PDA

View Full Version : L 300 Crossover Rebuilding-What Caps Would You Use?



Oldmics
10-17-2007, 09:35 PM
Gentlemen

I have a customer that frequents this site who has asked me to rebuild the crossovers for his L 300 speakers.

He is looking for better than stock preformance but using value engineering parts selection.

He has a brand new original pair of crossovers to compare against the soon to be rebuilt crossovers.

The original caps were mylar and he is leaning towards using polypropelyne caps.

I personally think this is the way to go for a slightly better than original sound quality at reasonable cost outlay.

What would you guys suggest?

And what would the sound quality differences be using the parts that each of you recommends,when compared to originals?

I know that everybody will have a unique perspective.I am just putting out some feelers at his request.

Best regards,Oldmics

macaroonie
10-18-2007, 01:23 AM
http://audioheritage.org/vbulletin/showpost.php?p=1042&postcount=3

4313B
10-18-2007, 04:39 AM
What would you guys suggest?Build him a charge-coupled pair. Having built the 4344/4345 and 4355 charge-coupled networks and having listened to the difference I have to agree with G.T. - "These days I don't even bother listening to anything that isn't charge-coupled."

These days I don't even bother building a network that isn't charge-coupled; And those old systems get a new lease on life.

CauYem
10-18-2007, 06:15 AM
Build him a charge-coupled pair. Having built the 4344/4345 and 4355 charge-coupled networks and having listened to the difference I have to agree with G.T. - "These days I don't even bother listening to anything that isn't charge-coupled."

These days I don't even bother building a network that isn't charge-coupled; And those old systems get a new lease on life.

What is "charge-coupled"?
Where can I get more in for about "charge-coupled"?
Thanks!

Earl K
10-18-2007, 06:34 AM
Where can I get more in for about "charge-coupled"?

Read the last post in this thread (http://audioheritage.csdco.com/vbulletin/showthread.php?t=3555) for an introduction to Charge Coupling™ .

Earl K
10-18-2007, 07:24 AM
Hi Oldmics


,,,,,,snip,,,,,,,,he is leaning towards using polypropelyne caps.

I personally think this is the way to go for a slightly better than original sound quality at reasonable cost outlay.

What would you guys suggest?

,,,,,,snip ,,,,,,,,

I know that everybody will have a unique perspective.I am just putting out some feelers at his request.

- After 4 years of dc biasing caps / I still think this topology is the best thing I've encountered since sliced bread . So ( never having heard the L300 ) I would still recommend Charge Coupling™ as an excellent starting point, for any new crossover build.

- For the sake of expediency ( and cost ), I'd start with biased Solens on the woofer and horn circuits .
- For the low entry price of doing comparitive studies, I'd try both biased Solens in the UHF circuit as well as unbiased Clarity caps ( on the 077s ).
- Since I never know what people will like / I'd offer up the two choices and then let the client decide what "floats his boat".
- ( ie; Some people prefer more of an Aphex style top end / of exaggerated UHF harmonics / while others don't find this musically appealing )


And what would the sound quality differences be using the parts that each of you recommends,when compared to originals?

- A ton more clarity offering a much deeper "look-see" into the music. I find this quite engaging.

addendum ;
- ( IME ) this ( the benefits of dc biasing ) is primarily due to the dc potential ( when held within the cap ) inducing electrical damping into "resonance prone capacitors" & providing for an electro-mechanical type damping on JBLs' "Over-Amped" transducers . ie; ( from my POV, ) most of these JBL transducers ( used in the 434x systems ) were originally designed for SR work and as such have way too many false dynamics built into them ( a type of loudness effect that's effectively damped down by Charge-Coupling ™) . Reducing both types of sonic aberrations ( which I call "splash" ) helps me hear further into the musical experience .
- I primarily like to emphasize these 2 benefits to dc biasing ( since I actually hear the effects being implemented ) .



:)

timc
10-18-2007, 09:19 AM
One question about CC. Is there a special reason 9v is choosen? Would you get an even better sound if you biased the caps to, lets say, 24v?

-Tim

Earl K
10-18-2007, 10:34 AM
One question about CC. Is there a special reason 9v is choosen? Would you get an even better sound if you biased the caps to, lets say, 24v?

Stick with voltages approved by JBL when using JBL transducers .

