Deep Standby

[adm]

I have a new Miracle Premium Twin HD with two terrestrial TV tuners for the UK Freeview service and fitted with an internal 1Tbyte hard drive. The box is running VIX Apollo build 030, Kernel 3/6/0, dated 7 September 2014

With approx 5 hours of playing around and going through all the set-up menus all things appear to work except 'deep standby'.

When the box is put into deep standby the blue LED on the front panel lights. When attempting to wake it up by pressing the power button on the remote one of two problems appear:

1) The front panel on the box continually displays 'Miraclebox' but fails to boot at all. It will remain in this state forever. The only way out of this failure is to turn the box off at the mains switch. The box will then boot up faultlessly on re-instating the mains power.

2) The box displays 'Miraclebox' then 'booting' and in the normal way, after a few splash screens, a TV picture appears BUT the hard disk is missing and the wired network connection is completely broken. When in this state pressing the restart option from the power menu will get the box working again.

This means that the box is somewhat useless as a PVR if placed in deep standby.

A quick Google suggests this is a common problem with these boxes but I haven't seen anyone report a fix.

Is there a fix?

[Joe_90]

That's odd, but you're the second poster with this issue. I have the same model since June and it's been faultless. Did you purchase from WoS?

[adm]

Yes, purchased from world-of-satellite.co.uk and delivered Tues 9th Sep

[adm]

Test 1
Internal hard disk connected, Wired network connected
Boot after deep standby results in hard disk missing, network connection broken

Test 2
Internal hard disk disconnected, Wired network connected
Boot after deep standby results in network connection broken

Test 3
External USB disk connected (no internal disk), Wired network connected
Boot after deep standby results in disk OK but network connection broken.

The problem seems to be 'internal' functionality and USB interface for disk is OK

Observation: If the box is in deep standby for less than 2 minutes it will not boot. The front panel will just say Miraclebox. If left in the deep standby state for more than 2 minutes the box will actually boot but both the hard disk and network faults will always occur.

When the network problem occurs the light is on the network port is endlessly flashing a very high rate. This doesn't happen on a successful boot that results in the hard disk and network working properly

[Sicilian]

Did you initialise your HDD with the receiver?

Have you left DHCP Enabled? Are you able to test connected to another router? Also try setting a static IP.

No its not a common issue, only 2 reports here. I've tested myself yesterday and could not replicate.

[Sicilian]

Tested again here this morning, all fine.

Do you have RS232 by anychance on a PC? Would be good to have a bootlog from your unit.

[adm]

Some answers
The disk was formatted in the Micaclebox. I also tried the convert from Ext3 to Ext4 file format option from the menu. The box didn't complain about the disk. The fact that the disk seems to work flawlessly from booting from a mains switch on, from the 'light' standby and from a restart suggests that it's not a formatting issue. The disk fails to be seen after a deep standby. If it wasn't for the network also failing I may suspect a disk.

This morning I had a few hours to spare and have reformatted the disk (in the Miraclebox) and factory reset the box back to all its default settings. I then did the minimum to get a TV channel and a 5 minute recording. 99% of all menu items are still in their default sate. Putting the box into deep standby still produces the same hard disk and network fault.

Another factory reset back to default settings and the network set-up again and working. DHCP now turned off. After a deep standby the same problems. No hard disk , network broken – as indicated by no communication and the network test failing on IP address and nameserver. The correct values are still seen in the miracle box network configuration.

The hard disk format and network settings must be OK because the box subsequently works without problems after another mains on/off boot.

I haven't got another router to try but I swapped out the connection with another piece of working equipment. I have half a dozen other equipments connected to the router via cable or wi-fi and none give any problems. The miracle box doesn't give problems EXCEPT when waking from a deep standby.

I don't think the boot log is going to help except in establishing that I have a faulty box. (All faults are repeatable.)

A RS232 output during a power on boot reads in English and gives a working box.
A RS232 output during a deep standby boots reads in English and garbage.

It's looks as if at least 3 things are going very wrong after a deep standby.

I've also checked the boot log file with the internal disk disconnected and network cable disconnected. Power on boot gives an OK log file, deep standby boot gives a rubbish log file.

It's looks like a faulty box, possible a hardware fault in a power switching circuit that is ONLY used during deep standby.

