AR3a Improved,Pictures from Woofer and Crossover

[jagcat]

Hi folks,

just finished disassembling of my AR3A-improved models.Woofer foam is gone as expected,ordered a set from simplyspeakers yesterday.Are there any special hinds from your experience,which way to do it or just make it step by step as simplyspeakers told me?

The two crossovers seems to come from different suppliers,is this possible? On the first is written by hand "AR3A-improved" an on the other one is stamped "Fleetwood industries" Would it be better to remove the A-B Switch for correcting the highs?

[mluong303]

Hi, Jagcat

Could you post picture of the second crossover as well? Since you have the Crossover removed, would you measure the three Inductors and find out the mH of each. I would leave the A/B switch alone since these rare AR-3a Improved crossovers are different from the original AR-3a or AR-11. Did you get a chance to check out all 6 drivers on your AR-3a Improved to see if they are working or have DC resistance reading?

Also measure the Capacitors as well. Very likely the 6uF, 10uF and 20uF are off. Take a reading and let Us know. You can consider Solen or Axon ( from Zalytron ) polypropylene capacitors as upgrade replacement. 250VDC rated Caps are fine, no need to pay extra for the 400VDC which will be physically bigger and difficult to place them on the original crossover board.

Thanks,

Minh Luong

[soundminded]

Image 2 clearly shows what appears to be a wire mesh screen over the openings in the basket except where the terminals are. I assume that the purpose of this mesh is to keep stuffing and anything else out. Has anyone who may have measured these drivers noticed any differences in performance to those without it? Does it reduce the amount of stuffing in the rest of the box to achieve 0.7 damping factor? Does it play a roll similar to the stuffing itself even in small part?

[tysontom]

>Image 2 clearly shows what appears to be a wire mesh screen

>over the openings in the basket except where the terminals

>are. I assume that the purpose of this mesh is to keep

>stuffing and anything else out. Has anyone who may have

>measured these drivers noticed any differences in performance

>to those without it? Does it reduce the amount of stuffing in

>the rest of the box to achieve 0.7 damping factor? Does it

>play a roll similar to the stuffing itself even in small part?

>

The screen was used on a great number of AR woofers from the early 1960s until well into the mid-1970s, and as you indicate was to keep the fiberglass out of the area under the cone. The screen is very porous and really has no effect on the woofer's "Q." The Q on the earlier AR-3as was actually 1.0, but later ones built in Norwood were 0.7 as in the case of the AR-10/AR-11.

--Tom Tyson

[dynaco_dan]

Hi there;

That is plastic door fly screen material, which replaced the old metal fly screen's of those realy old wooden doors on houses.

I bought a small role a few years ago, to use on any woofers that I had, without it already installed from the factory.

An excellent point is, it is non-conductive, and it won't rust, if it should touch a speaker terminal.

It does stop any loose wires or insulation from rubbing up again the woofers cone, which would cause a buzz.

[jagcat]

Hi Mingh, here are 4 photos of the second Crossover,

regards Martin, Hamburg

1086.txt

1087.txt

[mluong303]

Hi Martin

Thanks for the pictures of the second AR-3a Improved crossover. Both crossovers are original and from AR. The different being one used Sprague 150uF cap and the other one used different brand with aluminum case! Same thing happen to the #4 coils different on both crossovers. This kind of inconsistence and sloppiness is not unusual during the Teledyne AR era! The main reason was between the designers and marketing people who believe in all caps, coils, resistors and cables will have no audible difference as long as they read the same uF, mH, Ohms and 16-18 Ga. tinted wires or copper cables really won't make any difference! So the technicians just follow the lead and solder whatever they had on hands!

I am wondering if the coils # labeled different from the coils # between UK and US since the #14 coil looks to be the same size as the #9 coil in the US AR-3a which measured 2.85mH. When you start to restore your crossovers, could you disconnect each coil from the crossover network to measure the mH for the #14, #16, #4 and #15 Inductors.

Is there any One in the AR Forum have a clue of who was involving with the designing of the AR-3a Improved model? We all know Mr. Roy Allison + Mr. Chuck McShane for AR-3a and Mr. Victor Campo for the AR-10 Pi + AR-11... I would like to find out the Missing Link to fill in the Gap and I need all the Helps I can get...

