Im not quite sure where to post this but thought SS would fit ok.
Right first of all, what do you all think of isolation transformers, ie a 240-240 xformer. Putting this from the mains outlet and feeding the hifi after it, would it have any benafits a normal transformer in the powersupply wouldnt filter out??
Plus I am looking to build a good filter unit but dont really know where to start, have any of you any ideas or links to web sites with relavent information? or even schematics for the isotek devices, which I would purchase but I'd need the 2k varient and dont fancy the huge pricetag.
Cheers Matt
Right first of all, what do you all think of isolation transformers, ie a 240-240 xformer. Putting this from the mains outlet and feeding the hifi after it, would it have any benafits a normal transformer in the powersupply wouldnt filter out??
Plus I am looking to build a good filter unit but dont really know where to start, have any of you any ideas or links to web sites with relavent information? or even schematics for the isotek devices, which I would purchase but I'd need the 2k varient and dont fancy the huge pricetag.
Cheers Matt
Have a look at Jon Risch's website. It has some good info on line filters and a surge suppressor.
Regards,
Dan
Regards,
Dan
You should, but in practice you don't. The DC feeding the oscillator/amp is derived from the mains and suffers a fraction of the mains noise. I was really annoyed after noticing the mains cord of the regenerator is still clearly audible.
Most forms of mains filtering (iso transformers as well) are detrimental to the sound, at least to my ears. If it raises the mains impedance it can't possibly sound good. John Risch's filter seems better than most due to the very low resistance of the inductors and the minimal part count.
I've made up one of Jon Risch's line filters using Wima's FKP-1 capacitors and Hammonds 1538M13 chokes . I'm planning on installing it in my CDP in the near future.
Regards,
Dan
Regards,
Dan
It's a real shame that we have to invest effort with mains filters, but there are just so many hi-fi manufacturers who avoid this issue. The whole point of a PSU is to provide a clean, stable supply, no matter what the input is doing (within reason). A whole host of European test (part of the CE directive) were devised to help with this, but they can be circumvented so easily in audio.
We all know a decent PSU helps make audio sound good.
Many shun switched-mode PSUs, becasue they generate noise; however iof the designer can work with this and filter it, any mains-borne noise is not an issue.
Some companies like Linn are already using SMPSUs in their top-end equipment.
Class D amps are effectively SMPSUs and they are now receiving very good reviews.
When will the rest of the industry clean up its act? I'm fed up with reading report of a piece of hifi sounding different from one location to another, just becasue the mains power is different.
We all know a decent PSU helps make audio sound good.
Many shun switched-mode PSUs, becasue they generate noise; however iof the designer can work with this and filter it, any mains-borne noise is not an issue.
Some companies like Linn are already using SMPSUs in their top-end equipment.
Class D amps are effectively SMPSUs and they are now receiving very good reviews.
When will the rest of the industry clean up its act? I'm fed up with reading report of a piece of hifi sounding different from one location to another, just becasue the mains power is different.
The resistance of the inductors is generally trivial compared with the wiring back to the sub station! And I'm not sure why fewer parts would be better, when more stages could attenuate noise more.
Jon's filters seem to be quite good. For best effect, mains filters should really be mounted inside the hifi unit with wiring kept ideally under an inch long where possible. (Obviously this requires an appreciation of mains product safety wiring and is not recommended to anyone.)
My experience from measurements favours a ferrite-based filter followed by a wound-type common-mode filter.
Excuse me for budding in, but I just couldn't resist... 
Well, the directive only specifies limits (in Amperes) of the harmonic currents. For everything other than a power-amp, these levels are typically highter than the mains current consumption. Hence, these criterias are usually met without special attention to the subject.
If the SMPS is well designed, there may be noise left, but above audible frequencies. That's the clue... the frequency spectrum. You will never be without noise.. not even from batteries or capacitors - it's a choice of which noise is the least audible or annoying.
IMHO there's something left out of this comparison/calculation. You need to look at more than the route from your outlet to the nearest sub-station (transformer). When "seen" from your wall outlet, it's a parallel load of the substation's impedance and all the houses/appliances it supplies. (assuming that you don't live in the middle of nowhere, and have 100's of kilometers to the nearest transfomer).
Just my 2 cents...
Jennice
skippy said:
Well, the directive only specifies limits (in Amperes) of the harmonic currents. For everything other than a power-amp, these levels are typically highter than the mains current consumption. Hence, these criterias are usually met without special attention to the subject.