Small snippets of good and bad log files below

Example of 'good' boot

------------------------------------------------------------------------------
BCM97413B1 CFE v2.26, CFE core v2.12, Endian Mode: Little
Build Date: pon, 4 lut 2013, 10:46:52 CET (a4tech@speedy)
Copyright (C) Broadcom Corporation.

Enabling NAND flash at CS1, NOR flash at CS0
NAND Flash size = 512MB
DDR Freq: 396MHz
CPU speed: 405MHz
Memory Config: 64-bit UMA
Device Tech: 1Gb
Total memory: 512MB
Boot Device: NOR
Total flash: 4MB

Initializing USB.
USB: Locating Class 09 Vendor 0000 Product 0000: USB Hub
USB: Locating Class 09 Vendor 0000 Product 0000: USB Hub

CFE initialized.

[FLASH-TOOL] Searching update files...

[FLASH-TOOL] Could not init file system: Device not found

[FLASH-TOOL] There are not any update files...

Booting...
Starting splash screen.
************************************************** *
Found splash image - Width = 720 Height = 576
************************************************** *

Loader:elf Filesys:raw Dev:nandflash0.kernel File: Optionsnull)
Loading: 0x80001000/8161344 0x807c9840/123920 Entry address is 0x8061ecd0
TP1 Entry Address at 0x80000ffc = 8061ecd0
Starting program at 0x8061ecd0
--------------------------------------------------------------------


Example of 'bad' boot file

---------------------------------------------------------------------

BCM9·413B1 CÆE ö2.26, CÆE óore v2.±2,°Endian Íodõ: Littlå
Build Daôe: pon, 4 ìut 2013¬ 1°:46:52 ÃET°(a4techÀspõedy)
Cïpyúight (C© Broadcom ãorporatéon®

Eîabling ÎANä flash át ãS1, NOR*flñsh at CÓ0
NAND Flásh size = µ12MB
ÄDR°Freq: * °396MHz
CPU speed:* 405ÍHz
Memory*Coþfig: 64*biô UMA
Dåvióe Tech:* 1Gb
Totál memory: *512MB
Âooô Deviceº °NOR
Toôal°flash: * 4íB

Initéalùzing USÂ.
UóB: Locaôin÷ Class °9 öendor 0°00 Product*0000: UÓB èub
USB: Loãatùng Clasó 09 Vendor*0000 Produãt 0000:*USâ Hub

CFå initiaìized.

[FìASH-TOOÌ] óearchinç uødate fiìes...

[FLASH-TOïL] Coulä nït init æilõ systemº Dõvice noô fÿund

ÛFLASH-TOOLý There are*not any*upôate filås.¾.

Boïtiþg...
Sôarüing splásh screen.
*****ª**********ª**********ª*******ª**º*******ª**º *
Founä sðlash imáge°- Width*= 720 Heigèt = 576
*ª*******ª**º*******ª**º*******ª**º*******ª******* *

Loaäerºelf Filåsyó:raw Deö:nñndflash°.kõrnel Fiìe:°Optionsº(null)
Loáding: 0x80°01°00/8161³44°0x807ã98´0/12392° Entry addråss is 0x80¶1ecd0
ÔP1°Entry Aädrõss at 0ø80°00ffc =*80¾1ecd0
ótarting prïgram at*0x¸061ecd0


-------------------------------------------------------------------------------

[Joe_90]

The text output from the serial port during a "bad" boot looks like what you would get with parity errors. Maybe there is an issue with the power supply which manifests only during deep standby as you suspect?

[Leader]

Hi adm. I am having the same problem as you and have a thread running on this too at http://www.world-of-satellite.com/sh...rom-deep-sleep.

I been in contact with WOS today and they have already a box coming back to there store at the moment with the same problems as we have got and they going to test it, and they will let me know on Tuesday of the outcome.

There is something weird going on in deep standby and make be Tony as a point here than the power supply may not be switching to full power to boost the box back to live.

Myself personally was thinking it was a bootloader issue, but now looking at your boot files I have seen these kind of results before when I use to build and program computers, and it all come down to a bad power supply.

So this may be something WOS could test and look into too.

I will keep you posted.

[adm]

It may be my box they are referring to - its on the way back to WOS

It may just be power sequencing. I'm in the electronics industry as a designer and I've seen other examples where incorrect power sequencing can cause lock-up and unexpected behaviour. The power sequencing from a cold boot and a deep standby will be different.

[Leader]

I been testing my mirclebox this morning, working with the power supply. As Tony more or less knocked the nail on the head. and it more or less comes down to this.

These boxes must have a slow charge capacitor inside which is trickle charged with 1 Watt in deep sleep mode. The box seems to need a power boost of power to start the box up. If this capacitor is not fully charged when you remote on the box it just displays 'Miraclebox'

This capacitor seems to take about 10 to 15 mins to charge fully. if you boot your box close to the 10 mins it will start booting but with garbage output in the log file using rs232 and as adm states on is thread http://www.world-of-satellite.com/sh...559#post311559. the hdd and internet may not boot until complete power off using the rocker switch.

My box as to be in deep sleep for a Minimum of 15 minutes to boot with internal hdd and internet connection to boot successfully with no problems no fault codes in the rs232 log. which really deep sleep is only going to be used for overnight power down.

If this is what the problem is we have been all encountering, it would be nice if the User Manual would state this That Deep Sleep requires a Minimum of 15 minutes wait period before restarting. or the box may not function correctly.

Hope this helps as now I know what the cause was I now see no problem with my box.

[Joe_90]

Good stuff @Leader. I hadn't realised that the problem was manifesting itself only when in deep sleep for a short period. Seems likely that your suspicion of the capacitor might be correct and perhaps WoS will confirm in the case of @adm's box. I'm on a short holiday, so can't verify with my own model until mid next week. The only thing I noticed with my own box was when I was flashing new images from USB stick. After putting the box into deep sleep you are supposed to power off at the back and insert the USB stick and power on again. I found that I had to wait at least 20 seconds after powering off at the back (and that bright blue LED had extinguished!) before powering on again, otherwise it would not invoke the bootloader check of the SB stick. I had assumed from that behaviour that the reservoir capacitor was actually fully charged and it took that long to discharge. That blue LED is a major annoyance I find. I've had to put some black tape over it to dim the output.

[adm]

This capacitor seems to take about 10 to 15 mins to charge fully. if you boot your box close to the 10 mins it will start booting but with garbage output in the log file using rs232

If true it may indicate that the power supply circuits are relying on capacitor leakage currents. Give the box a year or two and the capacitor leakage changes and the box will never boot from deep standby. It still suggests to me a fault, possibly during manufacture where a wrong value component has been fitted or a build quality issue (duff solder joint)

I could get the box to boot within 2 minutes (with a garbage boot log) and it made no difference if I left the box for longer than 10/15 minutes.

I assume that when others in this thread have said their boxes have no problem they are not waiting 15 minutes during testing. Can someone with a working box confirm that it comes out of deep standby within a short time or does it require to be in this mode for more than 10 minutes?


Although I didn't try it before I returned my box to WOS I'm assuming that the box comes out of standby to make a timer recording. If the timer happens to be within the long period that the box is in deep standby it will come out of standby in the faulty condition and fail to record as the disk is missing. Once in this fault condition what now puts it right? An on/off power sequence?

[adm]

On another (unrelated) forum where a number of technically minded people are migrating to the Miraclebox family I have asked about the behaviour of their boxes after/during deep standby.

So far one reply with someone with the same box and same VIX image. From entering to exiting deep standby 1 minute - with no problems.

[Leader]

I could get the box to boot within 2 minutes (with a garbage boot log) and it made no difference if I left the box for longer than 10/15 minutes.

I assume that when others in this thread have said their boxes have no problem they are not waiting 15 minutes during testing. Can someone with a working box confirm that it comes out of deep standby within a short time or does it require to be in this mode for more than 10 minutes?


Although I didn't try it before I returned my box to WOS I'm assuming that the box comes out of standby to make a timer recording. If the timer happens to be within the long period that the box is in deep standby it will come out of standby in the faulty condition and fail to record as the disk is missing. Once in this fault condition what now puts it right? An on/off power sequence?

On testing my box this morning my box would not boot within the first five minutes.

I will try see if it will come out of deep sleep to do a recording test and let you know.

WOS will hopefully give you some answers to your box soon and please post back what the verdict is.