Tír Na HiFi

this royal mess is the cmos to lvds chip i struggled to find , abrax found or knew of it,
DS90LV017

https://www.mouser.com/ProductDetail/Te ... %252bLY%3d

anyway the old r1 soekris 1021 is singing again
however it is loosing i2s sync quarter sec every 5-6 mins with no pattern
sounds good though
even though initially i thought it lacked detail, fixed by adding ff reclock to t

my other soekris has the sd card player on it, its easily better

ill have to try the sd card player on both now to check if fpga fixed clock is better than programmable

on the scope the fixed clk lvds looks prettier than the programmable

http://www.soekris.dk/download.html


New firmware and filters available
4k taps

randytsuch wrote: ↑Mon Dec 07, 2015 6:00 pm

gstew wrote:Thanks Randy.

I was figuring that with the posted current requirement for the positive rail of .18a @ 10v, I could probably get away with a 200ma max current at 1.2v. Doesn't sound right, huh?

TIA!

Greg in Mississippi

The positive takes 180ma, the negative takes 60ma. The negative is only for the ladder, so figure the positive ladder is also 60ma, so you need 120ma for the 3.3 and 1.2v regs.

I'll make a wag that 80ma is used for the 1.2v, and 40 is for the 3.3v.

The 3.3V reg is a linear, but the 1.2V reg is a switcher, so you need to account that.
I'll assume 80% efficiency for this switcher, which should be close.

.08a x 10v = .8va
.8va * 80% efficiency = .64va
.64va / 1.2v = .53a =530 ma on the 1.2v.


I used a LT1764 because I had them, but it was really easy to do. You just need some wire, a cap, and the 1764, see my blog for the details. It is a 3a part, and maybe overkill, but it works
http://randytsuch-audio.blogspot.com/20 ... dding.html

Now 2 1/2 years after Randy's excellent refined estimate, here I am getting ready to FINALLY modify the regulators on my Soekris DAM1021. Since then, has anyone actually measured the current draw of the 1.2V rail on the DAM? I'm planning to start out with a 1A-capable regulator, but hope I can move to a .5A one after I measure the actual current draw. IF anyone can provide an actual, that'd save me some time!

TIA!

Greg in Mississippi

gstew wrote: ↑Fri Jun 01, 2018 10:16 pm

randytsuch wrote: ↑Mon Dec 07, 2015 6:00 pm

gstew wrote:Thanks Randy.

I was figuring that with the posted current requirement for the positive rail of .18a @ 10v, I could probably get away with a 200ma max current at 1.2v. Doesn't sound right, huh?

TIA!

Greg in Mississippi

The positive takes 180ma, the negative takes 60ma. The negative is only for the ladder, so figure the positive ladder is also 60ma, so you need 120ma for the 3.3 and 1.2v regs.

I'll make a wag that 80ma is used for the 1.2v, and 40 is for the 3.3v.

The 3.3V reg is a linear, but the 1.2V reg is a switcher, so you need to account that.
I'll assume 80% efficiency for this switcher, which should be close.

.08a x 10v = .8va
.8va * 80% efficiency = .64va
.64va / 1.2v = .53a =530 ma on the 1.2v.


I used a LT1764 because I had them, but it was really easy to do. You just need some wire, a cap, and the 1764, see my blog for the details. It is a 3a part, and maybe overkill, but it works
http://randytsuch-audio.blogspot.com/20 ... dding.html

Now 2 1/2 years after Randy's excellent refined estimate, here I am getting ready to FINALLY modify the regulators on my Soekris DAM1021. Since then, has anyone actually measured the current draw of the 1.2V rail on the DAM? I'm planning to start out with a 1A-capable regulator, but hope I can move to a .5A one after I measure the actual current draw. IF anyone can provide an actual, that'd save me some time!

TIA!

Greg in Mississippi

the 1.2v is for the fpga only so its power consumption will be in the data sheet im sure

nige2000 wrote: ↑Sat Jun 02, 2018 10:15 am <SNIP>
the 1.2v is for the fpga only so its power consumption will be in the data sheet im sure

It's not clearly stated in the datasheet... Looks like the answer is 'it's complicated':

https://forums.xilinx.com/t5/Spartan-Fa ... d-p/692375

Thanks for the try, Nige. Anyone measured it in this application?

Later!

Greg in MIssissippi

Why do you want to move to a 1.2V 0.5A regulator when you have 1A ones already - better quality or something else?

jkeny wrote: ↑Sat Jun 02, 2018 7:54 pm Why do you want to move to a 1.2V 0.5A regulator when you have 1A ones already - better quality or something else?

Hey John... it's a size & flexibility thing. The original regs I planned to used on the DAM1021 was a dual 4||LT3042 board by now-absent DIYAudio poster OPC.. it is basically a small power supply and both takes up a lot of real-estate AND is difficult to position for short runs to both the 3.3V & 1.2V insertion points. The later regs are from ldovr.com, a small single LT3045 (.5A) for the 1.2V (hopefully) & a somewhat larger parallel LT3045 (1A) for the 3.3V. Both are still MUCH smaller (and flatter) than the OPC dual supply.

I can fit the larger ldovr.com board over the center of the DAM's FPGA/SI514 area and the smaller over the area where the negative supply caps and switching DC-DC converter used to be (after their removal). They'll still leave me access and room for later mods such as replacing the FPGA output caps and future power supply trials such as LiPoFE4's & ultracaps.

So what I plan is to initially use the 1A parallel module for the 3.3V AND the larger OPC supply for the 1.2V and measure the current. IF the 3.3V draw is <.5A and the 1.2V is >.5A, I can swap voltages on the ldovr.com modules and swap them. IF both draws are >.5A, I'll leave the OPC 1.2V connected temporarily while I order another couple of the parallel LT3045 from ldovr.com... but shipping times are long, so I hope I can use what I have.

Greg in Mississippi

On a lark, I decided to put in the ldovr 1.2V regulator with my meter in-line to measure the current. I mostly have CD rips, with very little hi-rez.

When powered on, it read about 50mA. Playing, about 100mA. While I can see some activities that could possibly increase the current draw dramatically, such as loading new firmware or playing hi-rez, I can work with those situations when I come to them.

The meter is out, it's playing & breaking in, sounding good, but I expect it'll be better in a couple of weeks.

Then to the next set of experiments.

Later!

Greg in Mississippi

Maybe we need a power supply thread?

But since we don't have one, saw this post at diyaudio
http://www.diyaudio.com/forums/pc-based ... ost5459137

Ian is comparing a ultra cap he found to a123 lifepo4 batteries.

Here's the cap link
https://www.mouser.com/ProductDetail/Ma ... NaMAW3s%3d

As stated in the thread, if you need higher voltages you have to put them in series, and then you have to add the ESRs up, so seems like you don't want to put a bunch in series.

Randy

randytsuch wrote: ↑Mon Jun 11, 2018 3:51 pm Maybe we need a power supply thread?

But since we don't have one, saw this post at diyaudio
http://www.diyaudio.com/forums/pc-based ... ost5459137

Ian is comparing a ultra cap he found to a123 lifepo4 batteries.

Here's the cap link
https://www.mouser.com/ProductDetail/Ma ... NaMAW3s%3d

As stated in the thread, if you need higher voltages you have to put them in series, and then you have to add the ESRs up, so seems like you don't want to put a bunch in series.

Randy

Just some thoughts:

These supercaps are only 2.7V (as are most supercaps) so two in series needed for 3.3V output - Mouser cost per cap is €13 so it's costing €26 to get battery equivalent. LifePo4 batteries can be had for < €10 i.e 2.5 times the cost of LiFePo4 battery

Pretty sure you can't just connect a 3.3V Voltage reg to these as when fully discharged they would represent a short to the voltage reg & possibly burn it out but interested in Ian's further experiements