dvd players, and others use them for the longest time now....
that is where i harvest my smps psu with +-12 volts and +5 volts rails,
discarded dvd players have them in a separate module...
i used them in a ACA build, the smps modules side by side with the amplifier boards, and no issues on noise...
that is where i harvest my smps psu with +-12 volts and +5 volts rails,
discarded dvd players have them in a separate module...
i used them in a ACA build, the smps modules side by side with the amplifier boards, and no issues on noise...
Last edited:
Notwithstanding that a good SMPS can indeed be built, those with reservations about them may take comfort in the PSRR of the amp. The AmpCampAmp uses a mu follower style output stage so there is a constant current source involved.
all the while i was under the impression that it was a jfet buffered common source mosfet amp with a CCS loading...
here's one pecking order of rectifier diodes as told to me some time back by Stephie Bench. Hash chokes can also be used to emulate old electrolytic filter caps for guitar amplifiers as modern low ESR caps may modify their original "Tone"
It is possible to move the connection point of output C1 directly to drain of Q1 for pure CCS loading.
Which goes to show you never tested them for RF noise.
On one recent test we left a CB radio on (useful these days in one room), and a SMPS was turned on the kitchen - quite a little way away to run something else.
This blanked out all signal on the CB radio filling it with noise.
Quite instructive that was, so indeed our living spaces are full of RF mush, the air is filling with it thanks to Musk's stoopid satellites, and the forests of terrestrial mobile phone masts.
Even your car is now jammed with wifi and gsm mush.
I looked at the waveform of our 230V mains the other day,on a scope, which is also fed by a huge local windfarm,- probably a big combo of SMPS, and solid state SCR filled crap, cos it has to remain 50hz syncronous or the grid goes down and takes out all the trains out like it did last August in the UK!
Hey!
I said, that looks far more like some sort of square wave than a sinewave, no wonder some transformers are noisy!
That's full of harmonics.
As the mains looks like a sinewave nowadays with all the peaks chopped off (thanks of course to masses of,-
guess what (!) ALL THOSE SMPS at work), the last thing I would dream of is to put yet another massive noise source anywhere near a sensitive amplifier.
Using them of course for DVD players and who knows what, - well are you suprised by all those consumer grade appliances made in PRC, also generate hashes of noise and nasties?
Do you think those numpties care?
Why do I use choke input filters, stabilisers and some valve rectifiers?
It filters out quite a lot of the mush, and we haven't rolled out Huawei's 5g yet! 😡
You just wait and see how it looks when you have little 20ghz TX wired to every little bit of your house!
No wonder some people have started calls to STOP 5g dead with immediate effect. 😱
i can't complain about my amps using smps psu's,
i do not know what you did with yours, not interested....
i am satisfied with mine and you can say all you want...
i do not know what you did with yours, not interested....
i am satisfied with mine and you can say all you want...
Self satisfaction is always a great feeling.
I like consensus from multiple blind testing,- same reason why people are abandoning toroidial transformers, because they don't work anything like as well as E I ones.
I love this stuff with people "reinventing the wheel", then finding out, some 50-60yr old amp sounds loads better, and then can't explain why, look down their noses and say, "what's that old rubbish"?
We did that test yesterday.
Quite simply an ancient monobloc pair 1955 yr amps, kicked A.
Consensus was instant:-
It quite simply sounded 100x better and easier to listen to - no listener fatigue at all, V all these new, all singing all dancing stuff that's supposed to be so great and don't work.
More's the point, it sounded excellent no matter what speakers we connected it to. 🙄
Quite a quantum leap better than the monster I heard 2 weeks ago which had a power supply with 150 000 FARADS (not microfarads) as a PSU.
That sounded ugly, vicious and desperately harsh.
(actually I wish I could really explain the "magic" sound, but it's a weird and often unlikely combination of all sorts of stuff! I have 2 amplifiers that produce that effect, and they are pretty improbable candidates both of them).
