MicroSD Memory Card Transport Project

You have no idea what you are talking about.

Ah. Appreciate you, a real expert, pointing that out. Thanks!

A pack of four of these things can start an automobile.
In that case your car is in rather good shape. That A123 LiFePO4 cell has a capacity of 2.5 Ah (while a car battery is typically 80-100 Ah). With the typical startup current of a normal car starter motor, that battery might just keep the starter going for 5-10 sec, enough to start a car engine if it is in really good tune. But that is of course irrelevant to any audio use.

Tell me of four caps that can start an automobile.
I thought we were talking about audio here, not automobiles. Capacitors have a higher peak current capability compared to batteries - and that is what is important for dealing with peaks. Batteries have a greater charge capacity - that is what you need to start a car. Two rather different requirements, so one has no implication on the other.
 
I was talking about current delivery.

To be honest I am baffled why someone who does not own the device this thread is concerned with would bother reading it much less posting on it. I find this process works best when the respondents are actually working on the same thing.

What do you use as a source for digital music?

You are looking at the constant current spec not the "peak". I looked around and see four in series is not sufficient - eight in series-parallel is needed. Nonetheless ...

So being able to start an automobile says nothing about current delivery?

Oh, well ...
 
I was talking about current delivery.

I see. Well, the short-term peak current capability of the A123 battery and standard electrolytic caps is about the same. Internal impedance (ESR and ESL) is fairly similar at lower frequencies, and at higher frequencies you need smaller bypass capacitors in any case.

To be honest I am baffled why someone who does not own the device this thread is concerned with would bother reading it much less posting on it. I find this process works best when the respondents are actually working on the same thing.
Do you want to re-open the discussion about persons and their motivations, despite it being entirely irrelevant to the topic? More than happy to go there if you insist, but I don't think it is appropriate for this thread.

What do you use as a source for digital music?
Again, rather irrelevant to the thread, but my main sources are a Squeezebox Touch, an Intel NUC and a custom Raspbery Pi (both running Linux), all driving several different external DACs.

You are looking at the constant current spec not the "peak". I looked around and see four in series is not sufficient - eight in series-parallel is needed. Nonetheless ...
Peak current is specified at 120A (for 10 s).

So being able to start an automobile says nothing about current delivery?
Yes, it says something. You need to have a high enough current capability - which both the battery (barely) and the capacitor has. The reason the capacitor won't be able to crank your car is not insufficient current capability, but insufficient charge capacity - irrelevant for the audio application. But an expert like you knew that already, right?

I am sure you also knew that what matters in minimizing ripple is not actually medium-short-term (10s) current capability, but low internal impedance (ESR and ESL).
 
OK now you have me chuckling.

Good. Hobbies should be fun.

You are not being serious are you?
And why wouldn't I?

An electrolytic has the same current capability as the A123 26650 - I am not alone in wondering who makes this capacitor?
Panasonic, Vishay and many others.

As I wrote, A123 26650's have a maximum short-term current rating of 120A.

I think you are confusing the ripple current rating of electrolytic capacitors with maximum peak current. The ripple current rating (the one you usually see in data sheets) is the ripple current the capacitor can deal with on a continuous, long-term basis, usually limited by temperature issues. It is equivalent to charging and recharging your battery 50 times a second - how long do you think your A123 would survive in that kind of use?

The peak current is determined by the internal resistance. have a look a the internal resistance of a typical large electrolytic capacitor and your A123 battery. You will see that they are pretty much the same at low frequencies (again as I already wrote).

Batteries are not a replacement for capacitors. Batteries are a low ripple (but not always low noise) replacement for voltage regulators. With a battery, you don't need huge electrolytic capacitors to smooth the ripple, but you do need filter and bypass caps, especially if feeding digital circuits.
 
Last edited:
Well, we are keeping this thread alive by killing it, I fear.

All I know is that there is no comparison to the sound of my SOEKRIS DAC when using the typical voltage regulators using high feedback to give the illusion of low output impedance v. the A123 battery.

I know second hand that the A123s sound better than ultracaps in the same application.

Not with any expectation of perfection on my part; I know that is not possible but I do know what I hear.

Try some in your digital gear AND THEN try to explain what you hear. Numbers, graphs, etc, even words will be OK.

While you are playing around with placing A123 batteries in your DAC you should consider getting the SDTrans. It is a superlative device for digital playback.
 
All I know is that there is no comparison to the sound of my SOEKRIS DAC when using the typical voltage regulators using high feedback to give the illusion of low output impedance v. the A123 battery.

Glad it is working for you.

