S10 - power supply reseting issue explained - technical

[register1957]

Background: My new Openbox S10 (not a clone) would reset everytime I moved the dish.
Having an Electrical and Computer Engineering background I wanted to know why so I read the data sheets and application note for the SD4841 CURRENT MODE PWM CONTROLLER that runs the power supply.

Short answer is that the SD4841p was only designed to supply about 10W or power which is not enough power to run the electronics, LNB and a motor! The chip has internal current limiting that can not be altered with external components. Perhaps if they used the heavier duty SD4844 part it might have had just barely enough power to work.

So many of those that have reported seeing resets after a motor move - it is most likely the power supply that is badly under powered - not your firmware.

For those more technically inclined here are the links to the documents for the SD4841 chip

datasheet: SD4841 datasheet, Pinout ,application circuits Current Mode PWM Controller With Built-In High Voltage MOSFET

application note: http://www.micro-bridge.com/data/silan/SD484XP67K65_AN01Y.pdf

 

[pixl]

Reg,

There are problems with the S-10, but I think you are approaching the problem from the wrong direction.
The S-10 is rated to supply 500ma max at the lnb output. Tests I have done loading the S-10 with a variable power resistor show it is capable of this and then some.
A weakness in the S-10 and S-9 is a defective "resettable fuse in the lnb output.
See my post about this.
https://www.digitalworldz.co.uk/index.php?threads/283720/

 

[register1957]

Reg,

There are problems with the S-10, but I think you are approaching the problem from the wrong direction.
The S-10 is rated to supply 500ma max at the lnb output. Tests I have done loading the S-10 with a variable power resistor show it is capable of this and then some.
A weakness in the S-10 and S-9 is a defective "resettable fuse in the lnb output.
See my post about this.
https://www.digitalworldz.co.uk/index.php?threads/283720/

I should have said in my post - for those with resets not resulting from the poly fuse - sorry for the confusion. Not everyone with resetting motor issues has a bad poly fuse issue (as in my case). The poly fuse you are talking about is just one of the problems (and its great for people to know about it) - so my post focused on that issue of a poorly designed power supply.

One possibility is that you not have the SD4841 part I quoted or you are just lucky and working at the margins. The results you are getting are not by design of the SD4841 (that was my point in posting - ie that fundamentally the design was wrong for what it needed to do - the SD4841 should not technically be able to supply 500ma for the motor/LNB *from a 24V source* (it gets dropped by the LM317) and still have enough for the electronics and USB device on top of that. - it should not work - that it does in some cases is just luck of the draw.


In my own case I tested my supply directly on its 24V output - well before the poly fuse. Under 500ma load, monitoring both voltage and current, my supply will go into shutdown after a while - it is just at the very margin of working. First the output drops to about 21.5V then after 5 to 10 seconds it shuts down killing the 5V and 12V output causing the reset. I suspect that they changed the SD4841 at some point - is that the part you have? - if it is yours might just be working at the edge. The documentation for the SD4841 also agrees with my measurements - and my conclusion.

In summary if yours works you are just lucky and its not intended by design - unless you have another part, like say the SD4844 or a substitute by another vendor that has more current margin.

 

[pixl]

Ok, I see what what you are getting at. I looked in my OpenBox S-9 it uses a LD7550 PWM controller that has an output of 0.3 watts. But it appears to drive heat sink power device that then drives the transformer.

I don't have a S-10 here today to work with, but after looking at this photo of a S-10 I have to change my mind. I don't see this hardware. This is obviously a light design and I must agree, I don't see how it could work beyond powering the main board load.

.

 

[magore]

Hi,
Thanks for your reply! - that really helps - take a look at my pictures - I have a different power supply that yours

 

[magore]

Hi,
Here is a close up with the chip number visible


Here is the full image of the S10

 

[pixl]

In summary if yours works you are just lucky and its not intended by design - unless you have another part, like say the SD4844 or a substitute by another vendor that has more current margin.

