I have to admit this is going to take a leap of faith to use an smps in my system, i love the compact size given how small the whole amp could be, but the noise, out of band or not does worry me a little, call it superstition if you like, thats how i feel, mainly because i'm unfamiliar with quality smps, i only have experience of really really crappy ones that should be avoided at all costs for audio
Owen would it be possible to order extra sets of output transistor pairs. I would like to use a couple of 60-0-60 VAC transformers that I obtained from a VERY kind person. I would sleep better at night if I could parallel 3 or more pairs per channel. If possible I would like 10 pairs total, 4 kits already requested and 6 extra pairs of output transistors. You would probably get a better price than I ever could.
thanks in advance
thanks in advance
Would the SMPS be able to replace 60 or 80,000uF of good quality low ESR caps?
It's been well demonstrated that paralleling main reservoir caps leads to better, cleaner, louder, deeper bass. Due perhaps to the ever lower ESR at 20Hz, yes, not 120Hz or 100kHz, that paralleling confers. If you want to deliver current at low audio frequencies you need to parallel them as the ESR is MUCH higher down there. The much faster SMPS recharge may keep a small capacitor charged and the rails from sagging during loud listening, but can it deliver the strong clean low bass and warmth that a bank of caps with low ESR at 20Hz can give?
Further, very good grade caps can give a wonderful sweet and musical sound, especially to male and female vocal, that low cost caps sometimes cannot get even close to doing. Of course, system dependent.
My 12 Nichicon GS grade 4 10,000uF 100V (now called 'Super Through' I think) cost me a lot of money but were so worth it. (6 on each mono block. Just 40,000uF completely beat 80,000uF of Elna standards)
And it's not about being able to hold up the rails during hundreds of Watts and big slam, their delightful effect is there when just barely tickling the supply during lowish volume listening.
It really would be good if the SMPS could achieve this sweetness and subsonics without having to also incur the cost of such caps.
Thanks.
It's been well demonstrated that paralleling main reservoir caps leads to better, cleaner, louder, deeper bass. Due perhaps to the ever lower ESR at 20Hz, yes, not 120Hz or 100kHz, that paralleling confers. If you want to deliver current at low audio frequencies you need to parallel them as the ESR is MUCH higher down there. The much faster SMPS recharge may keep a small capacitor charged and the rails from sagging during loud listening, but can it deliver the strong clean low bass and warmth that a bank of caps with low ESR at 20Hz can give?
Further, very good grade caps can give a wonderful sweet and musical sound, especially to male and female vocal, that low cost caps sometimes cannot get even close to doing. Of course, system dependent.
My 12 Nichicon GS grade 4 10,000uF 100V (now called 'Super Through' I think) cost me a lot of money but were so worth it. (6 on each mono block. Just 40,000uF completely beat 80,000uF of Elna standards)
And it's not about being able to hold up the rails during hundreds of Watts and big slam, their delightful effect is there when just barely tickling the supply during lowish volume listening.
It really would be good if the SMPS could achieve this sweetness and subsonics without having to also incur the cost of such caps.
Thanks.
Personally, cooling fans are NOT acceptable. I hate the noise of fans.
I have one single 120mm 500rpm fan in my PC (which my music on stored on) and quiet though it is, the PC sets a background noise mush that I hate. I have a fanless SMPS, the Seasonic X400, fanless graphic card, HDD in a nexus drive away. No other noise sources or fans in there. Most of the noise is the HDD but the PSU makes a clicking buzz as does the mobo itself that are audible from a couple of metres away. SSD soon now the prices have dropped will fix some of the problem.
I wouldn't want to have clicking buzzing SMPS in the audio amplifiers.
I have one single 120mm 500rpm fan in my PC (which my music on stored on) and quiet though it is, the PC sets a background noise mush that I hate. I have a fanless SMPS, the Seasonic X400, fanless graphic card, HDD in a nexus drive away. No other noise sources or fans in there. Most of the noise is the HDD but the PSU makes a clicking buzz as does the mobo itself that are audible from a couple of metres away. SSD soon now the prices have dropped will fix some of the problem.
I wouldn't want to have clicking buzzing SMPS in the audio amplifiers.
Last edited:
Hi Guys,
So I've transferred all of 454Casull's boards over to Mark (Hi Mark!) which leaves things right back where they were. In fact, it's a little better since I can hand deliver Mark's stuff.
I will be adding the option for an opto on the MUTE pin at the request of QRikard. This will allow you to buy yourself an optocoupler and hook up an external mute button if you so desire. It costs nothing, and doesn't mess up the layout, so it should be an easy addition.
Cristi:
Sorry about that... I didn't realize it was a ZVS LLC design. I'm just a little concerned about the relatively high ripple numbers I've seen kicking about. I've built hard-switched full bridge supplies that had drastically lower ripple than the numbers I've seen on the SMPSX00 series, and I thought some of the same techniques might apply here, but I guess not with the differing topology.
What could be done to get the ripple down to the low mV numbers in the audio band?
mwmkravchenko:
When are we going to setup another Ottawa audio meet? We should plan something for later in the fall. I haven't seen any of the Ottawa folks in a while and it would be good to get everyone together and see some new projects! Did you get those line arrays finished up?
qusp:
I have heard some very good SMPS' on class A/B amplifiers, but it is very challenging to get them just right. They certainly have the potential to outperform a linear supply, but they're often hampered by cost cutting measures as you mentioned.
The simple fact that they can be easily regulated and can be made to eliminate 60Hz hum is a huge pro, but there are several cons that can quickly outweigh those to advantages if they aren't handled porperly.
multisync:
No problem at all with the extra devices. The more the merry when it comes to that sort of thing. I'll make a note, and when it comes time to order, I'll confirm numbers and pricing with you.
coolfox007:
Shipping to the UK will probably be about $15 but I'll have to get a more accurate quote when I get the total weight and size.
