The improvement was in the tweeter area.
It has been said that the sound though is coming from the cap with the largest value, but what I suspect the bypass cap did is to dissipate charge quickly.
If the cap holds charge this smudges the next lot of input.
If you have quick discharge then the cap is ready to receive the next lot of information free of previous input.
Technically a Cap should filter out given frequencies.
This is governed by the microfarad value.
The cap should otherwise be transparent (we know this is not the case as different caps give different results).
If the bypass cap simply adds the ability for the charge to dissipate quickly (reduce held charge) then this may be its value.
Just my theory.
The proof would be in trialing them.
My experience resulted in improvement when in parallel with electrolytic caps.
Just as your experiment resulted in electrolytic sounding better than poly.
i figured it was in the tweeter area.
i really doubt anyone will hear much difference in the woofer area.
as I said before and you just mentioned. the capacitor is an AC filter. It filters out freq.
for me as I think about it. It wouldn’t matter at all for the woofer if it’s poly or electrolytic. the only thing that would really matter is where the cap makes the cross at.
our ears are not that sensitive at lower frequencies.
yeah the woofer might go up to 2.5khz at the cross, but the leading notes of instruments and their harmonics are higher up. That to me is what gives them their character and timbre.
i would bet that most of the money should be spent on the tweeter and it’s section.
even cellos leading freq are up top. They give most of the character and timbre.
makes sense now when I see most speakers really give most emphasis on the Highs.
ok yeah of course you are gonna not have the woofers suck. But the tweeter most likely plays the most part in timbre and character. Unless the rest of the speaker design sucks.
i really doubt anyone will hear much difference in the woofer area.
as I said before and you just mentioned. the capacitor is an AC filter. It filters out freq.
for me as I think about it. It wouldn’t matter at all for the woofer if it’s poly or electrolytic. the only thing that would really matter is where the cap makes the cross at.
our ears are not that sensitive at lower frequencies.
yeah the woofer might go up to 2.5khz at the cross, but the leading notes of instruments and their harmonics are higher up. That to me is what gives them their character and timbre.
i would bet that most of the money should be spent on the tweeter and it’s section.
even cellos leading freq are up top. They give most of the character and timbre.
makes sense now when I see most speakers really give most emphasis on the Highs.
ok yeah of course you are gonna not have the woofers suck. But the tweeter most likely plays the most part in timbre and character. Unless the rest of the speaker design sucks.
This is fun tho you gotta admit. Lol
hopefully tomorrow I’ll be in the mood to take some mic measurements
gonna get some alligator clips tomorrow and make it so I can do things on the fly
god I wish the xo was external so I could tweak away easily
hopefully tomorrow I’ll be in the mood to take some mic measurements
gonna get some alligator clips tomorrow and make it so I can do things on the fly
god I wish the xo was external so I could tweak away easily
The components in last XO's I put together had their wire leads twisted together with pliers.
I never solder until I'm happy with the sonic result.
Technically you shouldn't rely on the solder for electrical contact.
This should be via the twisted wires exposed to each other.
The solder just encases the wires sealing them from the effects of the atmosphere and gives further rigidity and support.
I never solder until I'm happy with the sonic result.
Technically you shouldn't rely on the solder for electrical contact.
This should be via the twisted wires exposed to each other.
The solder just encases the wires sealing them from the effects of the atmosphere and gives further rigidity and support.
Yeah man.
clips will work just fine.
agreed that soldering should be the last step. But I’m pretty handy with a soldering iron. Can be undone easily.
will be fun
clips will work just fine.
agreed that soldering should be the last step. But I’m pretty handy with a soldering iron. Can be undone easily.
will be fun
I still suspect that all of this just has to do with slight deviations from values of capacitors rather than the innate qualities of them.
Theoretically , instead of jumping caps in the xo internally , could I put them on the woofer binding posts? Between the red and black of the bottom binding posts?
the woofers that is ?
so i can change them out quickly ?
thanks
the woofers that is ?
so i can change them out quickly ?
thanks
To start a discussion relating to the use of 0.01uF bypass capacitors in loudspeaker crossovers would be pitting faith against physics.
The science of electronics is based firmly on physics, making it difficult to understand for some.
That's why some so-called "experts" turn to subjective analysis of crossover components rather than to objective analysis.
Bypassing has it's roots in power supply rails for equipment where low impedance is required through the radio frequency range.
At radio frequencies a large capacitor also behaves like an inductor and presents a high impedance to signal flow.
A small bypass capacitor in a radio frequency application becomes the controlling element by lowering the impedance to signal flow.
HOWEVER:
Large capacitors in loudspeaker crossovers are working at audio frequencies where they remain in their range of operation and behave like capacitors. There is therefore no requirement from a physics perspective to bypass large capacitors in loudspeaker crossovers.
Although the differences between bipolar electrolytics and film capacitors are easily measurable, these difference become insignificant when you consider that the performance of the speaker is mainly determined by the characteristics of the drivers.
If you really like the sound of that electrolytic you need to measure what you've got to preserve or replace it over the years. Electrolytics deteriorate at either 1 year, 5 years or 20 years depending on the sealant used & amount of water initially present. You don't even know the capacitance; you have conflicting schematic & part labeling numbers. I suggest measuring capacitance @120 hz, capacitance at 10 khz, ESR, and inductance at a minimum. You could also measure the polyprophylene caps and see what numbers document the difference is that you can hear. This can be measured with a signal generator & a scope and some resistors, but it is not easy.
