 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
206.228.220.38
In Reply to: RE: My Adventures with Ferrite Clamps posted by rick_m on January 19, 2011 at 15:13:47
Thanks. I have one 20 amp dedicated line feeding my entire system, with 12/2 Romex wire. Two cryoed Hubble duplex outlets are fed by the single dedicated line.The Supratek preamp results were surprising to me - I didn't expect much but it was quite good actually. I was able to eliminate most of the tube hiss via filtration of the filament supplies. The preamp is dead quiet even with my ears right by the tweeters. There was a 100uF e-lytic across the filament wires after the filament supply regulator; I added a 3.3uF film and .1uF C0G ceramic, followed by a Steward #28 material ferrite clamp, which really brought down the noise.
Edits: 01/19/11Follow Ups:
Many people underestimate the importance of the filament supplies in preamp S/N ratios. Using a dual trace scope, I can see a ripple on the cathodes about 90 degrees out of phase to any ripple on the filament. Ripple as low as 20 mV on the filament is clearly visible on the cathodes. Using a series pass type filament regulator can lower the ripple down to 2 mV which adds significant detail to a tube preamp. If your regulator (LM7812, TO-3 case) is large enough, you can also add a very large cap, about 1K mF and lower the filament ripple to 2mV or so.
I suspect your issue with hiss stems from another factor.
Stu
All manner of noise on the heater supplies is coupled to the cathodes through this capacitance. Rectifier noise on the B+ travels through the common power transformer core (in amps with single power transformers) and stray capacitance to the heater windings and gets into the audio signal through the heater-cathode capacitance.
Using regulated DC only helps if the regulator bandwidth encompasses the majority of the noise present. Since rectifier noise is broad-band, the typical three-pin regulators won't be able to stop it.
Interesting about the filament and cathode ripple.
I don't really have an issue with hiss, I'm talking about a level most wouldn't even notice. Previously, it was barely noticeable with my ear right next to the speaker, now it's inaudible. In fact, I've now noticed a very slight hiss from the EQ unit between the pre and the amp when I put my ear directly against the tweeter; the drop in the preamp noise floor seems to have unveiled that. Time to go in there and see if I can knock that out as well.
I'm wondering if you were suffering from leakage in the filament power supply cap. My filaments in the preamp float at about 45 to 50 volts above true ground: this really helps noise rejection. Still, when I place a very large cap (1200 uF is my favored value) the S/N ratio is noticeably bettered. Trying a car battery to power the filaments was eve better in terms of quietness, Just not so convenient with the need to charge up the battery after every listening session (hint, use a Marine deep cycle model).
Stu
Again, very interesting.
Specifically two things, that you were hearing hiss as a function of filament filtration and that putting a bead on the Romex caused problems.
I presume that you are using indirectly heated cathode tubes, after all thoriated tungsten filaments just aren't common in small signal tubes...
In my day sonny the only problems we had from the filament supply was occasional hum. I presume that nowadays you are running them with DC. Maybe it's the power line's new and very unimproved noise spectrum that's the difference. Series impedance is usually vital in any sort of conducted noise problem. When I was in the avionics biz the saying was that the radio filtered the rest of the plane. Fact is, it was true if you didn't use a series choke.
The Romex thing is interesting. You had it clamped over the whole cable didn't you? With a single feed it simply should not have touched the differential currents and not made any sonic difference. Do you have any other earth ground paths besides the 'safety' power lead? Maybe a cable TV feed without an transformer or something? Curious...
Not that you have any onus to do anything further of course, I tend to run out of interest pretty quick once things start working well, what do they say, something like the good is the enemy of the excellent? But it's interesting, I've read other complaints about ferrites messing up dynamics but never understood how they could if applied properly. Still don't.
Ah the mysteries of the audiophile world...
Rick
Being a scientist myself, I tend to do my experiments with some rigor, although with the wife away this week double blind testing was not possible. I used big .35 and .50" ferrites on the Romex and the amp power cords. Taking them on and off several times it was pretty convincing to me that the music just sounded flatter with them on, much more so on the amps, as the Romex effect was more subtle. I did not expect that to be the case - perhaps having them around the ground wire as well as the hot and neutral was preventing the ground from functioning efficiently as a ground? From my understanding, putting them around a wire bundle like that will affect only common mode currents, not differential, so I was not expecting what I heard. I wonder if the ferrite was interacting somehow with the input section of my amps - they are 500WPC class D monoblocks (Gilmore Raptors) and have some proprietary "built-in power conditioning" that their technical material claims obviates the need for power conditioning. They are sealed with special clips and still under warranty so I'm not opening them up now to find out.
On the preamp side I clipped the ferrites on the the hot and neutral leads at the IEC connector inside the unit, and not on the ground. I did not hear bad effects in that case, but the preamp has a lower current draw than the amps, so whether that was due to not clamping the ground is up for debate.
The preamp features a Mullard GZ33 rectifier, Sovtek KT66 regulators, and Ken-Rad 2C22 linestage tubes. The phono section has EH 6922 and Philips E180F tubes as well. I only tested in linestage mode, as I do not presently have a turntable.
"music just sounded flatter with them on, much more so on the amps, as the Romex effect was more subtle"
That makes sense to me, especially if you are using unbalanced interconnects because some of the line-level signal return current may be flowing on the ground wire with HF stuff even parasitically on the hot and neutral.
But, if the Romex feeds the whole system and there are no other grounds then there shouldn't be any common mode signal current through it so you'd think it wouldn't affect the audio.
Out of curiosity you might try them on your speaker cables at the power amp. The cables make grand antennas and the speakers are nicely isolated so the signal is soundly differential mode. It's amazing what you can discover about a black box this sort of method and it's first approach used by EMC engineers to get a feel for what's happening and where.
Rick
I actually have RC filters on both ends of my Zu Ibis speaker cables. I made them using cascaded silver mica caps and TX2575 metal foil resistors, shielded with TI-Shield in a potted ABS box. I selected the resistor value to exactly match the characteristic impedance of the speaker cables for best RF transfer. The effect they have is very noticable; the sound is clearer, smoother, and more detailed with them in place.There is one other ground path in the system, namely through the TV coax cable, which has a ferrite clamp right at the cable box, which is plugged into the same Power Plant as the stereo.
Edits: 01/20/11
http://db.audioasylum.com/mhtml/m.html?forum=tweaks&n=170067
Thanks for the link. I'm curious why you bothered with the capacitors, a 71ohm resistor will dissipate virtually nothing with the average power feeding an 8 ohm speaker system used for home audio unless you're a major head-banger.
While terminating the line should take care of differential mode reflections, it likely won't do much for common mode RFI problems where your speakers and cable make a nice, end-loaded antenna so ferrite beads may still be your friend. But if you don't have a three element ham beam pointing your way from down the street causing gross breakthrough then it might not be of much concern... I don't now, but did and they saved the day.
There is no universal solution as the problems and environment vary.
Regards, Rick
It's a <4 ohm speaker system, Infinity RSIIb.
"Infinity RSIIb"
Hey, my speakers are quite similar to those, they are Infinity Renaissance 90's which years ago replace my Infinity Quantum 3's. I just use Kimber 4/8 PR to feed them (one run is twice as long so the resistances are the same) it measures flatter than other cables I've tried and doesn't seem to need terminations. I haven't measured their Zin but the old Quantum's were just all over the place as a fn of frequency and simply hanging a 10ohm resistor on them improved the sound (and barely got warm). My notion, quite untested, is that the improvement was probably due to limiting the impedance variations that the amps saw.
I'm rather bipolar (but hope to advance to cmos some day) when it comes to stereos: I love having GOOD SOUNDING music at home and feel that it's the most valuable technological thing in my life this side of books. But my other hemisphere seems fascinated by them as tricky systems rarely adequately defined and implemented. Both add to the hobby value...
Rick
Here are mine, with the outboard crossovers I built for them. I have brand new EMIM diaphragms from Apogee Acoustics in Australia - the new diaphragms are MUCH better than the stock units in terms of clarity and dynamics, sound like electrostatics, almost.
Nice!
I'd ask if the crossovers made much difference but maybe I don't want to know. One of the reasons I'm using the models I am is that their shape fits perfectly in the space where I want them. But. OK. Did the crossovers make much difference? I've never had these open or swept them so don't really have a clue what's what. But since I'm still happy with their sound...
Rick
Yes, a very big difference (sorry). I have replaced all the electrolytics with film caps: Solen on the woofer, Solen bypassed with Mundorf Supreme on the low mids, Mundorf Supreme bypassed with Duelund CAST on the high mids, and Mundorf Supreme bypassed with Audicap Theta on the tweeters. Iron core inductors are replaced (except for a 9.5mH on the midrange shunt circuit - too big!) with Northcreek 12 gauge on the woofers and low mids, and Solen Heptalitz 14 gauge on the high mids and tweeters. All wiring is Cardas Teflon insulated Copper Litz (11.5ga on woofers, 15.5 on mids, 17.5 on tweeters).
I also completely rebuilt the equalizer unit that came with the RSIIb's, which made even more of a difference than the crossover rebuild.
As I said, the EMIM diaphragms are the excellent Apogee upgrades; true ribbon design, Kapton film, built to last.
Sound is clean, clear, warm and wonderful. Bass is tight and deep. I am very happy with them.
Now I have yet another thing in the house that's either worn out or could be improved. And that's not counting me, likely the worst of the lot.
Good info, I'll save a copy against the day that these have problems or I get the urge to try and upgrade them. My wife has an extensive list of items with much, much higher priority just now so I anticipate enjoying what I have for quite a spell. Fortunately, that's easy.
Regards, Rick
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
