Hi,
In few kits or diy loudspeakers, people are using MKP and sometimes expensive audio MKP in // with the driver for their filter and damped by a resistor.
Does it make sense to use a low esr cap in such AC shunt arengement if at the end it sees a resistor ? What isn he benefice in spite of using a normal lythic non polar with lower ESR?
Is it snake oil or can it be heard VS a lythic non polar cap ?
What do you think or experienced ?
In few kits or diy loudspeakers, people are using MKP and sometimes expensive audio MKP in // with the driver for their filter and damped by a resistor.
Does it make sense to use a low esr cap in such AC shunt arengement if at the end it sees a resistor ? What isn he benefice in spite of using a normal lythic non polar with lower ESR?
Is it snake oil or can it be heard VS a lythic non polar cap ?
What do you think or experienced ?
Hi,
No shematic or see any Troels Gravsens' . I am talking about the low pass filter cap , i.e. the shubnt cap between the signal and the return path, i.e. between the + and the -.
No shematic or see any Troels Gravsens' . I am talking about the low pass filter cap , i.e. the shubnt cap between the signal and the return path, i.e. between the + and the -.
Maybe you are describing what is commonly called a Zobel network. The intention of this shunt circuit is to correct the semi- inductance of a speaker to maintain a flat impedance around the crossover frequency.
Can be a Zobel as a damped low pass of a second order or more. But what I do not understand is why one sould use a low esr cap while the resistor after or before has the opposite behavior. A simple lythic should perform as good if the total amount of esr is the same between a lythic and a serie resistor and a film cap and a serie resistor with the little more R value for the last cause the mkp has a lower esr than a lythic ?
I tend to agree with you, however N.P. electrolytics do not come with a tight tolerance and when aged, the ESR climbs. There are too many cases to be had of faulty electro,s and serious d.i.y. people avoid them. I have yet to hear an expensive MKP capacitor that improves over a similarly rated standard type ( same voltage rating ,same construction). the measurements do not lie. The best construction is still film and foil although there are now special metallized capacitors using thicker metal depositing , new polypropylene material and multiple connectors to improve the transition of charge currents.
Depends on how a-retentive one is about their sound quality. Years ago, I rebuilt a pair of KEF crossovers and - with all the Zobelling going on to make the entire speaker load appear "resistive" - I choose to do only the series signal path caps in polyprop.
I left what I felt were the burner caps as original NP electrolytic. Perhaps an OK decision if those networks were to flatten the looking into Z due to driver / cab resonances. Maybe not so good if the shunt network was to compensate for a rising VC winding inductance. So to decide, you need to know what the shunt network is for.
A resonance, might be able to filter out any "grit" the Al-El generates. A rising VC impedance is a wide band phenomena; perhaps you'll hear the sound of the parallel network as much as you'd hear it as a series element.
Hopefully I chose well. Those speakers are long gone, so, who knows. It behooves one to think about it. Cost went into overrun when I thought "just replace them all".
I left what I felt were the burner caps as original NP electrolytic. Perhaps an OK decision if those networks were to flatten the looking into Z due to driver / cab resonances. Maybe not so good if the shunt network was to compensate for a rising VC winding inductance. So to decide, you need to know what the shunt network is for.
A resonance, might be able to filter out any "grit" the Al-El generates. A rising VC impedance is a wide band phenomena; perhaps you'll hear the sound of the parallel network as much as you'd hear it as a series element.
Hopefully I chose well. Those speakers are long gone, so, who knows. It behooves one to think about it. Cost went into overrun when I thought "just replace them all".
Thanks, so apart the fact film caps do not drift there is no special reason to use them in shunt position because their low esr if a resistor is in serie. Was asking because sometimes I believe some like Gravsen use expensive caps there for the mid unit low pass...jantzen Sup.
So use the cheaper mkp. Does someone is able to hear a difference related to the resistor positioning here i.e. before or after the cap which is the same electrically while I assume a capacitor charging and decharging near the serie line to the driver to be more hearable when the resistor is nearer the return path than the signal path ?
So use the cheaper mkp. Does someone is able to hear a difference related to the resistor positioning here i.e. before or after the cap which is the same electrically while I assume a capacitor charging and decharging near the serie line to the driver to be more hearable when the resistor is nearer the return path than the signal path ?
I recaped a kef 104/2 already....quite a nightmare...
Hummm what you say make sense but difficult to know when refurbishing a filter but to copy or make R compensation if the lythic is swaped by a low esr mkp....
Hummm what you say make sense but difficult to know when refurbishing a filter but to copy or make R compensation if the lythic is swaped by a low esr mkp....
That was the one! Unbelievable coincidence; your question and us recapping the same speaker at different times too.
The question is not for this speaker but I remember the hard job it was to test and access several time to the filter to adjust the capacitance values... especially for the 100 uF caps and the shunt ones in the bass network to avoid a barel sound 🙂... very good but sensible loudspeaker... a Diva if you ask me 🙂
Basically they are called DC link capacitors and there are many manufacturers such as TDK, Cornell Dubilier, KEMET . For information on polypropylene film try https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7959306/
Last edited:
Not such a thing unless cable length measures in fractions of a light year.
Within a couple miles? ... forget it.
Thanks guys, so no fancy expensive audiophiles mkp caps in the shunt network if a resistor is a part of it, just basic mkps to get a precise capacitance value for convenience that will not drift.
Ok I will stay with my basic cheap Solen SCR or Jantzen cross cap for that job or any industrial film cap if cheaper 🙂 .... and no fancy decoupling or super resistor here.
Ok I will stay with my basic cheap Solen SCR or Jantzen cross cap for that job or any industrial film cap if cheaper 🙂 .... and no fancy decoupling or super resistor here.