according to Lazy Cat you need SMPS for best performance; please google Hypex website for SMPS 400. if you have power hungry speakers, you need at least 400-500W transformer for stereo channels; voltage should not exceed +-45V after rectifiers. I hope I could give you the best answer, a better one can be given only by LC himself, but it strongly depends on your load...
I don't intend too buy a PS, I intend to build it using Pete's boards. Maybe my amp will not sound as good as your amp but I will have fun finding out.
As L.C. had compared traditional PS Versus SMPS, everything else equal around, as i believe to his listening accuracy, as it correlate other various similar conclusions ( including mine ) on various amplifiers, in favor to SMPS, as they are less expensive, smaller, lighter, easier to build-in, have better efficiency, less power dissipation and generate less audible hum in case of poor grounding, i think there is no reason to not prefer SMPS.
Unless you have one over-sized linear PS apart that you want to use, for economical or other reasons.
Unless you have one over-sized linear PS apart that you want to use, for economical or other reasons.
Last edited:
Thank you metallicus69 for the answer!
But in my opinion about 100$ is too much. I thought about a PSU for round 30€?!
I have a SymAsym for big speakers which need more power but the VSAA I want to use with my diy Monitors CT227 MK2 ... So I dont need as much power as the SMPS 400 is offering!
But in my opinion about 100$ is too much. I thought about a PSU for round 30€?!
I have a SymAsym for big speakers which need more power but the VSAA I want to use with my diy Monitors CT227 MK2 ... So I dont need as much power as the SMPS 400 is offering!
Last edited:
Unfortunately they are not less expensive. Recommended Hypex SMPS cost c.100 Euros each + postage (two are needed!). It seems to me that SMPS are overpriced at this moment because they are fashionable, they should cost much less, especially when you consider that LC found out that perfectly new SMPS need elco replacement to achieve optimum performance and reliability. Only if one can make diy SMPS could that be cheaper option, but few diy-ers have capabilities and knowledge needed.
Mr. Evil/PMI capacitance multiplier is very good option at reasonable price. It is always possible to upgrade to SMPS and move cap multiplier to some other amp, so cap multiplier should be tried first...
I dont' know in your country, but in mine, big trasfos (torroidal or C) are sooo expensive. And, if you add soft start circuitry ?
About "two needed SMPS", i pretend one (double sized) can provide better results: more transient ability on each channel, and a better image stability, as a transiant decrease of V on one channel will be reflected to the other channel.
May-be some (L.C. ?) can make a comparison between Hypex and Connex SMPS (now, Cristi knows this topic).
May-be, too it would be a good idea to ask Hypex if they could provide as an option SMPS with the capacitance L.C preferred: it is a pity to have to change them and throw the brand new original parts.
I, too, love (and use) cap multiplier (mine is stabilized with Zener). May-be it can improve SMPS too ?
Last edited:
Hi all
More info for PSU, you have traditionally linear/PMI CM/reg SMPS/ureg SMPS and as i see all info reported back folks are happy with result of either type PSU, so probably because of good amp and thats beatifull.
About price se Hifimediy Store they sell both EU and US and have connex regulated SMPS for exampel S300R at 42,80€ x 2=very cheap.
Myself intend to test all 4 types before making final build and report back for others to have info. But info will first be available in couble of month when progress is done.
But try the link, they also have nice DAC and USB-Audio for diy'ers.
BR Ricky
More info for PSU, you have traditionally linear/PMI CM/reg SMPS/ureg SMPS and as i see all info reported back folks are happy with result of either type PSU, so probably because of good amp and thats beatifull.
About price se Hifimediy Store they sell both EU and US and have connex regulated SMPS for exampel S300R at 42,80€ x 2=very cheap.
Myself intend to test all 4 types before making final build and report back for others to have info. But info will first be available in couble of month when progress is done.
But try the link, they also have nice DAC and USB-Audio for diy'ers.
BR Ricky
I can buy 300VA toroid of very good quality for 40 euros. So, even if two separate cap multipliers are made, each with 300VA toroid, it could be still cheaper than single Hypex SMPS.
Last edited:
Hypex and Connex are businesses. They make products for sale, and by all accounts good ones. So, if they did not make money on what they sell (as they should), they would not be here for long. Hypex SMPS, and especially two of them, is an almost bullet-proof power supply which will probably run any speakers, or combination of speakers, until the VSSA heatsinks melt. 100+ EUR is not excessive, if you need a finished product that will do this.
I am trying to follow Shaan's concept of something anyone can build, using thru-hole parts, a soldering iron, and a multimeter, for power levels that correspond to normal-to-loud listening volume, for a few dollars... or EUR, if u prefer (Thanks to MrEvil, I am basically copying his circuit, and tweaking it a bit)
MrEvil's circuit design is a very clever twist on a Cap Multiplier, and has some of the best elements of the usual Cap multiplier, and a regulated linear supply w. feedback. It can be configured either for lower or higher power levels, and to some extent tuned to the demands of different types of amps (you have to read the MrEvil's very first thread for that).
There is essentially no noise or ripple at the output of the supply, down to a level where my 100 MHz analog/digital oscilloscope can measure with conventional 10X probes and in the presence of background noise (10-15mV). Test results and scope pics are posted.
We are also (for now) benefiting from professionally made boards in small quantity at a discount intended to attract new business customers, which the board house is kindly honoring even though the boards are obviously not part of my work.
