The links you put here are in another language and I can't read it! What I can see is that these CDE caps are not the same caps as Tony Gee talks about! These are the 950C series and the ones on Tony's page are the 940C
If you can, ask your friend what kind of phase shifts (elec. vs acous.) was he referring to so the rest of us can be enlightened.
OK, Thanks! I thought that my Google Chrome browser would have translated the page, but it didn't. I'm not sure if I can purchase the 950's here in the states. I guess I'll have to search around some to find them.
So the CDE 950's were the ones you used with the Vishay MKP 1837 .01uF
caps for bypass? You mentioned getting similar sound from these CDE as compared to the Mundorf Silver/Gold caps. I am assuming these CDE caps were in a speaker network when comparing?
I actually used Wima 0.01uF, 63V that seemed like a similar structure, because those were conveniently available. If you would like to know the exact model, I would need to check it up. Yes, they were in a speaker network in parallel with a resistor but in series with the live line used for some equalization.
I found some CDE caps here: Great deals in cde950 from World Famous Electronic Component | eBay Stores
Do you believe these are genuine OEM?
Do you believe these are genuine OEM?
It is really hard to tell, as mentioned before, when an OEM delivered some samples, they sounded different. I had to prove they were different with data. General application only requires that the published specs are met.
The 0.01uF parallel concept also works with signal level when you have large capacitors with small form factor. The difference is really small, but if you have good downstream equipment, you will hear it. The parallel cap concept I have first seen was in my Audible Illusions Modulus (probably IIC). Those days 1uF RelCap was the thing to use.
PS. Wasn't expecting PM, but cleared a few spots.
The 0.01uF parallel concept also works with signal level when you have large capacitors with small form factor. The difference is really small, but if you have good downstream equipment, you will hear it. The parallel cap concept I have first seen was in my Audible Illusions Modulus (probably IIC). Those days 1uF RelCap was the thing to use.
PS. Wasn't expecting PM, but cleared a few spots.
I see, OK thanks again! Perhaps I will try and contact CDE and find out where I might purchase the 950C's in the USA. I was thinking on getting the larger values I need, the try the 940C's in a .1uF. I can get the 940C's up to 3KV. I'm not sure if the higher voltage caps would be better or not?
Your mailbox is full again...lol 
You mentioned that using similar caps paralleled together is not a good thing. So lets say I need a value of 10.0uF in the tweeter circuit that is in series to the tweeter, wouldn't using (2) 5.0uF caps have better effects if all the specs on them were better than the 10.0uF?
You mentioned that using similar caps paralleled together is not a good thing. So lets say I need a value of 10.0uF in the tweeter circuit that is in series to the tweeter, wouldn't using (2) 5.0uF caps have better effects if all the specs on them were better than the 10.0uF?
It really depends on the capacitor design structure and material. Also how close the actual measured values are. Some people do not like using parallel capacitors because they experience smeared sound image focus. But I guess you have to find your own experience. I just try to avoid by default in the signal path, but it used in parallel with drivers, I generally am willing to try it.
The manufacturer also sent me another cap which they claim is also one on Tony's list, the structure consisted of mixed material made
specifically per customer requirement, it did not sound as clean as I like, and the roll was not as firm as we normally expect.
One reason I do not put the ordinary spec numbers as a high priority is that most do not relate directly with the sound quality we hear.
Sent from my iPhone using Tapatalk
The manufacturer also sent me another cap which they claim is also one on Tony's list, the structure consisted of mixed material made
specifically per customer requirement, it did not sound as clean as I like, and the roll was not as firm as we normally expect.
One reason I do not put the ordinary spec numbers as a high priority is that most do not relate directly with the sound quality we hear.
Sent from my iPhone using Tapatalk
Hi Carl, This is what my friend said:
Hi, The amount of phase shift in a capacitor is related to the frequency. If the frequency is considerably above where the capacitors impedance starts to rise then it is minimal. For audio that would likely be about 3 octaves. At the point you are most interested in (the actual crossover point) it will be significant. I'll use some numbers to illustrate. Say you want a simple 6 db up and 6 db down crossover....a typical first order one at 1000HZ. We can skip the low side as it will be inductive mostly. To get that crossover point will take a 20uf cap. It will have a phase shift of 90 degrees at that point. (the math will boggle you). So if you mix a 19 and a 1 uf to get it you now have two different 90 degree phase shift points one for the 19 (at about 1052) and one at 5,000. The simple formula for finding both the -6 db point and where the 90 degree shift will be is 0.16/C X R. C is in uF and R in ohms. R is the speaker impedance at that point (it assumes pure resistance) . So in effect you have two crossovers in parallel. The combination will do the 6 db part for power OK, but will have two phase shift points. Are they audible is the question. This has been thrashed many times and I'm not sure there is a definitive answer. I prefer to use a single cap to avoid the issue. My sense is that this can be a problem with other filters as well. In diy circuits particularly ones that need an odd value of capacitor the typical thought is to parallel as many smaller ones as needed to get it. I question if the same issue occurs. Premium gear uses only one cap for each such application. One area where it doesn't matter is in power filtering for B+. This is a bulk power application and phase is of no consequence. Actually paralleling several same size caps to get the value may be advantageous as it will usually have the effect of lowering the ESR. Kind of like paralleling the internal resistances. (not an exact analogy).
Electrical phase for a good quality audio capacitor is pretty much constant at -90 degrees from 10 to 20,000 hz. I can only assume fshz42's friend is referring to acoustic phase which is very complex in a 2-way speaker depending on the xfer function, driver spacing, etc.
IMO, the two crossover points that are a part of a bypassed cap are there by virtue of the cap values and I believe the sound benefits derived outweigh phase concerns which, the friend admits is still a topic of debate. If that's the case, then it's probably not significant and worth getting an ulcer over.
IMO, the two crossover points that are a part of a bypassed cap are there by virtue of the cap values and I believe the sound benefits derived outweigh phase concerns which, the friend admits is still a topic of debate. If that's the case, then it's probably not significant and worth getting an ulcer over.
Where you use a capacitor in an RC filter to create a single pole roll-off, you will find that the phase difference at F-3dB is 45degrees.
and the ultimate phase difference at many decades away from F-3dB is 90degrees.
But it's taken many posts from post306 for the question to be clarified to:
What is the phase of a single pole RC filter?
and the ultimate phase difference at many decades away from F-3dB is 90degrees.
But it's taken many posts from post306 for the question to be clarified to:
What is the phase of a single pole RC filter?
Last edited:
Quick question, Would these Vishay caps be the same as the ones listed on Tony Gee's page? These are rated at higher voltage at 160VDC, where the ones on Tony's site use the 100VDC. I would imagine they would be equal?
http://www.mouser.com/ProductDetail...yZMbqb1O1yeZE7T8VOLz1V6YwPVQD7Dm/XOYA2OCT/Q==
http://www.mouser.com/ProductDetail...yZMbqb1O1yeZE7T8VOLz1V6YwPVQD7Dm/XOYA2OCT/Q==
If it doesn't cost a lot more. Otherwise, the 160V cap will do the bypass job nicely because it passes only a small fraction of the total voltage going to the tweeter.
- Status
- Not open for further replies.