Jump to content
The Classic Speaker Pages Discussion Forums
ra.ra

AR-6 with zobel

Recommended Posts

Of the several AR-6 variations, I had been under the impression that this version (three-way tweeter switch, zobel circuit mounted on woofer) was a European model, but this pair I recently acquired has only the Norwood address on the labels. They appear to be dated from June 1974, since all drivers and labels have "7423" imprinted on them.

These speakers are more than a bit beaten up, but I remain confident that they are an excellent and viable restoration project. All of the drivers are functional and the simple crossovers are original and intact. The grille cloths show the usual effects of 40 years of smoke, mildew, and lack of maintenance, and the cabinets have experienced a bit too much humidity and nary a coat of oil. The tweeters are near perfect, but the woofers may present a few challenges.
The purpose of this post is not only to show this particular variety, but more importantly, to solicit advice on diagnostic and restoration measures. The pair has already been tested, dismantled, and partially measured, and a first pass has been dealt to various cosmetic issues. With my brief low level listening test, there was no discernible difference in tweeter output between the "normal" and "decrease" switch positions, but the "increase" position delivered as advertised.
On the cabinet crossover, there should be no reason to fuss with the switch or the resistors or the coil, so the only question is about the cap. It has often been stated these Sprague Compulytic caps often remain close to original value even after 40 years, so my question is this: should I de-solder and isolate and measure these 10uf caps, or should I just leave them alone? Also, after looking at the schematic again, can anyone confirm if this switch is an 'on-off-on' type?
About the woofers, I have a few questions. First off, why are they such a mess-o-goop? With this zobel circuit (24 uf cap and 10 ohm resistor), I'll assume the resistor is fine, but with the cap, should I: leave it as is; replace it; or isolate, measure, and replace only if necessary. Next, I've never seen an 8" AR woofer with this fiberboard ring (about 1/8" thick) between the metal basket and the foam surround - - to me, it looks original, and a dry-fit surround seems to fit just fine, but has anyone seen this before? Lastly, you might be able to see in the pics, but only one flat dust cap shows signs of previous surgery (suggesting a previous re-foam and shimming), but... any thoughts on this?

post-112624-0-43747800-1414465845_thumb.

post-112624-0-49381300-1414465866_thumb.

post-112624-0-99865900-1414466070_thumb.

post-112624-0-82856200-1414466367_thumb.

post-112624-0-58716200-1414466468_thumb.

post-112624-0-59356900-1414466584_thumb.

Share this post


Link to post
Share on other sites

Hi ra.ra

A few notes that may assist.

From the attached schematic the switch connections appear consistent with a 3 position (on.off,on) switch.

You can check the resistors/switch by measuring the resistances (0,3,6 Ohms) as shown in the pdf. (With the tweeter removed.)

I had a quick check of my AR6 with the parallel R/C components, like you, I could not hear a significant difference between the "normal" and "decrease" settings.

I have also attached a picture of a spacer under my Spider AND my Surround.

R

AR6-3 position switch-Zobel-rara .pdf

post-100645-0-83969300-1414496320_thumb.

Share this post


Link to post
Share on other sites

Thanks, RLowe, that's a really terrific response and I was hoping you might add something here. It occurred to me last night after posting that I neglected to include the appropriate schematic (which you originally drew and have posted again in your attached pdf), but I'll attach it here again just for easy access. And yes, the spiders on my woofers have the spacer as well - - - I had failed to notice this because they are so similarly colored - - thanks for the heads-up on this and at least I can proceed with re-foaming with a degree of confidence.

Your switch diagrams do help me understand this operation much better - - at least now I can see how this switch works and can follow the flow of juice thru the circuit. (This will also come in handy when I finally add a switch and complete my orphan MST project.)

Any thoughts on the original capacitors? On one of the zobels, the glue to the magnet has come disconnected, so I think I'll just de-solder and take them both apart to measure the 24 uf caps, and if necessary, replace with a similar NPE cap - - - the poly caps are just physically too large and too co$tly. Maybe I'll just do the same with the 10uf can cap (i.e. remove it from circuit and measure), but if these have drifted, I'd replace with a high quality film cap.

