Khorn crossover capacitors + harshness problem

[idahodewitt]

I just acquired a pair of 1978 cornerhorns with AA crossovers. I'd already disconnected and checked the capacitors with a multimeter in one of the x-overs, and the 13uf was off by more than 20%. I do not know what the other two are supposed to be. Does anyone have the proper uf rating for the two smaller can caps?

The speakers have a harshness problem in the upper midrange when I turn the volume up. I'm using a 14watt tube amp that sounds fine with my Von Schweikert monitors, but the khorns can't be turned up if I'm playing certain songs. Anyone with an AA x-over tried certain other caps and gotten smoother results than the stock caps? Any other simple updates or modifications that I can use on these for improved sound? I'd appreciate any suggestions. Thanks.

L.D.

[Guest " "]

don't think a multi meter will be much good on checking caps.

about your harshness problem....there are dozens of different subjects on this fourm about that....everything from changing caps, tweeters and even the conclusion by some that the issue lies with the upper band of the mid driver fixable by installing a p-trap..then theres incorrect phasing..bad diaphragms..etc..etcc

but you should plan your strategy carefully rather than making consecutive changes in a short period of time.

[Groomlakearea51]

http://forums.klipsch.com/forums/p/106045/1070106.aspx#1070106 should help out a bit.

[BEC]

First, 30 year old capacitors are bad. No good reason to test them.

Second, 30 year old oil filled capacitors are getting close to the age when they start leaking oil all over your cabinets.

Here is a schematic for the Type AA crossover.

Bob Crites

[HarryO]

1: Change the caps. I prefer the Sonicap from BEC. They're my #1 choice on the AA. If I'm keeping the costs WAY down I'll use the Jantzen from Parts Express. They don't sound as good as the Sonicaps but there's times when cheap is the only way I can go.

2: Some tubes don't match up well with some applications. The K-77/K-55/K-400 will tell you quick if they don't like your tube choices.

3: The diaphrams may well be showing their age. I've been changing out more diaphrams in the last year than ever before. There is a marked improvement in sound when changed in most cases. Some of the aged diaphrams do get nasty and crackly on top under a bit of volume.

4: And not particularly in any order. Check and clean all wiring connections and tighten up the screws holding the horn lenses to the motorboard.

5: Take a magnet and see if the screw holding the coil inductor is magnetic. If so, replace it with brass or stainless. They can also be glued to hold without screw.

6: Refer to #1. Bob (BEC) is right on. 30 year old caps are shot. If you look close I'll bet they're already leaking on the board. Bad caps = Bad sound. The last pair of AAs I did were 78 models out of Belles. When I changed caps the old oil caps weighed nothing. They were ALL totally dry but still played out. I've never seen all of them totally dry before.

Harry

[idahodewitt]

Bob,

Yikes! The two caps that are supposed to be 2.0uf tested at 2.47uf and 3.07uf. The 13uf tested at 15.8uf. When I buy new caps they usually test right on the money or very close.....these didn't. No leakage, but they're not where they're supposed to be, that's for sure. As soon as I get these khorns paid off (next month) I'll buy some Sonicaps. Thank you much for the schematic.

L.D.

[WMcD]

http://forums.klipsch.com/forums/p/46305/907814.aspx#907814

Let me suggest you try the P-trap described in this article.

Wm Gil McD

[idahodewitt]

Does the P-trap for the Heresys work the same for the khorns? That's a 0.25mh inductor parallelled or in series with a 4uf cap? This is connected in on the common side of the sqauwker, not positive? The reason I'm asking so many questions is that there's too much information to sort all of it out clearly. I also don't like to buy parts I can't return.

Let me know if you know what parts to use and how to hook them up. I'd sure appreciate it.

Larry D.

[boom3]

Make one change on one speaker at a time and then evaluate. Keep notes. The value deviation of the caps per se is not terribly significant, it looks like normal tolerances plus age to me. There is another, more significant aspect of caps that degrades over time; the dissipation factor, which, in very basic terms, is how much energy is wasted as the current flows through. That is one of the ways old caps can grunge up the sound.

Bob Crites makes beautiful replacement crossovers. Replacing them will probably make this concern go away. The upper midrange of the Klipschorns does bother some people, regardless of crossover. This has been explored on this forum and there are a variety of strategies to fix it.

[Groomlakearea51]

Sorta' related.... Some folks have said that the older Type A sounded better than the AA. When Bob (BEC) does the AA recap, he puts a "jumper" to bypass one of the caps so it can be used as either an A or AA. That might be worthwhile inasmuch as the caps are likely due for replacement anyways. You can also go the A/4500 route... To make that work right, you need to use Bob's CT-125 tweeter which is almost identical to the K-77 in appearance, but has an extended range. But that may the opposite "effect" of what you are looking for. I'd consider the A/AA route since caps are in order anyways. As others have said, one thing at a time. After caps.... change diaphragms, etc. etc. Bob can steer you correctly on the A vs the AA.

[WMcD]

Does the P-trap for the Heresys work the same for the khorns? That's a 0.25mh inductor parallelled or in series with a 4uf cap? This is connected in on the common side of the sqauwker, not positive? The reason I'm asking so many questions is that there's too much information to sort all of it out clearly. I also don't like to buy parts I can't return.

Let me know if you know what parts to use and how to hook them up. I'd sure appreciate it.

Larry D.

The P-Trap: The cap and inductor are in parallel. Then this combination is put in series with a lead to the midrange.

Traditionally, such filter components are placed in the positive side of the circuit. I should not make a difference though.

The P-trap is a type of circuit called a "tank." This is because the current circulates within it. Essentially it creates an open circuit at about 8000 Hz. This is the frequency range where some or all midrange drivers of this type has some sort of resonance.

Bob Crites has commented that not all drivers have this glitch and has put up a measurement curve showing this. OTOH, all the other curves I've seen, those published by Klipsch, have the glitch.

In my view, the glitch does not show up as too bad when the driver is on a plane wave tube. But the K-400 has, I believe, a lot of on axis gain above 3000 Hz. That is why the horn - driver combination works up to 6000 Hz per the classic set up. Also, the A and AA type crossovers use just a high pass filter in the mid. so there is nothing in the circuit (save for the additional P-trap) to block it.

I suspect that the newer crossovers for the K-Horn address the issue by having a bandpass filter for the mid. Klipsch keeps the schematic confidential. Also the upper crossover point of the filter is down at 4000 Hz and has a steep roll off. This means it reduces the glitch.

Best,

Wm McD

[wsu99999]

A good middle of the road cap is Clarity Caps. The PX version are cheap and are supposed to be better than other budget brands. I can't verify that but they sound fine in my LS's

[mark1101]

If your priority is listening to music loud I would consider replacing the lower slope networks with some higher or extreme slope nets. YOu could save money and not recap the stock ones. Doing that will only restore fine fidelty that dropped off over the years but it will not do anything at all for the harsh sound at loud volumes which is caused by the time mis-alignment of the (wide) overlap between drivers of those lower slope networks like A and AA. The network circuit is your problem not the components. A and AA are much better for lower listening levels as you may have already found out.

Gil has it right to minimize the issue with the P-trap. But you still have considerable mis-aligned driver overlap between the woofer and squawker as well as some remaining between the tweeter and squawker. If you took a visit to the ALK engineering website and read his white paper concerning extreme slope networks in his download section, you could understnad more about what I am suggesting here.

[Guest " "]

http://forums.klipsch.com/forums/p/46305/907814.aspx#907814

Let me suggest you try the P-trap described in this article.

Wm Gil McD

I could not find the P-trap schematic.

If I were new to this forum and had to read thru that battle of wit's thread, to better understand p-traps, I would not have a good impression of the forum. The thread seems more about time alignment.

Here's the P-trap as defined by Tom

"The P Trap is a .1mh coil and a 3uf
cap, in parallel, wired in series with the + terminal of an ATLAS mid
driver. The "P" stands for the name of a friend of mine who developed
the trap, an astute horny of long experience who found years ago that
many (but not all) of the Atlas mid drivers used by Klipsch exhibit a
rising response, a "flare" so to speak, ay 9khz. The P Trap is meant to
suppres said flare. No other trap is a "P Trap"."

[Dflip]

Not a schematic, but it's the "for dummies" version that Dean originally provided me and one Mike posted a while back. I hope this helps. It worked for me.

It is then hooked up to the positive side of the midrange horn.

[idahodewitt]

Thank you guys for all the info on fixing this issue with the K-horns. I'm about to purchase the caps to recap it, so I'll get the other parts as well. I really like these speakers, until I turn them up, so these two improvements should get these where I can really enjoy them.

Thanks, L.D.

[Deang]

I really like these speakers, until I turn them up...

The sound will improve when you change the caps out, however -- it won't solve the harshness problem when you turn it up loud. That problem is related to the horn.

[idahodewitt]

I understood that the purpose of the p-trap was to lessen the impact of the upper midrange spike that brought about the harshness. So, does the p-trap help this or not?

L.D.

[mark1101]

People have different opinions about this so called harshness. Honestly, I'm not exactly sure what you are hearing so we can't be sure which issue you are really pointing at. This is just my opinion but I think one would be hard pressed to even hear this 9K "glitch" that gets talked about. Ever see a picture of it and how far down that signal is? I would bet it is not your problem. What I speak of on the prevsious page maybe shouldn't be classified as harshness. SMEAR is a much better word and may be what you are having an issue with. If you listen to music LOUD most of the time, forget about caps and P-traps. They won't fix your issue. The Ptrap will help because it reduces the overlap on the squawker, but it doesn't fix it completely. You can reread my last post if you are interested in what the smear can be casued by. Extreme slope networks or time aligning the drivres are the only ways I know of to make the smear issue completely go away.

[idahodewitt]

The harshness I'm hearing could be better described as an over accentuating of the upper midrange that multiplies in loudness more than other frequencies when the volume is increased. I had the same problem with a pair of Klipsch Tangent 500 speakers. I sold them because of that very reason.

My last pair of speakers before the Khorns were a pair of Von Schweikert VR-1 monitors and they did not become unlistenable at loud volumes, though they also did not have quite the 'presence' of the Khorns at low & mid volumes.

3 Questions:

How extreme of a slope is needed to significantly diminish this smear?

What procedure is used to get the driver timing aligned?

Is all this beyond my capabilities A) in understanding, in equipment/tools needed, C) in funds required to get it 'right'.

I would like to get the cornerhorns 'right', but I'm not sure if it is possible for me to do. I may have to settle for 'better' rather than 'best'.

I guess I could just keep the volume down, then all problems are solved. Ha!

L.D.