I use dual bridges, but I use transformers with dual secondaries, which is much more flexible than just center tapped ones. Note that such a configuration allows one to use two positive regulators (or capacitance multipliers) for regulation, instead of the positive and negative one required for a center tapped setup (all you do is configure the output of the lower regulator to be connected to ground). You can also use chokes instead of resistors on the ground side as well. Consider the attached example which is my Aleph-X power supply (the regulators are the discrete capacitance multipliers from Mr Evil's site).
Attachments
Bah, I made a mistake, as I was thinking of another supply where I used two positive opamp-driven shunt regulators. Mr Evil's cap multipliers have both positive and negative: http://mrevil.pwp.blueyonder.co.uk/amp/amp4/sub/psu2.shtml
I have CAD files for the PCB if you want, though I took off the bridge and filter caps from the PCB since I wanted to use the CRC before it, and way larger caps.
Note that my Aleph-X supply is similar to the schematic I posted, just that the ground connects first in the +/- cap multipliers rather than the two positive regulators drawn. Part values are correct, but the snubbers on the secondaries are not finalized yet as I've not put the scope on and tested for ringing. I was thinking of adding the rail-to-rail instead of rail-to-ground cap on the end, but I'm not sure what value. mF is not a mistake but I mean millifarad.
Not show is what I put on the primary for almost all supplies I make: a MOV, gas discharge arrestor, mains filter with common mode choke and X and Y caps, fuse, a DC trap, and a thermistor.
I have CAD files for the PCB if you want, though I took off the bridge and filter caps from the PCB since I wanted to use the CRC before it, and way larger caps.
Note that my Aleph-X supply is similar to the schematic I posted, just that the ground connects first in the +/- cap multipliers rather than the two positive regulators drawn. Part values are correct, but the snubbers on the secondaries are not finalized yet as I've not put the scope on and tested for ringing. I was thinking of adding the rail-to-rail instead of rail-to-ground cap on the end, but I'm not sure what value. mF is not a mistake but I mean millifarad.
Not show is what I put on the primary for almost all supplies I make: a MOV, gas discharge arrestor, mains filter with common mode choke and X and Y caps, fuse, a DC trap, and a thermistor.
Getting a little back to PSU topologies
I am currently making a project with Dual bridge CRC supply for the output-stage and a separate regulated PSU for the input And VAS . I have made a PCB with Ground planes and a common star-point on amplifier PCB where all ground is connected. I run almost all my traces on the lower PCB side and thus the top layer is ground. I have now entered the final stages and was just about to send the PCB’s to prototyping. But Now I got the jitters. Maybe a ground plane is not such a good idea. Maybe I get in trouble and pollute the ground for my input stage.
How Do i upload a picture
I am currently making a project with Dual bridge CRC supply for the output-stage and a separate regulated PSU for the input And VAS . I have made a PCB with Ground planes and a common star-point on amplifier PCB where all ground is connected. I run almost all my traces on the lower PCB side and thus the top layer is ground. I have now entered the final stages and was just about to send the PCB’s to prototyping. But Now I got the jitters. Maybe a ground plane is not such a good idea. Maybe I get in trouble and pollute the ground for my input stage.
How Do i upload a picture
I have grouped the input and was rather tightly on the PCB and it would be easy to seperate the two gnd's and join them through a small resistor to prevent ground currents fron entering the input circuits. To me it must be a question of keeping impedances higher for the input cicuits than for the return path to the cap teriminals. You're right about my theroretical knowledge of this. I'am mechanical engineer and i do louspeaker design. But this is a gowing passion and a transistion of building Kits to making my own designs. I have been working on this for quite some time simulation,PCB layouting and so. And wilst I am quite confident in making the PCB's for the Supplies I do have a harder time making it for the amplifer. In particular i find it difficult to anticipate the currents through a Groundplane. So maybe your're right I should make the PCB without ground plane. But my feeling is that the Ground plane will provide the absolute lowest solder free return path impedance to the Cap's 2x30 mF pr Rail.
Michael Børresen
Michael Børresen
I'm not sure, that it's right place, but i'm wondered for a long time, why AC regenerators have positive effect to the amplifiers (not only) sound? Even if this regenerator provides comparable with the wall outlet sine quality, it seems effect is a positive whatsoever, e.g. passive LC line filters have not the same effect yet. I don't know if it's placebo effect at all, but i've heard it easily too. From the physics point of view, AC voltage garmonics+noise will effectively suppressed by the caps after rectification, so from what the effect appears generally?
IVX said:
I have often wondered, if the postulated effect from the mains is mains noise and harmonics that is reduced or improved by mains treatment, then has anyone ever:
Checked the noise and harmonics inside the amplifier on the rails before and after the mains treatment is applied?
Checked the output noise level and spectrum from the amplifier before and after the treatment is applied?
Bob
IVX said:
True if there were no ESR or ESL, nor that nasty 10 nano-Henries per inch parasitic inductance. The high frequency energy likes to get itself rectified by all those Base-Emitter junctions in your opamps, etc., etc.
