Total environmental cost is ignored by most. I saw a suggestion the most totally green car ever is an old Morgan. Pre-emission.
Made from a bit of steel, rubber and wood. No exotic materials, electronics etc.
Almost every one is still on the road. No pollution from scrap or building another one.
Only weighed about 1800 Lbs, so great mileage.
Maybe, because the did ride pretty rough, you did not have an incentive to make long drives. About two hours was my limit.
ok yeah lost in translation.
Valid questions. Some high end SMPS switching at 1MHz and more should be fast enough to regulate within the audio band. But even they have closed loops with compensation for stability which probably puts their response near the upper audio band. Industrial SMPS switching at 50kHz to 100kHz do not have response covering the entire audio band. Maybe up to a few kHz. I should know this but I don't... I always resort to datasheets to be sure anyways and dont spend too much time dwelling. (my work involves SMPS at times).
There is a limit to how much capacitance you can place on the SMSP output. But if the SMPS handles all the low end, the capacitance is only for the top end, and therefore doesnt need to be as much as linear supplies require.
Anyways, pulsed loads is the worst for SMPS, while constant loads are fine. Which is why SMPS is best for class-A, and probably the worst for class-D. But the class-D crowd is all about size and weight, so it's SMPS. The successful ones are probably using SMPS fast enough to respond within the audio band and beyond.
As to the lifeless sound... Subjective, but I tend to agree. Impressive sounding at first but boring in the long run. Lacking life or something. I kep going back to my flea powered tube amps, the 1-5W all tube builds. Of course requiring efficient speakers. But the music is involving and just giving that little extra that I'd almost call magic. Impossible to explain or documant and prove, since they certainly don't measure so great.
Last edited:
Now that is the wisdom I was looking for! Thanks.
I was aware they do not like much load capacitance. I dealt with some small DC-DC converters and they were very picky. The designer seems to have not read that section of the data sheet. A bit disappointing as this was in a many multi-million dollar super computer.
If we only knew what to measure what our brains think is musical, then we would have a transfer function to build to. So, we either build to specs assuming perfect is better, or build to ear and experience. Usually specs are decent. ( But folks on ASR would disagree 😈 )
The really slick power supply, if you know exactly your unchanging current, is a ferro-resonant where you modulate the core saturation. We had a 60A 5V supply and the active components were an op-amp and a 2n2222. Useless for audio of course.
I think I will keep to the sledge hammer I know best. Often knowing something offers better results than the core technology.
Linear power supplies have problems too. Consider the high frequency rectifier switching noise that commonly needs to be reduced using snubbers, and stray magnetic fields from the power transformer that can induce currents into wires, in the worst case disturbing the ground reference.
For analog audio circuits seems like it takes a lot more engineering effort to make really good SMPS, but its certainly possible.
Regarding potential linear supply problems, hexfreds are still nice for smooth commutation, and Torroidy audio-grade power transformers with electrostatic and magnetic shielding can be pretty effective too. Not the lowest cost choices unfortunately.
Regarding potential linear supply problems, hexfreds are still nice for smooth commutation, and Torroidy audio-grade power transformers with electrostatic and magnetic shielding can be pretty effective too. Not the lowest cost choices unfortunately.
. . and induce things in that other element: an inductor (good name for a device catching 50/60 Hz..)
I wonder whether one would have more or less switching noise at 50Hz compared to say 100kHz.
I am not fit with RF stuff.
I cannot filter and shield them properly, even with HF air-coil inductors and Mu-metal shields.
I have a HV switcher that when switched on will mask my FM radio completely.
At least I know how to deal with 50Hz.
Perfected through decades, nothing new or exotic.
The reason for industrial SMPS is to save cost.
Just as PWM amplifiers.
At work, where we need absolute best performance, we use linear amps and linear supplies to drive voice coil linear motors.
PWM only for less stringend applications.
But maybe Nelson would care to show us all how to do build auddio grade SMPS properly.
Cheers,
Patrick
if we are using a SMPS power supply like laptop brick power supply, what rating do we need to use when compared with idle draw of the circuit ? how much head room do we need to have for the SMPS power supply. for example if we have a amplifier with 80w power consumption , can we use 90w laptop power supply for it or do we need to get 120w or 150w laptop power supply for the amplifier.
Would the SMPS max. amp requirement be any different than with a linear supply? It might if you were to operate a class-A amp into "deep" class-AB where the peak rail current is more than twice the idle current.
what i meant was like what would be ok to continuously draw around 80-85w power from 90w laptop power supply or its better to to draw only certain % (for example say 60-70%) of rated power of the laptop power supply (SMPS)
My rule of thumb is about 50%. That way the supply can deal with peak cases into low impedance and will probably have a longer life span.
Yes, approximately. Put the scope probe near the transformer of the SMPS while it's running. Adjust the vertical sensitivity and the horizontal time base so you can measure the period of one cycle. The frequency is the reciprocal of the period. You don't need a direct electrical connection as the probe will pick up the stray magnetic field. It's not going to be a clean signal but you should be able to see a major repetitive pattern. And of course be careful, as there are high voltages present in a SMPS, even for a while after it's turned off. The attached photo is from a plastic brick SMPS, in the case. 5uS/div x 3 div = 15 uS/cycle or F = 66.7 KHz
Attachments
I came across several of these DC power supplies while I was looking around my pile of spare electronics.
Being ignorant, how do these look for supplying a clean low ripple DC power for a good DAC or small amp requiring good 10-15v? They're measuring 13.3V output and would be outboard of the DAC in its own enclosure. I think I also have one of these (bigger, same design) that is 24V/5A. Are these linear power supplies?
Being ignorant, how do these look for supplying a clean low ripple DC power for a good DAC or small amp requiring good 10-15v? They're measuring 13.3V output and would be outboard of the DAC in its own enclosure. I think I also have one of these (bigger, same design) that is 24V/5A. Are these linear power supplies?
Just adding to my post above (after editing time ran out), foot to say I was thinking about using the 24V version for the AMP Camp Amp? It looks almost the same as the one above.
This is a switch mode power supply. They are usually well regulated meaning the output voltage is constant with a variable load. It will have some switching noise on the output which may affect your project. Just give it a try.
There may also be a data sheet online if you search by the model number. That should have specs for the noise, ripple, etc.
There may also be a data sheet online if you search by the model number. That should have specs for the noise, ripple, etc.
That is a generic SMPS. It will probably power an amplifier, although the voltage is not really a good match for any of the FirstWatt clones that we like to build here. Whether the supply is suitable for audio use is entirely a roll of the dice, even if you add a filter of some sort to the output.
Yours looks like the board out the PSN6.5 Power Supply in Fig.1.