No highs from Klipschorns????

[bdc]

I have a pair of '74 Klipschorns with type AA crossovers (with zener diodes) and AlNiCo drivers and they sound absolutely fantastic. However, there doesn't seem to be much high frequency sound coming from them. The tweeters are working (confirmed by ear), but you really can't hear the highs. I've powered these with an HK430 and a Dynaco tube amp. I've used a pair of Fortes in the same room with the same equipment and there's definitely a lot more top end on them. I've tried that 'free crossover tweak' from the earlier thread too, without much difference.

[Randy Bey]

make sure the tweeters are working. They are often blown and the root cause of such a problem 95 times out of a 100. You can check them with an ohmmeter.

check polarity of all connectors at crossover. Make sure they are the same on both speakers.

Do you have other speakers to try on the same amp? If so, check if the other speakers have high frequencies.

[ricktate]

Cover the squawker with a piece of cardboard to insure tweeter is working also...loosen and re-tighten all connections,,,simple things first always....rick

[3dzapper]

bdc,

Add a 1mF capacitor across the two 2mF caps in the AA x-over. You will discover something wonderful.

See this thread in Odds & Mods for a detailed discussion:

"More extended highs for AA network"

http://forums.klipsch.com/idealbb/view.asp?topicID=41802

I have brought it to the top for you.

Rick

[sunnysal]

no klipschorn should sound like it has "ho highs" so something is wrong...forget tweaks and mods you read about on the forum for the moment and focus on seeing why you hear no highs...you say there is sound from the tweeters (both I assume) but I assume you mean it is too low in level and that you have decided this by comparing the sound of the same music through some confirmed working speakers. I also assume you used the same electronics with the other speakers to be sure that some other components (amp) in your stereo are not to blame for the lack of highs...if all that is true then you need to double check the wiring to the tweeters, screws and solder joints, to make sure everything is solid there. then I might give a good look at the x-overs, any caps look dried up or leaking? any thing smell burned? any thing look loose or rotten with the coils? get back to us with feedback about that and then we can focus better. tony

[Deang]

He stated he has tried the tweeter mod, and also confirmed the tweeters are working. On my Klipschorns, I have no trouble hearing the tweeters when I walk up the them.

After 30 years, and depending on their usage history and environmental exposure, it could be that your diaphragms are ready for replacement. According to Bob Crites, those things do wear out, and it effects their response. If they were my Klipschorns, I would consider sending him the tweeters to have the diaphragms replaced.

A secondary issue is related to the capacitors in the crossover. Capacitors also wear with age, and though they maintain their capacitance value, the ESR begins to climb. Increased series resistance means less energy to the driver, and more loss through heat. If you have the DIY spirit, you can buy and replace the capacitors yourself -- or buy replacement crossovers.

[Ryan C Inman]

n

[bdc]

I've retightened all the screw connections on the crossover (that is what I meant by 'free crossover tweak') but it did not seem to change anything. I ran a 20hz-16000khz frequency sweep and it sounds just fine, but for some reason it sounds a bit limited in the top end when playing music. I don't know what the condition of the capacitors are on the crossovers simply because I don't know what I'm looking at.

[Deang]

RI -- Makes sense. Maybe the shift in ESR is out of proportion with the shift in capacitance? I don't know. I forwarded your post to Bob Crites, since what I said came from my understanding of what he said to me during a phone call. I may have misunderstood.

BDC -- it is more than likely that the problem is being caused by worn diaphragms. Contact Bob C. for a more technical explantion for this. bobcrites@centurytel.net

[NOSValves]

BDC,

Have you simply tried using the tone controls on the HK to see if bumping the treble up actually works through the speakers ? Not saying to use the tone controls to compensate but just as a tool to see if the Khorn's will indeed produce more Highs. This could be some type of amp/source problem.

Craig

[BEC]

Ryan,

As far as the mechanism that changes the caps such that they increase in ESR and remain in spec for capacitance, I am not sure. Some places on the web suggest the "delamination of the foil" may be the mechanism. As far as measuring the effect, that is easier. One must preface this with the fact that It is impossible to have "new" 20 to 30 year old capacitors to do real comparisons so I don't know the "starting" ESR values on the old caps.

Best I can do is compare new caps to old caps of the same stated value. Here are the results of testing on 2 2uF caps I have replaced in a customers Klipsch Type AA crossovers and the two new GE caps I used for replacement. The crossovers that these came from showed a decrease in output to the tweeter on the spectrum analyzer before replacement and were back looking like they should after replacement. The two caps are used in series on the crossover so I tested them in series.

Old Caps.