Register,

Thanks for casting a new light on yet another weakness of these receivers. It might be interesting to do a before after loading test of the supply by putting in the SD4844 you suggested. I see it is rated for 16 watts and should supply the 9 watts that the lnb output would use at max 500ma. The photo Magore put up looks like the chip has overheated and discolored the pc board.

 

[slugworth]

old s10,judging from the plugin ethernet.
mine is newer and has the 4844

 

[pixl]

That is interesting. So I wonder if it is safe to suggest that the SD4841 should be replaced with a 4844.
It also explains why some S-10 are ok and others are troublesome.

Slugworth what are you doing up so late? It's almost midnight here in the US.
Must be 4am for you!



.

 

[slugworth]

That is interesting. So I wonder if it is safe to suggest that the SD4841 should be replaced with a 4844.
It also explains why some S-10 are ok and others are troublesome.

Slugworth what are you doing up so late? It's almost midnight here in the US.
Must be 4am for you!



.

I was playing with foreign firmware on my s10 and forgot to sleep
New England, not the jolly old one.

 

[slugworth]

That is interesting. So I wonder if it is safe to suggest that the SD4841 should be replaced with a 4844.
It also explains why some S-10 are ok and others are troublesome.

That assumes the 4841 was the only part changed,and not other resistors,caps,diodes etc.
Somebody with samples of each would have to compare.
I would hate to see somebody burn their house down.

 

[magore]

Hi,
Yes - the transformer and the feed back network are tied to the design - it would be interesting to compare all of the parts.

 

[slugworth]

it would also be interesting to see how the power supply reacts with different input voltages.U.S. vs Euro.

 

[fubarito]

So who's gonna try the swap first?? I do have an s10 but I am kind of hesitant on doing it myself. Hey Pixl where do you think I can buy the SD4844 in the USA?

 

[pixl]

I'll try calling Digi-Key next week and report back, or you could too. Don't bother with Digi-Keys web page, doesn't work that well, just call them and ask for a cross equivalent part. The people there are nice on the phone.



Electronic Components Distributor | DigiKey Corp. | US Home Page

or

http://www.digikey.co.uk/


.

 

[slugworth]

doesn't look too promising....

Find SD4844 Stock and Compare Prices Across the Most Reputable Distributors in the Industry.

 

[magore]

Hi,
Just a quick note - I changed my SD4841 to a SD4844 and it works - my dish motor runs fine with no resets.
I was able to obtain the parts from Shenzhen Jbest Electronics Firm via alibaba.com - searched for sd4844.
The power supply runs cool enough that I am not worried at this point.
As you mentioned however I would not suggest this as a general solution because we don't know if the transformer was down-rated for the lower power chip ( I think not in my case - but no guarantees for anyone else) Keep in mind that my background is in electrical engineering and I can do this on my one with some safety. Changing parts like this on your own will invalidate the electrical certification it has (if it actually has any) and this would likely invalidate your home insurance if it ever caught fire ( I am serious about that - you can look it up). My goal was to find out what happened to the design of this unit - clearly at minimum this part was changed.

 

[magore]

PS - Hi slugworth,
My comments about electrical safety below were not directed at you but everyone reading this - I really should have looked at the wording context a bit more. Thanks for your comments on this topic! - I hope my input also helps others trying to understand the various issues with their S10's. Looks like we have power supply and or polyfuse and or voltage settings on the LNB power (via LM317) etc as sources of trouble from what I have been reading so far.

 

[slugworth]

each power supply has a date of manufacture on it.
I thought I saw a picture of a power supply with the old SD4841 that was actually made after the power supply I have with the good part (09-2011) which to me makes no sense and would make it tougher for people with s10's.There would be no guarantee that a power supply made after a certain date had the correct part.

 

[slugworth]

I have swapped out power supplies on other older brands of receiver before without incident.
Actually,the s10 power supply would be similar to pc voltages except for the +24v