Last edited:
Having powered the SAME amplifier from both SMPS (12V ATX into DC boost modules), AND conventional linear supply, I can say there is no functional difference. The SMPS has better regulation (obviously). The cost is similar but the SMPS takes up more space in my instance, a purpose built supply wouldn't though.
Last edited:
The only sound I ever heard from a rectifier is when one arced over.
Other than that, I use enough filtering, and a stiff enough design, that I do not "hear" the rectifier change the sound.
I use both types of filters, choke input, and cap input.
Just my limited experience
Other than that, I use enough filtering, and a stiff enough design, that I do not "hear" the rectifier change the sound.
I use both types of filters, choke input, and cap input.
Just my limited experience
you are quite opinionated, not much substance i can use, but i can live with that...
Last edited:
And tho' I'm coming rather late to the party again, I like stout home-brew power supplies that use silicon rectification, tough little snubber networks, fast recovery discrete diodes, cap-input (tho' choke-input is also nice), a choke-first-HF-hash-filter, a quite-large subsequent reservoir, a long RC time-constant 90% voltage divider to set a reference and a simple “highly flat” MOSFET pass (series) regulator, driving yet another large reservoir capacitor.
This configuration optimizes a bunch of things in-toto.
Using the MOSFET regulator is at least as effective as 2 more C-L-C sections. And it regulates to a voltage that doesn't vary on the second-time-scale, but rather minutes. Barely sags by sustained signal demand and corresponding output power. No sag at all for high 'slam' output surges. (Diss-me-not! that was tongue-in-cheek!)
Its also quite cost-effective, quite reliable, and with a few protective what-to-do-when-the -amp-is-turned-off reversed MOSFET protection diodes, solid as a rock.
Seeing as it has no absolute-reference Zener, it also adapts to local line voltage differences from nominal — be they weather, distance, district, seasonal, whimsical or quixotic in nature. Just like using an unregulated high-capacity CLCLC supply, without all the mass (and iron/copper cost).
Moreover, with a remarkably low amount of fiddling, another XC (MOSFET-capacitor) section can be added to make an even-quieter mid-voltage B+ supply for the more sensitive front end VAS sections. One can even employ a common voltage doubler on the 12 VAC (6–0–6) filament to derive a respectable –28 or so bias supply, if that is desired. No real conduction, so lightest filtering is still plenty of it.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
This configuration optimizes a bunch of things in-toto.
Using the MOSFET regulator is at least as effective as 2 more C-L-C sections. And it regulates to a voltage that doesn't vary on the second-time-scale, but rather minutes. Barely sags by sustained signal demand and corresponding output power. No sag at all for high 'slam' output surges. (Diss-me-not! that was tongue-in-cheek!)
Its also quite cost-effective, quite reliable, and with a few protective what-to-do-when-the -amp-is-turned-off reversed MOSFET protection diodes, solid as a rock.
Seeing as it has no absolute-reference Zener, it also adapts to local line voltage differences from nominal — be they weather, distance, district, seasonal, whimsical or quixotic in nature. Just like using an unregulated high-capacity CLCLC supply, without all the mass (and iron/copper cost).
Moreover, with a remarkably low amount of fiddling, another XC (MOSFET-capacitor) section can be added to make an even-quieter mid-voltage B+ supply for the more sensitive front end VAS sections. One can even employ a common voltage doubler on the 12 VAC (6–0–6) filament to derive a respectable –28 or so bias supply, if that is desired. No real conduction, so lightest filtering is still plenty of it.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
for the life of me, i can not find a tube with transconductance that is more than a irfp240 mosfet can offer,
don't get me wrong, i like building with tubes first...
don't get me wrong, i like building with tubes first...
10,000 m℧ or better at 100 mA. Pretty awesome. Pity it isn't 450 VDS rated (only 200 VDS).
⋅-=≡ GoatGuy ✓ ≡=-⋅