SamyGO

LE40B653T5W. While TV on DVB-C channel, Picture mode Normal (default):

Energy saver set to: So, its almost triple energy consumption compared to standby mode.
One conclusion i made for my self: it is worth to use Energy-medium, because it reduces power consumption by ~40W (~28%)

juuso wrote:LE40B653T5W. While TV on DVB-C channel, Picture mode Normal (default):

Energy saver set to:

Code: Select all

OFF(~Auto)    140W 0.63A
Low           117W 0.53A
Medium        101W 0.45A
High          89W 0.39A
With LCD off: 35W 0.15A
StandBy       13W and 0.06A

Sh*t, 35W is quite a lot! I think my notebook has 5-10x more processing power running on 40W PSU.

Six months ago I got a spare B650 mainboard for free (well, almost: shipping was 12EUR, and I had to clone main EEPROM settings from the working one). Today I finally did some soldering, to power this mainboard from ATX power supply - it's working all right . I did some measurements: on idle ("no signal", exeDSP killed, DTV tuner/demodulator probably still operating - tuner temp is a little above ambient), this mainboard is drawing 1.1-1.4A from 5V line and 0.6-0.8A from 12V line (~15W in total).

Thermal dissipation on idle (from highest to lowest):
1) Chelsea chip (SDP93): heatsink is getting noticeably hot after 1-2 minutes
2) Various step-down DC/DC converter circuits on the mainboard
3) Arsenal chip
4) Tuner/DTV demodulator combo
5) DRAMs

Measured standby power consumption is shockingly bad (Samsung is quoting 0.3W for B650's, and 0.7-0.9W for B7000/B8000's). I really doubt micom (along with a couple of peripheral chips powered from A5V standby) are even capable to dissipate more then 3-4W in total.. Is Samsung PSU efficiency really that bad while on standby??

sbav1 wrote:Measured standby power consumption is shockingly bad (Samsung is quoting 0.3W for B650's, and 0.7-0.9W for B7000/B8000's). ...Is Samsung PSU efficiency really that bad while on standby??

I did not test if for myself, but I think standby power consumption is much greater with Anynet/HDMI-CEC enabled. Try to disable Anynet+ and measure again.

I don't have anynet devices so it is dissabled by default. I know about 0.3W (declared by Samsung) I should get in stand by mode and wondering why do I get 13W instead.
I`m using UPM PM777 tool for measurements. I don`t know how accurate it is, but i trust him.

I found some schematics for 37" B650 power supply (attached).

Quick & dirty, not-very-scientific approach: in my TV (LE37B650), with power cord unplugged, standby LED stays lit for about 45 seconds (35-40 seconds while receiving IR signals). Estimated energy stored in main capacitor (150uF, on "HOT" side) is ca 8J; another ~2-3J is probably stored in capacitors on "COLD" side (5.3V, 12/13V lines). That means total standby circuit[s] power consumption is most likely bellow ~0.2-0.3W (???).

Unfortunately, that's not telling us much about Samsung PSU efficiency for very low loads.. Personally, I don't quite believe in Samsung "0.3W standby" statement; their PSU design looks quite average.. Additional/integrated sub-unit for standby power will probably cost ~3$ more; I guess that was a big no-no for bean counters .

But how do i get this 0.3W...?
I get 13W instead. As i remember first time i bought this measurement tool a couple of months ago, it showed also same 13W.
Is here any service menu parameter to set to? Somewhere i red about some different power off modes. Don`t remember exact.

juuso wrote:But how do i get this 0.3W...?
I get 13W instead. As i remember first time i bought this measurement tool a couple of months ago, it showed also same 13W.

I guess our findings so far are quite inconclusive..

I asked some guys from engineering department ; it was like: blah-<something-understandable>-blah<something-understandable>-blah-blah:
1) until recently, sub-par PSU efficiency at low loads was "de facto" industry standard (!!!),
2) cheap power meters are not very reliable, especially for low & complex (non-resistive) loads,
3) "PSU on it's last legs" / bad capacitors (?).