Done - unless there's some subsequent moderatingOff topic: I'm being hounded here to place the 1000th post on this thread. Not bad. Some threads about new projects reach 3000 and still no hardware in sight
Done - unless there's some subsequent moderating
That would not be moderate at all. In fact, it would be rather extreme
This may have popped up before (sorry if it has), however...
From the smps600 datasheet I see the "auto enable amp" pin has disappeared, compared with the smps400.
The ncore400 datasheets states pulling nAMPON low will enable the amp 2 seconds after all error conditions clear.
Sounds to me like a "predefined" startup delay of 2 seconds after smps power on, and an automatic shutdown when power is lost/switched off, preventing turning on/off pops/clicks.
which, in it's turn, implies correct handling of smps standby mode (J1:1).
correct or hopelessly wrong?
From the smps600 datasheet I see the "auto enable amp" pin has disappeared, compared with the smps400.
The ncore400 datasheets states pulling nAMPON low will enable the amp 2 seconds after all error conditions clear.
Sounds to me like a "predefined" startup delay of 2 seconds after smps power on, and an automatic shutdown when power is lost/switched off, preventing turning on/off pops/clicks.
which, in it's turn, implies correct handling of smps standby mode (J1:1).
correct or hopelessly wrong?
Thank you Bruno for your response. A couple more points of clarification if I may
#7. DC protection. Would you rely on the NCore error circuitry to replace the usual DC protection circuit? It seems fast enough but triggering at 15v will still let a fair amount of DC voltage through (eg. if there is a DC offset on the input and no input decoupling cap) which could dissipate more than 30W into the speakers. So an error LED on the front panel would be a good idea. BTW - does the NCore use an input decoupling cap, I recall the UCD's did?
#8, Opamp regulation. A standard power supply should be sufficient (eg. a LM317 style).
#9. Thanks for the tip about avoiding a minimalist discrete design, which do seem to popular on this site (and they usually require a complex power supply that defeats their simplicity...). So basically the better option for the average amateur is a high quality op amp which if well implemented should be OK and not be the weak link.
#7. DC protection. Would you rely on the NCore error circuitry to replace the usual DC protection circuit? It seems fast enough but triggering at 15v will still let a fair amount of DC voltage through (eg. if there is a DC offset on the input and no input decoupling cap) which could dissipate more than 30W into the speakers. So an error LED on the front panel would be a good idea. BTW - does the NCore use an input decoupling cap, I recall the UCD's did?
#8, Opamp regulation. A standard power supply should be sufficient (eg. a LM317 style).
#9. Thanks for the tip about avoiding a minimalist discrete design, which do seem to popular on this site (and they usually require a complex power supply that defeats their simplicity...). So basically the better option for the average amateur is a high quality op amp which if well implemented should be OK and not be the weak link.
Looking forward to ordering Ncore in the very near future.We have had great success building custom amplifiers utililzing UCD Hxr modules in the past.Great sound! I was surfing Twitter recently and found this comment from Steve Guttenberg a well known reviewer in the US. Based on our experience with UCDs, and the comments here from Bruno and others on the sound of Ncore compared to other amplifiers, I am suspicious of "euphonic" effects of some sort. I am not familiar with Bel Canto so cannot provide any other insight. Here is the "Tweet" from Steve
"The nCore Hypex Class D amps are good, but my Bel Canto REF500S is way better, better as in less bright and sweeter overall".
"The nCore Hypex Class D amps are good, but my Bel Canto REF500S is way better, better as in less bright and sweeter overall".
7. The DC protection circuitry is the same as that of the HG supplies (other than being done in the microcontroller). Anything that couldn't happen then can't happen now. And vice versa. I like the idea of the warning output, I might do that in a later version. For now you'll have to find something yourself.
The Ncore has no coupling caps (on the maxim that "no cap sounds as good as no cap"), but you're free to add 1uF in series with both inputs (not just one).
8 Even without the 317 is fine.
OK I'm off for the holidays.
The Ncore has no coupling caps (on the maxim that "no cap sounds as good as no cap"), but you're free to add 1uF in series with both inputs (not just one).
8 Even without the 317 is fine.
OK I'm off for the holidays.
Cool. Makes for an easy standby-trigger for slaving it to a pre with (12v) triggers like the classe without having to worry about turning the amps on off. Don;t know whether it really saves that much energy, but hey... why not.
(i just realised that that was actually my first useful post in this thread. will shut up until i have any further useful q's
)
LEDs are high performance alternative to zeners
see http://www.diyaudio.com/forums/class-d/190434-hypex-ncore-96.html#post2830210
Bruno
a question about separation of smps and nc400
The physical design of the nc400 allows for some pretty funky case design, whereas the smps is a big (comparatively) rectangular lump
What are the limitations if any on going for a modular approach and mounting them in separate enclosures with an umbilical between them. Any problems of any kind with noise if the PSU interconnection goes between 2 cases?
Thanks
a question about separation of smps and nc400
The physical design of the nc400 allows for some pretty funky case design, whereas the smps is a big (comparatively) rectangular lump
What are the limitations if any on going for a modular approach and mounting them in separate enclosures with an umbilical between them. Any problems of any kind with noise if the PSU interconnection goes between 2 cases?
Thanks
@dallas: The Ncores Steve heard are the NC1200 demo's that are in Alan Silverman's possession. Alan's been demoing the amps around and so far Steve is the only person who didn't like them. I can only surmise that the Bel Canto's are pleasantly coloured. This is quite likely because Bel Canto's owner has a good pair of ears and, knowing the limitations of the actual amplifier hardware he has probably hammered the things into such a shape that whatever they couldn't do in terms of neutrality at least they did in terms of pleasantness. The Bel Canto vs Ncore comparison has been done elsewhere too and even someone who I'd expect to have been significantly biased towards the BC's called the Ncores the finest amps he'd ever spent time with. I think it's a fact of life that people have tastes.
@chris: I think the umbilical should work fine. Do keep the cable in a tight bunch.
@chris: I think the umbilical should work fine. Do keep the cable in a tight bunch.
The frequency response plot is given in the data sheet. It shows a ripple around 60kHz (corresponding to a local minimum in the loop gain, see the white paper) and for no reason other than ordinary time-frequency duality this means that the impulse response (and its integral, the step response) must have a corresponding wrinkle as well.
Isn't this getting the wrong end of the stick? Perhaps any armchair amplifier theorists might first come up with a hypothesis why the ripple in the frequency response should be a problem. Then we can see what to do about this particular hypothesis. Otherwise it's like asking a physicist why apples don't fall up, or up and slightly sideways, or sideways, or generally down but somewhat sideways. It's a great way of keeping him busy.
- Status
- Not open for further replies.