Minh Luong

http://www.classicspeakerpages.net/dc/user_files/1090.jpg

[Guest Bret]

Minh

>and the other one used different brand with aluminum case!<

Not that it matters, but just so you'll know. Every picture I see of a capacitor in this thread that isn't of a Sprague is of a Callins. The Callins are the ones I accuse of aging so badly; particularly the black with the red end-caps.

[jagcat]

Hello friends out there,

here's a plan for rebuilding a modified,more sophisticated crossover for my AR3-a improved.As a difference to the original version please look at the now 2 switches instead of the formerly 1 switch to increase or decrease the tweeter level. We are going to change this and place a switsh for also the midrange.Should get improved sound, as the pots in your original american versions seem to be dull over the time. What do you think of this? I will show you the completed parts when finally assembled, this will take some time as it is done after work.

Greetings from good old germany, Martin

1132.txt

[tysontom]

>I am wondering if the coils # labeled different from the coils

># between UK and US since the #14 coil looks to be the same

>size as the #9 coil in the US AR-3a which measured 2.85mH.

>When you start to restore your crossovers, could you

>disconnect each coil from the crossover network to measure the

>mH for the #14, #16, #4 and #15 Inductors.

>

The AR-11 and AR-10Pi used those four coils, and I suspect that the AR-3a-Improved is simply an AR-11-type crossover-iteration, which might also help explain the difference in the woofer-to-midrange crossover difference in the original AR-3a of 575 vs. the AR-11/10Pi/AR-3a-Improved crossover of 525.

>Is there any One in the AR Forum have a clue of who was

>involving with the designing of the AR-3a Improved model? We

>all know Mr. Roy Allison + Mr. Chuck McShane for AR-3a and Mr.

>Victor Campo for the AR-10 Pi + AR-11... I would like to find

>out the Missing Link to fill in the Gap and I need all the

>Helps I can get...

>

>Minh Luong

First of all, it was Roy Allison who designed the AR-3a, AR-5, AR-4x, AR-6 and AR-LST -- not Chuck McShane. McShane assisted Roy and contributed to the design of those speakers. McShane came from CTS in Paduka, and he had insight into driver construction and design. He was instrumental in the design of the stamped-steel, ceramic 200003 AR12W woofer, and was able to source the new cone, magnet and so forth, from vendors outside AR. The drivers were built by AR, but many of the new parts came from other vendors to AR's specifications. I believe that McShane coordinated those efforts.

Secondly, Victor Campos did *not* specifically design the AR-10Pi/AR-11; that series of speakers was designed through a joint-effort of many engineers and marketing people, including *primarily* John Bubbers, vp of engineering at the time, Robert Berkovitz, Victor Campos, Gerald Landau and Peter Dyke (marketing), engineers Alex P. deKoster and J Kates (I believe), Joe Green (testing), Joe Kay, Carl Logiudine (testing), not to mention president Marty Borish's inputs. There were probably many others that I have not mentioned. I suspect that the AR-3a-Improved came about as an off-shoot of the AR10Pi/AR-11 work that was done, but I am not exactly sure of the manufacture dates of the Improved. It was a European design, so the AR International people likely had a hand it its design.

--Tom Tyson

[mluong303]

Hi Tom

Thank you for your reply and correction to my posted info. I am glad to learn of so many additional names who were involving in the AR-10Pi project and the responsibilities of what Chuck McShane did for AR-3a...

By looking at Martin hand drawn AR-3a Improved crossver schematic, 2.6mH rings the bell and I recall AR-LST came out about the sametime uses this coil as well. I am wondering if the 525Hz crossover point came from this AR-3a Improved design which is not clearly published on paper? This clarified a bit about the different between the 575Hz and 525Hz crossover point of the AR-3a design? But how come AR-10Pi and AR-11 went back to 2.85mH again? Any thought?

Martin, the woofer cap may have been misread, please double check to verified if it is atually a 150uF cap? I don't want you to spend so much efforts and money and end up with a wrong Cap stuck in the woofer path!

Minh Luong

http://www.classicspeakerpages.net/dc/user_files/1133.jpg

[johnieo]

Martin:

Thank you for providing the inductor values. It would appear that the inductor numbering sequence used by AR-International differed from that used in the US.

Your schematic indicates the woofer crossover capacitor is 100 uF. Could you please check that value? Photographs of this crossover, where the label is visible, show it to have a value of 150 uF.

Regarding replacing the DPST switch with two SPST switches: that depends on whether you wish to keep it "authentic" or add another degree of freedom in setting levels.

