nah, but thats not what he intended.....
oh, you mean you hehe
nothing wrong with your question, it was obviously perfectly harmless, but its the sort of thing people get pretty religious about and switching supplies have traditionally been the devil
many an audio upgrade on this site starts with getting rid of 'that horrible switching supply'. in many cases I would totally agree, but i've been forced to reconsider that over recent months. previously you could safely say that the switching supply was not designed with audio performance in mind, but the latest generation have been and it showsyeah me neither, i'll play with linear regulated on the breadboard while the dps600 powered wire and the LPUHP power the 2 ways. that way direct comparisons can be made with the same amp.
yeah I figured as much, you are applying the same argument people put up against any voltage source amplifier with high damping factor and extending that to power supplies as well? IE you say that there can be too much control of the rails resulting in resonance issues.. correct? what are you doing here then?
mind you the paper is from 1984 when none of the tech under discussion existed
mind you the paper is from 1984 when none of the tech under discussion existed
Last edited:
Did you take the effort to read it (I guess not...)?
I have no further arguments apart from the fact that compression is not a function of only voltage sag caused by unregulated power supplies, but more so by dynamic compression of the loudspeaker.
Yeah, physics has changed a lot since 1984.....
Last edited:
C'mon guys, get some beer and chill. No need to argue, both of you.
From my perspective...lots of people don't like SMPS since they're not "analog" and many "audiophiles" consider this as a junk.
Also amplifiers like Hypex or B&O IcePower and similar don't rank same as their "analog" brothers.
I don't know if that is true or not but I do believe future is "digital" whether we like it or not.
No need to argue, both of you.From my perspective...lots of people don't like SMPS since they're not "analog" and many "audiophiles" consider this as a junk.
Also amplifiers like Hypex or B&O IcePower and similar don't rank same as their "analog" brothers.
I don't know if that is true or not but I do believe future is "digital" whether we like it or not.
C'mon guys, get some beer and chill.
Agreed... that quickly turned an open conversation into a hissy fit.
Let's get back on topic.
pieter t:
There is a valid argument in the fact that the paper you liked to is nearly 30 years old, and although physics hasn't changed, the quality of loudspeakers certainly has. As a very simple example, take a look at the Precision Devices PD1850. You're looking at only 1.6dB compression at a whopping 800W. That would equate to almost no compression whatsoever at the power levels we're talking about with this amp.
I also think your being a little generous with the 10% delta in rail voltage under full load. I've seem more than 10V of sag on 55V rails at full power into a 4 ohm load and that's with a 700VA toroid. That's near as makes no difference to 20% voltage reduction. That means the RMS output with the regulated supply will be 378 watts, and with the unregulated supply you'll get only 252 watts which is a significant reduction in output power. Far more significant than the power compression of the above driver.
All of this of course ignores the other benefits like lower noise, line regulation, etc... some of which are of particular importance on midrange and tweeter amplifiers where sensitivities are higher and noise can be far more audible.
I would agree that in non-subwoofer applications the power compression of the driver will likely exceed the compression effects introduced by an unregulated supply, but that's ignoring the bigger picture which carries with it far more benefits than just power compression. I get the feeling you've got a little bit of tunnel vision with this, and you're not considering the other benefits.
Regards,
Owen
Thanks for your serious and decent reaction.
The paper I linked to mentions larger diameter loudspeakers being less sensitive to compression, and the PD1850 is certainly a high quality unit in this respect. But let's face the fact that 18 inch woofers with 800W power capacity is not what the majority of us have at home. Rather 8 to at most 12 inch is "normal" for home audio, and besides the more or less "standard" HiFi loudspeaker has a sensitivity of some 90 dB instead of the 98 dB of your example, making them more prone to power compression. IMO dynamic compression at high power levels therefore is, despite the fact that bass units have improved over the years, still a factor in normal domestic situations.