I might agree with Galu that bypass caps at .01 uf are irrelevant to 18 uf caps, but there are enough 1 uf caps sprinkled around crossover imiages I've downloaded that the tenth of inductance parallel is doing something in the treble. In amplifiers, Peavey parallels two .22 uf mylar caps local to the power transistors with 10000 uf electrolytic 8" away. Peavey doesn't spend money where it can't be seen for no reason. Shows up in the HD spec @ 20 khz.
Last edited:
great write ups by you and Galu.
yeah I find all this fascinating and the first thing I need to do is , tomorrow get some alligator clips , put just legs where the caps are supposed to go on the internal xo and then be able to switch out the 18uf poly caps and electrolytic without soldering so I can pinpoint the differences i hear.
i really don’t believe that much in burn in. And if something does happen it’s very minimal.
and I completely agree that the best thing would be to be able to measure the electrolytics across a wide freq range to get a better idea of what they are doing in the crossover.
i read up on some stuff and saw someone say something along the lines that basically the differences between electrolytics and poly is where they are crossing over. Not their qualities. That makes sense to me.
if the a cap is fluctuating across freq in its capacitance it’s hard to trace where it’s crossing over. A poly cap which is more linear is easier to determine it’s behavior.
unfortunately I can’t see how I would be able to run all those tests on the electrolytic originally in the crossover.
maybe I could call someone local here who has the instrumentation to carry that out is the only thing I can think of
one thing is for sure , this is all very interesting and a great learning experience.
thank you guys for chiming in.
oh and I might add. I totally agree that long term it would be best to emulate the electrolytic if I like it more for it to retain the current measurement it has through a poly.
yeah I find all this fascinating and the first thing I need to do is , tomorrow get some alligator clips , put just legs where the caps are supposed to go on the internal xo and then be able to switch out the 18uf poly caps and electrolytic without soldering so I can pinpoint the differences i hear.
i really don’t believe that much in burn in. And if something does happen it’s very minimal.
and I completely agree that the best thing would be to be able to measure the electrolytics across a wide freq range to get a better idea of what they are doing in the crossover.
i read up on some stuff and saw someone say something along the lines that basically the differences between electrolytics and poly is where they are crossing over. Not their qualities. That makes sense to me.
if the a cap is fluctuating across freq in its capacitance it’s hard to trace where it’s crossing over. A poly cap which is more linear is easier to determine it’s behavior.
unfortunately I can’t see how I would be able to run all those tests on the electrolytic originally in the crossover.
maybe I could call someone local here who has the instrumentation to carry that out is the only thing I can think of
one thing is for sure , this is all very interesting and a great learning experience.
thank you guys for chiming in.
oh and I might add. I totally agree that long term it would be best to emulate the electrolytic if I like it more for it to retain the current measurement it has through a poly.
And to add some more thoughts.
to make this experiment more controlled I think when I build the external xo I will need to do the flowing
1) I am going to take the internal xo out of the speakers
2) i will take the internal leads from the drivers and connect them directly to the binding posts internally, then run wire from the binding posts externally to the outboard xo.
3) instead of just making the new xo with new parts , I might need to transfer the original parts to the external xo and test them one by one , by omission to see their influence on the overall sound. That is.….build the external xo initially with all the components from the current xo. Test each individual component with the new component until I can reproduce the same response. So that would mean, make the external xo with all the original parts. Take out the cap on the woofer, replace it with a new poly cap till I can replicate the same sound. then move on to the woofer inductor and do the same etc.
4). When I can successfully replicate the original board and components with the new ones can I proceed to do anything else.
an experienced builder of course wouldn’t need to do this.
what do you guys think?
to make this experiment more controlled I think when I build the external xo I will need to do the flowing
1) I am going to take the internal xo out of the speakers
2) i will take the internal leads from the drivers and connect them directly to the binding posts internally, then run wire from the binding posts externally to the outboard xo.
3) instead of just making the new xo with new parts , I might need to transfer the original parts to the external xo and test them one by one , by omission to see their influence on the overall sound. That is.….build the external xo initially with all the components from the current xo. Test each individual component with the new component until I can reproduce the same response. So that would mean, make the external xo with all the original parts. Take out the cap on the woofer, replace it with a new poly cap till I can replicate the same sound. then move on to the woofer inductor and do the same etc.
4). When I can successfully replicate the original board and components with the new ones can I proceed to do anything else.
an experienced builder of course wouldn’t need to do this.
what do you guys think?
I thought you had a signal generator or at least a PC with earphone jack, and a 2 channel scope. That, some load resistors is all you need. Phase shift voltage to current (voltage on load resistor) at frequency predicts capacitance. At higher frequencies inductance shows up in that. ESR shows up in current across load, too. Lot of calculation, reason I'm thinking of buying a CH500 meter from Japan.
There ought to be some software for PC that would do that with signal coming out earphone jack & load voltage on resistor coming in line level jack. Unfortunately those connections on PC have gone the way of the dodo bird. Everything now is hooked up through *****ng bluetooth.
Last edited:
no I don’t. All I have is an LCR meter.
the rest is above my head which you are mentioning. Ugh
im really new at this. Keep it in mind. Lol
the rest is above my head which you are mentioning. Ugh
im really new at this. Keep it in mind. Lol
I haven’t even got to the aircore inductors yet. Lol
wonder what effect those will have.
i measured the inductance and the DCR for those and ordered accordingly.
I suspect the audible difference again will be on the tweeter circuit
wonder what effect those will have.
i measured the inductance and the DCR for those and ordered accordingly.
I suspect the audible difference again will be on the tweeter circuit
Yes, crossover components have their greatest influence near the crossover frequency where their impedance is comparable to that of the circuit.
The impedance of crossover components will go very high or very low as we move further away from the crossover frequency, and their influence will diminish.
Please do NOT quote the entire post just above yours. It's not that kind of forum. (It's also in the rules)