Jump to content

Crossover A Klipschorn cap replacement help


Aoran994

Recommended Posts

10 minutes ago, Tom05 said:

Yes , I’ve got a set of AK2’s that I’d eventually like to convert to AK3 . I bought the 3636 a few years back  , never got around to doing the conversion . I see the 5mh across tap 3 what does that value need to be?

 

If you want a transfer function similar to the Klipsch AK-3 with the T4A, a 3.5mh inductor should be used with the 3636 across taps 0 - 3.  

 

  • Thanks 1
Link to comment
Share on other sites

  • Moderators
10 hours ago, KT88 said:

I have no golden ears and I am an average Joe in that sense.

My BC Type AA had Sonicaps polypropylenes installed. I replaced them with Polyester types (later in another thread I described my impressions for the better when using the very right capacitance values). I will not bore anyone with the thread linked below, but the difference between the Sonicaps and the Mylar types are night and day in my view and described in detail below in the link. Not only in the way the treble is performed but in an important change to the positive how the bass is stronger and the timing is much more musical.

 

And that's awesome, but you changed the sound from the original, which is absolutely your right to do. They don't sound the same as the lab standards/originals. It's of course a subjective thing, and people will have preferences.  It's different, and it's discernable. No one can say it is better, that's an individual thing and completely subjective. (You should have gone with Mundorf or Jupiter, everyone knows caps made in Germany are the best - j/k). 

 

The last thing someone who bought new (to them) vintage speakers is putting in something they know is going to change the sound from the original. First, you get them back to original, listen for 50 to 100 hours and then if you want to try something different, have at it. But do one crossover at a time, and A/B them.*

 

*If you have a preference, and it is Khorns that are being updated, you actually have to move the speakers to the opposite corners to determine if the preference is the XO network, or a corner preference. No two corners sound the same (it is counterintuitive, but it's true), and many people with have a corner preference, left or right, and so you have to switch the speakers to see if it is the XO, or the corner. 

Link to comment
Share on other sites

  • Moderators
8 hours ago, captainbeefheart said:

My point is that other non sanctioned vendor claims people should use the highest quality capacitor with the lowest DF possible. Dean and I are saying that's a bad move because it effects the transfer function, to match the original performance and transfer function you need a cap that behaves like the original ones, like Tecate.

And what I'm saying is that this is all great information and informative for those interested. After all, PWK got into this business after a classmate at Stanford told him that any loudspeaker will sound better if placed in a corner. PWK tried it, said "yep it sounds better." Most, or nearly all, would have stopped there. PWK set out to find out why it sounds better in a corner, and have an understanding of that at the deepest level. 

 

That's where his 4 core principles derived from.

 

For those that don't want the technical deep dive, the capacitors in the kits were selected because they result in the identical sound and performance as the original. That's what people who buy vintage speakers that have 20, 30, 40 year old caps should want. Get me to original so I can figure out from there where I want to go. It's the same on anything. You don't buy a 40 year old Marantz/McIntosh/[fill in blank] and go off the grid on caps do you? Don't you want the person who restores it to get it as close to original as possible?

 

So we got the ESR and DF low down. What about distortion? Do XO Network caps induce TMD, THD, IMD in the final sound, some higher, some lower?

 

From a manufacturer's Data Sheet: Technical specifications: "ERSE's Pulse X caps are a premium audiophile grade capacitor designed for quick transient response and ultimate musical performance. Every attention to detail was put into the PulseX capacitor line to insure premium performance. A premium grade of defect free aluminum metallized polypropylene film was chosen from an industry leading film supplier. Specially designed German winding machines with tension feedback control allow for precision repeatability during manufacturing. This industry first technique allows us to virtually eliminate mircrophonics and obtain a tight ±3% tolerance, with virtually no scrap. All leads are lead free (RoHS Compliant), oxygen free 99.99% pure copper. Each lead is attached to the body of the capacitor with ultrasonic welding. To further reduce equivalent series resistance and improve long term reliability, the leads are then soldered with lead free silver solder. PulseX caps are then wrapped in flame proof polyester tape and the ends are epoxy filled. Before final packing, each PulseX cap is tested with certified premium quality Wayne Kerr capacitance bridges to insure a ±3% tolerance centered on the nominal value. The PulseX caps display signs of a near perfect capacitor. Total harmonic and IMD distortion and phase distortion are virtually immeasurable, while all losses are near zero in every aspect. PulseX caps will allow your music to be accurate and uncolored in every way."