- JBL used 18 volts in the S9500 & M9500. The 475nd compression driver with the aquaplased titanium diaphragm seemed to have needed the extra damping provided by the larger voltage .

- I've never searched for ( & therefore, never found ) what dc voltage will overdamp a typical JBL compression driver .

- I've worked with ( & extensively listened to ) an Altec 288-8K for more than 3 years now , on a CCed Hipass circuit . I find that I can only use 3 volts of dc in cap pairs that are isolated from the driver by inline resistance ( such as an Lpad ) . Sometimes even 1.5 volts of DC in a pair of caps is too much ( damping ) when those caps are directly connected to the driver (ie; no inline resistance, as is the typical case of a HF bypass circuit ).

:)

Ian Mackenzie
10-18-2007, 01:31 PM
Oldmics,

There seems to be a number of members lately enquiring about wanting crossover upgrades for reasonable cost and even low cost.

There is no cheap / low cost upgrade by the time you add up your own time (incl (R & D) and all the parts incl new binding posts etc

If they are expecting you do to the R & D for a low cost upgrade that is worthwhile imho that is more difficult than a properly engineered outcome.

The CC network outcomes are well documents on the forums.

The less expensive options (non charge- coupled) with low cost budget capacitors are not worthwhile in that by the time you bypass these cheap parts with high quality bypass capacitors it will end up costing about the same as charge-coupling.

Some people pefer the more exotic and expensive capacitors. I dont think this model or your client belongs to that category.

Ian

Robh3606
10-18-2007, 06:02 PM
I have CC Mylars with bypass caps and PP without and either sounds very good. I would just go with PP and CC them.

Rob:)

timc
10-19-2007, 05:12 AM
Stick with voltages approved by JBL when using JBL transducers .

- JBL used 18 volts in the S9500 & M9500. The 475nd compression driver with the aquaplased titanium diaphragm seemed to have needed the extra damping provided by the larger voltage .

- I've never searched for ( & therefore, never found ) what dc voltage will overdamp a typical JBL compression driver .

- I've worked with ( & extensively listened to ) an Altec 288-8K for more than 3 years now , on a CCed Hipass circuit . I find that I can only use 3 volts of dc in cap pairs that are isolated from the driver by inline resistance ( such as an Lpad ) . Sometimes even 1.5 volts of DC in a pair of caps is too much ( damping ) when those caps are directly connected to the driver (ie; no inline resistance, as is the typical case of a HF bypass circuit ).

:)


Very good answer Earl. So just to clarify, The amount of voltage supplied to the caps, will determine how much damping will be applied to the driver? If so I sense a lot of experimenting in the future :D


-Tim

Earl K
10-19-2007, 06:10 AM
Very good answer Earl. So just to clarify, The amount of voltage supplied to the caps, will determine how much damping will be applied to the driver? If so I sense a lot of experimenting in the future

Note ; You're highjacking "oldmics" thread with questions that are very specific to application techniques for CCed circuits .

- Anyways, rather simplistically, "yes" . ( inline resistance plays a role )

- But as I said above, JBL has already done the R&D for their drivers, so I'm not sure why you want to bother with this ( assuming you're using JBL drivers ).

:)

timc
10-19-2007, 07:04 AM
Sorry for the hijacking:o:

It was a general question from my side Earl, I was just curious what the biasing actually did. I have discussed it with a few speaker builders and they have said it is just a "marketing hype" Good to hear they are wrong. :)

Anyway. Back to oldmics question. I have done something similar with my L65 crossovers. I used stock values on the caps and changed them to Solen. The coils i didn't change since i did not have an easy way to measure them. The result was a much smoother sound than the original, with less hiss. It also seems there is less "noise" in quiet parts.

I have also good experiences with Auricap and Mundorf Supreme Silver/oil. They are expensive. Worth the cost?.......if you can afford them without going broke.


-Tim

Thom
10-19-2007, 07:24 AM
Have there been blind A/B tests done with various high quality (not high quality against junk) capacitors. Is this really different than wires. Please, I'm not staking out a position here. I have no facts to do that on. It's an honest question. Also I saw that JBL, on it's top speaker I believe, used charge coupled for part and didn't for another part. Is there a belief that it is audible in one range and not in another?

