New improved Universal ALK network Version 6

[Al Klappenberger]

Sometime ago I tried to do a price cut on my original universal "ALK"
network. I made Kimber Kaps the standard brand of capacitor for the two 2.2
uF caps in the squawker / tweeter crossover section of the network and
offered Hovland Musicaps as an option. Well, everybody wanted the Hovland
option! I have learned my lesson. It's obvious that what my customers want is
the finest design built with the finest quality parts. So here it is! All of
the Solen "FastCaps" have been replaced with Sonicaps. These are the only
caps I know, besides the Hovlands, that are +-5%. The Harmony bypass caps are
also gone. The woofer crossover has also been upgraded with a lower value
inductor of thicker wire (#12) with a slightly improved L/C ratio. The woofer
inductor goes down to 1.3 mHy and the 39+1 uF combination becomes two 24 uF
in parallel totaling 48 uF. This will give the bass a bit more "slam". It is
a very minor thing though. It's is about equivalent to the DCR reduction you
would get by removing the woofer fuse "twice". The Sonicaps are expensive
though, so the price will have to be adjusted upward to $575 for a set of
two.

I will be keeping the Solen and Hovland capacitor combination in the kit
version. I do expect to change the woofer filter to the new inductor and
provide 24 + 24 uFd Solens rather then the 39+1 uF combination. I may also
change the 6.2 uf + 1 uF bypass to something else. I am still thinking about
that! In the version I build, it has been changed to a single 7.5 uF Sonicap.
This is the least critical part in the network and a Kimber Kap with its +-
10% tolerance is acceptable there.

Al K.

[Deang]

"It's obvious that what my customers want is the finest design built with the finest quality parts."

Parts develop reputations in the DIY and audiophile communities, and most of those people are willing to pay a premium for those parts which have earned and garnered high marks. I've built your original ALK with both Solen at the primary and with Auricap, and all I can say is that the sound takes a good step forward when the Solen at the primary position is replaced with something of higher quality. The Sonicap has an outstanding reputation, and I'm sure it's not without good reason. There is a direct relationship between cost and performance, and it's easily discerned in the listening room. People unwilling to accept this have closed off their minds, and have denied themselves the best of listening experiences.

[cjgeraci]

"There is a direct relationship between cost and performance, and it's easily discerned in the listening room. People unwilling to accept this have closed off their minds, and have denied themselves the best of listening experiences."

Could not agree more. Congrats Al, the latest version looks and should sound great.

Carl.

[Daddy Dee]

Al,

Congrats! Hope you can post some pics in this thread.

[awsjr]

hi Al K.....

if cost is not an issue, what are your thoughts of replacing the solen

caps in the es500T networks, especially the the 18uF's , 16uF and the

33uF...also what about replacing the caps in the es5800 ?

thanks Al S

[Erik Mandaville]

"There is a direct relationship between cost and performance.."

Sometimes there is, if the person in question happens to like the difference. I've known VERY experienced builders who preferred the sound of the less expensive passive part -- not because it was cheaper, but they simpler liked the sound of the part.

I have not closed my mind, and have some experience under my belt.

Erik

[Al Klappenberger]

awsjr,

I do intend to put Sonicaps into the signal path of all my ES networks. I plan to keep the FastCaps for the woofer filters and for the "notch" caps that have inductors associated with them. There is just no point in spending big bucks on fancy caps that you are not actually listening "through". All the caps that what you are hearing is actually passing through obvioulsy call for top quality parts. Woofer filters are so low in frequency requiring cap values so large that you just can't get fancy ones big enough. The caps in the ES400 woofer filter is 91 uFd. The biggest Sonicap value is 30 ufd. To get the 48 uFd I need for the woofer filter in the original "ALK" required two 24 uF in parallel! Here's the drawing of the ES400 with Sonicaps. Notice I will have to make the board wider to get them on.

[Colin]

-Looks like beautiful work Al

-You are still on my wish list

-Sometimes improvements make components twice as good, but cost 10 times more

[Erik Mandaville]

"-Sometimes improvements make components twice as good, but cost 10 times more?"

That is true.

And sometimes improvements cost less.

Erik

[sfogg]

"There is a direct relationship between cost and performance, and it's easily discerned in the listening room."

Sorry Dean... but statements like the above are what the snake oil industry thrive on. (NOTE: Before anyone jumps down my throat I am *not* saying you are in that industry.... )

If I took a bunch of cheaply constructed electrolytics, slapped a fancy label on them, made up some big fancy web site with a bunch of BS stories about the pseudo-science behind them, fooled some suckers into giving good reviews on them and sold then for $700 each.... are they going to actually be better sounding capacitors?