The larger filter cap before the active filter can be any good quality industrial capacitor as long as it meets the voltage requirements. As for the outputs cap, I have no cheap answer. If your VSSA has the onboard caps LC recommended (FG-series or similar), they are not as critical as other amplifiers, but... for best possible performance, those will cost as much as they would for any other supply, only help is that they can be a bit smaller... good caps will cost good money
I am trying to follow Shaan's concept of something anyone can build, using thru-hole parts, a soldering iron, and a multimeter, for power levels that correspond to normal-to-loud listening volume, for a few dollars... or EUR, if u prefer
(Thanks to MrEvil, I am basically copying his circuit, and tweaking it a bit)My point, exactly.
MrEvil's circuit design is a very clever twist on a Cap Multiplier, and has some of the best elements of the usual Cap multiplier, and a regulated linear supply w. feedback. It can be configured either for lower or higher power levels, and to some extent tuned to the demands of different types of amps (you have to read the MrEvil's very first thread for that).
There is essentially no noise or ripple at the output of the supply, down to a level where my 100 MHz analog/digital oscilloscope can measure with conventional 10X probes and in the presence of background noise (10-15mV). Test results and scope pics are posted.
We are also (for now) benefiting from professionally made boards in small quantity at a discount intended to attract new business customers, which the board house is kindly honoring even though the boards are obviously not part of my work.
Mr. Evil's circuit has an inherently gradual ramp up of output voltage (tested, pics posted). There is no inrush except at the first filter cap. No stress to the decoupling and filter caps on the VSSA boards, or series 10R. No replacement SMD resistors needed.
Moderate dips caused by transient power demand from the amp are suppressed, up to the level configured by the Vdrop adjustment (tested, & pics posted). Disturbances caused by larger dips are pushed down to sub-sonic range. This is one of the cool features of MrEvil's design, and you can tune this further for your amp, by changing the RC constant of the filter, in combination with the values of the large filter caps.
There will be no way to satisfy everyone. Same issues exist here.
The larger filter cap before the active filter can be any good quality industrial capacitor as long as it meets the voltage requirements. As for the outputs cap, I have no cheap answer. If your VSSA has the onboard caps LC recommended (FG-series or similar), they are not as critical as other amplifiers, but... for best possible performance, those will cost as much as they would for any other supply, only help is that they can be a bit smaller... good caps will cost good money
For the power levels you and I have been discussing, I doubt that two transformers are needed. And, to experiment with dual-mono power supply, I always intended a second board can be connected from the unregulated terminals of the first, so each channel has its own cap multiplier filter, and output capacitor bank. That is one reason why I used the TO-220AC diode and why there is a place for a larger rectifier heatsink. Dual Mono supply on a budget...
I do not intend to use two transformers. It was just an example for comparison. For people that have less efficient loudspeakers, big listening room and need a lot of power SMPS could be excellent solution, especially Connex with it's reasonable prices. But I do not have such requirements.
Two transformers will improve performance when the ground reference is locally on each PCB. Reason is that local decoupling current will slightly pollute each (supply) ground wire. Sooner or later these grounds will form a loop through signal ground.
Glen Kleinshcmidt has an (unfinished) article about this issue here. Well worth a read.
Glen Kleinshcmidt has an (unfinished) article about this issue here. Well worth a read.
I am somewhat familiar with that issue, and you are correct, two cap multiplier filters working off the same ground reference leaves a potential ground loop. However, with reference to figure 2, on my boards the input ground is connected through a small value resistor, which breaks that particular ground loop. I believe that is a common solution, although not a very elegant one.
In any case, this is easily diagnosed by listening to one speaker, and unplugging the other input at the backpanel RCA connector. If there is a hum, and it is reduced when the other connection is unplugged, good chance this is part of the problem. The author of that article does say as much, when he talks about testing a single channel.
I still see without any caps that the inrush transformer current can be huge in some cases (with one 1kW transformer only I managed to disconnect the fuses in my whole house) so a soft start is highly recommended even if the circuit after can handle the capacitors charging and other stuff. I would say you have to think about a soft-start circuit from 300-400W upward.
I did not believe until I felt it myself. it is not an expensive module, but it just adds to the total bill. In case of SMPS, it's included in the supply, therefore, an SMPS starts to be not that expensive as it seems.
One of my problems is I could not find an SMPS with 2 separate identical and isolated outputs for a true dual mono setup.
I see that LC uses 2 SMPS modules for exactly the same reason. This increases the price, but I also wonder if it's not possible that the two oscillators can interact and generate beat frequency or harmonics in the audio spectrum. Not sure if this is easily eliminated (or maybe the two frequencies can be synchronized somehow?).
As I did not experienced this kind of setups, maybe the ones that did can comment about, that would be highly appreciated.
I used a Coldamp SMPS in the past to power a stereo amp, but as I was adding more components in the chain finally I arrived at some ground loop hum (not huge, but listenable with the ear closed to the loudspeaker). I know, I don't listen in that way, but knowing such an issue is present irritates me.
Cheers!
I will add your name to the list. For the next few days, please do not send any money unless you were one of the handful of people who requested one of the first boards. The schematic, BOM, pics and test results are all in MrEvil's thread, starting around page 4:
Improved Cap Multiplier Thread
Any other requests for boards or information, please make a post there or send a pm. Thanks.