Least Controversial, Reasonably Priced AC Upgrades

Along with the Xentek's, another brand worth looking for is Topaz. You can find these on Ebay and in various surplus supply houses.

Outside of the KVA rating of the transformer, you should look at the capacitance levels too. The lower the internal capacitance, the better. The higher the capacitance, the more coupling between the primary ( dirty side ) of the AC transformer and the secondary side of the transformer. The more coupling that you get, the less isolation and more noise that you have feeding into your system.

On top of that, you should look at the weight of the unit too. The heavier that the unit is, the larger the core. The larger the core, the less potential for saturation. If the core saturates, even on momentary peaks, the transformer itself will generate distortion and feed it into the AC system.

For sake of comparison, the Xentek 3.0 KVA transformers that i have clock in at over 100 lbs apiece. These were custom made pieces i.e. 3 KVA windings using 5 KVA cores to reduce the potential for saturation. The stock iso transformers that come inside of the Tice Power Blocks and Titans are rated at 1.8 KVA and weigh about 40 lbs each.

As you can see, even if you were to double the rating and weight on the Tice pieces, you would still only end up with 80 lbs of transformer for 3.6 KVA. As such, the Tice cores are noticeably undersized for any type of heavy draw. Other than for the shear profit factor, this is another reason why Tice suggested using two cores in parallel via the Power Block / Titan combo. If one does not do this, it's quite possible to run into core saturation ( distortion ) and the lack of dynamics associated with "current starving" your amp.

As such, many of the negative comments that others have made about using iso's to filter their AC were not actual criticism of iso's in general, but more about the specific products and how they were implimented into their system. Just because something is made by a "high end" company, is "audiophile approved" and / or costs a lot of money does not mean it is "good" or even "suitable" for what you want to do. Sean
>

Lak's got the right idea. Using a larger transformer to filter the entire system has many advantages and it is what i would recommend. This keeps the system at the same voltage potential as all of wiring passes through the same transformer.

On top of that, this moves the transformer away from the listening area. Any / all transformers generate some form of noise & EM field, so the further it is away from your listening position and equipment, the better off that you are. Placing it in the room where your main AC system feeds your sub-panel is the best way to go, both sonically and in terms of ease of wiring. Listening to a very low level transformer hum in your room does nothing to help the noise floor that you just invested tall cash trying to achieve.

As far as selecting a transformer, follow the suggestions above and select a transformer that is rated for a measurably heavier load than what you'll be pulling from it. If you can keep the transformer from running above appr 60% of capacity, you'll never run into any problems with saturation or thermal induced problems. This also minimizes the potential for any type of dynamic constriction to take place during large swings in current draw / spl levels.

Once the AC is "cleaned" coming out of the mains feeding into the sub-panel, you have to keep RFI from entering the AC feedlines that will power your system. This can be achieved by using twisted pair lines from the sub-panel to the outlets. Using twisted pairs lowers the inductance and reduces the potential for your AC lines to act as a long-wire antenna. After all, filtering out all of the RFI at the sub-panel to have the AC system "re-infected" along the way makes very little sense.

The further the distance from the sub-panel, the heavier the wire that you want to use. 10 gauge should be sufficient for all but the heaviest draws or longest runs, but if you're running very high powered high bias amps into low efficiency low impedance speakers, you might want to try running 8 gauge. This is especially true if the runs are quite long i.e. the other side of the house from the AC sub-panel.

You don't need to run "mega-duty" cables for all of your outlets, but it is preferable to use the same gauge wire for all of the runs if possible. The use of even 12 gauge for all the dedicated runs would probably be a big step up for most every installation if one has been using the standard shared circuitry with other outlets in the house. Ten gauge is not that much more money though and if you're doing it now and want to do it right, use the ten gauge. An electrical contractor might tell you that this is "overkill", and for all practical purposes, it probably is for most systems and installations. For the extra few dollars that you spend using 10 gauge now, you'll know that you'll never have to worry about it ever again. Call it a cheap investment in the "peace of mind" factor : )

As far as how the wiring is routed and what you use to house the wiring, you'll have to check with your local inspector to see what will meet code. If you can use PVC rather than metal conduit, do that. Just don't forget that you'll have to run ground wire too as the twisted pair mentioned above only carries the hot and neutral conductors.

