• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Thiel crossover rehab

klamsoss

klamsoss

Be willing to fail and try again.
I'm about to rehab these Thiel CS2 crossovers. I'm experiencing what I think is rolloff above 16 or 17kHz although I don't have a way to measure. My comparisons are to B&W DM602.5 S3, wharfedale diamond 7.1, and B&W CDM1NT, all of which sound like they have more top end output.
I have verified that the tweeters are functional when directly connected(at very very low volume to prevent damage).
I have new caps, and resistors for ALL of what is in the circuits. The inductors won't be touched. Jim Thiel was a veritable genius and I don't want to modify his design.
My question is, do the orientations of the resistors and caps matter?
I have 33uf caps to run in parallel as replacements for the 66uf, and I have 6.2uf and 0.33uf caps for a parallel 6.5uf. I have 20ohm and 30ohm resistors for a series replacement of the 50ohm resistors. All of the new caps and resistors are Dayton audio/parts express, and the new caps are all film type except the 24uf, for those I have solen film caps. I even ordered 0.01uf bypass caps to replace the tiny silver 0.015uf polystyrene bypass caps seen in the pics. Obviously the new stuff wont just slide right into the original spots, but I dont want to screw anything up with my series/parallel arrangements on the rehab.
 
Register to hide this ad
If I understand your question correctly, the orientation of the resistors and capacitors doesn't matter because they're bi polar.
 
homebrew said:
If I understand your question correctly, the orientation of the resistors and capacitors doesn't matter because they're bi polar.
Click to expand...
Thanks for the reply, homebrew. I know the direction of the caps/resistors in the circuit doesn't matter unless the caps are polar, but I mean the physical orientation of them. I know physical orientation of inductors matters, but I don't know enough about speaker design (yet) to know if the orientation of the other parts actually matters. For example: If I were to physically relocate a capacitor or resistor to a different place on the board (while keeping it in the same position in the circuit), would that impact the function of the crossover?

I'm also unsure of if the board to which they are attached (glued? sticky tacky?) acts as a heat sink or not. My gut (and limited electronics education from engineering school 12 years ago) tells me "probably not, except maybe for the resistors", but I'm not 100% sure on that.

My plan was actually to tape the parallel caps together and get them as close to the original positions as possible, and similarly for the resistors, wrap one in electrical tape and then tape them together (still wiring them in series for equivalent resistance). That's the best way I can figure to keep things as close to the original design as possible, but would still necessarily change the physical position (not the circuit) of the parts.
 
Well, there isn't any interaction with capacitors like there is with inductors and the way they're positioned. I would recommend resistors be carefully positioned, not touching the circuit board or another resistor, because resistors generate heat. It's not uncommon to see over heated and burned resistors.
 
Homebrew is correct, but I'm not convinced of the OP's initial diagnosis of the problem with the sound. I doubt that a roll-off above 17kHz would be perceived as dullness. The part of the audio spectrum which would give the impression of brightness or dullness is much lower in frequency in my experience, maybe around 5kHz.
 
TudorTurtle, thanks for the heads-up on the app. I hadn't even thought of the fact that there probably IS an app for that. Do you recommend any particular one?

Rothwell and Homebrew, I'm 32, and don't have a history of going to a ton of concerts or working excessively loud jobs. My dad (approaching 70) definitely can't hear what I'm talking about.

I'd welcome input on possible alternative diagnosis for the issue!

I can say that I don't think it's a room, or upstream issue. I've tried these with 3 different amplifiers, multiple sources and in 3 different rooms, all with similar results. I would describe the sound "issue" as one that is apparent with the 'shimmer' of cymbals in rock and jazz, it cuts out a little of the 'hall' or "room" sound for recordings done in churches, symphony halls, etc. On the up-side, it also eliminates sibilance on some recordings that I have that are particularly heavy on that side of the spectrum. Here are some notes from a recent comparison done using a Rotel speaker selector. The only adjustment made was to slightly increase output for the less-sensitive Thiels. The speakers in question are B&W DM 303 bookshelf, Polk TSi300 floorstanders, and the Thiel CS2 :
______
With Alison Krauss and Union Station (multiple songs, from album New Favorite), the Thiels were CLEARLY the best speaker in every category. Voices were more natural, the timing was superb, and everything seemed to be there in the right amounts. The bass was proportional but hefty, and the ability to reproduce each voice or instrument very well, regardless of if there are zero other voices, or MANY other voices was impressive. In this area, especially, the polks were inferior.
The polks were also harsh, comparatively. Sibilance was exaggerated in an unpleasant way. Twang of strings, cymbal resonance and hard consonant sounds all blended to just become... noise...
The b&w were a bit better than the polks when it came to sibilance, but still slightly exaggerated, which makes me think that in part, at least, that is a characteristic of this recording. Their timing was better, but not on par with the thiels, and they couldn't compete with either floorstander in terms of impact.

The B&W had more shimmer than the Thiels with Stan Getz (Samba Triste). There was more breathiness in the sax, and more brush and shimmer in the percussion section. Not that the Thiels were bad, they just weren't as engaging and didn't sound as 'live'. The Polks, however, exaggerated the breathiness of the sax, so much so that it became harsh, almost sandpaper-like... moving from what I would describe as a "sound texture" into what I can't really call anything other than "noise"

With Sam McClain (I'm So Tired), the b&w really did very well outside of the bookshelf bass roll-off, [...] and had more top end than the Thiels (I speculate >17.5khz) which added a sense of space. Similarly with the polks, I heard more of 'the room', more 'space', and more of the after-tone from the ride cymbal... more of the "ring" rather than only the "ping" of the ride. Again, the Thiels sound staging is clearly better and the timing is flawless but they lack almost all 'room' sound/ambiance and what little splash/sizzle there is on the cymbals in this recording was almost completely gone.

With Richard Thompson (The Way That it Shows) the polks put him up front and compliment his voice exceptionally well. I didn't notice the same harshness, but did hear some of the timing issues. The Thiels made his voice sound distant and slightly weak, which is problematic. Again, their timing was unparalleled and staging was superior. The B&Ws sounded quite good but it was immediately obvious that the little bookshelves couldn't play the same game as the floorstanders.
_______

Maybe that helps convey/illustrate the sound situation I am experiencing with these speakers, maybe not. It sounds like I should get on it with some sort of frequency response app to really identify if there is a measurable roll-off.

Thanks for all of your help! I welcome further input!
 
You could possibly insert some kind of parametric EQ or graphic equalizer - either hardware or software - into the system temporarily and play about with the settings to find out which region of the spectrum needs a boost. Of course, the existing crossover will determine what can be done with a radical re-design.
 
Rothwell, That's actually how I arrived at the conclusion that it was above 17.5kHz that the rolloff occurred. I have a digital eq on my computer and when I cranked up the 16kHz band, it didn't fill out the "room" on hall-recorded music, or the "sizzle/wash" for cymbals.I should fool around with it more, though to make sure it's not an issue of harmonic destructive interference. (though I find it hard to believe that this is the first time in my life I am hearing recordings played back "the right way")
 
You must log in or register to reply here.
Back
Top Bottom