Crossover A Klipschorn cap replacement help

[Aoran994]

Hello everyone, i am from Italy and i got a pair of 1969 Klipschorn, i love them, right now i got an idea , changing the caps, they are for sure really old, as is the original crossover, the A one, could you help me in what to buy? I will leave the original transformers on ,thank you very much for your help , i love this speaker and i hope to make them sound better, right now i am changing home Soo i can put my hands on them, cheers!!

[KT88]

Congrats to your vintage Khorns, Aoran. Is it possible that you could take a picture of the crossover where the parts can be seen? May be you have already new caps from someone who owned them before you? Anyway, it would be of help. The original caps were motor run caps, quite big tin cans of the size of a cigarette package. The chance is there that they leak the oil over the decades. But in original state they were very good sounding.

[Deang]

Hermetically sealed paper in oil. This is actually a good question for @JRH

[captainbeefheart]

If you want to keep the original sound you need to choose a slightly "imperfect" capacitor. I.e. it should have a low quality factor or conversely a high dissipation factor.

 

Since dissipation factor is a ratio between ESR and X_c

 

X_c = capacitive reactance

 

ESR = Equivalent Series Resistance

 

DF = ESR / X_c

 

The reactance or impedance of the capacitor changes with frequency, as frequency increases Xc decreases as seen in the equation:

 

1 / 2*pi*f*c

 

f = frequency

c = capacitance in farads

 

The ESR is in series with impedance of the capacitor - as the impedance of the capacitor decreases with increase in frequency there reaches a point where the ESR becomes a problem and attenuates the higher frequencies. Simply, the ESR slightly changes the transfer function of the filter network by reducing high frequencies.

 

Say you have a capacitor with a DF=.1% and another with DF=.7%

 

2uF capacitor at 10kHz has an impedance of 7.95 ohms

 

The ESR between the two caps is

 

.001*7.95= .0079 ohms

.007*7.95= .055 ohms

 

 

Modern Polypropylene capacitors have a DF of .1% or lower, often lower, closer to .05%

 

The original caps in Klipschorn were  PIO but the vast majority have Polyester/Mylar film types like yours from 1969. Looking at a modern datasheet (linked below) for "Orange Drop" Polyester/foil type capacitor and it states maximum DF of .75

 

Also linked below will be the same "Orange Drop" capacitors but Polypropylene/foil type - scroll to page 8 for DF data and you'll see the maximum being .167

 

Paper is similar DF to Polyester and it's more linear. They sound very close, probably too close to even really say for certain but Paper caps might be slightly better sounding.

 

 

 

https://www.cde.com/resources/catalogs/225p.pdf

 

https://www.cde.com/resources/catalogs/716p.pdf

 

 

 

 

[captainbeefheart]

You can see in the graphs above that Polyester/Mylar has similar DF vs Temperature characteristics as Paper in oil types.

 

Look how Polypropylene, Polystyrene, and Teflon have extremely low DF at all temperatures.

 

 

It should be clear to keep the same sound, when replacing a Polyester/Mylar capacitor use the same type or Paper will give similar characteristics.

[Travis In Austin]

4 minutes ago, captainbeefheart said:

You can see in the graphs above that Polyester/Mylar has similar DF vs Temperature characteristics as Paper in oil types.

 

Look how Polypropylene, Polystyrene, and Teflon have extremely low DF at all temperatures.

 

 

It should be clear to keep the same sound, when replacing a Polyester/Mylar capacitor use the same type or Paper will give similar characteristics.

But little or none of that is how Roy and his team make sure the replacement parts have an identical transfer function. But I suppose it would be pretty close on paper.

[captainbeefheart]

19 minutes ago, Travis In Austin said:

But little or none of that is how Roy and his team make sure the replacement parts have an identical transfer function. But I suppose it would be pretty close on paper.

 

Of course one can install random capacitors into the filter networks and measure the output until you find the desired transfer function.

 

A capacitor's real world characteristics are due to the dielectric material used and construction. It makes sense to either simply stick with the same dielectric material and construction method.for similar results. Or ask yourself what properties of any capacitors will effect the transfer function? Obviously the capacitance determines impedance and that's the largest factor but ESR also plays a role.

 

I'm simply pointing out ESR matters and DF is a ratio between ESR and X_c so simply find any capacitor with similar DF characteristics and it should sound close.

 

 

[Aoran994]

11 hours ago, KT88 said:

Congrats to your vintage Khorns, Aoran. Is it possible that you could take a picture of the crossover where the parts can be seen? May be you have already new caps from someone who owned them before you? Anyway, it would be of help. The original caps were motor run caps, quite big tin cans of the size of a cigarette package. The chance is there that they leak the oil over the decades. But in original state they were very good sounding.

Hello thank you for the response, i have the old big can , i think that they have not leaked the oil i have to re-check, i didn't now they were very good thanks!

[KT88]

1 hour ago, Aoran994 said:

Hello thank you for the response, i have the old big can , i think that they have not leaked the oil i have to re-check, i didn't now they were very good thanks!

They were very good in the sense of the way how PWK used them to build up a circuit from parts he had available. 

 

Components and devices from one era can always be combined harmoniously and organically. For example, paper in oil caps or polyester caps with exponential horns like the K400, or spoked wheels with a Jaguar E Type. In retrospect, it sometimes looks like all the parts and materials were deliberately made. But mostly one always resulted from the other. And accordingly, it would not fit aesthetically and technically if a Jaguar E Type were to be driven today with carbon rims or a 1970 Khorn were to come with the most modern polypropylene caps.

I don't want to talk about nostalgia, that would be wrong and incorrect from today's point of view. No, we are talking about the innovations that made the earlier time modern.
For example, a Jaguar D Type was the first car with disc brakes and, apart from Bugatti, the first car with aluminium rims. The Khorn was the most compact and modern fullhorn speaker for the home and had the lowest distortion factor of all speakers you could think of. 