I know second hand that the A123s sound better than ultracaps in the same application.
My 35+ years with audio have thought me not to rely on unverified second hand opinions - just because my friend Joe thinks A sounds better than B doesn't mean I would have the same preference, and in fact there might not be any difference at all.

Try some in your digital gear AND THEN try to explain what you hear. Numbers, graphs, etc, even words will be OK.
I have tried rechargeable batteries long ago, and still do it in situations where I am measuring very small signals and any mains ripple would affect the results, but for clean power for digital circuits, nothing beats good bypass/buffer caps placed as close to each circuit as possible.

With the vary fast/short transient surge pulses from digital circuits, once you are more than 5 cm or so away from a circuit, the circuit board or cable impedance makes any differences in power supply quality (as long as you can supply enough average current, and keep voltage stable) pretty irrelevant. Unfortunately that is something self-thought "designers" with only analog experience don't really seem to understand. Once you have designed a couple of high-sped digital circuit boards, you pay a lot mor attention to track and lead impedances and local bypass caps.
 
Julf, do you have the ability to try the lifepo4 batteries in a dac? it would take less than thirty minutes.

Sure. It would require me to open up the DAC and figure out the right voltage, and then get the lifepo4 batteries, but that can definitely be done (albeit not appreciated by my wife).

On the other hand, I think it would be easier to try to find someone around here who is experimenting with power supplies for DACs anyway. I will ask around a bit!
 
Wondering if anyone here who is using the SDTrans will say what they use to transfer their files to the SD card.

I have used both a USB adapter and a SATA adapter. My perception is that the SATA device makes a much better transfer but I am starting to worry about the device I am using.

I realized this morning that AMAZON sent me the wrong one - I paid approx twenty dollars and they sent one they sell for about eleven dollars so I am returning this one in hopes it will not behave as strangely as the one I have used.

Takes forever to boot and seems to be even slower than the USB connected one. This, alone, does not bother me but it makes one wonder if the device is defective and might fail in the near future.

I detect noting wrong with the transfer. There is no comparison in the sound below 500 Hz. The lowest bass is in another league in comparison to the USB transfers. I used a better than average USB device. No computer supplied 5 volts, I used a linear supply. I, also, used a REGEN just to see if that would change things. I t amde a difference but comes nowhere close to the SATA. The SATA adapter is superior.

I am hoping someone who is using the SDTrans has found a really good SATA SD adapter, one that does not take forever to be recognized by the BIOS.

Of course, anyone who has had a good result with any hardware; your experience would be appreciated.
 
I am ashamed to say that when I connected the battery wires to the PCIe device I was not paying proper attention and burned the thing up so I never heard it at all.

One thing: it required this HUGE driver: the installation folder is 6 mB, so i figure I did myself a favor. Then I started worrying it was like that PCIe card that outputs IIS, which sounds like a neat thing, until you find it has USB junk in the middle of the chain.

SO the SATA seems to be the way with its direct path BUT there is something not right about it functionally. It should not make the machine take a couple of minutes to boot when without the adapter connected it boots in 15 seconds. Of course, this is another modified version of XP - might be a conflict because of minimizing? And being a HASWELL CPU and a recent MB for which drivers are not available for XP this might be the cause but I doubt it.

The PCIe thing cost sixty dollars so I have been hesitant to buy another one.

Trying to find a good SATA SD adapter but they all seem to be a little out of the mainstream. Also, looking for drivers.
 
Dear Bunpei,

I have bought the sdtrans384 after group buy, I would like to upgrade the power supply from batteries to the following external DC power supplies are available.
CN1 3.30V 2.97V~3.63V 50mA for NDK clock osillators (1nos of 3.30V rechargable 16650 size battery)
CN5 3.30V 3.00V~3.45V 150mA for MCU, Memory card (1nos of 3.30V rechargable 16650 size battery)
CN7 3.30V 3.14V~3.45V 150mA S/PDIF, I2S interfaces (1nos of 3.30V rechargable 16650 size battery)
CN10 2.50V 2.375V~2.625V 50mA FPGA (2nos of 1.2V rechargable battery in series)
CN13 1.20V 1.15V~1.25V 100mA FPGA (1nos of 1.20V rechargable battery)


but I found there is a wire through the pin with marks [E] and ! So I simply disconnect the wire and connect positive to and negative to [E] of battery power supply?!

bwt, I lost the manual from the package (Can you show me the donwload path for the softcopy) and would like to know which pin hole should connect to for power on&off button installation?

Thanks
by the way, anyone try the batteries power supply and what is the sonic performance, it will be nice if share your experience!!!!

Hi Rick McInnis, I am also using Soekris Dam with SDtrans, it sounds good
 

Attachments

  • FullSizeRender.jpg
    FullSizeRender.jpg
    161 KB · Views: 316
Last edited:
I sent you a note and can send you a copy of the manual.

You remove the pin entirely. The pin connects the output of the onboard regulator to the "part". You need to use the middle point as the input from the output of your regulator and the third place, the ground, will be connected to your regulator.

I think using off board regs for the FPGA stuff is probably not advantageous.

I will be using the 5 volts battery I am using to power the entire board (for the moment) to power those FPGA rails.

I will be using 3 - 3.3 volts A123 26650s for the three remaining rails. All require relays on their outputs since all rails should turn on simultaneously.

I am using a LARGE LiFePo pack at the moment to power the board. I will go to a pair of A123 26650s when I do the three battery mod.

I have removed the rows of electrolytic caps - the LiFePo batteries are not impressed with electroytic caps in their way!

I would stick with the A123s for the e.e volts rails. "Float charge" with a good regulator - best arrangement for digital audio I have found. Thanks to jkeny and nige2000 for this scheme!
 
I have found using a completely separate power supply for the IIS rail is worthwhile.

I had used the same raw supply for all of the 3.3 volts rails - each battery has its own regulator but shares the raw supply. I am powering the FPGA rails with a supply connected to the standard power input which also has its own raw supply.

Has anyone gone beyond this using completely separate supplies for each 3.3 volts rail?
 
Long time no seen. :) I have been only reading this forum for several years, but if I now dare... I have also purchased this unit and hoping to get it running before summer.

Comikitkit wrote that in the packakge there should have been manual?!?! Haven´t seen any in my package. Maybe someone could help me to get one?

I find it interesting to mod powersupply for this unit. Have given any thought to Twistedpear Audions Trident? Available as 3V3, 2V5 and 1V2, but what do you think their current supply capability?
 
Member
Joined 2003
Paid Member
Mine came with a CD. Manual was on that.

The Tridents would not be my first choices. Too little current for the main supply.

For tips on good supplies for digital processing, see these posts by John Swenson:

Sonore microRendu - Ethernet to USB output - Page 49

Uptone Audio JS-2 Power Supply and Linear Fan Controller Installation (Experience, Pictures, Results)

"Audiophile capacitors in Power Supply - A real advantage or not?

If you are looking at a single 5v supply, consider a Salas Reflektor.

If you want to do separate supplies for the different rails, consider Acko's AKD75s, IanCanada's TPS7A4700 regs, or OPC's Paralleled LT3042 boards... or LiFePO4 cells.

My 2 cents.

Greg in Mississippi
 
Last edited:
Long time no seen. :) I have been only reading this forum for several years, but if I now dare... I have also purchased this unit and hoping to get it running before summer.

Comikitkit wrote that in the packakge there should have been manual?!?! Haven´t seen any in my package. Maybe someone could help me to get one?

I find it interesting to mod powersupply for this unit. Have given any thought to Twistedpear Audions Trident? Available as 3V3, 2V5 and 1V2, but what do you think their current supply capability?

If you will send me a text message with your email address I will send you a copy of the manual if you have did not get the CD.

In my unhumble opinion I doubt there is much to be gained from doing anything with the LPGA rails. The action is with the 3.3 volts rails and I disagree with Greg (and he knew I would) that you should consider, no matter what regulator you choose to use, using an A123 26650 battery as an output capacitor on the reg.

I chose to use BELLESONs since they are small and I can place them across the individual batteries for a compact installation.

I had initially used one raw supply for the 3 3.3 volts regs but found, without intending to, that a separate raw supply for the IIS rail is worth the trouble.

I should try separate raw supplies for the remaining two rails and will eventually.

The FPGA rails are powered using the stock power input. I am using a pair of the above mentioned batteries in series for this.

You will need some kind of switch at the output of the batteries so you can turn all of the rails on simultaneously. I used relays since using a switch would have required lots of added wire.

There is something about these batteries and digital circuits - an affinity for lack of a better word. Thanks to jkeny and nige2000 for their excellent ideas.
 
Oh my... There indeed were a cd with the package and THE manual was found. Thank You both, Gentlemen!

I am definately aiming for 5 separate supplies but for conveniency of use/design, I had thought using only one lead acid battery. Either 6V or 12V with preregulation. So far I have seen it like this: 12V Optima > 20.000...40.000uF > 5 * pre-regulation > Tent/Trident/Sjöström/TeddyPardo/etc.

As I am wide open to any suggestions, I found points of John Swenson as valuable. Also I will need to take a closer look at smaller batteries. Maybe you would have some experiences to share, why prefering Lithium ion/iron over good quality lead acid?
 
Last edited: