AR Classic 18 info.

[AR303 and classic 18 user]

Hello,

I recently joined this site to find some info on the AR classic 18 model. After searching the site I found one post from someone who wanted to sell a pair about 3 years ago and another outdated reference to an ebay auction.

I have been the original owner of these speakers since day one in '92 when I bought them in Pemborke, MA.

The drivers are still in excellent condition. The butyl rubber surrounds used for this series seems to be holding up real well so far... But, I am wondering about the quality of the crossover components after 18 years of service. Should I be worried about this?

Based on my prior reading on this board, I should point out, that these model 18s are the floor standing 3 way classic version with dual 5.25 and 1 inch tweeter symmetical array, with the 8 inch woofer.

Regards.

[gary_wong5]

I have same question as you.

Your AR Classic 18s' are driven by what amplifier?

I found that my 2-way Classic 8s' are quite power hungry and only sounded the best when driving by Carver 4000t/TFM-45. combination.

[AR303 and classic 18 user]

I have same question as you.

Your AR Classic 18s' are driven by what amplifier?

I found that my 2-way Classic 8s' are quite power hungry and only sounded the best when driving by Carver 4000t/TFM-45. combination.

Thanks for the response Gary. I did not know that there were two way members of the classic series. I found your posts concerning this model very enlightning. I too have found the speakers to be agressive on the amp, this to me seems sensible given that the 18 classic is rated nominal 4 Ohms so they may dip to three here and again? Don't know I never have measured or seen measurements for this model.

I suppose its a topic for another thread but I can briefly say I drive my setup with a sunfire cinema grand signature 400. (With the classics as the center channel.)

Say have you or anyone else with classic series speakers ever taken the back plate off the speakers? I was going to attempt to do so but I don't know if I am in for any surprises. It would be nice to know if I should expect the crossovers to be attached to the plate or not? I am going to try to review the restoring 3a techniques as a start to familiarize my self with the correct procedures.

It seems from reading some of the info on this site that I doing this might at least enable me to determine what the current state of the crossover components is and then take it from there.

[AR303 and classic 18 user]

Well progress. I have managed to take the rear plate off the back.

Here are some observations so far:

Well the crossover appears to be mounter to the rear of the speaker cabinet, so I will need to take the woofer off as well to get at it. (I do not have a Torx driver set so I need to wait till I get one.)

The stuffing appears to be polyester. Given the discussions about polyester vs fiberglass on these pages I thought I would mention this. Does this mean the speaker would have been designed for polyester? or would it been designed for fiberglass and polyester used as a cost cutting measure later on?

I took a few pictures for reference and decided to share since info on this model seems to be limited.

[AR303 and classic 18 user]

Smaller versions of pictures.

Regards.

[AR303 and classic 18 user]

Well the Torx driver arrived. Its a T25 that fits...

So I was able to get the woofer off. My first impression is that everthing seems pretty typical, you can see the stuffing after the woofer was off. Also I was able to get the part number of the woofer, 1210152-5 923TNB. Anybody know anything about it?; I will try to search around for some specs on this driver.

Next I took out the stuffing which revealed the crossover. It seems to me it was installed upside down since there was very little slack to the other drivers remaining? When I put it back in I will install it 180 degrees to how I found it and this should make it easier to remove the midrange. I have not done this yet but I want to see the subenclosures (not revealed yet as I did not destuff the whole cabinet) and how they are built and also get part numbers for the other drivers.

I took out the crossover. All looks well. I have not measured anything yet. I need to go back to the capacitor reference on this site to try to determine what's up. These caps are marked as NP? I am not an EE so does this tell something about their type?

Anyway, more soon...

[AR303 and classic 18 user]

The latest info for this week.

I was able to obtain a Dayton Audio WT3. I have always been interested in speaker measurements and decided it was about time I had that capability on my own. I won't have much more time this weekend but it did arrive today so thought at the very least I should take it our of the box and measure something to see if it works. So I started with the speaker I did not take apart, so it is still 100% original never been opened etc. Here is the impedance and phase curves for this speaker. Not too bad the in box resoance seems to be at ~37 Hz right in line with the -3dB, 40 Hz spec. Should I expect the in-box resonance to be near or slightly below the -3dB point?