Obviously, the selection of outlets becomes a personal matter, so i'll refrain from making suggestions. Just make sure that the AC lines are connected to the outlets via the screw terminals, NOT the "inserts" in the back of the outlets.

Following this approach, you'll have a pretty phenomenal AC system as compared to what one would normally have using standard house wiring. If one were to stop here, i'm sure that most folks would be very happy with the results. After all, you've now got direct runs of heavy gauge low loss wire that has reduced potential to pick up RFI. All of the AC is filtered to the point of having a noise floor that is at least -120 dB's quieter / cleaner than it was before you started and the outlets offer a more secure grip / better connection to the power cord coming in from each component. The -120 dB figure was based upon using a "good" quality iron core iso transformer. The use of a "very high quality" ultra-isolation transformer can get you into the -140 to -145 dB region. If one chooses to use a toroidal design, you can expect to achieve somewhere around -60 dB's to appr -85 dB's of noise suppression.

Using the above figures as a baseline, let's do some math. The "good" iron core at -120 dB's of noise suppression offers over 4000 times more isolation than a "very good" toroidal at -85 dB's. To take that a step further, the "ultra isolation" transformer offers over 1 million times more isolation than that same toroid and slightly over 500 times more isolation than the "good" -120 dB iron core. Now do you folks know why i say what i do about toroids??? Using even just a "good" iron core iso is WAY more effective than using a "very good" toroidal based iso.

The use of additional filtering at the system or for individual components may also be beneficial, especially if one can filter the entire system at the component level. This keeps the "grunge" generated from one component from feeding back into the other components. This requires either a sophisticated line conditioner with isolated filters for each outlet or multiple individual filters for each component. Personally, i'm running 400 lbs worth of iso transformer at the mains with individual iso's for each individual line level component. Power amps are fed directly from the iso's via the sub panel. I'm very resourceful when it comes to finding pieces and parts at bargain prices, hence my ability to do this and not go broke. Buying large assemblies and then gutting them for individual parts has benefits : )

If one wanted to try something like this without going gonzo, i would try using some type of high quality filter to keep the digital gear electrically separated from the analogue gear. If you can do that, you'll be way, way ahead of the game. The money that you saved by not having to buy "mega dollar" power cords can now be used for component upgrades, to buy more music or see more live events. Sean
>

Volksigti:

A) Where can I buy retail-since I need it relatively quickly-the Square D Surge protector.

You should be able to find this at any electrical supply house.

B) Is the 10 gauge twisted pair wire available at Home DESPOT or Lowes?

I sincerely doubt it. You will probably have to find a wire distributor or a major electrical supply house for this type of product. I know that Anixter used to carry a 12 gauge twisted pair cable as normal stock, and that they could twist other cables to order with a minimum quantity, but other than that....

C) Is there any advantage to using PVC or Metal conduit for these 2 dedicated circuits, since here in the Dallas,TX area all houses have the wiring widing loose in the attic, with no conduit at all.

Lumping all of the cables together in one PVC or conduit run might be worse than having them all laid out separately. Isolating each run provides further potential for crosstalk between the circuits.

Having said that, this leaving the cabling exposed sounds crazy to me in terms of building standards. Mice, rats & squirrels tend to chew on cable jacketing. Having all of this open for gnawing with the potential for shorting and the resultant fire seems quite hazardous to me.

If you are going to use something to enclose the cabling, you might want to use very large PVC. This would allow you to run more cabling at a later date with great ease. While metal conduit can act as a shield, it can also cause distortion of the waveform and introduce further non-linearities into the system. As i've mentioned before, shielding is ALWAYS beneficial, but it is typically not implimented in the best fashion.

Kdl: The size of the Xentek being doubled means that it can pass measurably more current with far less potential for saturation / core induced distortions. Since core material is costly, the Xentek is probably a much better built / more expensive item.

As far as more hum & noise goes, this is why i stated that it was best to get the iso's out of the listening area. All transformers "hum" to some extent and the more power that you pull through them, the more noise that they will generate. Toroidals can be quieter in this regard, but they don't offer near as much isolation.

It is possible to "damp" and / or "absorb" the noise generated by a transformer, but one has to pay attention to not restricting the ability of the transformer itself to dissipate heat. How effective this can be done will depend on the approach taken. Damping the "bells" or "end caps" is typically okay, but you don't want to "damp" the core itself. Given that the "bells" tend to ring at a higher frequency ( that's why they are called "bells" ), damping them can reduce the overall spectrum of hum being generated.

Another trick is to pay attention to where the transformer is located. Given the low frequency of the "hum" involved, you don't want to place it anywhere that you would normally achieve bass reinforcement in a speaker. That is, not directly on the floor, not in a corner, not up against a wall, etc... Elevating it in some type of "vibration absorbing cradle" with a layer of properly vented sound absorption material over it may work wonders for those that have them in their listening room.

As far as the KVA rating goes, one needs to buy what they will need for their system. If one is going to run very low powered high efficiency amplifiers for the rest of their life, a smaller KVA rating may be fine. Given that most folks don't want to go through the trouble of hard-wiring something into their AC system only to have to re-do it at greater expense at a later date, "overkill" has benefits. On top of that, a larger transformer offers additional benefits in terms of lower distortion, a reduced noise floor, lower source impedance, etc...

Cincy Bob, here are some responses to your questions:

1) "splitting the line" is what i would recommend and is ideal. You can do this by splitting it at the output of the meter or forming a "Y" off the leads that feed the main box. Using the meter connections may be easier and produce slightly better results. It sounds like you've found an electrician that is "on the ball".

This presents a bit of a problem though if you intend to make use of a "whole system" type iso installation with more cost and work involved. That is, you have to have some type of fuse or breaker between the meter and the isolation transformers. This would mean installing some type of fusable link after the meter, a "jumper" to the transformer(s) and then another jumper to feed the A/V breaker box. You shouldn't have the iso transformers connected directly to the AC mains with no form of current overload protection. Given the possibility that a transformer could short out internally, you wouldn't want all the power that the pole transformer could feed into the shorted iso without some form of safety device between them. While a shorted iso transformer could start on fire, you can pretty much guarantee a BIG fire if it did so without any type of fuse or breaker in front of it.

2) The placement of the breaker box is not critical so long as one uses heavy gauge wiring at all points. Either way, you end up with the same length run. The key here is to keep voltage sag / series resistance down to a minimum. You can do so by maintaining high quality connections and a lot of surface area ( heavy gauge conductors ) throughout the entire circuit.

The other factor that could become involved in any long run is the use of a cable geometry that is resistant to RFI, especially if you are filtering at the breaker box itself. Otherwise, you have the potential to filter they system at the box and then have it "re-contaminated" along the run to the outlets. Even with a "low RFI" type of wiring, you still stand that chance, hence my above comments about further filtering at the component level.

3) If i can remember correctly, i think that PS Audio claims a reduction in noise of appr -40 dB's. As mentioned above, a "very good" toroidal design may achieve -85 dB's and a high quality "iron core" transformer will be somewhere between -120 and -145 dB's. If you want to know how much difference there is between the PS UO's and the iron core transformers, look at the comparisons between the iron core's and the toroidals listed above and double the figures. In comparision, the UO's do offer noise suppression, but compared to the "big monster" type transformers, they aren't nearly as effective. This is not to mention that the UO's may introduce other forms of distortion into the equation depending on the specific application that one is using them for.

The UO's might work okay as a "secondary filter" at the system, but they don't really offer any way to isolate individual components from each other.

As one can see, looking at the "big picture" on a project like this can be quite a task. If done properly though, it really will make a very noticeable difference in performance AND be cheaper than buying & trying tons of different AC filtering devices in the long run. This is not to mention that one is assured that each component that they have is working optimally in terms of being fed the cleanest power possible.

With the potential for your local AC provider to pump "broadband computer noise" over your AC lines in the future, i think that such a system will become almost mandatory for those seeking the ultimate in performance. Rather than try to "band aid" the system later with a bunch of add-on doo-dads that are quite costly and less effective, act now but plan ahead. It will be cheaper and more effective, both cost-wise and sonically, in the long run. Sean
>