Any opinion as to the best aluminum polymer caps
Someone suggested using these for the digital section power supplies - the SANYOs being the ubiquitous ones.
Are these all about the same or is there a better brand to use?
MOUSER and DIGIKEY both have lots of the PANASONIC version and no SANYO. Any reason to not use these instead of searching for the SANYO?
Someone suggested using these for the digital section power supplies - the SANYOs being the ubiquitous ones.
Are these all about the same or is there a better brand to use?
MOUSER and DIGIKEY both have lots of the PANASONIC version and no SANYO. Any reason to not use these instead of searching for the SANYO?
If you're looking for OSCON type solid polymer caps, I've found Nichicon NS series work well and are a fair bit cheaper than OSCONs
It seems that "OS CON" has become a generic label
So do you mean by that sticking with the SANYO or that the other brands are equivalent? For all we know they are all made in the same factory!
The PANASONIC are also labelled OS CON.
I agree the SILMIC are about as good as we can get for the analogue side.
Thanks,
So do you mean by that sticking with the SANYO or that the other brands are equivalent? For all we know they are all made in the same factory!
The PANASONIC are also labelled OS CON.
I agree the SILMIC are about as good as we can get for the analogue side.
Thanks,
I am starting with an empty board
So I have not made any comparisons. Just using comments others have made along with my experience with other circuits.
Whether it is worth the trouble to replace with SILMICs I hope to hear from others, also. Just for reassurance - I am going to use the SILMICs.
I have removed the standard regulators and will replace with the smallest package BELLESONs. Figuring the OS COMs for the digital side and SILMICs for the analogue. Going to use a separate raw supply for digital (3.3) and analogue (8) and not use the shunt supply since Brian Lowe does not recommend the use of a pre-regulator.
A little PCB surgery will be required to separate the 3.3 from 8.
So I have not made any comparisons. Just using comments others have made along with my experience with other circuits.
Whether it is worth the trouble to replace with SILMICs I hope to hear from others, also. Just for reassurance - I am going to use the SILMICs.
I have removed the standard regulators and will replace with the smallest package BELLESONs. Figuring the OS COMs for the digital side and SILMICs for the analogue. Going to use a separate raw supply for digital (3.3) and analogue (8) and not use the shunt supply since Brian Lowe does not recommend the use of a pre-regulator.
A little PCB surgery will be required to separate the 3.3 from 8.
I think OS CON is a generic designation meaning "Conductive Polymer Aluminum Solid Capacitors". Some people like the Panasonic better in digital circuits I have read. I have no experience comparing them.
I would not stick them just anywhere, as the data sheet mentions several conditions where they are *not* recommended, unless of course you are replacing an existing OS CON with another, then it's probably safe.
Computer motherboards (good ones) make liberal use of them.
I recognize them from many digital PCBs
And had never considered using them.
I have to admit I have never looked to computer motherboard manufacturers for component inspiration! But then someone here said they had used them within the digital section of the onboard DAC power supply.
They are not larger, or smaller, than regular electrolytics so they do not pose a problem or give an advantage in that aspect. They are offered in various levels of ESR - one wonders if you can go too low with a series regulator?
Thinking maybe it is best to play it safe with PANASONIC FCs ... Until I read something that steers me otherwise.
Thanks for the comments!
And had never considered using them.
I have to admit I have never looked to computer motherboard manufacturers for component inspiration! But then someone here said they had used them within the digital section of the onboard DAC power supply.
They are not larger, or smaller, than regular electrolytics so they do not pose a problem or give an advantage in that aspect. They are offered in various levels of ESR - one wonders if you can go too low with a series regulator?
Thinking maybe it is best to play it safe with PANASONIC FCs ... Until I read something that steers me otherwise.
Thanks for the comments!
Aww, not so much ...
One can never leave well enough alone!
I had the chance to meet Brian and his wife, Christina a couple of years ago. Very fine people. Christina has mastered SMD soldering and is now assembling the regulators. Just got 2 each of the 3.3 and 8's made with her steady hands.
