PSU Dissected for Repair

Discuss SoundBridge Radio-specific issues here. For generic SoundBridge discussion, please use the SoundBridge General Discussion forum.

Moderator: RokuLyndon

Re: PSU Dissected for Repair

Postby Tony Bravo » Sat May 08, 2010 1:10 am

This post has been extremely helpful for the repair I did. My PSU failed some seven months after delivery. I sent it to Germany for repair under warranty (transportation costs on my own). The repaired unit failed again after less than one year of operation. So this time I look for a definitive solution.

I opened the case in which the PSU is enclosed. Due to this damn black glue this proved to be rather complex. To avoid breakage of plastic enclosure I did spend almost TWO hours to open it. To get an appropriate access to wiring I literally destroyed the rubber pass-hole elements of wires.

I forgot about repairing the original PSU. The 19 Volts would provided but a former laptop PSU and 9.5 V from a run-of-the mill variable voltage 1.5 Amp power supply. Then inside, I cut all power wires, less the white one (ground).I drilled a couple of holes in the case, I passed thru them the 19 V and the 9.5 V wires and subsequently I performed the connection to the power wires. I make a test with everything open...and Roku Radio came to life! By the way I realized that tighteness of sound box is not that critical for the quality of sound so when I closed everything I did not use any silicone or other sealant. Only in the power pass-thru holes. By the way, I adjusted the low voltage input to just 8.8 V and everything works fine. I think this is a definitive solution for this disgusting problem. As said. the post was extremelly useful to guide me inside this "terra incognita" Regards from Madrid - Spain
Tony Bravo
 
Posts: 2
Joined: Thu Mar 26, 2009 9:06 am
Location: Madrid - Spain

Re: PSU Dissected for Repair

Postby Burkhardi » Sat May 08, 2010 9:20 pm

Tony Bravo wrote:By the way I realized that tighteness of sound box is not that critical for the quality of sound so when I closed everything I did not use any silicone or other sealant. Only in the power pass-thru holes.


Good job and I am glad you sealed the pass through holes or you would have sound quality issues and the woofer would be flopping around like a fish out of water. Just taking one of the four screws out that holds the woofer down can casue the woofer to have sound and suspension issues from the hole (from the air leak, not woofer being loose, even with three screws the woofer is held snug do to a gasket that is glued to the speaker basket. You can hear the speaker bottom out and the bass is not as tight just from the 1/8" hole. You do need to seal the hell out of the enclosure when you put it back together; you may have gotten lucky and the black sealant may have made a pseudo gasket when you you pulled it apart. I have even seen sound quaity issues from not resealing the portion of the port where the two portions meet in the inside (the port starts on the inside of the front pannel and then goes toteh iside of the read pannel and then out the side. There the two mate together is what I am talking about.

Some people just use a 12v supply for the whole thign and then you loose about 25% of your volume, you make a good choice going with a 19v and a 9.5 (I guess 8.8v in your case).

Dr. Godehard Guenther from Soundmatters (and founder of A/D/S ) and Roku put a lot of work and thought into the sound quaility and for a woofer that small, it really sounds amazing (better the an Acoustic Wave IMHO) and it also another reason when sealing it up is so important. same goes for the left and right satellite speaker assemblies, they are pretty incrediable too (ported as well).

Once agian, good job and regards, Matt
*Roku N1000 and Roku HD1000 (Rev B) via ethernet on a Samsung HLP5674W DLP in the living room*Roku M1000 via Wi-Fi in the bedroom*Roku R1000 in the kitchen*XP Pro SP2 running WMC and Slimserver (for SlimRoku) routed via WRT54GS
Burkhardi
 
Posts: 1900
Joined: Tue Jun 07, 2005 11:06 am
Location: Austin, Texas ...Y'all

Re: PSU Dissected for Repair

Postby shorty » Fri Jun 18, 2010 3:37 am

Hi all,
i just replaced the original power supply with a external one.
I just took 18V and used this small device from conrad electronic to generate the 9V.
http://www.conrad.de/ce/PrintPage.html?method=showDetailPrintPage&productcode=167809&details=1&showDetailPrintPage.x=37&showDetailPrintPage.y=13#

It works perfect. The load is about 0.5A. Next step is to take this one
http://www.conrad.de/ce/PrintPage.html?method=showDetailPrintPage&productcode=511212&details=1&showDetailPrintPage.x=66&showDetailPrintPage.y=16#

and build it inside the case instead of the original power supply. Think this should work also.
All together it will cost 24 Euro. Think this is the best solution so far.
I will post again if it works.