110V pk-pk is 38.87 VRMS is 378 watt/4 ohm;
90V pk-pk (20V sag) is 31.8 VRMS is 253 watt/4 ohm.
31.8/38.87 log20 is minus 1.74 dB; that is not "far more significant" but actually about the same of the PD1850 at full power.
My guess is that dynamic compression of the majority of home loudspeakers is more dominant than compression as a result of amplifier power sag, but I'd like to see more opinions. Actually, from what I read in that linked article, a "hot" voice coil causes less power demand from the amplifier because of it's raised impedance, so the compression affect is amplified.
I agree with the other technical benefits of power supply regulation, especially for voltage amplifying stages, but I could provide many links where unregulated supplies for driver and power stages were preferred sonically (however without stating that this will always be the case), and in the end we listen with our ears.
I am aware of tunnel vision; it's one of the most common issues in audio IMO. The fact that the regulated supply performs well on the bench is no guarantee that it is the best solution for all situations. My pick for a multi amp system would be regulated supply for bass, and unregulated supplies beyond.
Last edited:
To Pieter:
Hi Piter, just to understand better.
when you say that listening can be better with Unregulated psu, what type of regulated is reported the comparison?
All regulated power supplies I know, have a drop in voltage below the transients up to 20%.
linear supply with big capacitors offer a good behavior (no compression produces) only on signals that do not exceed 400mS in duration. after this time, drops.
Some measures seem perfect, not to see the faults of some amplifiers.
I agree that there may be a measure of the differences between the bench and listening.
I measured the ratio of non-linearity, at the output of an amplifier with a linear supply, and regulated DPS-500. if the input frequency of the amp, is fixed (400Hz) between 1w and 250w produces 18% error. if the input frequency changes, changes the ratio of error, even big.
Therefore, from a technical point of view, why do I need an amplifier that produces this variable compression to be sent to a loudspeaker, which already has its compression ratio?
you know how to measure the compression ratio?
I'm sorry but I must say one thing right. these SMPS can not be called "regulated SMPS", becouse he says nothing. then you made an entire reasoning based on what?
Regards
Hi Piter, just to understand better.
when you say that listening can be better with Unregulated psu, what type of regulated is reported the comparison?
All regulated power supplies I know, have a drop in voltage below the transients up to 20%.
linear supply with big capacitors offer a good behavior (no compression produces) only on signals that do not exceed 400mS in duration. after this time, drops.
Some measures seem perfect, not to see the faults of some amplifiers.
I agree that there may be a measure of the differences between the bench and listening.
I measured the ratio of non-linearity, at the output of an amplifier with a linear supply, and regulated DPS-500. if the input frequency of the amp, is fixed (400Hz) between 1w and 250w produces 18% error. if the input frequency changes, changes the ratio of error, even big.
Therefore, from a technical point of view, why do I need an amplifier that produces this variable compression to be sent to a loudspeaker, which already has its compression ratio?
you know how to measure the compression ratio?
I'm sorry but I must say one thing right. these SMPS can not be called "regulated SMPS", becouse he says nothing. then you made an entire reasoning based on what?
Regards
Last edited:
high heat, high mass tip, flux, good tweezers.
lightly tin the ground side pad first, add a touch of flux to the part, reflow the pad with some solder on the tip and slide the part into place, remove heat/tip after some solder flows up the part, hold... done. the other side is dead easy, just hole the tip at the junction of part/pad and flow solder.
lightly tin the ground side pad first, add a touch of flux to the part, reflow the pad with some solder on the tip and slide the part into place, remove heat/tip after some solder flows up the part, hold... done. the other side is dead easy, just hole the tip at the junction of part/pad and flow solder.
one could also argue that physics have very much changed in the last 30years, or rather our understanding has and the technology its applied through is another world; examining and manipulating things at the level of single sub-particles and tracking clouds of electrons and their relations via satellite through 1000's of km of rock across continents. the basic theories have mostly held true, but we use these new technologies, materials and modelling to mitigate the detrimental effects generation after generation, so factors that were once significant roadblocks are now nothing but things to be considered as performance increases.