Grilles are getting a facelift, cabs are getting some TLC, and the badges have a new spit shine. Pics to follow in coming days.

post-112624-0-85497700-1414507896_thumb. post-112624-0-36257100-1414507912_thumb.

Share this post


Link to post
Share on other sites

The can caps are probably still good but good idea to check. If you do need new caps, I'd suggest three 10uF surplus caps + one 3.9 or 4 uF from Madisound

10uF: https://www.madisoundspeakerstore.com/surplus-capacitors/10.0-mfd-polypropylene-cap-10mfdp/tyee/

3.9uF Carli https://www.madisoundspeakerstore.com/carli-capacitors/carli-mylar-3.9-mfd/

4uF Solen https://www.madisoundspeakerstore.com/solen-capacitors/solen-4-mfd-fast-cap-400v/

Or you could also use these 2uF caps. The leads are kind of short but if paralleled with the 10uF caps it should work OK: https://www.madisoundspeakerstore.com/surplus-capacitors/2.0mfd/

So if you used all surplus caps that would cost you $5 for both speakers! Can't beat that!

Kent

Share this post


Link to post
Share on other sites

Hi Rob,

My understanding of the electrical components, practices, and filters is insufficient to comment with any authority, but by definition, you may be correct. A very brief research effort appears to document a different arrangement of capacitor and resistor than what is shown in the schematic of post #4. My use of the term "Zobel" is merely a regurgitation of discussions I have read on this site and elsewhere regarding this particular AR-6 woofer assembly.

post-112624-0-60512500-1420656468_thumb.

Share this post


Link to post
Share on other sites

Ah yes, as depicted in the attachment the cap and resistor are in series and that can indeed be to correct impedance.

However I'm sure the diagrams I've seen for the AR6 show the caps and resistor both running parallel across the driver input. I might have read this wrong so will check.

Share this post


Link to post
Share on other sites

Yep, the AR-6 version B schematic does show the cap and resistor both in parallel with the driver terminals, and that is also how they were installed, as shown in pic from post #1. So, maybe not an authentic zobel?

Share this post


Link to post
Share on other sites

Yep, the AR-6 version B schematic does show the cap and resistor both in parallel with the driver terminals, and that is also how they were installed, as shown in pic from post #1. So, maybe not an authentic zobel?

Indeed, looks to me like a 2nd order crossover, though I'm not sure what purpose the resistor serves since as placed it will just drop the the driver impedance given it sits across the bass driver terminals. Would be worth running a simulation. I'll pass the schematic over to a friend with the necessary software.

Share this post


Link to post
Share on other sites

IRa.ra:

At this point, Iwould replace the capacitors. No elecrolytic made in those days was intended to last this long!

The Sprague and Compulytics made in the 1970s were well made with low ESR. If one looks in the old Handbook of Components for Electronics or the Electronic Engineer’s Master (EEM) available in your friendly nearby University library, one will see ESR values of order 0.016. The demand for switching high currents in capacitors while minimizing heat generation was the driving force behind these quality capacitors. Bridge measurements that I made in 2005-6 showed the capacitance values to be close to original value, but the ESR to have increased considerably. To evaluate these capacitors one needs a bridge that will measure both C and ESR in the octave or so frequency range near the crossover frequency of interest. Cant be done with large capacitors on a small L/C/ESR meter.

Cheers and enjoy refurbishing these--its a fun hobby.

John

Share this post


Link to post
Share on other sites

Relocating cap farther away from resistor would not be bad idea either. 10 ohm resistor will become very hot at higher listening levels, when stuffed in the fiberglass .

This 10 ohm parallel resistor and second order low pass filter does mean that AR6 must be something like 3-4 dB less sensitive than AR7... is this true?