C = 0.9761 uF

D = 0.004

ESR = 0.685 ohms.

New Caps

C = 0.9944 uF

D = 0.000

ESR = 0.122 ohms

The capacitance value of the old caps is low compared to stated value by about 2 1/2 percent. That is likely still in spec but since the caps are not marked as to tolerance, I don't know that spec. I would call these good based on capacitance for this use.

If the original ESR of the old caps I tested was as good when they were new as the measured value of the new replacement caps, I would say the old caps were bad because of the 1/2 ohm increase in ESR.

Bob

[artto]

This is a room response curve I recently took, at the listening position. Both Khorns & Belle Klipsch operating. Everything on all the speakers is original, except for disconnecting the woofer inductor & use of rope caulk to damp the mid horns.

[Deang]

Nice work, it is impressive. Still, I see HF response continuing to dip down after the crossover point, and you're about 10db down or more at 12Khz. Want me to order some caps for you?

[artto]

Dean, I'm not so sure that caps are a problem. They might be contributing to some of the roll-off. But in all honesty, the room was 'tuned' with a somewhat decreasing high-end roll-off to simulate what most concert auditoriums are like. Four of the six room surfaces are hard, but as you know, have plenty of diffusion. I think whats important here is the overall smoothness of the response, free of any major peaks or dips. In fact the low end is down only -5 to -8dB at 16Hz (had to be estimated because of the mic roll-off).

And actually, after looking at the mic manufacturer's response graph more closely, it appears the mic has a roll-off starting around 12KHz. Appears to be down about -8dB at 20KHz with a slight dip around 6KHz. I'm sure the mic is contributing to some of this. It's not a "calibrated" mic. I had to use the info from the manufacturer's website. It appears to be extremely flat 30Hz to 2KHz. Then gets a little bumpy after that which is obviously reflected in the room response curve.

I'll be posting more results and adjusted FR graphs on my thread in the Architectural Topic area when I get more specific info from the mic manufacturer and/or re-run the tests with a calibrated mic.

[Ryan C Inman]

n

[BEC]

Ryan,

I think we are really discussing the very premise for doing ESR testing in addition to Capacitance testing. I think the methodology was an attempt to determine, by measurement, the reason that some caps which tested good for capacitance were in fact bad in their circuits. I have been in electronics for around 40 years and had never, until recently, even had an easy way to measure ESR. I have, however, in the past found caps that for some reason needed replacement to make a circuit work right and still tested good on a capacitance bridge. Makers of ESR meters seem to suggest that ESR is the only way to ferret out bad caps using test equipment. Of course, they are also selling ESR meters. The fact that larger companies building test equipment are now adding the ESR function to component test meters (B&K Precision for instance) seems to indicate growing acceptance of ESR testing. As far as I know, the capacitor has to be at least a 1 uF to be tested for ESR. I think that has to do with the cap being large enough to pass the test pulse through it without taking on a charge from the pulse.

You are certainly correct that I cannot know the "new" ESR for an old capacitor, (I believe I also stated that in my earlier post). I would say, however, that the 1/2 ohm difference in ESR measured for the old caps is significant in the circuit where these are used. This circuit drives a tweeter with 8 ohm impedance. This extra 1/2 ohm will cause around 6 percent of the tweeters signal to be dissipated in the caps.

Here are another set of cap measurements. This one has really got me thinking and perhaps someone can tell me what to make of it.

This is a Dayton 2 uF oil filled metal can type cap with a part number of 4X424A. Don't know much else about it except it says on the case "oil capacitor - no PCBS". I bought it at a local electrical supply house as a possibility for use in crossovers. It is new but of course could have been on the shelf for some time. This test has convinced me not to use them.

C = 2.092 uF

ESR = 1.582 ohms.

Can you tell me how I would know not to use this one without ESR testing?

Bob

[Deang]

"Paper, next to Aluminum Oxide used in Electrolytics is the worst type of dielectric electrically. D and ESR figures will look poor compared to a modern plastic capacitor."

Well, the numbers won't look as good, but I don't think they should look "poor" in comparison. When comparing, the numbers only look poor if your baseline is derived from measurements using 30 year old cans. We may not know what the ESRs of the original caps were when new, but Bob's statement that the original crossovers "showed a decrease in output to the tweeter on the spectrum analyzer before replacement, and were back looking like they should after replacement" -- tells us as much as we need to know.

"The decrease in output to the tweeter baffles me though, could another component in the crossover have been responsible for the attenuation you noted?"

No. It's simple really. Bob replaced the old cans, and the output returned to normal as evidenced by the spectrum analyzer traces. So, even though the capacitance values were within tolerance, series resistance had clearly climbed beyond acceptable limits.