If the SMPS is well designed, there may be noise left, but above audible frequencies. That's the clue... the frequency spectrum. You will never be without noise.. not even from batteries or capacitors - it's a choice of which noise is the least audible or annoying.
skippy said:
IMHO there's something left out of this comparison/calculation. You need to look at more than the route from your outlet to the nearest sub-station (transformer). When "seen" from your wall outlet, it's a parallel load of the substation's impedance and all the houses/appliances it supplies. (assuming that you don't live in the middle of nowhere, and have 100's of kilometers to the nearest transfomer).
Just my 2 cents...
Jennice
Agree entirely. I was only thinking about 50/60Hz as being the main factor affecting the supply voltage to the hifi unit's PSU.
I'm thinking of susceptibility to main-borne interference. The directive has a very useful burst-noise test which can help enormously with designing a resilient PSU. My Naim CD player, for example, was designed so it didn't lock-up under this condition, which they considered to be a CE pass. I would rather it was designed so the noise didn't adversely affect the audio output, because the PSU would then be better in a normal domestic environment.
Skippy,
The burst transient test only requires things not to lock up (or loose data, for IT equipment). Some degradation of the performance is always permitted according to this standard (have a look at IEC/EN 61000-4-4), as it is not considered to be a permanently present noise source. If it is at ypur location, some appliance near you has a serious emission problem!
It is immensly expensive to make a PSU totally immune to this kind of disturbance, and personally I agree with Naim... rather spend the money on the music than on filtering.
Even one of the very well designed Phihong SMPS's I've tested recently does show slight influence during this test. Then again this was at industrial levels, but still...!
The bursts are really a series of 5khz (interval) peaks, but if you think about it, there will be harmonics of much higher frequency, that couple capacitively (is that a word ) to surrounding components. Thus, normal filtering is usually not enough.
Sure, it IS possible, but it's a trade-off. Would you buy HiFi gear with it's selling point being: "We spend our resources on industrial-grade transient immunity filtering", or would you prefer the resources being put into good audio circuits?
Just my 2cents
Jennice
The burst transient test only requires things not to lock up (or loose data, for IT equipment). Some degradation of the performance is always permitted according to this standard (have a look at IEC/EN 61000-4-4), as it is not considered to be a permanently present noise source. If it is at ypur location, some appliance near you has a serious emission problem!
It is immensly expensive to make a PSU totally immune to this kind of disturbance, and personally I agree with Naim... rather spend the money on the music than on filtering.
Even one of the very well designed Phihong SMPS's I've tested recently does show slight influence during this test. Then again this was at industrial levels, but still...!
The bursts are really a series of 5khz (interval) peaks, but if you think about it, there will be harmonics of much higher frequency, that couple capacitively (is that a word
) to surrounding components. Thus, normal filtering is usually not enough.Sure, it IS possible, but it's a trade-off. Would you buy HiFi gear with it's selling point being: "We spend our resources on industrial-grade transient immunity filtering", or would you prefer the resources being put into good audio circuits?
Just my 2cents
Jennice
In the case of my CD player it most likely loses data, so it would be a fail if you considered it to be IT equipment, which it almost is!
Actually to design stuff to be resilient to the burst-noise is not necessarily very expensive. I have been an industrial designer for some years and have done this. It generally needs care with system design, PCB layout and putting appropriate filtering where necessary. My experience showed that normal operation, with lower levels of noise than this test, was improved.
My Naim is an example of bad system design: the incoming mains cable is many inches long inside the case, and so a good radiator, and is tied to the transformer's secondary output wires for several inches - coupling guaranteed!
Keep up the great discussion!
skippy said:
Depends... The technical file for this procuct can state the degree of immunity, or allowed degrading of signal quality. However, since this will make no sense to most consumers, it's never stated in the normal product documentation, only in the technical file and "Declaration of Conformity". However, these are documents that normal customers are not by law able to get a copy of. Public authorities gan get a copy if they request it (so it must exist), but if customers get a copy, it's as a service. The manufacturer is not forced by law to hand it out to everyone who asks.
You know... In reallity the CE marking is like buying rubber band per meter lenght. There are strict rules for emission, but in terms of immunity, you can write yourself out of any problem, and still have a CE mark on the product. Also, only material covered my medical directive and certain power tools like chainsaws (mashinery directive) are by law subject to third party certification with regards to CE marking. All other appliances can be declared to conform by the manufacturer/importer (if manufactured outside EU). Thus, a lot of equipment may not actually meet the criteria if tested!)
Jennice
/QA Engineer @ Brüel & Kjær Sound & Vibration Measurement.
Slightly off topic here.........
Skippy,
If your a Naim owner you might find Andy Weekes webpage of interest.
