here.

Roy

Link to comment
Share on other sites

R-30 is more than twice as thick as R-15, and thus, one might reasonably suspect it is actually less dense. The ratings are for the assembly; we need to know the actual density to conclude anything. There's a toll-free number provided.

Home Depot used to carry the pink blown-in AttiCat, which is chunks, much like what came out of my AR3a's.

http://www.energysavers.gov/your_home/insu...m/mytopic=11530

The ratio of low/medium/high density FG batting for a 2x4 stud wall is R-11/13/15. Higher numbers are, of course, for thicker batts intended to be laid over attic rafters. So the ratio of high to low is about what is needed to put 50% more stuffing by weight into a cabinet. It does occur to me, though, that if Carl already started out using R-13 batts of the high density material, he may be SOL at his local hardware store.

HD used to rent the blow-in machines, as well, and when I was a kid long before the days of HD, my dad used to be able to buy bags of loose rock wool at our local HW store. The loose stuff seems to have fallen out of favor with DIY'ers, or at least with businesses selling to them, possibly due to everyone's worries about product liability following new data suggesting that prolonged exposure to loose FG particles may be just as hazardous as asbestos exposure. The batts are treated with binders to minimize shedding of particles.

Link to comment
Share on other sites

It does occur to me, though, that if Carl already started out using R-13 batts of the high density material, he may be SOL at his local hardware store.

Carl's measurements indicate that his AR4x's may, in fact, already be overdamped, depending upon intended use.

If he wants more density, R-15 will provide that, in the same space.

Others here need to become conversant with nearfield woofer measurements in closed boxes and submit their findings. It's slam-dunk simple, and can be done with an SPL meter, even....

Link to comment
Share on other sites

I haven't experimented with other R-types of OCFG. I do know that what I use weighs in about about 1 lb for each 5 ft of R-13 stuff 15 inches wide. That equates to a nominal density @ 3.5 inches thick (relaxed) of .55#/cu. ft w/o the backing paper. With 12 oz of this stuff crammed into a .65 cu. ft AR4x box, the density is about doubled to 1.15 #/cu.ft.

Link to comment
Share on other sites

Carl:

thanks for posting the 4x data; nicely done. It has been a while, but I recall the old rockwool to be denser starting material than FG. Apparently AR used it for a couple of years when FG was not as readily available or was more costly. Yes, it was difficulty to put 18 oz of FG in a 4x cabinet and for what ever reason I settled on #5 coil with 12 oz. FG. Have you measured that combination? As we know it is not the density of the starting material that matters. There are two variables in the wave propagation equations: fiber diameter (about 5 um for FG), and the fraction of the total interior volume filled with air, or porosity of the stuffing, about 0.982-0.992.

I was never sure if AR changed the stuffing at the same time it changed from cloth to foam woofer, or if the changes were independent. Historically, it would seem that several AR speakers had simiilar coil and stuffing changes in mid 1968. Those changes could have been to compensate for woofer construction differences, or it could have been a marketing-driven change in repsonse to competitive speakers.

Cheers,

John

Link to comment
Share on other sites

Interesting data Carl, we should question if it passes a sanity check.

First, do you have the impedance plots, anything out of the ordinary?

Where both under the same conditions of being measured through the

crossover?

Consider that a Qtc of 1.09, as compared to .95 is not large on a percentage basis,

and with regard to the amplitude sensitivity of the transfer function.

Link to comment
Share on other sites

Interesting data Carl, we should question if it passes a sanity check.

First, do you have the impedance plots, anything out of the ordinary?

Where both under the same conditions of being measured through the

crossover?

Consider that a Qtc of 1.09, as compared to .95 is not large on a percentage basis,

and with regard to the amplitude sensitivity of the transfer function.

Attached on the right side is the plot 1.09 Qtc test of the speaker with rock wool. This test was done with a dead tweeter installed and the 20 uF wires cut and original AP rheostat, which is why it looks like a stand alone woofer test. Also attached is the plot for the .95 test. Both look normal to me.

Yes, both tests were done with the woofer mounted in the cab with the rock wool and OCFG respectively and both had the #4 coils in the circuit.

I also attached a copy of an original AR4x woofer test. The curve is quite similar to my rock wool 1.09 Qtc result attached to my original post in this thread.

Admitttedly, the response was altered with the new stuffing. Is the sound better or worse? That's up to personal interpretation and tastes.

I speculate that AR deliberately boosted the Qtc to boost the low end response (Johnieo has mentioned this in a separate post as well) a bit in this little book shelf speaker - seemingly to the delight of numerous owners.

BTW, I couldn't access the pg numbers in your book link. I assume it was meant as a follow up to your prior post?

post-100237-1240487778.jpg

AR4x_woofer_and_tweeter_resp_by_AR.pdf

post-100237-1240489349.jpg

Link to comment
Share on other sites

Thanks for the impedance plots, looks reasonable.

