A conflict with ALK schematic

[Alexander]

I have noticed that in the ALK  Forte II schematic there is a conflict. The schematic lists a 36uf cap in the woofer section but if you look at the pic of his finished xo a 33uf cap is being used. All other values look to be the same otherwise. Could someone give us some insight as to why this might have been changed?  

[Wirrunna]

My advice is to email ALK and ask him.

 

Standard cap values are 33uF then 39uf.

36uf is an odd value. ALK probably had a 33uf on hand and used that for the first build.

[djk]

36 is not a standard value.

 

The original network use non-polar electrolytics, 10% types.

 

33 in a close tolerance film type is close enough.

[Wirrunna]

I just modeled the bass network in LT-Spice using both 33uf and 36uf and the -3db point changes from 692Hz to 684Hz respectively.

 

 

[JohnA]

Close enough. 

[Deang]

Within 10% of stated value in a non- critical position. 

 

Al uses 33uF for the Cornwall variant of the CornScala-wall Universal. Why use 36uF when you have a half dozen 33uF rolling around your bench. 

[USNRET]

Pardon please using the 10% deviation allowed in this post (forget tolerance stack up of the other components)
Required spec is 36uF = 32.4 to 39.6uF
by substituting a 33uF=29.7 to 36.3uF
the substituted 33uf falls outside of the specification requirement IF it deviates to the maximum amount (or to some extent lessor) allowed on the low side

I know jack about this but I do know blue print specs and tolerances and this caught my eye. You can relax the 10% deviation if you want to but you have to red line the drawings. Yea, ya'll are talking sound but this substitution doesn't get installed on my airplane.

[mark1101]

When you build these things you could talk to the supplier and make sure they hand match the caps to the exact values you need (and write the values on the caps).  I did this with the sets of ESNs I built and this eliminates the tolerances altogether.

 

Most of the time they will do this for asking, especially on an expensive purchase.  In order to get a 33uF cap they might send you a 36 with 33.1 written on it.  I have a few caps in situations like this but the values are what matters.

 

It's funny how some will say "it's OK"........10% off is acceptable.......Well it might be OK.  But it is unacceptable to me.  Your networks will not be identical...........In my book 10% is WAY OFF.  If you are doing this, why not do it right?  It's not a bit more difficult.

 

Why does one speaker sound slightly different than the other?  Well I don't know..........but at my house it's definitely not because the networks are not identical.

 

 

[Deang]

If both sides are using the same value, and those values are within 1-2% of each other - they will sound the same.

 

On a low order filter, 10% variance is completely acceptable.

 

Any difference you hear related to FR, will be because of your room, not because the capacitors aren't perfectly matched. 

[CECAA850]

11 hours ago, USNRET said:

I know jack about this

Me too.  I believe both boards will have the same cap which will be within 10% of the plans but identical to each other, at least that's how I interpreted it.

[JohnA]

On 6/27/2017 at 7:23 PM, Wirrunna said:

I just modeled the bass network in LT-Spice using both 33uf and 36uf and the -3db point changes from 692Hz to 684Hz respectively.

 

 

 

If this is accurate, THIS is why a 33 is plenty good enough in this place.  It makes a 1.17% difference.  A nothing burger.