Earl K
10-19-2007, 10:20 AM
Hi Oldmics,

- There's an old thread from Regis that's worth a read ( for your client & yourself ) . Click Here ! (http://audioheritage.csdco.com/vbulletin/showthread.php?t=6183)
- Regis bypassed his stock L300 crossovers with small value PolyPropylene & Polystyrene caps .
- This simple approach might represent all that your client needs to hear & experience, to feel the change has been worth the investment.


:)

4313B
10-20-2007, 09:18 AM
Stick with voltages approved by JBL when using JBL transducers.G.T. used 9V because it was convenient. JBL never delved into various voltages. G.T. said Ed Meitner also uses diodes which he thinks could be cool since one would then never have to change the battery.
- JBL used 18 volts in the S9500 & M9500.G.T. said Japan asked for that after a special K2-S9800 was done that way for a certain famous reviewer.

If so I sense a lot of experimenting in the future :DWell, have fun then! :)

Also I saw that JBL, on it's top speaker I believe, used charge coupled for part and didn't for another part. Is there a belief that it is audible in one range and not in another?I don't know, we just do the whole network and worry about other things.



That's one limitation DIY folks don't have, they can worry about everything and try whatever they want. :p

Ducatista47
10-20-2007, 10:52 AM
G.T. said Ed Meitner also uses diodes which he thinks could be cool since one would then never have to change the battery.

I would be most interested to know more about this.

Clark

Earl K
10-20-2007, 01:45 PM
G.T. used 9V because it was convenient. JBL never delved into various voltages.

- I use 6 volts dc on the PLLXO before the amp and anywhere from 1.5 to 3 volts dc on the passives after the amp .
- Therefore, from a certain POV, in total, I'm using pretty standard operating voltages on my stuff ( though, apportioned differently ).


G.T. said Ed Meitner also uses diodes which he thinks could be cool since one would then never have to change the battery.


Hmmmm,

- I've thought some about a diode regulated current feed / though I haven't yet, concluded much of anything .
- Perhaps an inline combo, ( paralleled resistor/diode ) would offer the constant feed/dissipation ratios that I'm looking for .
( Something like ; Extremely High Value "R" with the diodes' voltage rating chosen to be less than half of the final DC potential ) .




:)


psst; Oldmics, Solens' at 9 volts !

Hoerninger
10-20-2007, 02:26 PM
G.T. said Ed Meitner also uses diodes which he thinks could be cool since one would then never have to change the battery.

Many years ago KOSS used a diode-capacitor network to provide the polarisation voltage for their electrostatic headphone. The voltage rsp. very low current were extracted from the audio signal (speaker out).

The situation for charge coupling is comparable as the current is extremly low. The circuit may be simpler as only a few volts are needed.

I hope this was helpful.
___________
Peter

S20RVERT
11-09-2007, 12:51 PM
All this sounds great. I want to update my LX5's and was looking around the site for help. Now that I read about polystyrene and polypropylene I want this sort of upgrade. My Chicago JBL service center in says he gets his caps form JBL if they have them or finds the best avaliable. Can anyone help me with a Chicago service center that does these modern upgrades? If not Chicago, in the USA? Thanks

Domino
11-30-2007, 05:56 PM
Well guys I was seriously considering charge coupling the crossovers in my L300's. About a year ago I changed out the capacitors in the crossovers. I installed Solens and have never been real happy with the harshness in the mids and highs. So I decided to try one more set of capacitors in the passive configuration first. This time I installed Audicaps by Audience. I have about 6 hours on them now and all I can say is WOW! What a difference. I have owned these speakers since new, purchased back in 1977, and they have never sounded so good. Sheer liquidity in the mids and highs. I can not recommend these more. When I asked them about break in they said that after about 6 hours they will start to attain their sound and be totally broken in after about 100 hours. They are right on target. They are not cheap by any means ($45.31 for a 16uf cap) but they are well worth it!

toddalin
11-30-2007, 06:13 PM
Well guys I was seriously considering charge coupling the crossovers in my L300's. About a year ago I changed out the capacitors in the crossovers. I installed Solens and have never been real happy with the harshness in the mids and highs. So I decided to try one more set of capacitors in the passive configuration first. This time I installed Audicaps by Audience. I have about 6 hours on them now and all I can say is WOW! What a difference. I have owned these speakers since new, purchased back in 1977, and they have never sounded so good. Sheer liquidity in the mids and highs. I can not recommend these more. When I asked them about break in they said that after about 6 hours they will start to attain their sound and be totally broken in after about 100 hours. They are right on target. They are not cheap by any means ($45.31 for a 16uf cap) but they are well worth it!