What does the curve of the midrange indicate? The AR spec says the midranges are in their own sub-enclosure and have a -3dB point of 200 Hz. I am assuming the second peak in the impedance curve is for the midranges. The resonace from the curve would seem to be at ~250 Hz. How should this data be interpreted? Does having two parallel midranges effect the measurements? or is it reasonable to assume that the midranges can respond below the enclosure resonance?

Also the nominal impedance does seem to be 4 Ohms as advertised by Ar with a couple of dips to 3.5 as I was speculating about above.

Cheers.

[AR303 and classic 18 user]

I had some time over the week to start playing around with the WT3 tester. I also was able to remove the 5.25 midrange from its enclosure. Attached are the photos. The part number of the midrange is 1210150-5. Also can be seen is the sub enclosure of the midrange unit. Once I saw its construction I realized I had an easily measureable volume and hence a pretty good case study for my first speaker to measure. So I hooked it up to the woofer tester.

Here are the parameters I was able to measure:

Fs = 49.80 Hz

Re= 6.83 Ohm

Qms=2.12

Qes=0.64

Qts=0.492

Sd= 66.5 cm2

Vas = 8.0 L

Xmax= ? Is 2.0mm a reasonable estimate?

Le=0.87

I then was able to repeat measurements under different box conditions, empty, stuffed with the AR stuffing and then stuffed with fiberglass. No surprises here there are other threads in the group that show the same trends fiberglass giving the most damping. The graph shows the measurements.

Well now that I had the thiele-small parameters I wanted to simulate the system to see how it would compare to the measurements. I obtained the latest version of Unibox and setup a model for the AR midrange using the parameters that I measured above. I am using a volume of 0.9 L for the simulation model as this is as close as 0.93L that I calculated for the enclosure which is the enclosure volume of 1.2 L with allowances made for the volume taken by the speaker magnet and motor assembly. There is also a graph attached that show the results of the simulation. The stuffing was modeled by adjusting the Qa parameter in Unibox. The fiberglass is modeled with a value of 5 which corresponds to the heavy fill item in Unibox. The AR stuffing required a number of nine for the adsorption Qa.

I not sure how to tell if the numbers are reasonable or not, obviously the simulation Fbs are skewed a little higher than the measured values and the Qtcs are a little lower however overall the attached results seemed to fit the best?

[AR303 and classic 18 user]

Now that I seem to have the process down for the measurements I decided to tackle the woofer measurements. Its up above but I will repeat the part number for the woofer is 1210152-5. Here are the parameters I was able to measure:

Fs = 28.26 Hz

Re= 3.65 Ohm

Qms=2.81

Qes=0.55

Qts=0.459

Sd= 205.9 cm2

Vas = 68.0 L

Xmax= ? (Anybody care to guess??)

Le=0.61

I had to guess at the enclosure volume, I have decided on 30.4L, which I actually calculated from unibox by simulated the woofer in an empty box with no stuffing and matching this to the measured data. After a few iterations the numbers seemed to converge on 30.4 L for the box size with a Vas of 68 liters for the driver. This seems reasonable for a driver with this Fs as I compared to a couple of other make drivers with similar (but higher say 33 Hz Fs). If this isn’t appropriate way to measure these parameters, please let me know. (But, it’s going to be rather difficult to actually calculate the box volume due to all the odd angles of the cabinet construction.) I did however, to test the method out, use this same method to “guess” at the midrange enclosure which yielded a result of the correct volume 0.9 L, which I was able to calculate reasonably accurately.

As before are attached the measured data and the simulation results from unibox. Curiously, or consistently the Fbs for the measured data are slightly lower than those predicted by Unibox. I thought maybe before this effect was caused by the small volume of the enclosure I was trying to simulate. Has anybody else seen this effect?