As far as PIOs being "the worst type of dielectric" next to electrolytics, I'd just like to point out that 'D' and ESR don't tell the whole story either. I find 'K' much more interesting, and still think the various sonic signatures of capacitors are linked to this.

"Dielectric constants vary with temperature, voltage, and frequency making capacitors messy devices to characterize. Whole books have been written about choosing the correct dielectric for an application, balancing the desires of temperature range, Temperature stability, size, cost, reliability, dielectric absorption, voltage coefficients, current handling capacity (ESR). (Ivan Sinclair wrote a nice book on passives; unfortunately, it is out of print. This points to the fact that our universities are no longer teaching this material)."

[WMcD]

I think BDC is hearing something which I do also.

My Forte II and Quartets seems to have a very sparkely high end. This is in contrast to my home built units using the T-35 which is like the Klipsch K-77 unit. Unfortunately I don't have them side by side to confirm.

The Stereo Review, analysis of the Forte reported exceptional good high end.

My guess is that the Forte tweeter is indeed superior. It looks to be be more horn based rather than relying upon diffraction effects.

My tests of the T-35 with a calibrated mic on the LMS system tracks Art's results. Above 10 kHz, it is rolling off. Therefore perhaps Art's data is not inaccurate because of uncertain qualities of the measurment system.

I'm a bit undecided on what combination of qualities of the tweeter makes for good subjective performance. It is easy to mentally correlate lack of response at 15 kHz (measured how) with lack of zing. But most of us can't hear 15 kHz, or even 10 kHz.

Rather, the issue is in the octave from 6000 Hz (or lower) to 12000. This is the area in which we can hear things. It is similar to the often observed fact that if you want impressive bass, crank up 120 Hz rather than 60 Hz.

It is my thought that the Forte tweeter is much better than the K-77 and there is no way that easy tweeks are going to improve it.

Gil

[bdc]

----------------

On 4/29/2004 11:40:16 PM William F. Gil McDermott wrote:

I think BDC is hearing something which I do also.

My Forte II and Quartets seems to have a very sparkely high end. This is in contrast to my home built units using the T-35 which is like the Klipsch K-77 unit. Unfortunately I don't have them side by side to confirm.

The Stereo Review, analysis of the Forte reported exceptional good high end.

My guess is that the Forte tweeter is indeed superior. It looks to be be more horn based rather than relying upon diffraction effects.

My tests of the T-35 with a calibrated mic on the LMS system tracks Art's results. Above 10 kHz, it is rolling off. Therefore perhaps Art's data is not inaccurate because of uncertain qualities of the measurment system.

I'm a bit undecided on what combination of qualities of the tweeter makes for good subjective performance. It is easy to mentally correlate lack of response at 15 kHz (measured how) with lack of zing. But most of us can't hear 15 kHz, or even 10 kHz.

Rather, the issue is in the octave from 6000 Hz (or lower) to 12000. This is the area in which we can hear things. It is similar to the often observed fact that if you want impressive bass, crank up 120 Hz rather than 60 Hz.

It is my thought that the Forte tweeter is much better than the K-77 and there is no way that easy tweeks are going to improve it.

Gil

----------------

Are you saying that the Khorns are supposed to sound this way? My Khorns are the only ones I've ever heard so I don't really have anything to compare them to besides the Fortes. Between the two speakers, there is a dramatic difference in the sound. With the Khorns, I can turn the treble knob almost all the way up and it doesn't quite have those 'sparkly' highs of the Fortes. If this is the case (that the Khorn's highs are weak), then what do you suggest should be done to improve them?

DeanG: As per your suggestion, I've emailed Bob Crites asking about his opinion on the diaphragms.

[John Warren]

----------------

On 4/29/2004 11:40:16 PM William F. Gil McDermott wrote:

I'm a bit undecided on what combination of qualities of the tweeter makes for good subjective performance. It is easy to mentally correlate lack of response at 15 kHz (measured how) with lack of zing. But most of us can't hear 15 kHz, or even 10 kHz.

As I have mentioned before, what you cannot hear is very important. Bandwidth correlates directly with transient response, larger the bandwidth means better transient response and that means in the reagion of the specturm where you can hear.

Bandwidth, efficiency, dispersion, response linearity, power handling are the first parameters that I consider looking at tweeters. The impedance modulus, Fs and it's relationship to the crossover point are of concern when trying to integrate it into an existing system.

It is my thought that the Forte tweeter is much better than the K-77 and there is no way that easy tweeks are going to improve it.

You are correct.

Gil

----------------