Unfortunately the GB is full at the moment, but you're the first on the waiting list which means there's a very good chance you'll get what you're looking for if you're patient. There are always one or two people who sign up and drop out when it comes time to pay for the boards.
I'll keep you posted.
Cheers,
Owen
So I've transferred all of 454Casull's boards over to Mark (Hi Mark!) which leaves things right back where they were. In fact, it's a little better since I can hand deliver Mark's stuff.
I will be adding the option for an opto on the MUTE pin at the request of QRikard. This will allow you to buy yourself an optocoupler and hook up an external mute button if you so desire. It costs nothing, and doesn't mess up the layout, so it should be an easy addition.
Cristi:
Sorry about that... I didn't realize it was a ZVS LLC design. I'm just a little concerned about the relatively high ripple numbers I've seen kicking about. I've built hard-switched full bridge supplies that had drastically lower ripple than the numbers I've seen on the SMPSX00 series, and I thought some of the same techniques might apply here, but I guess not with the differing topology.
What could be done to get the ripple down to the low mV numbers in the audio band?
mwmkravchenko:
When are we going to setup another Ottawa audio meet? We should plan something for later in the fall. I haven't seen any of the Ottawa folks in a while and it would be good to get everyone together and see some new projects! Did you get those line arrays finished up?
qusp:
I have heard some very good SMPS' on class A/B amplifiers, but it is very challenging to get them just right. They certainly have the potential to outperform a linear supply, but they're often hampered by cost cutting measures as you mentioned.
The simple fact that they can be easily regulated and can be made to eliminate 60Hz hum is a huge pro, but there are several cons that can quickly outweigh those to advantages if they aren't handled porperly.
multisync:
No problem at all with the extra devices. The more the merry when it comes to that sort of thing. I'll make a note, and when it comes time to order, I'll confirm numbers and pricing with you.
coolfox007:
Shipping to the UK will probably be about $15 but I'll have to get a more accurate quote when I get the total weight and size.
Unfortunately the GB is full at the moment, but you're the first on the waiting list which means there's a very good chance you'll get what you're looking for if you're patient. There are always one or two people who sign up and drop out when it comes time to pay for the boards.
I'll keep you posted.
Cheers,
Owen
Hi Owen,
How does the prototype behave at turn-on/off?
The lack of output protection also means that there is no external means (e.g. relay) for delayed turn-on and fast turn-off. Given that this is a single output pair MOSFET design, those who are concerned about DC can use supply rail fuses. If DC offset from a source is bothersome, then use input coupling caps.
If the LME49830 has no turn-on/off thump, that's great. If there is some thump, then how about the simple circuit shown in Fig 6 of AN-1849? It's designed for the LME49830 in mind. Like R38 in your schematic, the zener value will need to be sized for each builder's supplies. Hopefully, we don't need the extra complexity....
Thanks!
How does the prototype behave at turn-on/off?
The lack of output protection also means that there is no external means (e.g. relay) for delayed turn-on and fast turn-off. Given that this is a single output pair MOSFET design, those who are concerned about DC can use supply rail fuses. If DC offset from a source is bothersome, then use input coupling caps.
If the LME49830 has no turn-on/off thump, that's great. If there is some thump, then how about the simple circuit shown in Fig 6 of AN-1849? It's designed for the LME49830 in mind. Like R38 in your schematic, the zener value will need to be sized for each builder's supplies. Hopefully, we don't need the extra complexity....
Thanks!
Hi Guys,
The list has been updated one last time to add AndrewT. Sorry for the omission Andrew, and thanks for checking.
IanAS:
You are correct, so you are the first person on the wait list. If anything becomes available, I'll let you know.
Arius:
If you use the resistor value listed, then the mute stays active long enough to let the rails to stabilize before the output is enabled.
There is absolutely no click or pop on start-up with the prototype, and on shutdown it just keeps playing until the filter caps are drained. Not entirely desirable, but there are no clicks or pops on shutdown either. Overall it is very well behaved.
I really dislike the National PSU as it's overly complex and doesn't address the most logical PSU arrangement which is a higher voltage regulated front end with a lower voltage unregulated supply for the mosfets. I'm not sure why they didn't suggest that, as it's by far the most practical approach. Building a fully regulated supply for the whole amplifier is absurd unless you're outputting less than 50W into 8 ohms, and the fully unregulated supply really does leave a lot of room for improvement.
Regards,
Owen
The list has been updated one last time to add AndrewT. Sorry for the omission Andrew, and thanks for checking.
IanAS:
You are correct, so you are the first person on the wait list. If anything becomes available, I'll let you know.
Arius:
If you use the resistor value listed, then the mute stays active long enough to let the rails to stabilize before the output is enabled.
There is absolutely no click or pop on start-up with the prototype, and on shutdown it just keeps playing until the filter caps are drained. Not entirely desirable, but there are no clicks or pops on shutdown either. Overall it is very well behaved.
I really dislike the National PSU as it's overly complex and doesn't address the most logical PSU arrangement which is a higher voltage regulated front end with a lower voltage unregulated supply for the mosfets. I'm not sure why they didn't suggest that, as it's by far the most practical approach. Building a fully regulated supply for the whole amplifier is absurd unless you're outputting less than 50W into 8 ohms, and the fully unregulated supply really does leave a lot of room for improvement.
Regards,
Owen
I'll take his! I'd like a set of PCB and parts for a stereo SE amp. Is that OK?
when can i pay?