Single Capacitor Two Way Crossover -- distortion??

[Blumlein 88]

Hey hey - here we have two universes in conflict!

To me it is obvious that milezone is deep in Japanese hifi culture where listening impression is all that matters. All other guys in the thread are thinking in the academic box of mathematics and physics. This dilemma can not be solved, forget it!

milezone - obviously you won't find the best sounding capacitor (or resistor as well) by asking at this forum! Don't waste your time.

Other guys - it just might be that masses of haphazard peaks and dips of interferences in treble, cone resonances and extra distortion can sound better to milezone or others, subjectively. That is an old trick with high-end hifi - remember whizzer cones in fullrages, Manger transducer, BMR, panel speakers, supertweeters of ´70s etc.

Yeah I remember. I remember people adding piezo supertweets. I remember the headache inducing results if you listened to them. I REMEMBER. I'm trying to prevent it from happening again.

What is the line from the Electra movie with the most beautiful Jennifer Garner. Stick (the old blind master) tells Electra:
"Some lessons can't be taught, Elektra. They must be lived to be understood."

Maybe this is one of those. Asking advice you aren't capable of taking, then you'll just have to live it and learn.

 

[Doodski]

Yeah I remember. I remember people adding piezo supertweets. I remember the headache inducing results if you listened to them. I REMEMBER. I'm trying to prevent it from happening again.

What is the line from the Electra movie with the most beautiful Jennifer Garner. Stick (the old blind master) tells Electra:
"Some lessons can't be taught, Elektra. They must be lived to be understood."

Maybe this is one of those. Asking advice you aren't capable of taking, then you'll just have to live it and learn.

Hehe. Reminds me of playing about with piezo transducer rings from a old submarine. We put a couple of hundred watts into one and it was singing loud and "clear." Sooper tweeter de jour.

 

[March Audio]

Hey hey - here we have two universes in conflict!

To me it is obvious that milezone is deep in Japanese hifi culture where listening impression is all that matters. All other guys in the thread are thinking in the academic box of mathematics and physics. This dilemma can not be solved, forget it!

milezone - obviously you won't find the best sounding capacitor (or resistor as well) by asking at this forum! Don't waste your time.

Other guys - it just might be that masses of haphazard peaks and dips of interferences in treble, cone resonances and extra distortion can sound better to milezone or others, subjectively. That is an old trick with high-end hifi - remember whizzer cones in fullrages, Manger transducer, BMR, panel speakers, supertweeters of ´70s etc.

Whizzer cone here https://www.stereophile.com/content/rethm-maarga-loudspeaker-measurements

............and yet the controlled research shows that people dont like masses of haphazard peaks and troughs and off axis response that isnt smooth or doesnt match the on axis...........

So I dont think that is a convincing argument.

 

[Blumlein 88]

............and yet the controlled research shows that people dont like masses of haphazard peaks and troughs and off axis response that isnt smooth or doesnt match the on axis...........

So I dont think that is a convincing argument.

Come on now. Think about it again, it will make sense in its own sort of way. Its the idea of the thing one must deal with more than what the thing is in these kinds of cases. The other world is the one where the idea is too strong. The reality is a secondary consideration that only dawns on one at a later time in that universe.

 

[Blumlein 88]

So @milezone give us the details. My off the top of my head suggestion is thinking of paralleling some capacitors. If you parallel two the "crossover point" is at half the frequency. So maybe let us help you choose a value which you can get 4 of. Start with one, add one, listen, add two listen etc. etc. You'll find the point you like it best.

 

[March Audio]

Come on now. Think about it again, it will make sense in its own sort of way. Its the idea of the thing one must deal with more than what the thing is in these kinds of cases. The other world is the one where the idea is too strong. The reality is a secondary consideration that only dawns on one at a later time in that universe.

When I say "convincing argument" Im referring to the scientific reality rather than the sometimes overwhelming influence of the other bias factors that come into play.

I have no doubt some "think" they hear better sound when it is not the case.

 

[March Audio]

I thought it was an interesting idea though I admit I was and am a bit skeptical. Still willing to give it a try though... what's the worst that could happen haha? The speaker is a Mark Audio MAOP 11. I may also purchase a Fostex FE168NS to compare the two. The tweeter I was considering is a Visaton G25FFL or for something fancy a Seas T35002. I once owned a pair of FE206EN speakers playing full range with Ribbon tweeters and a single capacitor filtering at around maybe 8000hz I recall. I liked the results quite a bit and am curious to revisit the idea hence my mentioning it in this thread.

In any case it's interesting to me how the cultures of Japan and the US differ so much and it's reflected in their cultural preferences and even things like diy speaker designs. The world would be a very boring place without that sort of variety of preference.

Well you spend money that could be put towards a good speaker

 

[Doodski]

I think for a tweeter R @ 20kHz values of ~0.8microF@8R, 1.5microF@4R and 3.3microF@2R would be a good start. Pick standard available values from there. What is the R of the tweeter?