I'm looking for some Solens. What values do you have? :o:

Earl K
11-30-2007, 06:38 PM
This time I installed Audicaps by Audience. I have about 6 hours on them now and all I can say is WOW!

Did you mean to write "Auricaps" ? Anyways, I'm glad you like them / they do consistently get good press .

FWIW & IME ;
(i) Placing two 6 uF caps back to back, ( with a net value of 3 uF ), results in a different sound than a single 3 uF cap ( from the same manufacturer ) .

(ii) Two caps back to back holding a DC charge ( but disconnected from the DC source ) will sound different than the same pair holding zero DC charge .

(iii) A Pair of caps holding a DC charge ( but not connected to the DC source ) will sound different after the DC source is connected .


(iv) The amount of DC voltage held within the cap pair represents another difference in sound .

(v) The big bad Solen signature sound of old / does disappear when they are wired up in series, & when those serial-pairs hold a DC charge .

<. :)

4313B
11-30-2007, 08:27 PM
They are not cheap by any means ($45.31 for a 16uf cap) but they are well worth it!Pick your poison...

Domino
12-01-2007, 01:29 AM
My mistake. They are the Auricaps. I may still try charge coupling in the future but for now I'm gonna stick with these.
The Solen values that I have are (2) 16uf with a .47uf across each. (2) 1uf, and (4) 1.5 uf.

4313B
12-01-2007, 04:59 AM
My mistake. They are the Auricaps. I may still try charge coupling in the future but for now I'm gonna stick with these.It's probably best that you just stick with what you have after spending that kind of money on capacitors.



For those who want to stick with the non-biased topologies you might want to know that both Greg at JBL and Jeff at Sonicap suggested using the RelCaps with JBL compression drivers back when we were discussing such things.

Here's yet another page on capacitors for all of you folks out there with those terrible sounding JBL's running bypassed capacitors for the past 28 years:

http://www.10audio.com/sonicap_oimp_multicap.htm

Oops, it appears the RelCaps aren't all that after all. Oh well. :dont-know

I am sure engineers have a very good reason, and a couple of them have tried to educate me on this subject. I respect the science and electrical theory on this subject, and my technical background helps me to understand it fairly well. But there is one small problem: the bypasses all sound bad! They add a quality that at first sounds like an increase in air and detail, but after a couple of hours becomes an intrusive harshness and discontinuity in the upper treble. Remove the bypass: all of the detail is present but without that grating and annoying sound. The high frequencies are cleaner, smoother, and much more enjoyable. It doesn't matter if the bypass cap is Teflon®, polystyrene, or common polypropylene, the results are very similar. And to be avoided.

It's no wonder why JBL's are so maligned by the GP and I would definitely suggest that everyone remove all those bypass capacitors from their stock JBL networks. It is very easy to just snip the leads to the bypasses without having to rip everything apart. Those folks with JBL's older than ~ 1980 are "safe" because those networks didn't use bypass capacitors. That's probably why the older JBL's sound so much better than anything after ~ 1980. For those with Everest II's and any of the K2 Series, any decent speaker repair shop should be able to build you a nice pair of normal networks without the biased and bypassed capacitors so you can start to enjoy your systems. :yes:



Better yet, just go totally active and play with the much smaller and cheaper capacitance values in the active filters instead. :thmbsup:

Chas
12-01-2007, 06:24 AM
I'm looking for some Solens. What values do you have? :o:

Todd, FYI:

http://www.solen.ca/v1/

Ian Mackenzie
12-01-2007, 09:42 AM
As also noted in the article there are no blanket recommendations.

On bypassing I think it depends on when the capacitors were made and what they are. ie mylar.

The older metalised mylar lacked high frequency detail and bypassing helps restore that. It was better than nothing in other words. With the more modern caps you need to be a lot more careful.

That guy used too large a bypass cap. 0.1 is way to large with 8.00uf so I would take that aspect of the article with a grain of salt...he really had no idea. A value 0.01 would have worked better.

But I have encountered the cheese grater effect using metalised polypropolyne to make up the exact value with a Hovland.

It really was horrible.

My resident favourite no holds barred capacitor is the Mundorf Silver Oil.