Anyway also interesting in the measured data I believe you can see a standing wave resonance at around 180 Hz which disappears when the AR stuffing is added. I did not have enough fiberglass to try the cabinet stuffed with this material. However the Qts of the system with the AR stuffing is ~0.7 so I may not do this.

That’s all for this week. I still need to get the tweeter part number trace the crossover and oh yes finally measure the caps (the original goal). Oh the fun of a hobby and the while you are at it effect…

[AR303 and classic 18 user]

I was able to trace out the crossover today. L3 I was not able to determine as the label is underneath the inductor. Also what is T1? You can see from the pictures the large rectanglar object marked only "T1 made in mexico".

AR18Classic_Crossover.pdf

[AR303 and classic 18 user]

Oh I forgot, the caps seem fine so I have left them alone for now...

[briodo]

I was able to trace out the crossover today. L3 I was not able to determine as the label is underneath the inductor. Also what is T1? You can see from the pictures the large rectanglar object marked only "T1 made in mexico".

Nice documentation on the refurbishment. It's great to see an older system coming back to life!

I assume the T1 you mention is a non-polarized capacitor wired in series with the tweeter. I'm not sure why it was labeled as T1, but every AR schematic I have seen has something like a 3 to 8 uf cap wired in series with the tweeter (values are for example not the rule).

I am in the process of rebuilding an AR9LSi with a similar crossover circuit card, but the caps are hot glued between the bottom of the card and the back cabinet wall. I do not have the measurement tools to check them but am going to replace "just in case". My experience doing an AR9 cap replacement was startling.

Also, the plots are impressive. I recently downloaded an app on my Apple Ipod Touch called "AudioTools" which includes a real time audio analyzer and works with the Iphone. I have much to learn about audio measurements and want to share my thanks for taking time to post your results. This helps me understand how the tools can be applied.

Brian

[AR303 and classic 18 user]

I just realized I never posted the tweeter part number. Its 1210148-0.

Installation is pretty simple as can be seen.

Brian, I am glad you enjoyed the post. It was fun to put together and cool to see first hand the stuff up until then I had only read in book come to life in front of me.

Well all is back together now and ready for more years of listening.

[AR303 and classic 18 user]

Here is some more info on the crossover.I was able to obtain an RLC meter and measure values for those components that were unknown in the previous schematic. I also found some corrections and the attached image is corrected. I thought it would be of interest to list the measurements since degrading cap performance is a regular topic here. The 90's era caps used in this model seem to be holding up real well so far. Below is shown the specified value followed by the value that I measured.

                        Specification               Measured
L0                       3.40 mH                     4.3 mH  1.3 Ω
C2                       50 µF                       52.4 µF
R17                      6 Ω                         6.5 Ω
C3                       100 µF                      105 µF
C12                      100 µF                      107 µF
L15                      0.405 mH                    0.351 mH 0.705 Ω
C13                      24 µF                       28.2 µF
R14                      10 Ω                        9.6 Ω
R11                      0.5 Ω                       0.69 Ω
C10(T1)                    ?                         8.3 µF
L8                         ?                         0.322 mH 0.672 Ω

T1 does indeed appear to be a capacitor as had been previously suggested.

[AR303 and classic 18 user]

The investigation into these speakers continues...

Since the last post I have been able to get measurements of the raw drivers in the enclosure without crossover parts. Attached are the results. Here are a few notes on the setup. Its proably not ideal but works well enough for my purposes...

The mike used was a radio shack analog sound meter that was calibrated to the radio shack spec in the manual.

SPL data is relative only and was not calibrated to an absoulute level.

All are measurements taken at 1.0 V rather than standard 2.83 V. (Did not want to risk tweeter...)

Soundeasy MLS measurement system with realtek 2 channel sound card.

The data is near field and far field data (1 meter) merged together to form the complete response.

The observable bump at around 1 kHz on the midrange is interesting and falls in line with my listening perceptions of these speakers when directly compared to the AR303. More soon as I import this data into the crossover tool to get full system response.