Starting with a fresh board - all of this is straightforward. Removing the existing reg chips is simple especially if you are already used to removing SMD stuff.
Now to see if I can get it to work! Not because of what I am doing but because I am the one doing the work.
One can never leave well enough alone!
I had the chance to meet Brian and his wife, Christina a couple of years ago. Very fine people. Christina has mastered SMD soldering and is now assembling the regulators. Just got 2 each of the 3.3 and 8's made with her steady hands.
Starting with a fresh board - all of this is straightforward. Removing the existing reg chips is simple especially if you are already used to removing SMD stuff.
Now to see if I can get it to work! Not because of what I am doing but because I am the one doing the work.
If I'm right, Panasonic took over Sanyo (at least the caps division) last year. On their site you can find the Sanyo Oscon, under a Panasonic header : Business | Industrial Devices | Products | Passive & Electromechanical
Panasonic bought a majority share in sanyo in 2009. So I would expect the panasonic branded oscons to be from the same production as the sanyos.
It is worth trying other caps for analogue on the dac boards. I noticed the muse are not so open and clean sounding as others like BG and silmic.
But is also a matter of balancing the sound in your audio set, so some might prefer the muse ( I don't).
The problem I am facing now is that the 4 boards are soldered together and it will not be possible to desolder all muse caps without damaging the pcb. So I think I will order some new boards from Doede soon.........
It is worth trying other caps for analogue on the dac boards. I noticed the muse are not so open and clean sounding as others like BG and silmic.
But is also a matter of balancing the sound in your audio set, so some might prefer the muse ( I don't).
The problem I am facing now is that the 4 boards are soldered together and it will not be possible to desolder all muse caps without damaging the pcb. So I think I will order some new boards from Doede soon.........
Last edited:
I have the same experience, however both with spdif and usb.So, here I am, happy owner of a 2 deck DDDAC 1794S.
Building was easy, but now I have a minor problem.
When I skip a track, using SPDIF, I hear some pops / statics.
When I use the USB it's perfect.
It happens with all kind of SPDIF sources.
Is this normal behaviour, or can I do something about it (apart from using USB)
Thanks,
Jeroen.
Spdif is much louder (sometimes very loud) though.
It seems to occur when switching between 16 & 24 bit. If it stays the same bit rate there is no pop.
It pretty much makes the spdif unusable and the usb pops are pretty annoying as they come in unexpectedly in a playlist.
I run with cinemags connected to pos & neg, with rca outs (with black on negative and white on gnd). Should I try to switch to common & pos?
Any other options? Or areas where I could have made a mistake?
My DDDAC "simple kit" is now assembled and burnt for 150 hours.
I started listening yesterday, after some 140 hours of burning.
I am very pleased with whai it gives, but I feel the bass are restrained, and there could be more dynamics. It's just like if I forgot the hand brake.
I'm just about to perform the PSU mod, but wondering what to do next for smoothing the expected "harshness" (can't find a proper word). I've read some of your mods, but is there any consensus on which cap to swap on PSU, to fulfil the PSU mod ?
I started listening yesterday, after some 140 hours of burning.
I am very pleased with whai it gives, but I feel the bass are restrained, and there could be more dynamics. It's just like if I forgot the hand brake.
I'm just about to perform the PSU mod, but wondering what to do next for smoothing the expected "harshness" (can't find a proper word). I've read some of your mods, but is there any consensus on which cap to swap on PSU, to fulfil the PSU mod ?
The sound you are hearing is for part due to the muse caps.
Use silmic on the power supply instead of the standard caps. This makes it smoother but also more detailed in the midrange. You can experiment with leaving out the wima bypass on the power supply, these make a noticable difference and can expose harshness.
Also: smaller values of caps makes for more speed and dynamics in the mid and high registers.