[Deang]

Since ESR increases with frequency, it also acts as EQ in the circuit. The polypropylenes I've measured are all pretty much flat lining, as the chart shows.

 

The Tecate polyesters I've measured so far are averaging .44 ohms at 1kHz. That's quite a bit of resistance in that circuit, and we haven't even addressed the autoformer and coils yet.

 

Unless you're hearing is impaired, the increase in brightness is pretty noticeable, which is why I always knocked the midrange down a dB. I understand that's not the same as using the right parts.

[Aoran994]

1 hour ago, KT88 said:

They were very good in the sense of the way how PWK used them to build up a circuit from parts he had available. 

 

Components and devices from one era can always be combined harmoniously and organically. For example, paper in oil caps or polyester caps with exponential horns like the K400, or spoked wheels with a Jaguar E Type. In retrospect, it sometimes looks like all the parts and materials were deliberately made. But mostly one always resulted from the other. And accordingly, it would not fit aesthetically and technically if a Jaguar E Type were to be driven today with carbon rims or a 1970 Khorn were to come with the most modern polypropylene caps.

I don't want to talk about nostalgia, that would be wrong and incorrect from today's point of view. No, we are talking about the innovations that made the earlier time modern.
For example, a Jaguar D Type was the first car with disc brakes and, apart from Bugatti, the first car with aluminium rims. The Khorn was the most compact and modern fullhorn speaker for the home and had the lowest distortion factor of all speakers you could think of. 

nice comparison i loved it

[Aoran994]

i will try to follow the thread, im not technical at all, i know that they have certain value for example i saw in the bob crites website that he use a 13uf sonicap and another little one that i dont know in the Crossover A

[henry4841]

Discussion of caps never ends on this forum. All modern caps from reliable sources will function quite adequately in a crossover network. Best to stick with what Klipsch recommends if one has to change the capacitors that came with your speakers. They are what the engineers deem best sounding vs cost for your speakers. Lots of splitting frog hairs about differences in capacitors and how they function on forums. And yes using different brands may change the sound subtly but that is the case changing any other part in your speaker and amplifiers for that matter if your hearing is good enough to hear the difference. That is if the part being changed is a quality part from a reliable source. The difference is going to be subtle at best. 

 

 

[Deang]

Paper in Oil capacitors are now found on eBay, and you would have to parallel quite a few to get the 13uF value. An acceptable substitute are the Klipsch OEM polyesters that are currently used in their current production loudspeakers.

 

There are pinned posts at the top with resources, like schematics, and authorized vendors (like me) for parts and work. I have original OEM parts and could build you a new pair for $529.00. This would allow you to keep the originals intact for their historical value.

[henry4841]

As Dean and I know this thread can go south very quickly as most discussions of capacitors do on this forum. Just listen to Klipsch, Klipsch vendor for Klipsch parts and authorized dealer, Dean, if one wants to have have modern caps or new crossovers. 

[geoff.]

…so paralleling ANY capacitor will make the resultant sound brighter than a single value?

 

Could that be a little of the bypass capacitor magic?

 

[henry4841]

19 minutes ago, geoff. said:

…so paralleling ANY capacitor will make the resultant sound brighter than a single value?

 

Could that be a little of the bypass capacitor magic?

 

When you parallel capacitors they add to the total capacitance. Parallel a 10uf with a 2uf and you have made a 12uf capacitor as far as the circuit sees. Bypassing generally means when a technician adds a low value film cap to a high value electrolytic capacitor. Not something done in crossover networks being the values are low enough to find a film cap with the correct value. 

[Deang]

37 minutes ago, geoff. said:

…so paralleling ANY capacitor will make the resultant sound brighter than a single value?

 

Could that be a little of the bypass capacitor magic?

 

 

Bypassing apparently has its use in power supply circuits, but in AC filters I'm pretty sure it's been debunked. 

 

As for ESR, the difference with these polyesters is 2/100's of an ohm.

 

 

[captainbeefheart]

Dean do you have a 2uF Polypropylene you can test at 1kHz?

 

Can you test at 10kHz and 20kHz also?

 

Get a heat gun or freeze spray and watch ESR change as you heat/cool the mylar capacitors while watching the meter.

 

 

4 hours ago, henry4841 said:

All modern caps from reliable sources will function quite adequately in a crossover network.

 

Not really no.

 

In a perfect world a capacitor is a capacitor and they are all the same, in the real world that's not true and why we discuss capacitors. The reason is different dielectrics have different properties. Look at the graphs again, Polypropylene, Polystyrene and Teflon all will exhibit very low DF at all temperatures, it's almost ruler flat across a very wide range of temperatures. The Mylar and all Paper capacitors have very large changes in DF with temperature, and in general at average temps higher than the other mentioned. As frequency increases to the point where impedance of the cap is very low the ESR from the higher DF will come into play much more with a paper/mylar type than with the others mentioned.

 

4 hours ago, henry4841 said:

And yes using different brands may change the sound subtly but

 

The differences in dielectric and construction properties is what will yield different results and not so much manufacturer. Two manufacturers making similar construction caps from the same dielectric will show very similar properties that won't be audible.

 

As an engineer brands are in the back of our minds. We choose a capacitor by the dielectric for a particular application. Once we know that we are using say a polystyrene cap for it's properties in an application then we may choose a reliable manufacturer for good measure.

 

First and foremost it's always dielectric properties that dictate what cap to use in a particular application. That's why you see all different types of caps in devices, they all have their little niches and tradeoffs.

 

 

 

[billybob]

All cool and everything but, has the OP's questions been asked and answered?

Thanks!