In the same network I have the Sonicap Gen 1 shunted on the horn network. I am not happy I accepted that recommendation. The point being that while in direct A/B you may guess two capacitors are close what goes on inside them can cause a conflict when used in the same filter, either bypassed or in another location of the filter.

You should always use the same capacitor through the network if you can. Particularly the mid and high frequency filters.

The Mundorf Silver/Oil is Over Kill in most systems unless you need to compare just how bad other capacitors sound.

My suggestion is what you dont know won't hurt you. Don't A/B and you will believe what ever you put in there will be the Bee's knees.

Conclusion: Of the three capacitors evaluated in this application, the Mundorf silver/oil was objectively the best performer. It provided the best clarity, resolution, and transparency. However, if I was doing this project without the agenda of evaluating different capacitors, I would use the Sonicap Gen I because the small improvements I heard when substituting the Mondorfs were really only apparent on direct comparison. The Sonicaps offered the same level of musical enjoyment for far lower cost.

Robh3606
12-01-2007, 10:32 AM
Why are you guys putting any stock in what's on the internet?? Who gives a damn what someone else thinks. Any Tom, Dick or Harry can put up a web page.

Build up a biased set for a small 2 way like the L20t's and use that to experiment with. When you find your favorite your done.



Those folks with JBL's older than ~ 1980 are "safe" because those networks didn't use bypass capacitors. That's probably why the older JBL's sound so much better than anything after ~ 1980. For those with Everest II's and any of the K2 Series, any decent speaker repair shop should be able to build you a nice pair of normal networks without the biased and bypassed capacitors so you can start to enjoy your systems. :yes:


LOL

Rob:D

speakerdave
12-01-2007, 10:46 AM
Unfortunately some people are not going to get the ironical sarcasm.

David

speakerdave
12-01-2007, 10:48 AM
. . . . Any Tom, Dick or Harry can put up a web page. . . .

Yeah--an off-the-shelf crossover maker who's being asked why his crossovers are not bypassed, for example.

David

toddalin
12-01-2007, 12:17 PM
Todd, FYI:

http://www.solen.ca/v1/

Thanks Chas. I usually buy them at Parts Express, but these are the values I need and figured I could save some $$$.

4313B
12-01-2007, 04:18 PM
Unfortunately some people are not going to get the ironical sarcasm.Some people evidently miss alot more than just ironical sarcasm. :D

MJC
12-04-2007, 08:17 AM
Build him a charge-coupled pair. Having built the 4344/4345 and 4355 charge-coupled networks and having listened to the difference I have to agree with G.T. - "These days I don't even bother listening to anything that isn't charge-coupled."

These days I don't even bother building a network that isn't charge-coupled; And those old systems get a new lease on life.
I would think even newer systems, like the PS, would be improved by replacing the stock xo with CC, which I plan to do, sometime in the future.

4313B
12-04-2007, 08:30 AM
Yep :yes:

Another forum member built a pair of networks outboard since they are pretty large to stuff in the stock PT enclosure.

http://audioheritage.org/vbulletin/showthread.php?t=9250

External filters can be quite nice.


Why are you guys putting any stock in what's on the internet??Because everyone knows that if it's on the Internet then it must be true. Why would anyone bother posting anything untrue? That would be totally weird. :screwy:

pioneer
04-17-2008, 11:21 AM
Gentlemen

I have a customer that frequents this site who has asked me to rebuild the crossovers for his L 300 speakers.

He is looking for better than stock preformance but using value engineering parts selection.

He has a brand new original pair of crossovers to compare against the soon to be rebuilt crossovers.

The original caps were mylar and he is leaning towards using polypropelyne caps.

I personally think this is the way to go for a slightly better than original sound quality at reasonable cost outlay.

What would you guys suggest?

And what would the sound quality differences be using the parts that each of you recommends,when compared to originals?

I know that everybody will have a unique perspective.I am just putting out some feelers at his request.

Best regards,Oldmics

Hi -
I see your schematic .... looks interesting. It looks/sounds like the caps and inductors are critical. Can you send me a parts list ?

Thanks, Jay

cooky1257
04-17-2008, 03:15 PM
So, if as described earlier in the thread(and to paraphrase) the cc'd network damps out some of the false dynamics present in some older drivers am I right in assuming cc'd passive is better than going fully active?
Cooky