Regarding my dac boards: I will remove the shunt regs so they will need to be disassabled. I also have an extra motherboard lying around so I will probably build it together again and couple to one of my other streaming devices in my hobby room
Something to consider
I am thinking, due to problems I had when I first started this project, that as elegant as Doede's architecture is if maybe it could be improved for those who cannot leave things alone.
Instead of stacking all four boards atop each other I am planning on placing two boards on top and two on the bottom and using separate lengths of wire from each board to the point on the mainboard.
I will make a loop of silver wire on both sides of the mainboard for terminating two top and bottom boards. Each DAC board will run its own length of wire to this point making disassembly much simpler. Yes the wire lengths will be longer than to the second board as Doede has configured but less than the length to the fourth board so I think it will be a benign compromise.
Sounds like you made the same "mistake" as I as far as removing electrolytics - I tend to feel better about bending over the leads to touch the trace instead of depending on solder for the connection. One can use an EXACTO knife while the joint is hot to bend the lead back up with minimal damage being careful not to keep the board heated too long. If my first attempt is not successful I let all cool back down before making another attempt.
Splitting the supplies to the DAC board is simple. Just cut the trace close to the 22R resistor, not the trace the resistor is connected to - look at this: http://www.dddac.com/documents/dddac1794_nos_ver30.pdf,
and it is easy to see where the cut is made. Just leave a little extra lead from the + capacitor (wish they were numbered on the board but they are not) leading to the 3.3 V regulator and connect the digital power supply line to that.
Grounding will take care of itself - the ground point will not have changed though it could be advantageous to separate the digital and analogue grounds but I am (almost) sure it is impossible without having lots of PCB surgery and adding wires. If someone knows better ...
All of this is based on using the BELLESON regulators which prefer having raw supplies. Has to be better splitting the two supplies (doesn't it?!). I am not proclaiming the BELLESON to be the best way. They are very good and I like the fellow who designed them. I pay the same price as anyone else for them. In this case I think they will be just fine and personally I am uneasy with the small shunt regulators. I have no experience with them but am skeptical they really work as shunt regulators. Shunt regs waste lots of energy and I cannot see how the amount of energy that NEEDS to be wasted with a shunt can be in the small package without destroying itself. I feel there is very little margin with these and they are on the thin edge of being pass-through devices that do very little regulating at all
Leaving out the chokes on the DAC board and using a choke within the raw supplies (in my case the .320H/10R/600 mA TRIAD choke that is inexpensive and decent enough for the job. Using the DUNCAN PSU DESIGNER tool it predicts with retaining the onboard filters on the DAC, though I am going to use smaller resistors and a simple CLC filter after the transformer, a vanishing low level of ripple, 400uV for what it is worth, after all it is a model. This is for the analogue supply. The digital supply, lacking the extra onboard filters, will likely not be quite as clean, unless I decide to add them.
The same "digital" raw supply will power the mainboard which will also be using one of the BELLESONs.
I am thinking, due to problems I had when I first started this project, that as elegant as Doede's architecture is if maybe it could be improved for those who cannot leave things alone.
Instead of stacking all four boards atop each other I am planning on placing two boards on top and two on the bottom and using separate lengths of wire from each board to the point on the mainboard.
I will make a loop of silver wire on both sides of the mainboard for terminating two top and bottom boards. Each DAC board will run its own length of wire to this point making disassembly much simpler. Yes the wire lengths will be longer than to the second board as Doede has configured but less than the length to the fourth board so I think it will be a benign compromise.
Sounds like you made the same "mistake" as I as far as removing electrolytics - I tend to feel better about bending over the leads to touch the trace instead of depending on solder for the connection. One can use an EXACTO knife while the joint is hot to bend the lead back up with minimal damage being careful not to keep the board heated too long. If my first attempt is not successful I let all cool back down before making another attempt.
Splitting the supplies to the DAC board is simple. Just cut the trace close to the 22R resistor, not the trace the resistor is connected to - look at this: http://www.dddac.com/documents/dddac1794_nos_ver30.pdf,
and it is easy to see where the cut is made. Just leave a little extra lead from the + capacitor (wish they were numbered on the board but they are not) leading to the 3.3 V regulator and connect the digital power supply line to that.
