14.7psi is ~atmospheric pressure. If the cap is vacuumed before sealing then you have applied <=~14.7psi until the sealing leaks.
Any electrical attractive force is going to generate less than one atmosphere of pressure.
You are right to suggest that the pressure <<<12000psi.
If the air is already removed by vacuum, then all that left to compress is the dielectric and the plates themselves.
There must be another mechanism to explain what you are hearing.
Like maybe all the seals do leak and it is air compression effects that are being heard.
Any electrical attractive force is going to generate less than one atmosphere of pressure.
You are right to suggest that the pressure <<<12000psi.
If the air is already removed by vacuum, then all that left to compress is the dielectric and the plates themselves.
There must be another mechanism to explain what you are hearing.
Like maybe all the seals do leak and it is air compression effects that are being heard.
Does a DC bias alter the distortion form a cap? Maybe, maybe not, but it is a well defined, easily measured variable.
Does a DC bias alter the sound? Maybe, maybe not, but this is a separate question and in audiophile land there isn't any acceptable means of measuring this. I place the emphasis on acceptable because any differences can be measured. Audiophiles just don't except definitive measurements as being definitive.
Does a DC bias alter the sound? Maybe, maybe not, but this is a separate question and in audiophile land there isn't any acceptable means of measuring this. I place the emphasis on acceptable because any differences can be measured. Audiophiles just don't except definitive measurements as being definitive.
I'm shooting for this friday but it might end up being next week.
Going to set up for easy swap for, with and without DC bias.
Having two friends over to comment on any audible difference when I make the change. Plus they wont have any idea what I'm connecting.. They have been good at hearing changes in the past
I'm Curious to see what happens
Joel
Going to set up for easy swap for, with and without DC bias.
Having two friends over to comment on any audible difference when I make the change. Plus they wont have any idea what I'm connecting.. They have been good at hearing changes in the past
I'm Curious to see what happens
Joel
Hi,Joel Wesseling said:
Don't forget the caps take some time to charge (depending on the value of the resistance) and discharge (could be very long!).
Plus without DC bias two caps in serie are worse than one cap of halved value.
So you will have to build two crossovers to really be able to change the settings on the fly.
pos said:
I can see the logic behind that statement but reality may be different based on Menno's paper re signals below the crossover (going a bit on memory here) where the reactance is increasing relative to the load (kind of equal voltage divider). If so, the effect is shared between the caps. But I am not speaking with any certainty.
Joe R.
Hi Joe and any other crazy people who have added batteries to there crossovers or wonder if they should..........
To day i spent about 5 hours removing all the parts from my external crossovers to neaten them up and box them amongst other things.
So pets do not destroy them and my girlfriend does not pull strange faces at the spiders nest of crossover wiring on the lounge room floor.
any how the picture below is the result.
the four easy to see hi viz yellow caps which have DC bias present on them are 4.7uF metal poly 250V which replaced an expensive Solen tin foil (this brought a tear to my eye)
the two 9V batteries you can see at the bottom of the picture in series are in series with a 100k resistor (1W carbon) the input resistor for amp safety is 1k carbon (1W)
the battery snaps holding the two 9V batteries are easy to unclip to do A-B testing.
I did some listening after about of ambient tv noise to break in the caps this is probably unecessary but i dont know or care.
the caps without bias certainly did not sound bad the elsinores still sound excellent.
i tried the caps on a few tracks without bias listened intently to the treble which sounded very clean.
so i clipped the battery snap on and listened again...
was it better????????
unclipped the clip discharged the cap by shorting the clips
listened again - could the treble be slightly cleaner ?
clipped the cap on again... wound the volume on Metallica's "One" (sorry to my neighbors) there is plenty of cymbal crashing and guitars all happening at once.... no breakup...... tried foo fighters "let it die" no treble break up....... tried armin van buuren's album "Imagine" also good...
with bias it sounds to my ear that the treble is slightly better....
listened to a whole heap more stuff.