Does anyone know how many dB change result?

Cheers,

[RoyC]

Minh and Martin,

Which tweeter did the 3a Improved use? Was it the original 3a version or the AR-11 type?

Thanks,

Roy

[mluong303]

Hi Roy

It is the front wired tweeter. T, M & W are the same drivers as the original AR-3a stocks.

Minh Luong

[mluong303]

Hi Martin

Two things you may want to double check for yourself in your AR3a Improved crossover schematic.

Since you are in Germany, you may want to check out the Rifa or Wima capacitors instead of the usual Solen caps.

Good Luck,

Minh Luong

http://www.classicspeakerpages.net/dc/user_files/1134.jpg

[dynaco_dan]

Hi there;

Simple question here.

In the schemetic there is a number of inverted v's.

What number does this represent?

1 or 4 or 7?

[mluong303]

It is number 1

[jagcat]

Hello friends,

at first, I want to thank all of you, who are taking part of my problems and are so superkind with their help. I believe,all of you are really convinced of the quality of these AR-speakers and like a child waiting for santa claus I'm looking for the day of first connection them to my amp.

You've asked for the figures of the caps, here they are:

/4 (woofer) 2,60mH

/5 (mids) 0,15mH

/3 (mids-dividing) 0,90mH

/6 (tweeter)0,10mH

Now you see 7 simulations at 6 Ohms, bringing an impreesion of how

the switches might work and the overall character of the crossover.

Sorry, I don't know how to get the pictures directly in this post...

Martin

1163.txt

[dynaco_dan]

Hi Martin;

Could you please translate the major words to English, please.

Thank you.

[mluong303]

Martin,

Two things I want to bring to your attention:

1). The Tweeter, Midrange and Woofer are all 4 Ohms drivers. So what is the advantage of using 6 Ohms for R2 here?

2). There is a missing Cap C3 in this Midrange Schematic, see picture.

The UK numbered Coils for the AR-3a Improved crossover are:

#13 = .88mH

#14 = 2.63mH

#15 = .16mH

#16 = .105mH

Minh Luong

http://www.classicspeakerpages.net/dc/user_files/1167.jpg

[mluong303]

Correction:

the #13 coil indicated in Martin's crossover schematic actual labeled #4 which happen to be the same #4 as in the US that measured .88mH.

Minh Luong

[Pete B]

>Hello friends,

>at first, I want to thank all of you, who are taking part of

>my problems and are so superkind with their help. I

>believe,all of you are really convinced of the quality of

>these AR-speakers and like a child waiting for santa claus I'm

>looking for the day of first connection them to my amp.

>You've asked for the figures of the caps, here they are:

>/4 (woofer) 2,60mH

>/5 (mids) 0,15mH

>/3 (mids-dividing) 0,90mH

>/6 (tweeter)0,10mH

>

>Now you see 7 simulations at 6 Ohms, bringing an impreesion of

>how

>the switches might work and the overall character of the

>crossover.

>Sorry, I don't know how to get the pictures directly in this

>post...

>Martin

Hi Martin,

Thanks for your post as it's interesting to compare the different versions of the early AR-3a.

The accuracy of your simulation will be improved by using the complex input impedance for the drivers. It looks like your using SPICE or a similar electronics package, the speaker design tools make this easy.

You could use the model shown here inside the green box for the woofer that was previously posted by richd:

http://www.classicspeakerpages.net/dc/user_files/803.txt

Note that he pointed out that it was for the driver in free air. You should reduce the tank circuit inductor to model the decrease in compliance in box to obtain an Fc of 41 Hz, and decrease the tank resistor to model incresed losses due to stuffing to match the height of the resonant peak. Do not change the cap as this represents the moving mass which does not change significantly in box.

You could also use this model for the mid and tweeter load, decrease the tank L and C by a factor of 10 to raise Fc by 10 and get in the ball park for the mid driver, further adjust to match Fc, Qtc and impedance height. VC inductance should be reduced to match the input impedance HF rise. It is iterative but not too difficult. This could also be done for the tweeter.

It would be helpful if members posted measured input impedance for the early drivers.

Also the DC resistance of the XO inductors should be included for better accuracy.

I only looked quickly at your posted circuit however I'm not clear on what are changes from the original circuit.

Pete B.

[Pete B]

I didn't mean to stop this thread, my point was just a suggestion.

Pete B.