The direct relationship which is 'easily discerened' sometimes only seems to be there when the listener also happens to know exactly which part is the more expensive one and which one they are listening to.

The far more interesting thing (to me anyway... not to those selling snake oil) is if the listener can 'easily discern' the difference when they can't let price, reputation, bias, and pre-conceived notions influence their listening. If you are ever interested in really trying this let me know and I'll lend you the equipment to listen this way...

That is not to say that the more expensive parts/component/etc may not in fact be better, in many cases they are. Just to say that that is not always the case.

Shawn

[Erik Mandaville]

Truly.

I'm not saying Dean is in that camp necessarily, either. I'm responding to the generalized aspect of the statement. More expensive is only better if the person paying for or implementing the upgrade think it to be so. If one were take the Teac 3 channel digital amplifier as an example, I know of many other components in that same power output range that cost very, very much more. Is the performance subsequently also that much better than the Teac? Who can decide other than the person doing the comparing -- with the fincancial investment out of the picture. There is a strong psychological component.

Erik

Erik

[Al Klappenberger]

Guys,

There is a direct relationship between cost and quality, GENERALLY. The relationship is sort-or expnential. It's called the law of deminishing returns! The more you spend, the less you get for your money as the price goes up. The Solen caps are unquestionably the best bang for the buck, but the Hovlands are the best I know of. In between are the Sonicaps. All the other brands are +-10% tolerance. That puts them out of the running in my book no matter how they sound. Kimber Kaps, for example, have a reputation of sound good, but just try to get sets of matched pairs within 1% like I do. You wind up with a lot of rejects that have to wait for the next batch to find one of matching value.

Al K.

[J.4knee]

Al,

Re: your V6 ALK's are you going to put the woofer ckt change on your web site. My networks are your kit versions and I have at least a passing interest in the woofer ckt change.

R/Jim

[Al Klappenberger]

J.4,

Yes, I will eventually update the download information. For now though, the change is simple. The 2.4 Mhy inductor becomes 1.3 mHy of #12 solid wire. It is nearly the same size as the old inductor. The 39+1 combination becomes two 24 uF caps in parallel. These can be Solen Fastcaps.

Al K.

[jwc]

Al.

How come you don't consider Solen's F & F capacitors that have a +/- 3% tolerance. Dean put those in my ALK B's and I thought the best part of that network is the tweeter.

From the website:

SOLEN FILM & FOIL CAPACITORS

Tin Foil Polypropylene Film Capacitors
Capacitance Values per voltage:
SN .010µF, 1200Vdc, Tol. ±3%
SM .022µF to 1.0µF, 630Vdc, Tol. ±3%
SE 1.5µF & 2.0µF, 400Vdc, Tol. ±3%
SY 2.2µF, 250Vdc, Tol. ±3%
SB 2.7µF, 3.3µF & 3.9µF, 150Vdc, Tol. ±3%
SA 4.7µF, 100Vdc, Tol. ±3%

jc

[Deang]

Look at the part of the filter Al modified -- you can't get F&F in values that large. Even if you could, who would pay for them? In the tweeter/squawker section, it's simply a choice. When you get to that level of quality, the differences are technically and audibly miniscule. The fact that Al has about 200 of those boards precut and drilled plays a big factor too -- the Hovlands lead lengths accomodate the layout.

[jwc]

Yes. I should have done a little better with my question. I was wondering about the use of the 2.2 uF instead of the Hovlands. I forgot about the size. The hovs are long and skinny. The Solen F & F are shorter and fatter.

[popbumper]

Al:

Some of the value changes you are suggesting correlate to those that you use in your Cornwall replacement circuit. I am wondering if then I should apply these changes in value ("2.4mH vs 1.3mH inductor", and "24uF +24uF" vs. "39uF + 1uF" caps) to the Cornwall circuit as well? Please advise, thanks!

Chris

[Al Klappenberger]

Jc,

Dean nailed it when he pointed out that the Solen film and foil caps are to short. The Hovlands have the proper leads. The boards are cut and drilled for me on a CNC machine and it costs to modify the layout. The other parts would have to be moved too. It's just not worth it. I considered the Solen F&F caps for use in the ES5800 network to replace the 2.2 uF FastCaps there but one of the cap values in it is not available from Solen. I can get that value in a Sonicap though.

Chris,

No, don't use the 1st order values in the 2nd order netwrok of my Cornwall "B" design. The modification to that network I posted a while ago already includes the thinking applied to this network. It's already been done!

AL K.

[Erik Mandaville]

"The fact that Al has about 200 of those boards precut and drilled plays a big factor too -- the Hovlands lead lengths accomodate the layout."

Sure, and that makes very good sense to me, and I would probably do the same. However, this is completely different from the basic assumption that there is a direct correlation between cost and performance.

Erik