I dont really like A-B test after you do some work on your system, psycho acoustics play a role i think....
nevertheless. i think the treble sounds better with DC bias
the only way i can think to describe it is it sounds slightly more natural less harsh without any loss of volume. I would say slightly less distorted. it is hard to gauge with the treble already being so good with the elsinores... but at high volume i think the trebel does not break up as easily instruments dont seem muddled or anything like that.
any how i would encourage other people to have go and see what they think
-Dan
To day i spent about 5 hours removing all the parts from my external crossovers to neaten them up and box them amongst other things.
So pets do not destroy them and my girlfriend does not pull strange faces at the spiders nest of crossover wiring on the lounge room floor.
any how the picture below is the result.
the four easy to see hi viz yellow caps which have DC bias present on them are 4.7uF metal poly 250V which replaced an expensive Solen tin foil (this brought a tear to my eye)
the two 9V batteries you can see at the bottom of the picture in series are in series with a 100k resistor (1W carbon) the input resistor for amp safety is 1k carbon (1W)
the battery snaps holding the two 9V batteries are easy to unclip to do A-B testing.
I did some listening after about of ambient tv noise to break in the caps this is probably unecessary but i dont know or care.
the caps without bias certainly did not sound bad the elsinores still sound excellent.
i tried the caps on a few tracks without bias listened intently to the treble which sounded very clean.
so i clipped the battery snap on and listened again...
was it better????????
unclipped the clip discharged the cap by shorting the clips
listened again - could the treble be slightly cleaner ?
clipped the cap on again... wound the volume on Metallica's "One" (sorry to my neighbors) there is plenty of cymbal crashing and guitars all happening at once.... no breakup...... tried foo fighters "let it die" no treble break up....... tried armin van buuren's album "Imagine" also good...
with bias it sounds to my ear that the treble is slightly better....
listened to a whole heap more stuff.
I dont really like A-B test after you do some work on your system, psycho acoustics play a role i think....
nevertheless. i think the treble sounds better with DC bias
the only way i can think to describe it is it sounds slightly more natural less harsh without any loss of volume. I would say slightly less distorted. it is hard to gauge with the treble already being so good with the elsinores... but at high volume i think the trebel does not break up as easily instruments dont seem muddled or anything like that.
any how i would encourage other people to have go and see what they think
-Dan
Attachments
And more than two years down the road we have this article on enjoy the music:
Charge Coupled Crossovers Article By Jeff Poth
Charge Coupled Crossovers Article By Jeff Poth
It makes the point I was going to make:
The benefits are more likely with inexpensive capacitors - I am current using Jantzen Red Superior caps as the single most important component in the Elsinore Project Speakers and no bias.
But if money is tight, here in Asutralia use the yellow 250V MP caps available from Jaycar - and you will get really good results for less expense.
The Jantzen caps are wound incredibly tight (they designed their own machine to get the right result they wanted) and thus mechanically stressed damped internally. Lesser capacitors don't have that benefit.
Cheers, Joe R.
The benefits are more likely with inexpensive capacitors - I am current using Jantzen Red Superior caps as the single most important component in the Elsinore Project Speakers and no bias.
But if money is tight, here in Asutralia use the yellow 250V MP caps available from Jaycar - and you will get really good results for less expense.
The Jantzen caps are wound incredibly tight (they designed their own machine to get the right result they wanted) and thus mechanically stressed damped internally. Lesser capacitors don't have that benefit.
Cheers, Joe R.
Tnteresting topic. Unfortunately the image file is deleted because this file wasn't upload here. "Photo-Bucket", "Image Shack" and similar portales are no good choice. Who can upload the image here?
I have a doubt regarding consider I have a 3rd order crossover for the tweeter so we have cap in series and inductor in shunt and then cap in series. So when we are biasing the cap then as per the above articles and discussions the caps to be doubled and then put in series to get same value and add a 100k to 1M resistor in between the joint where two caps are connected together.
Now I have a question that can I use all the capacitors used in the crossover with only one +20VDC? consider we have midrange and also tweeter with many capacitors so do we need to use multiple +20V sources? or can we use one global Voltage source?
I will be using these with Raal ribbons hence I believe the difference can be perceived well.
Now I have a question that can I use all the capacitors used in the crossover with only one +20VDC? consider we have midrange and also tweeter with many capacitors so do we need to use multiple +20V sources? or can we use one global Voltage source?
I will be using these with Raal ribbons hence I believe the difference can be perceived well.
DC biasing works for me, but..
To "rhythmsandy", the drain from the battery is minor due to the applied resistors meaning that one DC source will be sufficient.
Regard effect of dc bias or not I have experimented with dc bias for my JBL Ti250, but for so far, only on the low pass filter of the low frequency driver. Adding 9V bias made a small subjective impression of more accurate bass. Increasing the voltage to 18V potentiated this impression further and I lived happy with this setup for a couple of years. With the 18V bias, the amps (Zapsolute 118W Class A) control over the 14" LE14H drivers seemed to increase, indicated by less cone movement during high volume. Then, recently, I put in one more battery, biasing the two 47uF Clarity caps to 27V – this made a big difference! The bass / midbass changed to a tighter, more dynamic and musical version concerning micro dynamics and tone – a surprising change was that guitar solos, as well as male voices, subjectively were more involving and listenable. Inspired by this I pushed the bias further to 36V – what happened? Big disappointment, the bass / midbass seemed to lose both volume and precision – it now blurs.
I will keep the 36V bias a couple of days just to ensure that the decrease sonically not is temporally. However, the result indicates that biasing change the voicing of my system, but it is dependent by the charge. Have anyone else similar experiences? I am tempted to speculate also if implementing bias change the speaker – amp interaction regarding dampening factor.
In rest of my 250s network I have replaced the old stock capacitors with Mundorfs Supremes with Supreme Silver in Oil and Silver/Gold in Oil for the bypasses (without DC bias). Before adding DC bias on the bass filter, one 24uF Sonicap Gen 1 did the crossover job.
To "rhythmsandy", the drain from the battery is minor due to the applied resistors meaning that one DC source will be sufficient.
Regard effect of dc bias or not I have experimented with dc bias for my JBL Ti250, but for so far, only on the low pass filter of the low frequency driver. Adding 9V bias made a small subjective impression of more accurate bass. Increasing the voltage to 18V potentiated this impression further and I lived happy with this setup for a couple of years. With the 18V bias, the amps (Zapsolute 118W Class A) control over the 14" LE14H drivers seemed to increase, indicated by less cone movement during high volume. Then, recently, I put in one more battery, biasing the two 47uF Clarity caps to 27V – this made a big difference! The bass / midbass changed to a tighter, more dynamic and musical version concerning micro dynamics and tone – a surprising change was that guitar solos, as well as male voices, subjectively were more involving and listenable. Inspired by this I pushed the bias further to 36V – what happened? Big disappointment, the bass / midbass seemed to lose both volume and precision – it now blurs.
I will keep the 36V bias a couple of days just to ensure that the decrease sonically not is temporally. However, the result indicates that biasing change the voicing of my system, but it is dependent by the charge. Have anyone else similar experiences? I am tempted to speculate also if implementing bias change the speaker – amp interaction regarding dampening factor.
In rest of my 250s network I have replaced the old stock capacitors with Mundorfs Supremes with Supreme Silver in Oil and Silver/Gold in Oil for the bypasses (without DC bias). Before adding DC bias on the bass filter, one 24uF Sonicap Gen 1 did the crossover job.
Not directly but I know some of the vintage JBL users have found that higher voltages seem to have the subjective effect of damping that leg. Strange. I'd be more inclined to give some credence to that than to a lot of the drivel cranked out about capacitors since I can't see any thought process / reasoning that would cause expectation bias along those lines. Not initially anyway, although obviously the second the subject is raised, expectation bias for or against is introduced in the reader, human nature being what it is. I occasionally use DC biasing myself, but not with a battery; Greg Timbers, IIRC, said he couldn't hear any difference between 9v and 18v, but I don't know how often he tried higher than 9v levels, or if it was just the original K2.
Last edited:
Given that quite a few people on this thread have expressed significant reservations, you do not appear to have an expectation bias 'opposite to most others'. Just expectation bias, which everybody has to an extent, degree and nature varying.
Personally, I was & am largely neutral on the subject, so presumably of a similar level of bias to yourself. The theories advanced for it, while plausible, do not stand up well, and personally I'm not a believer in fancy capacitors (nothing wrong with high quality 'standard' MKPs). OTOH, I thought it was interesting enough to try out in a spirit of curiosity, since it's not expensive. TBH, I didn't expect to find any notable changes one way or the other, but I tried it out on a handful of systems and it did seem to result in a small difference in certain conditions (higher efficiency setups), but this was not invariable. That's as far as I go. What the mechanism might be, I don't know, nor am I going to write a load of meaningless subjective waffle or recommend everybody goes out and tries it. I'll simply observe in my experience this is one of the few cases which at least in certain cases appears to result in a small change in behaviour. YMMV as ever.
Personally, I was & am largely neutral on the subject, so presumably of a similar level of bias to yourself. The theories advanced for it, while plausible, do not stand up well, and personally I'm not a believer in fancy capacitors (nothing wrong with high quality 'standard' MKPs). OTOH, I thought it was interesting enough to try out in a spirit of curiosity, since it's not expensive. TBH, I didn't expect to find any notable changes one way or the other, but I tried it out on a handful of systems and it did seem to result in a small difference in certain conditions (higher efficiency setups), but this was not invariable. That's as far as I go. What the mechanism might be, I don't know, nor am I going to write a load of meaningless subjective waffle or recommend everybody goes out and tries it. I'll simply observe in my experience this is one of the few cases which at least in certain cases appears to result in a small change in behaviour. YMMV as ever.
DC biasing works for me, but..
Following up on this - thing have changed, bass is not gone as my first impression was - I'm not sure if it is a little bit dampened - or if the cross over frequency is changed to a lower one as would happens if the capacitance increases with higher voltage on the capacitors. Then a dip in the midbass region could explain the first impression. The experience after a couple of days is more spacious midrange (can also be explained by a dip) and overall dynamics - which is strange to me. I still speculate if the charge coupling have some impact of how the power supply in the amplifier "see" the speaker load? My amps have a zero feedback design maybe more exposed for changes of speaker behavior? However, I have no problem believe others who reports changes in sound when they charge couple their networks, whatever the reason may be. I have been in the scientist game to long to reject a phenomenon if it's not measurable at the first time. I we see a phenomenon we cannot explain, than more than once the history shows that maybe we looked the wrong place or did inadequate measurements to understand.
Following up on this - thing have changed, bass is not gone as my first impression was - I'm not sure if it is a little bit dampened - or if the cross over frequency is changed to a lower one as would happens if the capacitance increases with higher voltage on the capacitors. Then a dip in the midbass region could explain the first impression. The experience after a couple of days is more spacious midrange (can also be explained by a dip) and overall dynamics - which is strange to me. I still speculate if the charge coupling have some impact of how the power supply in the amplifier "see" the speaker load? My amps have a zero feedback design maybe more exposed for changes of speaker behavior? However, I have no problem believe others who reports changes in sound when they charge couple their networks, whatever the reason may be. I have been in the scientist game to long to reject a phenomenon if it's not measurable at the first time. I we see a phenomenon we cannot explain, than more than once the history shows that maybe we looked the wrong place or did inadequate measurements to understand.
Calling our erstwhile moderator.
Hi Dave,
It occurs me that this topic should be a thread on its own as most DIY'ers can try this and it can be applied to all order High Pass crossovers, both to Tweeters and also Midranges in 3-Way systems (see below). What do you reckon? This is not just an Elsinore tweak.
Sorry Dave, but full-range guys miss out, but maybe they have had an advantage all along. Talk about the last laugh, maybe they had the last laugh first.
Maybe the topic should read "DC Bias in Crossovers (mostly Tweeters)."
An externally hosted image should be here but it was not working when we last tested it.
Joe R.
Well, I like to put a zobel circuit on fullrange to damp rising inductance, so
I think , we fullranger can still take advantage out of this
Hahahahhahahahhahaha
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.