|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
207.200.116.196
I have a technical question. I just bypassed the output caps, 4.0uf, in my preamp with a .1uf cap and the sound was very tilted, or shelved., if you will. I believe this to be a symptom of some type of phase anomaly. I took out the .1uf caps and then bypassed with a .01uf value and the frequency was much more even and balanced. Can someone explain this phenomena?
Follow Ups:
Instead of bypassing the cap, change it with a different value. I think that 4.0uF is an awfully large size for a coupling cap. Especially since you are driving a tube power amp which probably has an input impedance of 100K or more.If it were my preamp, I would replace the 4.0uF with a 0.47uF paper in oil cap. You can get a very nice copper film paper in oil coupling cap in the 0.47 value without breaking the bank. It will not need a bypass either.
The smaller size capacitor should be plenty big for deep bass, and also small enough to be quick and articulate. Just make sure that you have the space inside for a paper in oil coupling cap. They can get large and heavy.
Based on my calculations, a .47uf cap into my 100K ohm monoblocs would put the -3db cut-off at 34 Hz. Isn't that a little high? Some have said that too high a cut-off can cause "phase anomalies".
1/6.28RC where r is ohms and c is Faradsso that's 1 divided by the total of (6.28 x 100,000 x .00000047)
or 1 divided by (6.28 x .047)
or 1 divided by 0.29516
or 3.38HzI think I got that right. So, if you used a .047uF cap it would be 34Hz, a .47uF cap would be 3.4Hz, and 4.7uF would be 0.34Hz.
When you say "too high a cut-off can cause "phase anomalies" I honestly don't know what you are referring to. Maybe someone else can explain this to us. I do know that all capacitors and inductors induce a 90 degree phase shift by nature. I don't think there is anything you can do to change that.
Thanks for the calculations. I guess I put the decimal point in the wrong location. As to the "phase anomalies", I'm not sure really what that means, only what certain people have told me when placing the cut-off point too high.
Ultimately bypassing of the signal caps, which I ve tried many times in my pre, never produced long time musical involvement. Also my friends experiment with larger values Hovland signal capacitors 2-4uF and in the SRPP pre 2uF-600v fit perfectly as a buffer, while I am very happy with my 0.47uF-600v. So maybe just like in case of German WIMA MKP or ROE metalized poly capacitors caps below 1uF with 600v rating made magic in the signal path. Larger value 100-200v Hovland capacitors in speaker’s application also tend to sound a little too heavy and burled.
Maybe you should try once again Hovland but this time 0.47uF-600v particularly if yours original 4uF Hovland capacitors were 400v.Be careful this is just my personal experience and opinion; therefore fell free to experiment with smaller cap values in buffer of your pre without bypass.
for your suggestions. Actually, the 4.7uf Hovlands are 100V.
Thanks for the recommendation. Yes, I am using VAC monoblocks amps with 100K ohms input impedance. My tube preamp has an output impedance of 600ohms. I have tried to read through all the threads pertaining to what would be the perfect value output cap, but after reading so many differing opinions, it has become frustrating for me. I have read that the output cap should be as small as possible without going too small to cause phase problems. Some say to find a cap value that gives you a -3db roll-off at at least 5 hz, regardless of where your speakers roll off.
Take a look at the coupling capacitor values in other equipment if you can, especially tube gear. Look at the caps that go between tube stages in an amplifier, or the output of a DAC or preamp. Most coupling caps are usually around the 0.22 to 0.47uF range. That's it. There's a good reason for that too. It's usually enough capacitance. And what about transformers? Transformers don't go down into the 5HZ world. Yet you have trannies coupling the power to your speakers. So do I. I wouldn't worry about it.I think the only reason a manufacturer would use a 4.0uF cap is to guarantee ultra low bass performance into a VERY low impedance power amp. Probably a pro-audio situation or a powered speaker. You are miles away from that with your VAC.
Even then, do you really need to amplify 5Hz? 15Hz?? Not really. You're probably better off rolling the lows off somewhere in the 20Hz region anyway. Your goal is to hear music and not subteranian artifacts. Your amp and speakers will be happier that way too.
Just remember to get the correct VOLTAGE value when you replace a capacitor. You can use a higher voltage rated replacement cap, but never lower.
