Care for AR tweeters

[prrtraincow]

This is a follow-up from my other thread about having to replace my AR5 tweeters.

I have replaced them with originals I got off Ebay. I replaced the 3 bipolar electrolytic capacitors in the crossover circuits with polypropelene caps (got them from Parts Express).

All went well. They play great. My concern is why they even blew in the first place. I have seen many mentions on this forum about how "fragile" these tweeters are. What I want to know is,

how "fragile" are they, and what should I do to protect them?

Do they fail because of their age? Is there something inherent in their design that means they just can't be expected to last 20 or 30 years without failing?

OR

Do they fail because they are very sensitive to clipping? I drive them with a Dynaco SCA 80 solid-state 40 watt/channel power amp and a Dynaco PAT-4 preamp. I know from past experience that they can definitely stand a lot more power than 40 watts, and they sound better on a 100 watt amp. But is this significant as far as protecting them?

What aftermarket other-brand tweeters do folks use to replace these when they give up on the AR originals and want something more durable?

Thanks, everybody.

[KlausDK]

Hi

You can protect your speakers with fuses as recommended by AR. In the library there is a document named "Fusing AR Speaker systems". The recommended Fuse for AR-5 is the Bussman Fusetron dual element FNM 8/10. You can find both fuses and fuseholders on ebay. This is highly recommended for speakers with the older tweeter type like the one used in AR-5, 3a and LST.

BR Klaus

[soundminded]

As old electrolytic capacitors age, changes in the dielectric cause capacitance to increase lowering the crossover frequency to the tweeter the series capacitor defines. This allows more lower frequency energy than was intended to reach the tweeter increasing the risk of damage or burnout. In addition to fusing the tweeter as Klaus advised, replacing the series capacitor as well is a good idea. There is some controversy about whether or not more expensive polypropylene and other exotic capacitors are superior to inexpensive electrolytics. PeteB said in carefully conducted blind tests nobody could hear any difference and I believe him. I use the inexpensive electrolytics Parts Express sells myself. Be sure to get the non polarized type if you do. BTW, if you hear distortion from them at loud levels, you are risking damage if you haven't damaged them already. They are not intended as "rock monitors" and using them at those levels is almost certain to result in damage.

Replacement tweeters from other manufactures with an equivalent has been a problem for those who want to restore old AR speakers. No tweeter on the market has ever come close to matching the AR 3/4" dome tweeter for dispersion. Some of us have urged Ken Kantor to use his influence at Vifa to persuade them to reverse engineer and manufacture an equiavlent tweeter but so far, they haven't. There are some fine tweeters from many manufacturers available although not exactly like AR's. Some of the better ones come from Vifa, Scan Speak, Morrel, even North Creek. Get a Madisound and a Parts Express catalog and you will find a wide variety of choices.

[prrtraincow]

>As old electrolytic capacitors age, changes in the dielectric

>cause capacitance to increase lowering the crossover frequency

>to the tweeter the series capacitor defines. This allows more

>lower frequency energy than was intended to reach the tweeter

>increasing the risk of damage or burnout.

I wondered about this. I would expect the capacitance to change as electrolytics age, but I've never been able to really convince myself whether that means it will go up or go down, or whether it's even sure to go one way or the other.

>replacing the series

>capacitor as well is a good idea. There is some controversy

>about whether or not more expensive polypropylene and other

>exotic capacitors are superior to inexpensive electrolytics.

>PeteB said in carefully conducted blind tests nobody could

>hear any difference and I believe him. I use the inexpensive

>electrolytics Parts Express sells myself. Be sure to get the

>non polarized type if you do.

I replaced them with polypropelenes mostly because I expect they will age better than electrolytics, not because I expected them to sound any better. And yes, of course they are non-polarized because they are not electrolytics.

>BTW, if you hear distortion

>from them at loud levels, you are risking damage if you

>haven't damaged them already. They are not intended as

>"rock monitors" and using them at those levels is

>almost certain to result in damage.

Yup, I agree. But they'll play pretty darn loud off this 40-watt/channel amp without distorting.

Thanks for your info. Dispersion is one thing that would be hard to match. Parts Express has some Vifa silk dome and other dome tweeters that will fit in there, are 8 ohm, and have a quoted freq range of 2000-20000 Hz. Since the crossover is at 5000 on these, that should be more than adequate. Aside from the dispersion characteristics, I'm more concerned with the behavior of the crossover components with a difference driver in there, and with the flatness of the response of the actual tweeter. Is the resonant frequency important to match in a tweeter?

Thanks for your replies

[soundminded]

I don't think the resonant frequency matters if it is low enough meaning 1/2 the crossover frequency or lower. You want one with as flat a response and as wide a dispersion a possible but it should match its electrical characteristics as closely as possible as well. this means inductance, resistance, and sensitivity. To the degree they are different, the crossover frequencies, slopes, and loudness won't match the original. If you are not concerned, you could substitute another crossover design at about the same frequency and use an L-pad to trim its loudness. I think the tweeter operated from about 5khz up with a cutoff slope of 12 db per octave below that but you'd better check the archives or someone more familiar with it than I am.

