wrt54 v1, too much voltage

[suckithard@gmail.com]

I've got a linksys wrt54g v1.0 that I've burned by using the wrong ac
adapter. I accidentally used a 10v dc adapter. It ran for ~10 minutes
before I heard multiple pops, and a smell of burning chips. Now, if I
plug in the original adapter nothing happens, the leds don't come on.
Cosmetically I don't see anything wrong with the board or components.

I've read the thread on the fried wrt54g because of ac. But I didn't
see a solution. From that thread, it seems that the buck converter
chip, and the diodes are probably the culprits. Is there anyway to
test them? or are there other components that seem likely too?

[Jeff Liebermann]

On Fri, 23 May 2008 12:04:52 -0700 (PDT), suckithard@gmail.com wrote:

>I've got a linksys wrt54g v1.0 that I've burned by using the wrong ac
>adapter. I accidentally used a 10v dc adapter.

Poof. The v1.0 mutation runs on 5V 2A adapter.

>It ran for ~10 minutes
>before I heard multiple pops, and a smell of burning chips. Now, if I
>plug in the original adapter nothing happens, the leds don't come on.
>Cosmetically I don't see anything wrong with the board or components.
>
>I've read the thread on the fried wrt54g because of ac. But I didn't
>see a solution.

No solution. I can identify the components that are fried, which can
be replaced if you like.

>From that thread, it seems that the buck converter
>chip, and the diodes are probably the culprits. Is there anyway to
>test them? or are there other components that seem likely too?

It's easier and cheaper to replace them.
Duz your v1 look like the v1 in the photo?
<http://www.linksysinfo.org/images/Content/Autopsy/WRT54G-mblg.jpg>
If ues, then it's an RT9202 regulator with external FET's to increase
the power output.
<http://www.richtek.com/www/Docs/DS9202-08.pdf>
I'm not sure about the numbers on the FET's

Or does it look like this?
<http://802.11junk.com/jeffl/pics/wireless/WRT54G/slides/wrt54g-v1.1.html>
If yes, it's an AP1501-33:
<http://www.diodes.com/datasheets/AP1501.pdf>

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558 jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com jeffl@cruzio.com
# http://www.LearnByDestroying.com AE6KS

[Kim Clay]

On Fri, 23 May 2008 12:04:52 -0700 (PDT), suckithard@gmail.com wrote:

>I've got a linksys wrt54g v1.0 that I've burned by using the wrong ac
>adapter. I accidentally used a 10v dc adapter. It ran for ~10 minutes
>before I heard multiple pops, and a smell of burning chips. Now, if I
>plug in the original adapter nothing happens, the leds don't come on.
>Cosmetically I don't see anything wrong with the board or components.
>

If the 10V adapter was a linear device (transformer inside) then its
output voltage may have been 12-15V easily. Or more!

>I've read the thread on the fried wrt54g because of ac.
Good job doing a search first!!

>But I didn't see a solution.
Replace
Find a used v2, v3, or v4 or a newer GL version.
Or a Buffalo WHR-HP-G54 if you can find one.

> From that thread, it seems that the buck converter
>chip, and the diodes are probably the culprits.
Yes - the smps & the TL7705 supervisory chip is all I could find that
ran directly from the +5V input.

>Is there anyway to test them?
Not easily.
Repair is probably possible but not worth the hassles!!

>or are there other components that seem likely too?
TL7705 - it might have survived - see above.

It could have been worse I suppose - the v1.0 only had a 125MHz cpu.
I think all other versions have at least a 200MHz.

Having said that, my v1.0 has been running dd-wrt nicely in an outdoor
enclosure for 9 months now.

kc

[suckithard@gmail.com]

>Duz your v1 look like the v1 in the photo?
><http://www.linksysinfo.org/images/Content/Autopsy/WRT54G-mblg.jpg>
>If ues, then it's an RT9202 regulator with external FET's to increase
>the power output.

It's definitely a 1.0, with the minipci card.
I found that the RT9202 isn't produced anymore, but I also found an
identical replacement.
I plan on ordering it, and the diodes from mouser.com. But the diodes
are a bit of a problem. I have no idea what specification they are.
All I know is that they are surface mounted. Do you guys know their
specs? or do diodes even have specs? I've only worked with the typical
generic ones from mechatronics labs, and I've never blown any of
those.

Also, are the FETs possible culprits too? Are they sensitive to higher
voltages?

>Not easily.
>Repair is probably possible but not worth the hassles!!

With the prices I've seen on ebay, repair may be the cheapest for me.
and I've got to break that new soldering iron in sometime. :)
I've been running newer versions of the wrt54g but they don't seem as
stable as the original.