You can also experiment with an inexpensive cap first just to see if the bass suddenly disapears.
My first modification of pre was substitution of generic 2uF buffer cap with 0.47uF Hovland musicap without any loss in bass and with great results. My advice is to try 0.47uF V-CAP teflon or realcap TFT. I believe that Quiet Earth catch the point, smaller non oil teflons are just great. Currently I used 0.1uF realcap TFT - original value, as a coupling cap, in my pre with mentioned 0.47uF Hovland musicap in the buffer position. Old configuration used custom made tin-foil styroflex but realcap teflon was much better. Maybe you are not a fan of oil cap just like me.
My preference for coupling caps is the Audio Note copper foil paper in oil cap. The silver foil PIO cap is outrageously good, but you have to be careful where you use it. It's a question of balance. And break in.
I'm sure that the teflons that you mentioned are very nice too. It's just a matter of personal preference. But either way, 0.47uF seems to be the magic number for most situations.
Milos, can you tell me what the input impedance of your amplifier is? Is it a tube amp?
My power amp is solid state with input impedance of 50K. Also very important detail is cathode follower - ecc82 in my pre, hence the output impedance is below 150 ohm. Also my pre fit well my friend’s 20w Hiraga power amp with 22K impedance.What pre, power amp and speakers you have?
Regards
Milos
Milos, I just did a calculation and based on the 50K ohms input impedance on your amp, using a .47uf preamp output cap would give you a -3db cut-off of 67.7 Hz. Doesn't that seem a little high for your Heybrooks?
Hi Sherod, thank you did you take into account the low output impedance of my pre - 100 ohm. Please post your calculation
Milos,
No, I didn't take into account your low 100 ohms output. I'm not sure what formula to use to take that into account. My preamp output is 600 ohms with amp input being 100K ohms. Could you post your method for calculating this -3db cut-off? I would appreciate it. I am using a chart provided me by Chris at Vh Audio who has the V-cap.
I am also not sure what formula to use but 100 ohms output is also important. Where did you find chart by Chris from Vh.
Milos, here is a recent thread I started about preamp output cap value. The next post gives the formula which is used in the chart that Chris sent me. I e-mailed you a copy of the chart, but I'm not sure if you received it.
Thank you Sherod I forget that I have 4uF input capacitors in my power amp, therefore total was 4+0.47....4.47uF.
Milos,
I'm not sure if this is the same thing. If so, then we have to start over and go back to my original post where I currently am using a 4.0uf cap in my preamp output. Thanks for all your input, though.
My preamp is the Audio Horizons TP 2.o. It has Siemens CCa(E88CC) in outputs. It is rated at 600 ohms output. My power amps are VAC Musicbloc monoblocks, rated 70W RMS, push-pull, KT-88 outputs and E82CC driver and splitter. Input impedance is 100K ohms. My speakers are Reference 3A Dulcet monitors.
My Heybrook HB-100 speakers also have main driver which is directly coupled to amplifier with simple high pass filter on the treble unit with only two capacitors. My Heybrook is little bigger, frequency response is 40Hz to 20 KHz, while your Reference 3A have 48 Hz to 20 KHz.
Also I believe that your speakers, because of mentioned direct coupled main driver, are very neutral, direct and transparent, so you will hear everything. I have also very hard time to make my speakers sound excellent and finally with very sophisticated pre and power amp they have some real magic in terms of speed, openness and dynamic.
Calculated cut-off at with 0.47 capacitors at 34 Hz, in the smaller room, with your speakers could bring some benefit because of yours speakers natural limitation in the deep bass.
From my personal experience I believe that you will have articulate bass without any problems after modification with V-CAP or Realcap TFT 0.47, 600V buffer capacitors.
Milos,
Thank you for your observations. Yes, the Dulcet speakers are very transparent with their simple one cap crossover as well as being time and phase aligned. They are extremely revealing of everything in the chain. I will try the .47uf V-cap teflon. If it doesn't work out, I can maybe use this value in another location. I appreciate your consideration.
Ok Sherod,
I will be very happy if 0.47 made magic. I ve never tried V-cap but Realcap TFT after 400 hours in my pre start to sound glorious.
Please post your experience with 0.47.P.S. Extremely revealing speakers like ours are great but painful.