Yes there are some plots in that book, I loads fine for me, I wonder if

others can read it.

Link to comment
Share on other sites

Thanks for the impedance plots, looks reasonable.

Yes there are some plots in that book, I loads fine for me, I wonder if

others can read it.

That page is not available in what I see, either; I suspect that access is randomized. Unfortunately, it is clearly identified as copyrighted material....

Link to comment
Share on other sites

  • 3 weeks later...
I've just completed the restoration of two AR4x speakers Ser.#FX75842 & FX785854. The tweeters were dead and thanks to Roy C. I was able to replace them with used but good units. The remainder of the restoration consisted of the usual 20 uF cap replacement and cabinet exterior renewal.

The pots didn't clean up well so I replaced them with new Ohmite units.

These were original, unopened speakers so I decided to do some additional tests on the stuffing and replacement thereof. I first did a WT test on one of the speakers and got a Qtc of 1.09 and followed up with a close mic response test. Below is a ATB-PC response chart of the as yet, unopened 4x speaker. The results show a hump in the 80-150 hz range with a slow tapering off of the response up to 1 khz and then a rapid drop beyond that due to the series inductor. I have seen this same response shape previously in a different set of 4x's with similar Qtc's.

Both speakers were opened and the rock wool removed and weighed in at 9.6 and 7.8 oz. This was suprisingly low based on my previous work and also what John O'Hanlon has published in the past. http://www.classicspeakerpages.net/IP.Boar...cabinet+damping

I found the rock wool had been installed mostly around the inside edges with very little located behind the woofer.

Both speakers had inductors marked with the number '200'. This indicated 200 turns or, a #4 coil which is expected for ser. no's as low as these.

Following John's recommendation on stuffing 4x's with #4 coils with 18 oz. of FG, I weight out 18 oz of OCFG removed from a roll of st'd R-13 paper backed insulation. Of course, I removed the backing paper prior to weighing. When I stuffed the cabinet with the OCFG I found I could only put in 12 oz without jamming more in with excessive force. So, I decided to install the cloth surround woofer which I had resealed with some new material Roy is developing. The Qtc measured at 0.95, Fs=60 hz. A close mic response test was run which resulted in a response shape with no hump and the bass extending lower with an F3 of about 50 hz (see resp. chart marked AR4x ser FX75842 close mic'd woofer 12 oz. OCFG).

I ran one final set of response tests on the completed restoration which included off axis measurements which, I believe, have not been shown here before.

Carl,

I think your response curves, especially off-axis, dovetail fairly closely with the factory anechoic curves. For example, your 60-degree off-axis curve appears to show about -15 dB or so at 18-20 kHz, and the factory anechoic curves of the speaker are in the -12 to -15 dB range at 60-degrees off axis at 20 kHz. You may be slightly underdamped, but otherwise those look pretty good to me! Nice job!

--Tom Tyson

Link to comment
Share on other sites

I think your response curves, especially off-axis, dovetail fairly closely with the factory anechoic curves. For example, your 60-degree off-axis curve appears to show about -15 dB or so at 18-20 kHz, and the factory anechoic curves of the speaker are in the -12 to -15 dB range at 60-degrees off axis at 20 kHz. You may be slightly underdamped, but otherwise those look pretty good to me! Nice job!

And what factory anechoic curves would those be, please? Not ones for "the speaker" as you suggest, surely.... :rolleyes:

Link to comment
Share on other sites

Carl,

I think your response curves, especially off-axis, dovetail fairly closely with the factory anechoic curves. For example, your 60-degree off-axis curve appears to show about -15 dB or so at 18-20 kHz, and the factory anechoic curves of the speaker are in the -12 to -15 dB range at 60-degrees off axis at 20 kHz. You may be slightly underdamped, but otherwise those look pretty good to me! Nice job!

--Tom Tyson

Thanks Tom. Actually, those were measured in my "den-achoic' room :rolleyes: with the test speaker sitting atop an AR3a

Link to comment
Share on other sites

Thanks Tom. Actually, those were measured in my "den-achoic' room :rolleyes: with the test speaker sitting atop an AR3a

If you look at your off-axis curves and compare to the factory chamber curves, you can see the similarities. I do think that your stuffing might have resulted in too-little damping around resonance for the woofer, however. Note that the factory system only has around a 2-3dB rise at resonance, but I think your bass response at that point had a bit more output. Conditions are very different, even though you used the close-mike technique, but you should be able to replicate the 2Pi woofer curve reasonably closely down close to resonance. Adjusting the amount (adding) of fiberglass can change the acoustical damping, which can bring down any rise at resonance. Overall, you did a good job I think.

--Tom Tyson

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...