[nfmisso]

>Do they fail because they are very sensitive to clipping?

They are VERY sensitive to clipping. Clipping causes high power high frequency signals to be sent to the speaker - all of which goes to the tweeters.

There have been more tweeters blown by low power amplifiers than anything else - all due to clipping.

I would suggest you try an Adcom GFA545, 545 II. 5400 or similar amplifier to drive your speakers. You will greatly reduce the chance of blown tweeters, and the lower impedance output stage may improve the bass response.

I suggest Adcom because they are relatively inexpensive, reliable and easy to find. There are many other amplifiers out there that are also suitable.

[RoyC]

The AR-5, like the 2ax and 3a, used a simple 6db/octave crossover for the tweeter circuit. A 4uf cap for the AR-5 and AR-2ax, and a 6uf cap for the 3a, were all that stood between the tweeters and oblivion. The increased protection of parallel tweeter inductors showed up in later models.

Roy

[prrtraincow]

>The AR-5, like the 2ax and 3a, used a simple 6db/octave

>crossover for the tweeter circuit. A 4uf cap for the AR-5 and

>AR-2ax, and a 6uf cap for the 3a, were all that stood between

>the tweeters and oblivion. The increased protection of

>parallel tweeter inductors showed up in later models.

>

>Roy

I don't think so. The AR5 has inductors in the crossover, at least 3 of them. I've seen them. And I'm quite sure it uses 12 db per octave rolloffs.

You are correct that the 4uF cap is the primary thing that protects the tweeter, but why shouldn't that be enough? Unless the cap is bad,

it should start rolling off at 5000 Hz, so there is really no way that any appreciable energy down below, say, 2000 Hz, is getting to that tweeter.

I'm inclined to think it's just clipping that does these in. That's how the first tweeters blew (which I replaced in 1981). I stupidly lent my AR5s to a communal "stereo" for a party, and we were driving them with a cheap crap Panasonic 50 w/channel receiver. Cranked up to

party volumes, it probably started clipping all over the place. And when the tweeters and one midrange blew, it all seemed to happen at once because the stereo seemed to immediately "shut off". But in fact, it was just that all the treble disappeared. When I looked

at the receiver, I noticed that somebody flipped a "high boost" switch on the front panel - I'd bet that instantly blew the tweeters and midrange. And the (*@_)(* who flipped it probably never confessed.

At the time, I knew I had played these AR5s with a Dynaco Stereo 400, and I figured if they could take 200 watts/channel, they could stand anything. Of course, the Stereo 400 is a really NICE amp, and I'm sure it never clipped at all no matter how loud I played it with just AR5s on it.

Anyway, after I replaced them, I used the speakers for the next 25 years without incident, until just recently when I was refoaming the woofers and noticed that the tweeters weren't even playing (I haven't been using the stereo much for months, so I wouldn't have noticed).

People on this forum have suggested I use a beefier amp to avoid this problem, and I probably will. At the levels I play them at, I doubt I'll have a problem even with my 40 watt/channel Dynaco. But I still have that Stereo 400, and I'll probably start using that anyway. It always made these 5s sound like a million bucks.

[RoyC]

>I don't think so. The AR5 has inductors in the crossover, at

>least 3 of them. I've seen them. And I'm quite sure it uses

>12 db per octave rolloffs.

Not the tweeter...There are no inductors in the tweeter circuit.

The 3 inductors in the AR-5 are as follows:

-3.83mh (#10) is in series with the woofer

-.16mh (#2) is in series with the midrange

-2.3mh (#11) is in parallel with the midrange

>You are correct that the 4uF cap is the primary thing that

>protects the tweeter, but why shouldn't that be enough?

It probably is for most folks (including myself), but it doesn't provide the protection of an electrical circuit with a steeper slope. Modern tweeters' power input ratings are based on 12db per octave slopes.

Roy

[prrtraincow]

>

>Not the tweeter...There are no inductors in the tweeter

>circuit.

>

Well, whaddya know. I stand corrected, Roy. Thanks for pointing this out. I more glad than ever that I put new polypropelene 4uF caps in there (as well as replacing the other caps).

Since refoaming the woofers, new caps, and new tweeters, I am once again in awe of how beautiful these speakers sound. I go into modern electronics junk shops like Circuit City and think, "you people have GOT to be kidding." Of course, you could still hear fine equipment in a true audio salon. Those places are few and far between now. But "back in the day", any decent electronics retailer had plenty of quality audio equipment, even in the modest price ranges.

[RoyC]

>Since refoaming the woofers, new caps, and new tweeters, I am

>once again in awe of how beautiful these speakers sound.

I agree...The AR-5 was a genuine sleeper in the shadow of the AR-3a!

Roy