Grounding will take care of itself - the ground point will not have changed though it could be advantageous to separate the digital and analogue grounds but I am (almost) sure it is impossible without having lots of PCB surgery and adding wires. If someone knows better ...
All of this is based on using the BELLESON regulators which prefer having raw supplies. Has to be better splitting the two supplies (doesn't it?!). I am not proclaiming the BELLESON to be the best way. They are very good and I like the fellow who designed them. I pay the same price as anyone else for them. In this case I think they will be just fine and personally I am uneasy with the small shunt regulators. I have no experience with them but am skeptical they really work as shunt regulators. Shunt regs waste lots of energy and I cannot see how the amount of energy that NEEDS to be wasted with a shunt can be in the small package without destroying itself. I feel there is very little margin with these and they are on the thin edge of being pass-through devices that do very little regulating at all
Leaving out the chokes on the DAC board and using a choke within the raw supplies (in my case the .320H/10R/600 mA TRIAD choke that is inexpensive and decent enough for the job. Using the DUNCAN PSU DESIGNER tool it predicts with retaining the onboard filters on the DAC, though I am going to use smaller resistors and a simple CLC filter after the transformer, a vanishing low level of ripple, 400uV for what it is worth, after all it is a model. This is for the analogue supply. The digital supply, lacking the extra onboard filters, will likely not be quite as clean, unless I decide to add them.
The same "digital" raw supply will power the mainboard which will also be using one of the BELLESONs.
Last edited:
Thank you Supersurfer. It is always reassuring to hear a clear summary.
I went to DIY just because it looks grat, and I have soldering skills. Beside that, I don't have one hint of electronic knowledge, so I have to tweak after someone...
So next step is PSU mod + capacitor switch, and I/V resistor replacement.
Hope it'll free the bass and loose the brake
By the way, THANK you Doede, the buils was as easy as it could be, and everything worked at first start.
I went to DIY just because it looks grat, and I have soldering skills. Beside that, I don't have one hint of electronic knowledge, so I have to tweak after someone...
So next step is PSU mod + capacitor switch, and I/V resistor replacement.
Hope it'll free the bass and loose the brake
By the way, THANK you Doede, the buils was as easy as it could be, and everything worked at first start.
Hi Rick,
I wanted to test the dac in its original form before changing the components, that's why the muse is soldered in. This, of course, has its downside
The shunts work very good, they get pretty hot, you can bearly touch them. I have them running for about a year now, my dac is always turned on, and they still work great. They are set at 40 and 35mA, so they waste quite some energy in heat. That, among others, is the reason I have my dac still mounted, open space, on a board.
Splitting the digital power is easiest by letting out r22 and just connect your 3.3v supply on the r22 hole. If you like you can add r22 on your raw power supply.
I finished my new raw power supply today with a very nice 200va transformer and two lundahl chokes. The chokes have been running last week with another trannie and I soldered the big trannie in today.
The shunts on the dac boards do really not need any pre-regulation. It sounds quite a margin better with the new power supply!
I will post a picture soon.
Regards,
I wanted to test the dac in its original form before changing the components, that's why the muse is soldered in. This, of course, has its downside
The shunts work very good, they get pretty hot, you can bearly touch them. I have them running for about a year now, my dac is always turned on, and they still work great. They are set at 40 and 35mA, so they waste quite some energy in heat. That, among others, is the reason I have my dac still mounted, open space, on a board.
Splitting the digital power is easiest by letting out r22 and just connect your 3.3v supply on the r22 hole. If you like you can add r22 on your raw power supply.
I finished my new raw power supply today with a very nice 200va transformer and two lundahl chokes. The chokes have been running last week with another trannie and I soldered the big trannie in today.
The shunts on the dac boards do really not need any pre-regulation. It sounds quite a margin better with the new power supply!
I will post a picture soon.
Regards,