[Jeff Liebermann]

On Fri, 23 May 2008 19:41:23 -0700 (PDT), suckithard@gmail.com wrote:

>>Duz your v1 look like the v1 in the photo?
>><http://www.linksysinfo.org/images/Content/Autopsy/WRT54G-mblg.jpg>
>>If ues, then it's an RT9202 regulator with external FET's to increase
>>the power output.
>
>It's definitely a 1.0, with the minipci card.
>I found that the RT9202 isn't produced anymore, but I also found an
>identical replacement.

Where? I'm interested.

>I plan on ordering it, and the diodes from mouser.com. But the diodes
>are a bit of a problem. I have no idea what specification they are.

Look on the schematics on the data sheet:
<http://www.richtek.com/www/Docs/DS9202-08.pdf>
Panasonic makes the MA732 diode on the schematic. Looks like a fast
shottky diode.
<http://www.datasheetcatalog.com/datasheets_pdf/M/A/7/3/MA732.shtml>

>All I know is that they are surface mounted. Do you guys know their
>specs? or do diodes even have specs? I've only worked with the typical
>generic ones from mechatronics labs, and I've never blown any of
>those.

Yes, diodes have specs. There are quite a few different types.

>Also, are the FETs possible culprits too? Are they sensitive to higher
>voltages?

It's highly likely that both FET's are also blown. I had numbers off
of those from the board, but I don't have V1 board handy.

>With the prices I've seen on ebay, repair may be the cheapest for me.

Ok. However, I value my time. What I do is buy the parts, stick them
to the unit, and wait until I have some of that mythical spare time to
work on it. Then, I do perhaps a dozen repair jobs at once. Setup
and cleanup time plus finding the parts is what kills me, not the
parts cost.

>and I've got to break that new soldering iron in sometime. :)
>I've been running newer versions of the wrt54g but they don't seem as
>stable as the original.

Duz the new version have a version number? I'm running v24 RC6.2 on
perhaps a dozen local wireless hot spots and business customers, with
zero issues. I haven't tried RC7 yet.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558 jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com jeffl@cruzio.com
# http://www.LearnByDestroying.com AE6KS

[suckithard@gmail.com]

> Where? *I'm interested.

The FAN6520. From comparing the spec sheets I believe this is a
replacement for the rt9202.
<http://www.fairchildsemi.com/ds/FA%2FFAN6520A.pdf>
I'm ordering the FAN6520AMX from mouser, 512-FAN6520AMX.

> Look on the schematics on the data sheet:
> <http://www.richtek.com/www/Docs/DS9202-08.pdf>
> Panasonic makes the MA732 diode on the schematic. *Looks like a fast
> shottky diode.
> <http://www.datasheetcatalog.com/datasheets_pdf/M/A/7/3/MA732.shtml>

I'll order a some of those too. But looking at the data sheet it has
the forward voltage at around 0.3-1v. Maybe I'm reading the circuit
wrong, but doesn't the wrt54g feed 5v through them?

> >Also, are the FETs possible culprits too? Are they sensitive to higher
> >voltages?
>
> It's highly likely that both FET's are also blown. *I had numbers off
> of those from the board, but I don't have V1 board handy.

From your previous thread you said:
>The associated FET's are 9965 or 9985. I have not been able to find sourcesfor any of these.
I'm not sure what the numbers mean. A quick google but didn't produce
anything useful.


> Ok. *However, I value my time. *What I do is buy the parts, stick them
> to the unit, and wait until I have some of that mythical spare time to
> work on it. *Then, I do perhaps a dozen repair jobs at once. *Setup
> and cleanup time plus finding the parts is what kills me, not the
> parts cost.

You sound like you are very very busy.

> Duz the new version have a version number? *I'm running v24 RC6.2 on
> perhaps a dozen local wireless hot spots and business customers, with
> zero issues. *I haven't tried RC7 yet.

I mean the newer versions of the router. I've got a v5, and v6. They
seem to need reboots more often than the v1. I'm not sure which
version of the dd-wrt they are running, maybe a firmware update is in
order. Those two were harder to get dd-wrt on, so once I had it
running I didn't try updating.

[Jeff Liebermann]

On Sat, 24 May 2008 07:49:37 -0700 (PDT), suckithard@gmail.com wrote:

>> Where? *I'm interested.
>
>The FAN6520. From comparing the spec sheets I believe this is a
>replacement for the rt9202.
><http://www.fairchildsemi.com/ds/FA%2FFAN6520A.pdf>
>I'm ordering the FAN6520AMX from mouser, 512-FAN6520AMX.
>
>> Look on the schematics on the data sheet:
>> <http://www.richtek.com/www/Docs/DS9202-08.pdf>

Looks like a good substitute. However, I didn't look at it very
carefully and may have missed something.

>> Panasonic makes the MA732 diode on the schematic. *Looks like a fast
>> shottky diode.
>> <http://www.datasheetcatalog.com/datasheets_pdf/M/A/7/3/MA732.shtml>
>
>I'll order a some of those too. But looking at the data sheet it has
>the forward voltage at around 0.3-1v. Maybe I'm reading the circuit
>wrong, but doesn't the wrt54g feed 5v through them?

