 
|
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 |
207.200.116.66
Al Sekela and I have been experimenting in our very different systems
and listening environments with R-Cs for parallel AC filtering (and
R-Cs for speaker cables, but that's another post). These are among many
AC tweaks Al has cooked up and I have duplicated in different degrees
under his tutelage. Since they work similarly in our disparate
situations, I'm confident that the changes are real and significant
improvements.
The first one in question is the common parallel AC filter using an array of X-rated caps declining in values by a tenth (.47uf, .047, .0047) and
120 Ohm resistors that he has described in other posts on Tweaks Asylum
previously. We have each used different 2 prong plugs, different cap
values (I tried doubling and tripling the above), and placed them in
different locations adjacent to the audio system outlets, as well as
around the house - all of which are sonic improvements and cumulative.
The experiment last week was driven by a desire to see if Allan Maher's
suggestion of the use of grounded/earthed TI-Shield would improve
sonics. It was inspired by the recent post by Chris (cdc) on Tweaks of his use of grounded TI-Shield on his speaker cable R-Cs per Allan's
suggestion.
This was a single blind test. Al knew what he was playing with and I
did not. Test CD was Al's favorite Shirley Horn "You Won't Forget Me".
Al's system is a Wadia 861 driving a pair of Gilmore Raptor monoblocks
and Magnepan MG-20s. Suffice it to say that it's very revealing and
quite sensitive to changes in cables, tweaks, etc.
We tested the following configurations:
1) No extra filtering in play on the audio system outlets.
2) Al's honking big transformer/R-C-on-secondaries filter (he's posted
this before) in play alone.
3) R-C parallel filter in 3 prong plug with disconnected ground wire (only on the first test).
4) R-C parallel filter in 3 prong plug with carbon fiber sleeve around
R-C and disconnected ground wire (only on the second test).
5) R-C parallel filter with R-C surrounded with microsorb and TI-Shield
and shield grounded.
Note that the R-Cs were all burned in for at least a week on Al's system before the tests.
The results were consistent over 2 trials separated by about 1.5 hours, with the change between tests noted above and the second not being blind on either side, and Al claims that they mirror his own trials alone previously.
The order of improvement in sonics is as listed above, from least to
most.
The TI-Shield is definitely a significant improvement over all other
configurations with these filtering schemes. Improvements noted were:
lower noise floor; more palpable and pleasing string reproduction in
piano and double bass punched up in presence; vocals receded into
proper soundstage place, became more nuanced and fluid; sax timber
improved and became more pleasant. Overall, a critical listener would
appreciate all the improvements quickly; as with all these more subtle
changes, I'm not sure a non-audiophile would get it, but that's not who
we do these things for!
On return home last weekend, I mocked up my own test. My R-C parallel AC filters was brand new and not burned in (though the one it replaced had burned in for several weeks). My test started with my existing double value R-C AC filters that had been burned in, then with those removed, then with a newly designed R-C. I used a new R-C array in a 3 prong plug with no ground wire attached and no shielding; then followed with a grounded TI-Shield and microsorb surround on the R-C. The results were the same as previously at Al's, with the grounded and shielded yielding superior sonics very noticably as described above.
Now here's the kicker: I've been reconfiguring my R-Cs for speaker cables to accept these grounded shields, so I now added boxes, microsorb, and grounded TI-Shield. The difference from Chris' experiment (other than that I used smaller boxes for the speaker ends and one large box with one large shield for the amp end) is that I tied the shield grounds as a star ground directly into the AC parallel R-C filter I just described. That required using 14 AWG from the plug to a set of Euro termination blocks into which I tied the 12 AWG ground wires from the speaker and amp side R-Cs. This also allowed me to test with and without this new grounding scheme.
Again, the results with the shields tied to ground were noticably better than without. The strongest result were with the grounded AC filter and with the speaker end grounded R-Cs, less so (or already less distinguishably so) with the R-Cs at the amp end of the speaker cables. At this point I may simply have reached some limit of my ability to discern the differences with these tweaks, but they are definitely there.
In fact, the whole setup consistently sounds cleaner grounded, TI-Shielded, with R-Cs on the speaker cables and AC parallel R-C as well. Background is significantly blacker. No other way to describe it than the word "clean". Actually more like CLEEEEEEEAN! It's not sharp, fatiguing or analytical, just dynamic, full range and very clear and pleasing on all sources (FM and CD). I like this!
Thanks to Al for the R-C work, Allan for the grounded TI-Shield work,
and Chris for the stimulating post on speaker cable R-Cs using these
tweaks!
Follow Ups:
As per Chris and Alan, the earthing wire gauge does make a difference to the sound. As per Al and all of the above, these things do sweeten up further with a week or more of burn in. I"m really having fun here!I just started replacing my older R-Cs on AC cables with these Ti-Shielded newer versions and so far no adverse affects. I'll report on this later as I go along and they burn in.
Meantime, here's a crappy picture of the first one with the speaker cable grounds crudely tied in. (I realized after posting that this photo showed only 2 ground wires from the speaker/amp R-Cs and there are actually 3; one had come loose!) What you should see if you were here is: 3 ground wires coming from the speaker cable R-Cs (one is in a combined box at the amp end, rather than 2) wirenutted into 3 14 AWG ground wires from the AC R-C filter; a white PVC plumbing fitting as the base of an eventual housing; inside the fitting is a Ti-shield and microsorb strip surrounding the X rated R-C filter array of 3 C values as described previously. What you cannot see is that the TI is grounded into the same plug as the AC filter below with a 12 AWG ground wire, nor can you see the 3 prong plug at the bottom.
I like housing these things in the plumbing fittings for safety and aesthetics, but you don't really have to do that at all.
Hi,
Thanks for sharing your great work.
I am interest in this project and would like to consult you on the followings :
1. Is the TI shield easy to cut and shape ?
2. How do you connect the 12 AWG ground wire to the TI shield ?
3. Is the TI shield placed around the R-C filter in an open tube form ?
4. What is the thickness of the microsorb material used in your project ?
5. Do you think this 'TI shield' project can be applied to other parts of the audio system such as DAC and speaker X'over ?
Thanks
YC
A smaller iron won't provide sufficient heat. Clean the surface carefully: start with Brasso if it is corroded, and remove the residue with a powerful solvent such as Goof-Off (sold for removing dried latex paint spatters). Use some liquid electronics grade soldering flux and a good solder such as WBT. Clean off the residual flux when the joint is cool.
:-)
1) Yes, I cut mine with ordinary scissors, shape it by hand. It's just a thin sheet, mostly copper, so it forms reasonably easily.
2) First I punch a hole in the TI, then I crimp roughly the stripped wire end through the hole and onto both sides of the TI. Then I solder it (not the easiest to do), then I close a flap of TI over the soldered wire joint. So it's really both soldered and physically held in place. Sometimes I add electrician's tape.
3) Yes, and you don't need to overdo it. My big caps are about an inch wide and that's about all that's really needed of the TI, an inch strip or slightly larger. Make sure you do the Microsorb first, because you'll discover that you need extra TI due to the width and bending involved, plus the flap over to solder a complete tube together. I'm not actually sure you even need a complete tube - you clearly don't on the speaker cable R-Cs at all, BTW - but I'm using that.
4) I'm using the quarter inch version, the thinnest. Al's using the larger size I think. It doesn't appear to be a major factor here at all.
5) Have no experience, but thinking of these applications. Somebody else in the archives did post that it does help in these places, as well as with torroids. I'd try the microsorb and TI layer, but not sure about grounding in such applications at all and wouldn't know how to do it properly myself.
Another consideration: TI-Shield is highly conductive. So I wrap the ones in my AC filters with a layer of electricians tape, having discovered the magic of a short in tight places! In the speaker cable R-Cs you can see that Chris advised a spacer that's non-conductive - cotton in his case and Microsorb in mine - for that very reason.
Hi,
Thanks for your kind reply.
Do you solder the overlap areas of the TI shield to form the open tube ?
TI shield seems to be a wonderful material. Chris used TI shield on one side of the RC filter and got the improvement. How about the influence by the electro and magnetic field on the RC filter from the side that not protected by the TI shield ? This is the point that I cannot understand.
YC
Shielding reflects energy, while damping absorbs it. Contrast a mirror with a piece of black velvet. If you were to build a dark room, would you line the walls with mirrors or black velvet?
TI-Shield in this application is acting as a damping material. RF noise travels along wires and through components as electric and magnetic fields. The center layer of TI-Shield is a high-permeability alloy that provides a low-reluctance place for magnetic fields to go. Since these fields are time-varying, they induce eddy currents in the copper outer layers. The copper has some resistance, so these eddy currents dissipate energy from the RF noise.
A shield, such as a thick piece of copper, would reflect the RF energy much more than it would absorb it. Shielding, to be effective, requires complete coverage of the circuit with the shielding material, and then you still have the RF energy bouncing around inside.
Damping, on the other hand, does not require complete coverage. This is why Chris' design works. I should also point out that I do not know if the cylinder shell of TI-Shield is optimum for the AC damping networks: it may be overkill, or too little. It is clear, though, that you do not want the TI-Shield piece to be able to vibrate in response to room sound. Putting a vibrating piece of high-permeability material next to a circuit invites severe microphonic behavior. The Microsorb or equivalent acoustic damping material is very important.
Hi Al,
Thanks for your explanation on the concept of noise energy reflection versus absorption. I learn a lot from you.
That helps. I'll stop incorrectly using the term "shield" when I mean "damping", though the manufacturer obviously capitalized on it.
At long last I'm starting to get it!
I was trying to understand how a one sided shield would work, when what I understand of Faraday cages requires a virtually complete surround. Never got an answer that I could fathom, but did find that you simply have to experiment to know what works and what doesn't. Some of these shielding schemes can be easily overdone for audio purposes (ERS, carbon fiber, etc.). What works where and why is still a mystery to me other than what I've tried myself.
The TI in cdc's post is one-sided, but his is for a low current speaker cabling system. I'm duplicating that to good results. He experimented with this, so between his results and mine, I'd say follow that recipe for speaker cabling.
However, I thought you were asking about my use of it in AC filtering R-Cs. In that case, Al's results and mine were with a TI tube, open on bottom and top. This is a different application with different kinds of caps (X rated for AC vs. silver mica for speakers), different values of caps and resistors, that has to carry the full current of your AC input at 120V. In this case the full tube of TI doesn't seem to be any problem, though I can't honestly say I experimented with other configurations here.
Yes, I solder the overlap of the TI tube together, but not like a joint for the hull of a ship! Just enough to make secure electrical contact. I don't try to solder the little flap I hand close over the drain wire where it's soldered into the hole in the TI, though. That seemed overkill to me.
As to solering, I got a 40 Watt iron and used the liquid flux as Al suggests. This wasn't particularly easy for me to solder either, but I'm really a complete novice at soldering.
Hi,
Thanks for the input.
It seems the optimum usage of the TI shield involves many factors. I will purchase some TI shield and microsorb to do some experiments.
One detail to add is that the outlets used for my system have carbon fiber slabs cemented to the cover plates. These add a minor increment of improvement over the case with no cover plate. I confirmed this by making more filters for my other circuits after the reported test.
This test was conducted on a dedicated AC circuit that feeds a single Oyaide Ultimo SWO-GX+ outlet. The Wadia 861 is plugged into one outlet, and the filter into the other.
It is important to allow the R-C filters with TI-Shield to 'burn-in' for about a week. When I compared freshly-made filters to my existing setup, I found the 1 KVA transformer and some two-wire filters to be superior. However, after a week of exposure to the AC, the TI-Shield versions became more effective than any of the others. The burn-in can be done on any AC outlet.
While Al and I both use carbon fiber cover plates (or over our cover plates), as I recall the test at his house that I heard was on the one without the cover. In my own house the place where I tested the TI=shielded R-C that was grounded was also on an outlet that was not covered with one of my carbon fiber plates. So I'm not clear if or how the CF was affecting any of the results consistently, other than that we've both proven to our satisfaction for some time now that CF plate and CF sleeving on plugs and on cables is another effective RF tweak.
For those who are curious, the implementation in Al's and my system are different; they are essentially the same elements, but not exactly the same design in many cases. Al solders handsomely, while I often skip soldering because when I do it it's quite amateurish. Plus Al has the dedicated circuitry and I don't. And our rigs are completely different, with Al's on the high end and mine on the low end of high fidelity. Al has planars and monoblocks in a small room, while I have monitors and a sub in a large room driven by a receiver/integrated unit. We are on electrical grids that are separated by almost a 2 hour drive too.
I make this point because I see that many of these tweaks we bother to report work in completely disparate situations, so I accord the effect to the tweak itself. That IS important for the rest of you!
Great info. Thanks to both you and Al for taking the TI-Shield tweek for Chris speaker R/C and trying it with the AC filters.
I use A/M PE filters and I also use isolation transformers on my digital and another for my preamps.
Just to clear things up! Are you no longer using your Transformer and are just using the TI-Shield A/C filter on the circut where the transformer used to be.
Or are you now using both your transformer and the TI-Shield A/C filter on the same line? Have you tried it with both on the same line?
I have tried a few different filters but always with my isolation transformers on the same line and the results were great.
The tests seem to say you omited the transformer in the early part of testing. Have you now omited the transformer in your system altogether in favour of just the tri-Shield by it's self ?
Thanks for your posts, always interesting and informative!
Bart is incorrect: all the dedicated audio circuits for my system have carbon fiber cover plates. These four circuits are grouped in two dual-gang boxes, and each box has its own cover plate. These plates are made from thick slabs of carbon fiber laminate Bart found on the Internet. The test was done on the dedicated circuit for the Wadia 861, which is fed from the box with a cover plate fabricated to accept larger plugs.
I removed the other cover plate to install two additional filters (one for each power amp circuit), as the plate for that box had smaller holes. I let the filters burn in for a week, then compared the sound without the cover plate to that with the cover plate reinstalled (I had drilled out the holes to fit the filter plugs in the mean time). There was a small improvement with the cover plate reinstalled, so most of the improvement the two of us heard in the original test was likely due to the filter by itself.
The transformer I swapped out in the original test was the damping device for the Wadia circuit. It was a 1 KVA, 14-volt filament transformer with R-C loading on the secondary. It was my approach to the parallel-inductor (Hammond choke) tweak. Previous tests of unplugging this beast showed that it improved the system sound considerably compared to the case of no damping on the AC circuit. The TI-Shield R-C damper did not work as well as the transformer when the R-C damper was freshly made. However, after a week of burn-in, it worked better than the transformer, as reported by Bart. I removed the transformer permanently after our test. I now have three similar TI-Shield R-C dampers installed, one for each dedicated audio circuit in use. I plan to make more for use elsewhere in my house.
nt
I'm still using my little transformer with the R-Cs on the secondaries that Al calculated for me. It's on the same line as the new R-Cs with the grounded TI-Shield.
Al took his very large transformer out during the testing sequence when I was at his house. I don't know if Al put his back or not.
IME it's not a matter of one or the other, the TI R-Cs or the transformer. I still find that all these types of parallel AC filtering devices that I've worked with (all mine are DIY) act cumulatively. Have not reached the limits of their value and synergy in my system, nor has Al said that to me either.
Some of you guys with Alan Maher's PE system and Hammond chokes seem to be finding the same: that they're cumulative in impact and that you can use a lot of them together.
When I'm done replacing and exploring with grounded TI-Shield on R-Cs, I'm going to experiment with it elsewhere, as suggested by other posters months ago.
nt
Big thanks to you guys for the circuits, theories, experiments, revisions, observations, etc. That's what it is all about!
Cheers, Mike
Because of the sheer amount of these speaker filters installed in my main system (like 14 or so of them), I have three star grounds points. I did not notice that star grounding the filters improved sonics. It was just easier for my setup. I did experiment (A/B) to see if I could hear a difference, and I could not. The best results I have experienced when taking a single ground wire from the wall to a junction point is to use 10ga. wire. Then from the junction box out to the filters I used 12ga.
If you do experiment with different gauge wire with the TI-Shield, You will probably notice the following (I did), and the differences are not subtle:
* thinner then 12ga. and the sound is leaner
* thicker then 12ga. and the sound is thicker/warmer and will mask finer details of the presentation.
* 12ga. being the most neutral across the audible spectrum and the most effective for my system. I was told by more then one person (including Alan) experienced with TI-Shield that the TI-Shield needed 12ga. to be most effective. This is what got me wondering and experimenting with different gauge wire.
The above is regarding the speaker filters, not AC filtering. This is still being experimented with.
Chris
I followed your and Alan's advice on the 12 AWG THHN ground wires from the TI-Shield. Unfortunately, the plug will not accept more than 2 of those and I needed 4 wires in for this config. So only the last 3 inches are the 14 AWG, but all leading to it (a total of 35' of wire) is 12 AWG. I was lacking another place to wire in the grounds, so this is what I had to go for, but figured that tying into the AC R-C would be interesting in and of itself. It is!
Do you think those last 3" make that much difference?