 

Virtually immeasurable means they add THD and IMD distortion, but very low. Compared to what?

 

Others claim same thing, like Jantzen: "High quality and affordable entry level PP capacitor made from high quality foil, produced and metalized in Finland. This capacitor is virtually distortion free and has a neutral sound profile."

 

Maybe Sonicaps are high in distortion? Produce a sound that some prefer. You go to other Forums and they say tube amps are preferred by some (assuming they can even really tell a difference) because the high distortion softens this, or does that, which covers up shortcomings in their speakers and that's why they prefer the sound. This may come to a shock to some, but there are entire Forums built around this premise. Anything to the contrary is pure subjective hooey. 

 

So I guess some caps have more distortion than others? How much does the distortion that ends up coming out of the speaker relate to the "preference" of one brand or type of cap vs. ESR or DF? 

Link to comment
Share on other sites

7 minutes ago, Travis In Austin said:

And that's awesome, but you changed the sound from the original,

 

The original types had vastly different construction than the new Tecate types, yet share the same dielectric material. It stands to reason that if the new Tecate types can give the same transfer function as the large metal can originals then it also stands to reason that there are other mylar types out there that would function just the same.

 

Since there are no measurements for us to look at then none of will ever know if it was indeed close enough or way off. Since they were Mylars he used, I'd wager they were more close than not.

 

  • Like 1
Link to comment
Share on other sites

  • Moderators
43 minutes ago, Deang said:

You might want to read his post again. 

Why? What did I miss?

 

Which post? His post here. Or his post about Sonicaps in his LS?

 

I'm not smart enough to understand 90% of this, I never read his other thread. I could read some of this 10 times it isn't going to sink in. It wasn't until your post where you re-capped what [we what I did there?] what you were trying to convey that is sort of clicked in - a little bit. 

 

I would settle for someone just explaining why a capacitor acts in a way where it lets sound above a certain frequency (or is it below) get through to the transducer. How much of of change in the value of the capacitor results in how much of a change in the "cut off" value. 

Link to comment
Share on other sites

  • Moderators
5 minutes ago, captainbeefheart said:

 

The original types had vastly different construction than the new Tecate types, yet share the same dielectric material. It stands to reason that if the new Tecate types can give the same transfer function as the large metal can originals then it also stands to reason that there are other mylar types out there that would function just the same.

 

Since there are no measurements for us to look at then none of will ever know if it was indeed close enough or way off. Since they were Mylars he used, I'd wager they were more close than not.

 

Ohhhhhh, well that's different. Nevermind. 

Link to comment
Share on other sites

  • Moderators
6 minutes ago, OO1 said:

Tom ,  FYI   ......klipsch Parts sell a kit  for the AK-5   , 

klipsch CS   800-554-7724, Option 1  /9A.M.-4P.M. (EST)  ,  

Awesome, something he isn't looking to do.

 

I'm guessing that means they don't offer something that he is looking to do? 2 to 3? Or a 3 to a 4, but they offer a 4 to a 5?

 

They can't offer a 5 to a 6 as I recall because it is beyond switching out parts, need to make cuts, etc. 

 

 

Link to comment
Share on other sites

35 minutes ago, Travis In Austin said:

So I guess some caps have more distortion than others? How much does the distortion that ends up coming out of the speaker relate to the "preference" of one brand or type of cap vs. ESR or DF? 

 

Dielectric non-linearity causes distortion. Hysteresis causes distortion in some caps. Aluminum Electrolytic and Ceramic are some of the most non-linear types of capacitors.