Regards,
Dan
Skippy,
If your a Naim owner you might find Andy Weekes webpage of interest.
Regards,
Dan
current harmonics
High bandwidth balanced-secondary isolation transformers (Equi=Tech and the like) noticeably reduce, among other things, equipment generated current harmonics.
I've been experimenting with cleaning my mains supply using a patchwork of available options. The experiment began by purchasing a PS Audio P600, which improved the overall sound of my system. I then added a 650VA balanced isolation transformer after the P600 (feeding a 200W amp, among other components) and noticed a large improvement, pretty much as described in the Stereophile review. I then added a 2kVA balanced transformer, which Equi=Tech are selling for near to cost, to feed the P600 and things further improved.
I suspect the latter improvement derives from nullifying, perhaps among other nasties, current harmonics generated by the P600.
Part of what's happening here seems to be this. The P600 (or any electronics component) draws current in spikes (low power factor). The current spikes create voltage drops across any resistance and induced counter-phase voltage spikes when the current in-rush abruptly halts. Some of those voltage nasties are cancelled by the balanced transformer. Very audible.
High bandwidth balanced-secondary isolation transformers (Equi=Tech and the like) noticeably reduce, among other things, equipment generated current harmonics.
I've been experimenting with cleaning my mains supply using a patchwork of available options. The experiment began by purchasing a PS Audio P600, which improved the overall sound of my system. I then added a 650VA balanced isolation transformer after the P600 (feeding a 200W amp, among other components) and noticed a large improvement, pretty much as described in the Stereophile review. I then added a 2kVA balanced transformer, which Equi=Tech are selling for near to cost, to feed the P600 and things further improved.
I suspect the latter improvement derives from nullifying, perhaps among other nasties, current harmonics generated by the P600.
Part of what's happening here seems to be this. The P600 (or any electronics component) draws current in spikes (low power factor). The current spikes create voltage drops across any resistance and induced counter-phase voltage spikes when the current in-rush abruptly halts. Some of those voltage nasties are cancelled by the balanced transformer. Very audible.
Although impractical for power amps, it seems to me feasable and reasonable solution (at least for the fanatics) for preamps, CD Players and similar devices that move line-level signals around, would be to run them from rechargable battery pack(s). I'm not up on current technology, but I think there are better batteries for this available than the old NiCads. A quick web search seems to indicate there are many DIY plans and app notes out there for battery chargers - including the constant trickle kind. J. Carr's power supply book covers this too I think.
It also ocurrs to me that you cound get some benefits of the approach in the case of a power am if it were one like Jan Dupont's Lynx where the IS section is an opamp. Run that from the battery and the rest from an ordinary PS. The IS is the most suseptable to PS noise anyway.
It also ocurrs to me that you cound get some benefits of the approach in the case of a power am if it were one like Jan Dupont's Lynx where the IS section is an opamp. Run that from the battery and the rest from an ordinary PS. The IS is the most suseptable to PS noise anyway.
Hi, has anyone here tried building an external mains filter for their equipment? I was just wondering, as I have just compleeted one this afternoon (although I haven't had a chance to test it yet, as I forgot to buy the right mains cables when I was in the shop the other day, oops) based on the deign you can find on the decibel dungeon website (here) which is supposed to remove a lot of the noise on the mains such as you have been talking about.
However, it will not solve the problem of the current clipping on the mains caused by running a large amount of power through a mains loop into various transformer supplies, so if this is your main concern it may be of little use.
However, it will not solve the problem of the current clipping on the mains caused by running a large amount of power through a mains loop into various transformer supplies, so if this is your main concern it may be of little use.
Attachments
Why would you use high bandwidth transformers...?
I don't get it... You're running 50 or 60 Hz, and then use high-BW transaformers, connected to complex filters.
The easiest way would be to use transformers with low bandwidth (if such a thing exists, really), to let them do some more filtering of the HF you want to kill, instead of just removing ground loops...
Jennice
I don't get it... You're running 50 or 60 Hz, and then use high-BW transaformers, connected to complex filters.
The easiest way would be to use transformers with low bandwidth (if such a thing exists, really), to let them do some more filtering of the HF you want to kill, instead of just removing ground loops...
Jennice
I actualy picked up the whole thing from a guy selling the kits on ebay (well, the PCB and components to go on it), but now that I have seen it, I think that it wouldn't be too difficult to come up with something better by yourself.
Also, if you look closely at the picture below you can see how the main tracks have been routed for the PCB if you want to try and coppy the design for yourself.
Also, if you look closely at the picture below you can see how the main tracks have been routed for the PCB if you want to try and coppy the design for yourself.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.