A few pictures of my work:

Generating 9V from 18V.
Image

Just took little nails to get in contact. Don´t try this at home!
Image

Working with 18V and 0.5A load.
Image

If anyone in Germany has trouble with a dead power supply, you can contact me for more details.
shorty
 
Posts: 2
Joined: Wed Jun 16, 2010 3:53 am
Location: Munich - Germany

Re: PSU Dissected for Repair

Postby justme222 » Fri Jul 30, 2010 2:31 pm

Hello,
I discovered this EXCELLENT thread after being asked to repair my brothers
SB R1000. Unfortunately he had taken the SB to a repair shop that did a bit of \
a butcher job seperating the 2 power supply circuit boards.

They broke off one of the blue cirular ?( surge surpressor's) and there also appears
to have been some other component connected to the same spot.

The numbers that I got off of the blue components is CY /HT 102M 400V-250V X1 Y1

I would be greatly appreciative if someone could help me identify this component
and also the missing component.

Thank you.

Michael
justme222
 
Posts: 1
Joined: Fri Jul 30, 2010 2:22 pm

Re: PSU Dissected for Repair

Postby AnonymousCoward03 » Sat Jul 31, 2010 2:42 pm

wilf1651 wrote:Shorty

I took your advice and bought the component from Conrad - this does work but I seem to need to "short" the 19 v and 9 v outputs to get the device started. Once connected up, the output from the Conrad transformer reads only 2v until you short the two and kick start it. Did you have the same experience? How did you fix it?

Excellent thread.

Best wishes

Craig



Hi ! I just tried to repair my roku's PSU today with this stepdown converter and an external 18v PSU and i have EXACTLY the same problem.

The voltage is OK (9V) when i unplug the molex-like connector (the white one) but under load, the voltage fall to 2V. And if i "short" the 18v and 9v like you said, it works.
But, is there a solution ? I am a total noob in electronic :/


I am glad you posted this today, because i was about to trash everything in rage. And thx to everyone on this thread, very helpfull.
AnonymousCoward03
 
Posts: 1
Joined: Sat Jul 31, 2010 2:32 pm

Re: PSU Dissected for Repair

Postby Burkhardi » Sat Jul 31, 2010 3:50 pm

Sounds like it's one of the surge supressors.
I have these, check your PM in box for a message, Matt

justme222 wrote:Hello,
I discovered this EXCELLENT thread after being asked to repair my brothers
SB R1000. Unfortunately he had taken the SB to a repair shop that did a bit of \
a butcher job seperating the 2 power supply circuit boards.

They broke off one of the blue cirular ?( surge surpressor's) and there also appears
to have been some other component connected to the same spot.

The numbers that I got off of the blue components is CY /HT 102M 400V-250V X1 Y1

I would be greatly appreciative if someone could help me identify this component
and also the missing component.

Thank you.

Michael
*Roku N1000 and Roku HD1000 (Rev B) via ethernet on a Samsung HLP5674W DLP in the living room*Roku M1000 via Wi-Fi in the bedroom*Roku R1000 in the kitchen*XP Pro SP2 running WMC and Slimserver (for SlimRoku) routed via WRT54GS
Burkhardi
 
Posts: 1900
Joined: Tue Jun 07, 2005 11:06 am
Location: Austin, Texas ...Y'all

Re: PSU Dissected for Repair

Postby S80_UK » Sat Jul 31, 2010 5:15 pm

AnonymousCoward03 wrote:
wilf1651 wrote:....I took your advice and bought the component from Conrad - this does work but I seem to need to "short" the 19 v and 9 v outputs to get the device started. Once connected up, the output from the Conrad transformer reads only 2v until you short the two and kick start it. Did you have the same experience? How did you fix it?....

Hi ! I just tried to repair my roku's PSU today with this stepdown converter and an external 18v PSU and i have EXACTLY the same problem.

The voltage is OK (9V) when i unplug the molex-like connector (the white one) but under load, the voltage fall to 2V. And if i "short" the 18v and 9v like you said, it works.
But, is there a solution ? I am a total noob in electronic :/

I have not bought one of these devices, but since I am not a noob when it comes to electronics :wink:, the problem seems pretty obvious to me...

The 9 volt converter from Conrad is rated at 1 amp. Since it is a smart device it very probably limits the voltage and current if anything tries to exceed that rating. It is also likely that the current required on the 9 volt rail for the radio exceeds 1 amp while the unit starts up. The display in particular requires a lot more current when cold than once it has been on for a few seconds. The original supply could probably handle this surge. The 9 volt converter most likely is struggling. By shorting it (definitely NOT a good plan, by the way) you are relieving the converter of this start up surge, and after that it can handle the steady state current once things have powered up.

The correct solution would be to use a converter of a higher current rating so that it could handle the switch on surge without limiting the voltage and current.

Kind regards,

Les.
Roku M1000, M1001, M2000, R1000, Roberts WM-201, Stream 83i
S80_UK
 
Posts: 981
Joined: Wed Oct 25, 2006 3:11 am
Location: Cambridgeshire, UK

Re: PSU Dissected for Repair

Postby S80_UK » Sun Aug 01, 2010 7:40 am

Interesting idea...

You could. But it would be better to have the diodes only pull the voltage up to around 7 volts. That would take care of the start up surge. Then the regulator should be able to do the rest and pull the rail up to 9 volts, and at that point the current through the diodes will be more or less zero.

The other factor is that the diode chain will dissipate heat (current times volts) while the current flows through the diodes. (Not a lot of heat but some.) If the diodes to not pass any current once everything is up to voltage, then the regulator can do its job, and since it is a high-efficiency switched-mode regulator it will not be dissipating the same amount of heat.

(Switch mode regulators operate by taking in a certain amount of power at one voltage and releasing that same power at a different voltage with only small losses due to internal inefficiencies. Conventional regulators, diodes, resistors, etc cannot do this and reduce voltage by dissipating significant amounts of heat.)

Try a chain of about 10 1N4001 diodes or similar. See if the radio will start without the Conrad module in the circuit. If so, see what the voltage is on the 9v rail. Add or reduce the number of diodes to get somewhere around 7 - 8 volts. Then put the module back in circuit. The radio should then start up normally and operate with a correctly regulated 9 volt rail.

Hope this helps.

Les.

PS - I doubt that this will have any bearing on the quality of the display, but you never know.
Roku M1000, M1001, M2000, R1000, Roberts WM-201, Stream 83i
S80_UK
 
Posts: 981
Joined: Wed Oct 25, 2006 3:11 am
Location: Cambridgeshire, UK

Re: PSU Dissected for Repair

Postby S80_UK » Sun Aug 01, 2010 4:24 pm

Hi Craig,

Interesting that two devices did not completely solve the problem. My German is pretty poor, so I ran the item description through the Babel Fish...

"Step DOWN" - Converters with far input voltage section. The converter does not need a Mindestlast and no output condenser. It had an internal soft start (8 ms), is short circuit proof and pin compatible with 78xx-Spannungsreglern. The regulator controller makes even with high ambient temperature its maximum output current available without reservation."

So it has a "soft-start" circuit. That is probably what is tripping it up. The diode solution should work. You could probably also use a zener diode of suitable current and voltage rating to drop the voltage, but I'd go with the chain of diodes as probably cheaper and certainly more robust.

Les.
Roku M1000, M1001, M2000, R1000, Roberts WM-201, Stream 83i
S80_UK
 
Posts: 981
Joined: Wed Oct 25, 2006 3:11 am
Location: Cambridgeshire, UK

Re: PSU Dissected for Repair

Postby FrankB » Mon Aug 02, 2010 7:04 am

Hi,

If you are so inclined, I've repaired 2 radios using http://search.digikey.com/scripts/DkSea ... V7809-1000 successfully. I'm not sure why this would work where the Conrad doesn't but I've been running one radio since late May and the second for just a few weeks. So far so good.

Frank
FrankB
 
Posts: 108
Joined: Mon Sep 19, 2005 8:16 am

Re: PSU Dissected for Repair

Postby g0pkh » Tue Aug 03, 2010 2:33 am

Hello all

I agree totally with Les.

The Conrad device is NOT suitable as a replacement regulator for the Soundbridge Radio.

Sometime ago, a very kind chap offered me his radio for spares, after he tried a repair it himself. Using this module.

Whilst it looks very promising on paper, it cannot properly supply the load the Radio requires.
I have done a lot of testing on the radio which he gave me. The conrad device was only outputting around 2V on load.
However when I loaded the module up with a 10 Ohm resistor, It could supply 8.5V
With a 10 Ohm load at 9V it should be able to supply 900ma
which is less than what it is rated at. However even then it dropped to 8.5v. So clearly the device is not up to the job.
Quite why the output fell off totally to 2V when supplying the Radio is a mystery, and Les's explanation is the most likely,
the module probably includes a current foldback overload protection system.

If you are going to repair these radios using this idea, then a far more suitable regulator needs to be used.
Look for something rated at at least 1.5A.

Otherwise spend time stripping down the original PSU, removing all traces of the glue
both on top and underneath the board. Treat it to new caps and ALL components around the feedback circuit particularly those covered with glue, and problem solved.

Regards

Pete
Soundbridge and Photobridge repairs offered.
contact via PM or hillyp02-roku@yahoo.co.uk
g0pkh
 
Posts: 346
Joined: Sat Jan 21, 2006 12:04 pm
Location: Essex Nr. Stansted Airport.. UK

Re: PSU Dissected for Repair

Postby shorty » Tue Aug 03, 2010 12:26 pm

Hi everyone and sorry for the late answer,

i took the power supply from conrad http://www.conrad.de/ce/PrintPage.html?method=showDetailPrintPage&productcode=511212&details=1&showDetailPrintPage.x=66&showDetailPrintPage.y=16# and i had the same issue that the device did not start up.

Originally there was a resistor between the 18V and the 9V in the connector as you can see in this picture:
Image

In my new PS i also took a few resistors (each 0,5W) about 400-600 Ohm and switched them between 18V and 9V. I started with one resistor. Then test if the device starts up if you plug it in. If not, take another resistor and switch it parallel to decrease the value (or replace it). Then test again and so on. I don´t know the exact value of my parallel resistors but you can try it out. I built everything together so that i cannot look quickly inside to tell you and i don´t want to disassemble my device if i don´t have to :wink:
It works for over 1 month now.

Shorty
shorty
 
Posts: 2
Joined: Wed Jun 16, 2010 3:53 am
Location: Munich - Germany

Re: PSU Dissected for Repair

Postby S80_UK » Tue Aug 03, 2010 3:31 pm

There are two problems with using resistors to "fix" this problem.

1) they dissipate significant heat which is not ideal.
2) the problem with the start up failure is that it happens because of the need for a higher than average start up current. In theory it is possible that as the current through the resistor falls, the 9 volt regulator's output be pulled up above 9 volts, which may not be good for the regulator or for the radio.

I would not recommend anyone to use that method. Roku chose to, but they did it for a different reason - to slightly lessen the current through their power supply's regulator circuit - and they did it with the knowledge of that circuit and the radio that is was powering.

Good that your radio works, but I could not recommend your modification.

Les.
Roku M1000, M1001, M2000, R1000, Roberts WM-201, Stream 83i
S80_UK
 
Posts: 981
Joined: Wed Oct 25, 2006 3:11 am
Location: Cambridgeshire, UK

How critical is the 9.8V line level?

Postby jgtec » Wed Aug 04, 2010 7:10 am

Hi!

I have just joined this forum: what a great resource this is - Thanks!

I have read many of the posts in this thread and I am a bit confused. Here's my question - I am sure someone will be able to answer this for me.

Can I just use a standard 9.0V regulator IC or do I need to use an adjustable one and set it for 9.8V?

Cheers, John
Cheers, John

2 x M1001, 3 x M500 and 1 x ailing RB1000
jgtec
 
Posts: 11
Joined: Wed Aug 04, 2010 6:53 am

Re: How critical is the 9.8V line level?

Postby S80_UK » Wed Aug 04, 2010 11:57 am

jgtec wrote:Can I just use a standard 9.0V regulator IC or do I need to use an adjustable one and set it for 9.8V?


Hi John, and welcome to the forum.

The voltage is not critical, ans so a standard 9 volt regulator will be fine so long as it can deliver the current, withstand the input voltage, and so long as it can dissipate whatever heat is needed. If it's a standard linear regulator it will get quite hot (depending on the input voltage)and will need to be coupled to some kind of heatsink which must be able to get the heat away. Don't just put a heatsink in an enclosed space, since it can only cool the regulator if the heat has somewhere to go.

That's the reason that some of the discussion above concerns switched mode regulators - they dissipate much less heat and for applications like this a heatsink is generally not needed.

Regards,

Les.
Roku M1000, M1001, M2000, R1000, Roberts WM-201, Stream 83i
S80_UK
 
Posts: 981
Joined: Wed Oct 25, 2006 3:11 am
Location: Cambridgeshire, UK

PreviousNext

Return to SoundBridge Radio General Discussion

Who is online

Users browsing this forum: No registered users and 3 guests