I, too, have several parallel AC R-C filters in my system as well as one of Al's transformer with R-C on secondary dohickies (though mine is an ant compared to his elephant).
Yes, I think that if you changed the THHN gauge that you are using will make a difference. How much, I would have to duplicate to be sure. The point is you are terminating at 14ga, not 12ga. So if there is a bit of performance gain wouldn't be worth it to try?
I know what you are saying regards to some AC male plugs only being able to accept 10ga. I recommend taking 10ga. from the plug and branching out to 12ga to each TI-Shield termination. If you do not have any 10ga. wire which is really inexpensive to purchase, use two 12ga. to simulate the 10ga.
Try it and let me know what happens. Based on my experiments, you should hear a difference.
Chris
When you go larger gauge for the ground wire, 12ga and thicker, the presentation/sonics get richer and more musical. TI-Shield is very cool stuff.
Chris
Tuning with a drain wire? Who'da thunk it?
Chris (and Alan Maher) both contend that the gauge of the earth/ground wire from the TI-Shield definitely makes a difference in the sonics, with both claiming that the thicker you go toward 10 or even 8 AWG, the thicker the sound that results.
I figured that the 3 individual 14 AWG wires (each about 3.5" long), that terminate the 12 AWG drain wires (total 35') from the TI-Shield on the speaker R-Cs into a star at the ground pin of the AC plug, would equal about a 10 AWG wire if totalled together. So I did that to see if Chris was right (as Alan had suggested too) that the gauge of the end of those grounds would determine the leanness or heaviness of the sound.
What I did was simply to join all of them together with one wire nut. About the only difference would be that the star point would be 3.5" from the ground pin and the effective gauge would be much greater.
I'll be damned if it didn't add real weight to the mids! Then I noticed that it was also beefing up the bottom end. So I was trying to critically listen to see if any of the detail and clean sound had gone south. So far they don't seem to have. Overall it's thicker, but as Chris suggested, more "musical". Can't say as I prefer this yet, but it's different enough to warrant an extended test.
I now have a choice of 3 effective gauges: with one 14 AWG wire (terminating each of the 12 AWG drains), or with 2 together making it more than 12 AWG, or with all 3 together making it around or greater than 10 AWG.
Why wouldn't I now experiment with the in between value and report that to all of you? Well, my wife has been staying home this evening making lots of noise, so critical listening is dead for tonight. But you can be sure I will do this quick and dirty experiment and settle on a final configuration.
Who'da thunk it? Not me for sure!
almost daily too! lol
ummm, told ya...lol Now you have to let the TI-Shield and filter stabalize, this takes about 10 days. I found 12ga. to the TI-Shield most sonically pleasing. 10ga. to the TI-Shield and the presentation gets too thick and masks information. 14ga. and the presentation was too thin and missing information. Felt/seemed like the TI-Shield with thinner then 12ga. is not working at full potential. And you do not need that much of the TI-Shield, really cool stuff.
To think, all this from a simple filter. Cool huh? :o)
By the way, anyone reading this post, this is why in the DIY Project I mention that if you deviate from the recipe in any way, the effectiveness of the filter will suffer. Everything in the recipe makes a difference. I tried so many variations of this filter that I got sick and tired of fooling around with them.
Best Regards,
Chris
Yep, can't disagree. But you know us tweakers: gotta try it our own way... LOL
My situation was a bit different and I was trying the idea of integrating the speaker cable R-Cs with the already effective AC filter R-Cs.
Sadly, most plugs do not accommodate much wire beyond 12 AWG. Pushing 2 of those into a ground pin connection is a real squeeze, let alone 4, which would have been impossible. So I was improvising. Learned something from the experiment anyway, including to listen to others who've done it before...
I also have my grounds for these speaker filters star grounded. But plugged into a more sophisticated cap array filtering. I never thought of mentioning my AC filtering schemes...hmmm
Ready for an upgrade? Go to 12ga. for the ground wiring. As briefly noted in the DIY filtering project, different gauges of wire used "will" determine how effective the TI-Shield is. I tried many configurations and wire gauges from 16ga. to 10ga. using THHN wire, the Home depot stuff. I found that 12ga. was the most audibly pleasing. In the last 10 or so months of experimenting with TI-Shield, to get the most from TI-Shield requires 12ga. or thicker wire. Try 12ga. and combinations that give you thicker then 12ga. and be ready for surprises.
Let me know how it all turns out bro...
Chris
| 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 ]