Milos,
I will be glad to post my impressions of .47uf cap. Yes, our speakers are very difficult to extract that "magic", but once achieved, it is well worth it. I fully agree with you.
All caps ('lytic that is) have a resonant frequency at which they electrically drop out of curcuit. In order to prevent this the 'lytic should be bypassed using a small NON-INDUCTIVE cap such as Silver Mica, stacked film, or ceramic in descending order of performance. This bypass cap then shunts the very high frequencys to ground, preventing the 'lytic from reaching it's resonant frequency. In my experience a bypass cap much exceeding 0.01 uF begins to impinge on the audio frequencys with detremental affect. Hosfelt Electronics (www.hosfelt.com) is an excellent source for SM's at very reasonable prices.
Quite a number of the esoteric audiophile caps tend to alter the sound detrimentally when bypassed. You may just hearing just too much of a good thing.I would suggest that you make an inquiry of the manufacturer as to whether the cap is designed for and should be bypassed.
DaveT
Hi Dave,
The manufacturer doesn't recommend bypassing this 4.0uf preamp output cap as he says that will "brighten" the sound too much. The cap he puts there is a Hovland Musicap, a very musical cap, but IMHO, not the most revealing in the frequency extremes. I initially experimented with replacing it with an equal value V-cap OIMP( oil impregnated) and after break-in of the cap, there is more musical information being passed, but the overall sound is a tad too warm and rounded. I was hoping that bypassing with an equally good teflon would balance the sound by taking the extra warmth and roundness(softenness) away and adding the proper neutrality and natural edge that my ears are craving.
I use a 1uF VH OIMP in my prepre and found the sound also a tad too warm.So I bought the teflon V-cap of .01uF for bypassing the OIMP hoping to get a better balanced soud especially in the frequency extremes.I have about 100 hours on this and feel that I can hear a bit more air in the highs.Hope things will still get better with more time on them.
Ewe,
I have read that the V-cap Teflon takes up to 200 hours before it actually shows its sonic capabilities. Please keep me posted on its progress.
One problem with bypassing caps with smaller units is that the inductances may interact with the capacitances in the other units.Every capacitor also has some sonic character. If your 0.1 uF caps sound bad by themselves, the combination with the existing 4.0 uF caps would likely also sound bad.
Some caps are made with steel parts, for example. Do your 0.1 uF caps show attraction to a strong magnet?
Al, sorry I didn't mention the brand of caps I'm currently experimenting with. The larger 4.0uf cap is the V-cap OIMP (oil impregnated metallized polyproylene) and the bypass caps are V-cap TFTF Teflons.
A good rule of thumb is to have at least a 100 to 1 ratio between cap values. Why? Because the self resonance of the cap, related to value, changes as the square root of the value, so 100 to 1 in value is 10 to one in resonance range. Less separation means interaction between the caps. It can be shown with a computer simulation.
Sorry to get a bit more complicated .
What is the situation when you have parallel same sized caps..ie 2 x 100 Mf + 200 Mf ? or more?...do you bypass with 1 mf or 2 mf?.etc.
I can even see a reasons for going lower than the 1Mf or have I got it wrong...You can go higher that 2MF.
There is no HARD rule. You just should not put the values too close together, because they most often will interact. 100;1 is a good start, but in your case, maybe .1uF is better.
...(and, of course, lots lower than that) when bypassing their 'styrene coupling caps. I recall they love the sound of 0.15µF bypasses on values as small as 1s and as large as 4s.Don't know if they're doing that with/to their Teflon-film caps.
Probably every combination (of brands, types, and values) has a different sound, so what works for A plus B doesn't work as well for X plus Y. Sure am glad I'm no GEA! :-)
-------------------------------------------------------
Tin-eared audiofool and obsessed landscape fotografer.
http://community.webshots.com/user/jeffreybehr
Hi and thanks for explanation.
Does this also go someway towards answering my previous thread re bypassing at 50 to 1 (half decades etc) which sonivally caused some dullness at certain frequencies and or possibly highlighted others.
Based on my experience, I have heard what you are talking about when using bypasses that fall under the 100 rule. Some call it "phase anomalies". I'd like to see in a computer graph what this looks like, but nonetheless, I can still hear it.
This post is made possible by the generous support of people like you and our sponsors: