Wouldn't it be wiser to convert the HF into heat (by terminating the line with a snubber ) instead of reflecting it back into the mains? Or maybe a combined approach, shorting and snubbering?
@AndrewT:
The bigger EPCOS snubber and damper foil types seem to be up to the task:
http://www.epcos.de/web/generator/W...principals=none_21none_21_20_21_20_21_20.html
@AndrewT:
The bigger EPCOS snubber and damper foil types seem to be up to the task:
http://www.epcos.de/web/generator/W...principals=none_21none_21_20_21_20_21_20.html
John's right. Even a simple cap across the line can make a difference. Don't get too lost in the particulars. Look in your junk box, find a .1 or .22uF cap, and try it. If you think you hear a difference, try paralleling another. You don't have to go as high as 5 or 10uF to get some benefit from this.
Polyester will work, but polypropylene is better. I have not tried polystyrene vs. -ester or -propylene, but I imagine it would be worth trying. I don't have enough money (nor do I know the right people to get free samples) to try Teflon. Even if I had Teflon caps on hand, I'd probably use them in circuits first, rather than line conditioning. Just be sure to use something with good high frequency characteristics--not electrolytics. Electrolytics are non-existent at RF.
If you don't have caps with a sufficiently high voltage rating on hand, then put a lower voltage one across the transformer's secondary. That works, too. Again, it can be a permanent part of the circuit...your little secret.
I don't have schematics for all the top-rated circuits available to me, so I can't check to see, but I wonder just how often a piece of equipment reviews better than a competing piece of comparable equipment simply because of a 50 cent cap that lowers line noise.
There's a certain irony in the fact that as my circuits have grown simpler, my power supplies have grown more complicated. Some sort of a Yin/Yang principle involved, I suppose.
Grey
P.S.: Nelson has suggested large electrolytics in series to keep DC out of your transformer. Somewhere in the Dungeon, I've got some high voltage caps, but I keep forgetting to look for them. Fortunately, I'm not yet suffering from DC on my AC lines...at least not too often. Considering the deterioration in AC quality at my house over the last ten years or so, it won't be much longer.
Polyester will work, but polypropylene is better. I have not tried polystyrene vs. -ester or -propylene, but I imagine it would be worth trying. I don't have enough money (nor do I know the right people to get free samples) to try Teflon. Even if I had Teflon caps on hand, I'd probably use them in circuits first, rather than line conditioning. Just be sure to use something with good high frequency characteristics--not electrolytics. Electrolytics are non-existent at RF.
If you don't have caps with a sufficiently high voltage rating on hand, then put a lower voltage one across the transformer's secondary. That works, too. Again, it can be a permanent part of the circuit...your little secret.
I don't have schematics for all the top-rated circuits available to me, so I can't check to see, but I wonder just how often a piece of equipment reviews better than a competing piece of comparable equipment simply because of a 50 cent cap that lowers line noise.
There's a certain irony in the fact that as my circuits have grown simpler, my power supplies have grown more complicated. Some sort of a Yin/Yang principle involved, I suppose.
Grey
P.S.: Nelson has suggested large electrolytics in series to keep DC out of your transformer. Somewhere in the Dungeon, I've got some high voltage caps, but I keep forgetting to look for them. Fortunately, I'm not yet suffering from DC on my AC lines...at least not too often. Considering the deterioration in AC quality at my house over the last ten years or so, it won't be much longer.
Either my AC is worse than I gave it credit for, or yours is better, in the sense that you have to go to more extreme measures to get results. I got noticeable improvements from much smaller values. In any event motor start caps are indeed cheap. I hadn't thought about using them, but will look into what I can get my hands on.
Cheap improvements are welcome any day of the week.
Grey
Cheap improvements are welcome any day of the week.
Grey
If you're putting caps across the AC line, it's a good idea to use caps that are rated for that kind of service (X1 and X2). These are designed with heavier dielectrics and hipot tested to withstand reasonable line voltage spikes and surges, and also have self-healing characteristics (within reason). Switching power supplies have become so common that tese caps are widely available new or surplus. The suggested bleeder is a good idea, though a Meg or two is generally sufficient. Use at least 1/2 watt for voltage withstand, maybe 2 in series for Europe. Any cap across the AC line should be located after a fuse, for obvious reasons. BTW, if a cap is used, why not go the whole way and use a line filter (canned or discrete)?
Here in Oz our city power is also being loaded up with more noise and junk and power conditioners are now usually beneficial, and often necessary.
Like John, we have been attaching the motor start caps across the 240v/50Hz powerlines particularly for power amps and then using the more sophisticated L-C-L-C type filters for the rest of the gear. For some reason, using the full filters on the power amps seems, in many cases, to "take the heart out" of the sound (much discussion about this - few conclusions) but simply adding that big start cap works everytime, as John says - and simple to try.
Some years ago, Prometheus posted an article about "tuning transformers" and this has become a standard procedure for us. The benefits are extraordinary and are applied to everything in the system - cd transport, dac, pre,amps (especially the F3), etc.