Last edited:
FloridaBear:
Indeed... follow qusp's instructions and you should be fine. Don't be afraid of lots of heat, just try not to linger for too long.
Also, please tell me you're not using a fine tip on your iron... If so, swap it for a larger chisel tip.
You can also be thankful that you didn't solder the heatsink on there before those front end components... qusp will tell you how much fun that was
Cheers,
Owen
Indeed... follow qusp's instructions and you should be fine. Don't be afraid of lots of heat, just try not to linger for too long.
Also, please tell me you're not using a fine tip on your iron... If so, swap it for a larger chisel tip.
You can also be thankful that you didn't solder the heatsink on there before those front end components... qusp will tell you how much fun that was
Cheers,
Owen
nah its not that, though it wouldnt help. Its about the caps right in front of the LME decoupling the supply pins, only possible to solder them at all when the heatsink is there because of the amount of SMD ive done and also soldering the parts on the underside while balancing the board on the sink was pretty fun too. 2nd 2 I did normally, big difference. I didnt solder the sink to the board, I actualy hardly ever do that, mostly the parts dont need it and if they do I never do it till after initial testing, do you think I should for this build opc?
I would suggest it once you're done testing and you know the LME is good.
That heatsink isn't really made for a part like the LME (few are really) so I think with a hard enough hit or drop the heatsink could potentially pop off. It's always a good idea to fix it in place with a little bit of solder on the clip pins once you're happy with it.
Cheers,
Owen
That heatsink isn't really made for a part like the LME (few are really) so I think with a hard enough hit or drop the heatsink could potentially pop off. It's always a good idea to fix it in place with a little bit of solder on the clip pins once you're happy with it.
Cheers,
Owen
AP2,
If Owens test on the dps-500/DS work out, and I suspect they will, I will purchase the unit or units (unless Owen wants to keep) at least that way you won't loose out on expenses. Owen I have not forgotten - a couple bags of Kona for you when I get it.
As to the dreaded regulation issue I would not think you could compare Linear power supply regulation to SMPS power supply regulation. It would be interesting to see them side by side.
Bill
If Owens test on the dps-500/DS work out, and I suspect they will, I will purchase the unit or units (unless Owen wants to keep) at least that way you won't loose out on expenses. Owen I have not forgotten - a couple bags of Kona for you when I get it.
As to the dreaded regulation issue I would not think you could compare Linear power supply regulation to SMPS power supply regulation. It would be interesting to see them side by side.
Bill
Hi,
To achieve this speed of response and low harmonic product on the induction of the supply rail (especially when used with Class D),
I have developed a new philosophy.
The Regulator has an analogue output stage , as Class A, with rise time of 150ns.
The power generator is a similar stage in class AB, with a special sync, and particular shape of the signal on the inductive load. (Trafo)
So ... can not be called "regulated SMPS as other standard"
with standard pwm regulation, is not possible obtain this performances, apart very poor spurius at output of amplifier, just at 30% power.
DPS-500 uses the same design philosophy.
I have compared with linear regulated during the design, yes, there is a difference, the DPS-600 weighs 400 grams.
Regards
Last edited:
Hi,
I have compared with linear regulated during the design, yes, there is a difference, the DPS-600 weighs 400 grams.
Roberto,
The 400 grams is a physical parameter; I admit that weight is an important factor.
But I understand that you agree that equally good results (by measurement and sonically) can be obtained with high quality regulated linear supplies.
Like regulated SMPS's, regulated linear supplies can be made in very different qualities.
When you are able to read French I can give you a link to a 30 years old (but still basically valid) article where two unregulated supplies are compared with 14 different regulated supplies; I would not be surprised when you would conclude that much in this article will confirm your own experience.