I think that the problem is more critical in preamplifiers and probably doesn't matter one iota for a power amp -- that is perspective of inbound noise and RFI from the mains. On the other hand, it is prudent to assume that RFI can be generated by the rectifiers and tank inductance and work its way backward down the audio food chain.
FWIW I have started using Corcom EMI suppressors. I don't use solid gold cables but whack up the orange ones from Home Depot.
I have looked at the power line harmonics with the spec analyzer function in my scope, seem to recall that Audio Amateur did a couple of articles on this in the early 1990's with a power quality analyzer. You can see the quality of the power change as the day goes on -- particularly when the Soprano's over in Montclair fire up their hair-blowers before they pomade themselves.
Using an RF probe on my scope, there was a bit of HF stuff that showed up until I added the usual EMI filter built into the IEC connector (common mode choke, X and Y capacitors). There was a big reduction from the primaries to the secondaries, so I now choose EI cores all the time. There was still a bit not only on the rails, but pretty much everywhere in the amplifier, so I guess coupled through the air.
I already had snubbers on the secondary so it wasn't diode ringing noise, and unlike that, it didn't seem to be really modulated by the mains frequency. It wasn't being picked up by the power cable since mine are shielded (not boutiqute, standard Belden stuff), so somewhere between the pole pig transformer and my outlet, my mains wiring is picking up garbage.
I already had snubbers on the secondary so it wasn't diode ringing noise, and unlike that, it didn't seem to be really modulated by the mains frequency. It wasn't being picked up by the power cable since mine are shielded (not boutiqute, standard Belden stuff), so somewhere between the pole pig transformer and my outlet, my mains wiring is picking up garbage.
I don't believe, that i could note some differences in the output FFT, but i've (actually a lot of people) heard the differences surely. Other simple reason is a bigger peak voltage if unclipped sine in use, therefore low ESR regenerator required, strange, that this fundamental specification, would be too rarely read from the regenerator manual, rather can be found abstract promising like "low output impedance", or Exact power claimed "extremely low output impedance =500mOhm", hm, but my wall outlet not worse too, so AC regenerator improvements (mostly) it's an ignorance based myth?
MikeBettinger said:
Innaresting, always thought this was an interesting observation.
Seemed of parallel interest to the concept of using a sinewave generator and an amplifier to supply power to an amplifier; both already using power transformers and amplifiers, both affected by the same mechinisms that allow line noise to blow through the supply into the signal path.
Obviously my sense of curiosity is misplaced.
Someone please bump the thread and get my name off as the last person to post.
Toroids vs EI
Some have argued that toroid power transformers sound different (as in sometimes not as good as) than conventional EI core transformers. Seems like it might have had something to do with the toroid's higher bandwidth or closer primary-secondary coupling letting mains garbage through. Any thoughts or observations on this?
Bob
Some have argued that toroid power transformers sound different (as in sometimes not as good as) than conventional EI core transformers. Seems like it might have had something to do with the toroid's higher bandwidth or closer primary-secondary coupling letting mains garbage through. Any thoughts or observations on this?
Bob
Only in the negative - how about the toroidal better coupling the internally-generated noise back int the low-z of the mains network? It's interesting that most of the talk on DIY boards is in terms of 'incoming', but within my (very modest!) experience, at the sharp end the equipment powered is often noisier. Being able to 'sink' that noise can really help...
I have played with self-built regenerators in an effort to understand what really goes on; the first time I tried this I wrote it up here (context - UK, 240VAC/50Hz):
http://www.acoustica.org.uk/other/mains_regen.html
OK... so it's some years ago and I've learnt a lot since, but I believe the basic issues identified stand.
I have played with self-built regenerators in an effort to understand what really goes on; the first time I tried this I wrote it up here (context - UK, 240VAC/50Hz):
http://www.acoustica.org.uk/other/mains_regen.html
OK... so it's some years ago and I've learnt a lot since, but I believe the basic issues identified stand.
Ok, there are bunch of pics: main_unloaded.png, main_loaded.png (about 300w trafo-diodes-40000uF-res), stax_da80m.png (output FFT from the stax-da80m class A amp, feeding from AC main, which presented in the main_loaded.png), stax_da80m_perfect.png (same case, just feeding from the regenerator, see regen_main_loaded.png). Well, i see no advantages on the output FFT if regenerator in use, please note, THD and output impedance 100-200 times less then typical regenerator can provide, i can't imagine any differences on the FFT if Stax will eat AC from the typical regenerator with 2-3% THD, but many people heard the EFFECT nevertheless. I'm ready to swear, that i heard the EFFECT also, and on the CD players, or DAC even. Placebo? Maybe.. 
marus said:
Definitely classic so far, but the thread has also included issues of grounding. I think there may have been a couple of posts awhile back that involved smps, and I don't think anyone would object to a discussion involving smps. A couple of power amplifier companies, for example Halcro and Chord, assert that smps have sonic advantages in high-end gear. Many others would strongly disagree and argue the other way.
Bob
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.