 

When testing capacitor distortion I had to use quite a bit of signal across the capacitor. In a crossover we want frequencies of interest to the driver to pass through unimpeded while frequencies we don't want we impede and shunt away back to the source. What this means is the frequencies that may show distortion are the ones that we attenuate away anyway. The frequencies we want to reach the driver will not create enough of a potential across the capacitor and so no distortion.

 

This is 600Hz at 70v across the capacitors

 

You can see the hysteresis with the Aluminum Electrolytic and Tantalum.

 

 

 

cap9.jpg

cap8.jpg

cap15.jpg

cap4.jpg

cap3.jpg

cap2.jpg

Link to comment
Share on other sites

  • Moderators
16 minutes ago, captainbeefheart said:

 

Dielectric non-linearity causes distortion. Hysteresis causes distortion in some caps. Aluminum Electrolytic and Ceramic are some of the most non-linear types of capacitors.

 

When testing capacitor distortion I had to use quite a bit of signal across the capacitor. In a crossover we want frequencies of interest to the driver to pass through unimpeded while frequencies we don't want we impede and shunt away back to the source. What this means is the frequencies that may show distortion are the ones that we attenuate away anyway. The frequencies we want to reach the driver will not create enough of a potential across the capacitor and so no distortion.

 

This is 600Hz at 70v across the capacitors

 

You can see the hysteresis with the Aluminum Electrolytic and Tantalum.

 

 

 

cap9.jpg

cap8.jpg

cap15.jpg

cap4.jpg

cap3.jpg

cap2.jpg

I think the Tantaulum looks the best, so it will sound the best right?

Link to comment
Share on other sites

4 minutes ago, Travis In Austin said:

I don't even know that the lines on that O Scope mean. Linear?

 

I should have explained what you are seeing, sorry.

 

Linear, line - basically the straighter the line is the more linear the cap is. With the ceramic you can see how as the line isn't straight, it has a curve to it.

 

With the Tantalum and Electrolytic the hysteresis is so bad it hides the non-linearity.

 

So nope Tantalum isn't the best, it would have high distortion. But who knows, people love amps that produce over 1% THD below 1 watt so someone might like the high distortion effect. 

  • Like 1
Link to comment
Share on other sites

  • Moderators

Wait, wait, wait. Does the type of capacitors used impact [can't think of word/tech term] insertion loss? [A higher drop in voltage/output given same input?] If so that results in a drop in efficiency/sensitivity right? That's breaking a cardinal rule, that would drive up distortion independent of the caps. 

 

Or are we talking too low to measure?? Can't be. You change caps you change transfer function. Change transfer function you change sound. You change sound, but you don't know how you change the sound. So you might as buy Bose or Beats headphones because the sound clearly doesn't matter to you, only that it's coming out. 

 

OR

 

I remember hearing Klipsch ___________ when I was 16 to 25) and I was blown away. I want these to sound like when I bought them, or I want these to sound like when I hear them back when I was _______. 

 

Then you need this kit/builder to update your XOs, that will get you the original sound. How do we know? They were tested, and measured acoustically, and double checked by their sound (something PWK always insisted upon). 

Link to comment
Share on other sites

1 minute ago, Travis In Austin said:

Wait, wait, wait. Does the type of capacitors used impact [can't think of word/tech term] insertion loss? [A higher drop in voltage/output given same input?] If so that results in a drop in efficiency/sensitivity right? That's breaking a cardinal rule, that would drive up distortion independent of the caps. 

 

Yes. That's exactly what DF is, the higher the DF % the more lossy the capacitor will be. That larger drop in voltage at the output across the ESR.

Link to comment
Share on other sites

@Travis In Austin attenuation, or insertion loss is different than distortion.

 

With a loss, or attenuation - the shape of the waveform at the output is identical to it's input, only smaller.

 

Distortion means that the waveform has changed it's shape, the output waveform is not the same original waveform that was at the input.

Link to comment
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.
Note: Your post will require moderator approval before it will be visible.

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...