Best Regards

Kimmo

Share this post


Link to post
Share on other sites

Relocating cap farther away from resistor would not be bad idea either. 10 ohm resistor will become very hot at higher listening levels, when stuffed in the fiberglass .

This 10 ohm parallel resistor and second order low pass filter does mean that AR6 must be something like 3-4 dB less sensitive than AR7... is this true?

Best Regards

Kimmo

Hi Kimmo,

I don't think the sensitivity drop would happen.

The 2nd order crossover will just sharpen the roll-off at the crossover point (and alter phase but thats a separate consideration).

Rob

Share this post


Link to post
Share on other sites

If you calculate AC resistance without DC losses in pass band at 1000 hz and 500 hz, you will find how much there is attenuation due crossover. Woofer is fed directly in AR7, so there are no crossover losses and woofer impedance should be quite similar to AR6 with crossover.

1000 hz figures. 24uf cap (ac resistance 6,6ohm at 1000 hz) parallel to 10 ohm resistor and 8 ohm woofer makes 2,6ohm across woofer terminals. This forms voltage divider with 0,88 mH coil (5,5 ohm AC resistance) with total impedance of 8 ohm at 1000hz. When there is 1V at input terminals, there is approximately 330mV across woofer terminals.

500 hz figures. 24 cap ac resistance is 13 ohm at 500 hz... 13 ohm parallel to 10 ohm and 8 ohm tip the scales at 3,3 ohms at woofer. This forms again voltage divider with 0,88 mH coil with AC resistance of 2,7 ohms at 500 hz. Total impedance is 6 ohm. With 1 volt input you will have 550mV across woofer terminals... so there is slightly less attenuation, but still more than 5dB.

Unless I have mistaken... these figures show that there are no free lunch here. If you want to improve performance with steeper slopes... you will loose sensitivity. Same was true when AR5 was added to line to improve AR2ax performance..

Best Regards

Kimmo

Share this post


Link to post
Share on other sites

If you calculate AC resistance without DC losses in pass band at 1000 hz and 500 hz, you will find how much there is attenuation due crossover. Woofer is fed directly in AR7, so there are no crossover losses and woofer impedance should be quite similar to AR6 with crossover.

1000 hz figures. 24uf cap (ac resistance 6,6ohm at 1000 hz) parallel to 10 ohm resistor and 8 ohm woofer makes 2,6ohm across woofer terminals. This forms voltage divider with 0,88 mH coil (5,5 ohm AC resistance) with total impedance of 8 ohm at 1000hz. When there is 1V at input terminals, there is approximately 330mV across woofer terminals.

500 hz figures. 24 cap ac resistance is 13 ohm at 500 hz... 13 ohm parallel to 10 ohm and 8 ohm tip the scales at 3,3 ohms at woofer. This forms again voltage divider with 0,88 mH coil with AC resistance of 2,7 ohms at 500 hz. Total impedance is 6 ohm. With 1 volt input you will have 550mV across woofer terminals... so there is slightly less attenuation, but still more than 5dB.

Unless I have mistaken... these figures show that there are no free lunch here. If you want to improve performance with steeper slopes... you will loose sensitivity. Same was true when AR5 was added to line to improve AR2ax performance..

Best Regards

Kimmo

I'll be modifying one of my 7s this afternoon so we can do a before and after comparison.

The losses calculated at the top of the bass driver pass band are somewhat mitigated by extending the output of the tweeter by using the larger 10f capacitor, and that must also be factored into calculations.

Rob

Share this post


Link to post
Share on other sites

Yes... calculation was not complete analysis of the system. As attenuation is still approximately 3 dB at 200 hz and impedance will drop close to 5 ohm here, some loss of efficiency might be expected.

It came to mind that one reason for using 10 ohm resistor might have been to increase tweeter output relative to woofer level.

As you are handy with measuring tools... let us know what you will find.

Best Regards

Kimmo

Share this post


Link to post
Share on other sites

Rob

What kind of results you got from before and after comparison?

Best Regards

Kimmo

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...