I think the article is now in his "Lycos" (?) site and it refers to a practice that the old time amateur radio guys used in their rigs for extra performance - suggested reading. .... jh
Like John, we have been attaching the motor start caps across the 240v/50Hz powerlines particularly for power amps and then using the more sophisticated L-C-L-C type filters for the rest of the gear. For some reason, using the full filters on the power amps seems, in many cases, to "take the heart out" of the sound (much discussion about this - few conclusions) but simply adding that big start cap works everytime, as John says - and simple to try.
Some years ago, Prometheus posted an article about "tuning transformers" and this has become a standard procedure for us. The benefits are extraordinary and are applied to everything in the system - cd transport, dac, pre,amps (especially the F3), etc.
I think the article is now in his "Lycos" (?) site and it refers to a practice that the old time amateur radio guys used in their rigs for extra performance - suggested reading. .... jh
The only thing that worries me is placing a cap not rated for direct across-the-line operation across the mains without some sort of protection in case a line transient or a defect nails it. Some of the motor run types do have internal safety disconnects, but I don't know if they'd be effective in that sort of situation. I think it would be prudent to at least use a cap overrated in terms of voltage.
People worry and worry. Every motor start cap that I have seen and used is FUSED! Internally. The metal case allows for easy heat dissipation, but they don't get warm. You can ground the case to the 3rd wire. I usually use 600V caps, but 400 would be OK in the USA. You can also put a gas discharge tube in parallel with the 10 uf cap. This WILL protect you from lightning. Today I found that I own about 5000 gas filled tubes. Want to buy some?
I remeber a guy who used a polyester cap for the same purpose. Unfortunately, together with triac based wake-up amplifier circuit (he invented ), that produced pretty high dv/dt. The polyester cap did not get it and exploded once. Oh wow, what lovely environment inside and outside amplifier! 
Check your max. dv/dt rating, guys
Check your max. dv/dt rating, guys
I take John's capacitor advice as what it is; A very good advice. Thank you for this.
The potted mains filters, besides capacitors across the phases, also have series inductors. Is the reason you consider these potted filters no good that the inductors saturate, do you think? If so, non-saturating inductors would do, and give even better rejection of RF.
RK
The potted mains filters, besides capacitors across the phases, also have series inductors. Is the reason you consider these potted filters no good that the inductors saturate, do you think? If so, non-saturating inductors would do, and give even better rejection of RF.
RK
Whilst I don't doubt John's comments on the effectiveness of big input caps, I just wanted to point out that in Europe X and Y rated caps are mandatory for connection across the mains because they are guaranteed to fail safe. Suitably high working voltage is important as well, but not as important as the fact that they will not put voltage anywhere it shouldn't be if they go bang, (or even start leaking voltage a little).
Common mode filters in theory are not substantially affected by the high differential mode current drawn by a power supply capacitive input filter. They are commonly used in input filtering for switching power supplies that have a terrible input crest factor (not even the impedance of a transformer between the AC line and the input caps), and do their job well. Most of the small canned filters will have only a common mode inductor inside (in addition to any capacitors), as they can be made fairly compact for a given current rating . Of course, as with anything, there are quibbles and caveats. Some common mode inductors are wound on sectioned bobbins, whch provide line-to line safety creepage distance, reduce intrawinding capacitance (raising resonant frequency and filtering effectiveness), and thirdly, introducing substantial leakage inductance between the two windings, thus adding some differential mode filtering capability. These inductors need to be chosen with a current rating in line with the RMS draw of the supply, or they will saturate. If chosen with this in mind, they are effective.
Differenetial mode inductors are the ones you really need to worry about, as they do see every bit of the current spikes drawn by a capacitive input filter.. However, if they are made with powdered iron cores, they are extremely difficult to saturate if done right, due to the inherent characteristics of the material. A common filtering combination for high power supplies is a toroidal common mode choke (low leakage, compact size, high current capability) and a powdered iron differential mode filter (compact size, high resistance to saturation).
I do EMI testing on switching power supplies all the time, and believe me, you can tell right away if a filter component saturates by watching the EMI scan - properly chosen, they don't.
One thing I don't understand, though, is this report of inductive filters "squeezing the life" out of an amp's sound. I can't argue pro or con, as I know nothing about the filter, the amp, or how they were used - I've just heard the assertion in passing.
Differenetial mode inductors are the ones you really need to worry about, as they do see every bit of the current spikes drawn by a capacitive input filter.. However, if they are made with powdered iron cores, they are extremely difficult to saturate if done right, due to the inherent characteristics of the material. A common filtering combination for high power supplies is a toroidal common mode choke (low leakage, compact size, high current capability) and a powdered iron differential mode filter (compact size, high resistance to saturation).
I do EMI testing on switching power supplies all the time, and believe me, you can tell right away if a filter component saturates by watching the EMI scan - properly chosen, they don't.
One thing I don't understand, though, is this report of inductive filters "squeezing the life" out of an amp's sound. I can't argue pro or con, as I know nothing about the filter, the amp, or how they were used - I've just heard the assertion in passing.
- Status
- Not open for further replies.