Higher value caps question

This is part of the reason I don't tell folks to go swapping parts willy-nilly in well designed and working products. Also, "recommended" and "trustable" have no bearing on the outcome of modding, especially when money or ego are involved. Design quality, part quality and physics have more to do with it.

Caps are not linear devices and because I am a designer, I can tell you first hand that the people who design these products do their prototyping with the parts that you see on the board and then qualify and QC all with those parts.  In other words, they make their design work as well as it can with those parts.  If they felt like they needed to hit a higher goal, they would have made a different design with different parts.  Some of the most loved and highest quality designs have parts that would make modders sick to their stomachs if they knew were being used.  Just food for thought.

Let me give you 4 factoids:

1. Wima caps have some sort of great reputation in audio modding circles, but it's not founded. Wima caps have a hit/miss rep in the design world, and I would trust the panasonic caps over Wima every single day of the week. The lower end wimas (the ones that modders tend to use due to price) are especially poor. Honestly I can't figure out why folks like them so much, as there are much better options out there for a lot cheaper.  The name is easy to remember I guess.

2. Ceramic(NPO) and polypropylene(PP), PPS and polyester(PET) caps have roughly the same dielectric absorption rate depending on the size and quality of the dielectric material, so the main reason most modders give for changing out electrolytic caps for film caps doesn't have much credence for the most part. Most differences that are heard are placebo effect. Many, many studies have shown this but modders and hopeful folks alike tend to ignore what they don't want to hear and only focus on making money or getting a better product for less money.

3. The difference in PCB trace impedance from board to board is usually more than cap ESR differences unless you are using tight "impedance controlled" PCBs, which cost about double what a normal cheap PCB does.  Going from electrolytic to film/ceramic WILL have a difference, but going from ceramic to film will have very negligible difference in most places.  Most audio designs are low enough impedance that the designer doesn't really care about PCB trace impedances.  ESR has become the new buzzword for modders because lots of folks have become wise to the fact that dielectric absorption is outrageously overblown as a problem.

4. (the dirty secret) Electrolytics actually have more flat ESR across frequency than the other types do.  If the designer accounted for that, then the electrolytic cap is the flattest frequency option you can use.  WHAT'S THAT??  Yep, I said it, the electrolytic is the best option in most cases due to the flatness of it's ESR.  I always choose bipolar electrolytics for most of my personal audio work, although people blast me all the time because it's not part of the modding cabal handbook.

That being said, if the caps were X7R dielectric or of poor quality, then moving up to NPO ceramic or PP/PTFE could be warranted, especially in EQ circuits, however with the impedances involved and the audio frequencies involved, the difference is minuscule at best.

So on to your issue. You changed to lower ESR caps, which should have a decreasing series resistance as the frequency increases. In other words, they are non linear high pass filters, so it would make sense that you are possibly hearing more highs, rather than less lows.  As in #4 above, you might have created a very gently sloping HPF depending on the I/O impedances of the audio circuit.  this is why the majority of people who do this think they are getting more "detail" out of the caps, when in fact they are just getting a boosted high end.

Larger values mean lower cutoff points.. But you knew that already.. Did you know that typically, a higher value cap in the same size package intrinsically has lower ESR already?  that means that you probably just cut the ESR of the cap at higher frequency by double or more.  It's still fractions of an Ohm, so it's likely less than 1/10th of a dB of difference, but add that up over a number of caps and you could hear a slight difference.

Did you know that the ESR of NPO/PP caps is so low that most designers have more issues with the PCB material permittivity and trace impedances rather than cap ESR?

That's right, the layout matters more than the caps, unless it's a poor design and then caps matter just as much. Changing caps in the audio path, might actually make things worse in some cases where the design is marginal. Remember, the design was done with the parts that were present on the board. Change those parts and who knows what will happen.

Anyway, you also didn't reference a schematic, nor explain the circuit at all, so it leaves me guessing.

I'll get the schematic and mark which caps I changed and to what ESR. That's really insane..... I'm taking classes to be an EE, but i'm still in the first 2 years of general ed (wish I would had got my AA when I was young) so I havn't learned some of these things. I had no idea by changing up the caps to a lower ESR would make the a hpf.... here's what I know, the entire board was populated with 10uf saxmon caps, all of those I switched to 22uf elna similic's. I don't think that was my problem, I think what happened was when I strayed off he path of advice and swapped all the caps out in the amp control board with the same value, but drastically lower esr values. 

I'll post both schematics so you can see what I mean. Man I was hoping you would respond to this thread, I'm hoping you might be able to help me get some of the bottom back. 

Here you go svart

www.dropbox.com/s/g3865cyvf463thr/HS80M_HS50M_Schematics.pdf

Quick side question...
Did you measure all caps before soldering and do both monitors have the same issue?
Just to be sure we get on the right track?

I want to say, that there is still lowend, but it's been decreased from where it was when they were stock. I will say that since I've changed the caps the high/high mid's are incredibly better, before they are almost not listenable because of some sort of high - high/mid hash sound that I couldn't stand. I honestly believe svart is correct that somehow the wima's are causing this issue. I've only replaced three of the .01 values on the input board.

I replaced them all to begin with, and I had a complete loss of low freq's. I removed a couple, specfically where LF is noted on the PCB, and I got lowend back. But, there is something still not right, and it seems like the HPF has been altered in some way by these types of wima's. I'm not sure if it's the wrong material of dielectric in these particular wima's and when I was given the suggested part he just went for the first .01 film wima that popped up or what, but something is not right, and I believe it's to do with the film wima's. I went up to radio shack and bought a few metal film caps of the same value, if you guys think it could be in the wima's I can start to replace them one by one until it feels like it's coming back. The input board takes me about 3 seconds to get out and getting the caps out is a quick process too.

Before I do that, I want to see if you guys feel it actually could be them causing the issue before I take the monitors down, that's really the only hard part, getting them off the stands haha.

Thanks.

If you read svarts hints about caps, well, you see, that it might make a very marginal difference if any to replace ceramics with film caps. Even in EQ circuits it is normally no functional difference but just a tad of taste if you like film caps better in the trebles or ceramics better in the mids.
So my guess is you did replace ceramics with wrong value films...first idea see above...

D'oh, i didn't even see the (MY), oops. Yes, makes no sense.

Welp..... I stand corrected, I did replace c140/c141 with the wima's..... 

You might have upped the lowcut freq due to lowering the value from 0.1uF to......?
0.01uF?

Yeah that's what I did.....I got some metal film .1 caps and it fixed the problem..... I feel like a dummy, but I'm not sure where the hell I was supposed to put these .01's at.... everything that is .01 looks to be SMD's.....

I completely took out the Wima's and replaced them with .1 metal film polyester caps I got from radio shack. Now that I know the exact values of what I need I'll order them from Mouser. I'd still love to know where the .01's were supposed to go..... the only values that calls for .01 is SMD's on the other side of PCB.

They've definitely got their lowend back now, and it's a lot tighter as well. The high end problem that these are known for stock is gone and they sound quite nice now. I've got some opa2134's that I want to replace but the PCB is really weak so I'm going to send the input boards to someone that can replace them without puling the traces off. Just replacing the electrolytics there were a couple that lifted a little and I had to go in and use a trace pen to fix it. That was with my iron at like 680 too.

No no, I didn't replace them all. I only replaced c140,c141, and c146 with the wimas to begin with. So those are what I replaced last night with 100nf polyester metal film.

The problem with this board was they bent the legs of all the caps over, so once you start to remove it you have to stay over it for a little bit, I had someone tell me to turn the heat down pretty low so that I could work the part out without messing anything else up around it.