 

[Blumlein 88]

Well you spend money that could be put towards a good speaker

The money is buying an experience which is part of the journey.

Psychology has been telling us for some time money spent on experiences is far more satisfying than money spent on objects.

 

[milezone]

So @milezone give us the details. My off the top of my head suggestion is thinking of paralleling some capacitors. If you parallel two the "crossover point" is at half the frequency. So maybe let us help you choose a value which you can get 4 of. Start with one, add one, listen, add two listen etc. etc. You'll find the point you like it best.

I believe 1uf will provide a filter cutoff of 20khz at 8ohms. I was thinking of starting there. I don't have high expectations for this speaker necessarily. I have heard the Mark audio drivers playing simple music and was very impressed. I'm not sure how they'd fare with more complex material. Typically that's the major flaw of fullrange drivers in my experience -- is that they lose steam rather quickly when playing complex music. If my experimentation mimics this finding I'll ditch them at a moments notice. But aside from a pair of Genelec speakers, they're what I have on hand right now, and I look forward to experimenting a bit even if the results are undesirable.

 

[March Audio]

The money is buying an experience which is part of the journey.

Then there is no point asking for advice. Just get on and do it and find out the results for yourself

Also wont the experience and journey be a lot more satidfying if you end up building a speaker that works well and sounds good?

 

[milezone]

I think for a tweeter R @ 20kHz values of ~0.8microF@8R, 1.5microF@4R and 3.3microF@2R would be a good start. Pick standard available values from there. What is the R of the tweeter?

Thanks. 8 Ohm for the Visaton tweeter which is where I'll start.

 

[Doodski]

I believe 1mf will provide a filter cutoff of 20khz at 8ohms.

You're on the right track but it's micro Farad and not milli Farad.

 

[March Audio]

I believe 1mf will provide a filter cutoff of 20khz at 8ohms. I was thinking of starting there. I don't have high expectations for this speaker necessarily. I have heard the Mark audio drivers playing simple music and was mightily impressed. I'm not sure how they'd fare with more complex material. Typically that's the major flaw of fullrange drivers in my experience -- is that they lose steam rather quickly when playing complex music. If my experimentation mimics this finding I'll ditch them at a moments notice. But aside from a pair of Genelec speakers, they're what I have on hand right now, and I look forward to experimenting a bit even if the results are undesirable.

Thats probably due to all sorts of complex distortions caused by cone break up and resonances - another reason full range drivers dont work.

 

[Blumlein 88]

Then there is no point asking for advice. Just get on and do it and find out the results for yourself

There is an in between space where technical knowledge is helpful in making the experience something other than a total useless thing even if the final result is sure to be iffy at best.

Here is a simple online calculator.
http://www.erseaudio.com/First-Order-2-Way

It suggests just below 1 microfarad at 20 khz. So with my other advice, I'd suggest buying at least 4 of whatever 1 microfarad caps one wishes. If one isn't suitable, then parallel two if not good, then 3 and if not good then 4. All four should put the crossover region around 5 khz.

Ask the guy in the shop if he suggests tantalum, or mylar or some other type capacitor. I'd probably try some oil caps if it were me.

 

[March Audio]

There is an in between space where technical knowledge is helpful in making the experience something other than a total useless thing even if the final result is sure to be iffy at best.

Here is a simple online calculator.
http://www.erseaudio.com/First-Order-2-Way

It suggests just below 1 microfarad at 20 khz. So with my other advice, I'd suggest buying at least 4 of whatever 1 microfarad caps one wishes. If one isn't suitable, then parallel two if not good, then 3 and if not good then 4. All four should put the crossover region around 5 khz.

Yes of course we have all learnt from experimentation, but its probably a good idea to read some speaker designs books before embarking on the project so you dont make a complete hash of it.

 

[Blumlein 88]

Yes we have all learnt from experimentation, but its probably a good idea to read some speaker designs books before embarking on the project so you dont make a complete hash of it.

That is why he is asking us. So he doesn't have to do that, so he'll know what values work, and so he can see how it strikes him when he experiences it.

 

[Doodski]

That is why he is asking us. So he doesn't have to do that, so he'll know what values work, and so he can see how it strikes him when he experiences it.

Yeah it would be at best a 2 month read and study to have it all hashed out properly. By the time he attacks electrical principles and then inductive/capacitive reactance fundamentals and then hashes out all the maths and all the tidbits in between. That's a lot of stuff for a newbie to do without proper formal instruction.

 

[solderdude]

Ask the guy in the shop if he suggests tantalum

LOL

 

[Mnyb]

The problems will dwarf what kind of capacitor he gets unless really bad and not suitable for a crossover , so stopp fidgetting over capacitor brand and worry more about implementation .

Cone breakup and whizzer cone , parts of the same problem , som FR drivers tries to "controll" cone breakup which invetiable will happen in a full range driver and then the wizzer cone mechanically takes over . A whizzer cone is that extra cone sitting on the dustcap on dome FR drivers