Electrolytics sound fine

[fredex]

That's a matter of physical location, not connection.

Not what I was thinking. When I said, "If you draw the circuit of salas's xover, swapping the cap's position...." I meant in the circuit diagram, not physically.

With the cap and coil physically fixed in position, there is an electrical difference between the two connection methods that you won't understand by just looking at a circuit diagram.

My point is, a circuit diagram usually doesn't include stray couplings, if the diagram did include the cap's 'proposed' coupling to earth the diagram would better reflect reality.

 

[SY]

Perfectly electrically fitting healthy tubes must not make any difference for brand or vintage scientifically speaking.

I've never measured two tubes of the same type that showed exactly the same results. So I truly don't follow this.

With the cap and coil physically fixed in position, there is an electrical difference between the two connection methods that you won't understand by just looking at a circuit diagram.

Any chance you could illustrate what you're talking about? If the physical position doesn't change, neither does the coupling, but I'm ready to be shown the error of my ways. An order-of-magnitude comparison between any proposed coupling and the effect of moving either part a millimeter or so would be helpful as well.

 

[fredex]

Any chance you could illustrate what you're talking about? If the physical position doesn't change, neither does the coupling, but I'm ready to be shown the error of my ways. An order-of-magnitude comparison between any proposed coupling and the effect of moving either part a millimeter or so would be helpful as well.

I'll try with words. First forget about shifting anything physically in either case. We just use short wires to change the connections. The long cable to the amp (antenna with RF on it) can either go to the coil first or the cap. If that connection goes to the coil first the RF current flowing through the cap to ground will be attenuated by the coil's inductance.

Hypothetical, yes.

Edit: To clarify the cap's outer foil is capacitively coupled to earth, the external connection to the cap is to the inner foil.

 

[SY]

So, you're saying that there's no return current, it just... vanishes?

 

[fredex]

Sy
Not saying or suggesting.

 

[Skorpio]

So, you're saying that there's no return current, it just... vanishes?

It seems clear that you have never tested an electronic circuit for EMC?

Two quick ones for you:

-Do you know the function of a ground plane?

-If you don't have a ground plane, does this means that RF current stop flowing?

 

[Salas]

I've never measured two tubes of the same type that showed exactly the same results. So I truly don't follow this.

We don't talk tolerance SY. We talk different brands or vintage adequately manufactured to spec including tolerance, which is a spec itself. You thought lab matched when I talked NOS mania VS new Asian stock?

 

[Salas]

P.S. Don't try to argue on something I got in a certain crossover. You will end up nowhere. Either its repeatable when you build a band pass which has an air coil next to a Black Gate N type wired in series, or it is not. This is hands on. Its a tip. Not a claim. The big difference is that I have the experience and you the theory. I can be totally wrong and got fooled subjectively. It happens all the time. Or not.

 

[SY]

I can be totally wrong and got fooled subjectively. It happens all the time. Or not.

In all seriousness, I compliment you on your honesty. Most people in this niche cannot even admit the possibility that their perception could be an illusion.

 

[SY]

It seems clear that you have never tested an electronic circuit for EMC?

Two quick ones for you:

-Do you know the function of a ground plane?

-If you don't have a ground plane, does this means that RF current stop flowing?

In order, you would be incorrect.

Yes.

No, but it's irrelevant to the discussion of crossovers, and even more so in high level, low impedance circuits. And it again completely begs the question about the inherent symmetry violation of the claim.

 

[Salas]

of the claim.

Not a claim, a report and just a thought. Don't get it that literal.

 

[SY]

Fair enough!

 

[ShinOBIWAN]

Salas, shut up. Your boring me.

Look at this way my friend, you've now got another topic you know not to bring up again.

 

[Salas]

 

[ShinOBIWAN]

Salas, as always your a good sport.

 

[elac310]

The inductor before, feeds less high frequency to the cap, than having the cap first feeding the inductor. The bandpass maybe the same.

Although the discussion dates back to some time ago and participants preferred to close it...
I'd like to confirm Salas' impression/hypothesis, although my conclusion in respect of my own applications is the opposite.

I have myself experienced that the order of L and C can make a difference. My experience is based on loudspeaker crossovers where the midband involves a high and a low freq. filtering. I found on a couple of occasions that the cap before the inductor has a beneficial effect. Whether it's because the cap preserves the large coil's windings from some electromechanical low frequency stress or for other reasons...

Chris.

 

[system7]

You gotta wonder if we aren't regressing in HiFi. Old discrete transistor preamps were class A with all electrolytics correctly biased. Modern opamps are class AB often with a vaguely biased 200uF electrolytic in the -ve feedback circuit. Not good since electrolytics are highly non-linear without forward bias.

We looked at this simple single-rail zero-feedback Mosfet amp recently:

An externally hosted image should be here but it was not working when we last tested it.

That output capacitor of 4700uF might look a bit of a horror, until you realise that it hardly gets stressed by any high power factor at bass frequencies, and like every other electrolytic in the circuit is correctly biased. Apparently it sounds really fine.

Lynn Olsen is of the opinion the exotic silver and oil polypropylene capacitors sound better because they form less of an oxide layer than your regular metallised films like tin and aluminium. Maybe.