I don't know. I haven't traced out the power circuit wiring and the
reverse engineered schematics previously posted are un-readable. I
don't think so, because in the data sheet schematic, the diode is in
the feedback circuit and doesn't carry the full +5VDC load.

>From your previous thread you said:
>>The associated FET's are 9965 or 9985. I have not been able to find sources for any of these.
>I'm not sure what the numbers mean. A quick google but didn't produce
>anything useful.

Those were the numbers on the FET's. I'll try again (later).

>I mean the newer versions of the router. I've got a v5, and v6. They
>seem to need reboots more often than the v1. I'm not sure which
>version of the dd-wrt they are running, maybe a firmware update is in
>order. Those two were harder to get dd-wrt on, so once I had it
>running I didn't try updating.

I've had mixed experiences with stablizing V5 and V6 routers. When
they first appeared, they both would hang erratically, requiring a
power cycle. I installed DD-WRT v23 on one or two V5 routers, which
did nothing. About 3 months ago, I tried again with a friends V5
router using DD-WRT v24 SP6.x. That worked and the router has been
stable ever since. However, he wanted more features, so we traded a
V5 for my V4 router, which has more ram and runs a more feature
infested version of DD-WRT. With such limited units and testing, I
can't claim that V24 fixes the hangs, but it sure looks that way.



--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

[Kim Clay]

On Sat, 24 May 2008 07:49:37 -0700 (PDT), suckithard@gmail.com wrote:

>> Where? *I'm interested.
>
>The FAN6520. From comparing the spec sheets I believe this is a
>replacement for the rt9202.
><http://www.fairchildsemi.com/ds/FA%2FFAN6520A.pdf>
>I'm ordering the FAN6520AMX from mouser, 512-FAN6520AMX.
>
>> Look on the schematics on the data sheet:
>> <http://www.richtek.com/www/Docs/DS9202-08.pdf>
>> Panasonic makes the MA732 diode on the schematic. *Looks like a fast
>> shottky diode.
>> <http://www.datasheetcatalog.com/datasheets_pdf/M/A/7/3/MA732.shtml>
>
>I'll order a some of those too. But looking at the data sheet it has
>the forward voltage at around 0.3-1v. Maybe I'm reading the circuit
>wrong, but doesn't the wrt54g feed 5v through them?

Here I have duplicated a small schematic from my May 23, 2007 post:
(use fixed pitch font)
+------>7705 chip
|
Jack1(+)--L2---F1---+-----+-----+-----+-----|>|--->SMPS(RT9202)
(-) | | | | 0.188V
| +-----+ | | |
| 1000uF| _|_ --- --- ---
L1 16V --- /_\ --- --- ---
| 105° --- | | | |
_|_ _|_ _|_ _|_ _|_ _|_
- - - - - -
ID= DS1 CK1 CA3 C3 D1

Device Zss smt smt smt Zss
ID 33 chip chip chip 33

Match the device name to the input ckt shown in this pic:
<http://web.archive.org/web/20040912091847/seattlewireless.net/~mattw/photos/linksyswrt54g/gallery/fullsize/IMG_3425.JPG>

D1 had Vfwd of 0.188V when I measured my working unit in 2007.
I don't remember if the measurement was taken using a VOM diode test
or if it was taken under actual operating conditions.

Looking at the above pic DS1, D1, & D2 are all the same.
At least DS1 should have survived. Maybe even D1 survived.

I am about 99.9% sure of my ascii schematic.

>
>> >Also, are the FETs possible culprits too? Are they sensitive to higher
>> >voltages?

FETs quite likely are blown.
I never found a data sheet for the FETs, or what they might cross to.

>>
>> It's highly likely that both FET's are also blown. *I had numbers off
>> of those from the board, but I don't have V1 board handy.
>
>From your previous thread you said:
>>The associated FET's are 9965 or 9985. I have not been able to find sources for any of these.
>I'm not sure what the numbers mean. A quick google but didn't produce
>anything useful.
>
In the above pic you can see the RT9202 gate drive leads from pins 2 &
4 going through some low R (<100R) gate drive resistors R5 & R6 to
another SOP-8 device which must be dual FETs.
One of the FETs gate is pin-4.
Trace a few more pins & maybe post a question over in
sci.electronics.repair - someone should know of a cross.

>
>> Ok. *However, I value my time. *What I do is buy the parts, stick them
>> to the unit, and wait until I have some of that mythical spare time to
>> work on it. *Then, I do perhaps a dozen repair jobs at once. *Setup
>> and cleanup time plus finding the parts is what kills me, not the
>> parts cost.
>
>You sound like you are very very busy.
>
>> Duz the new version have a version number? *I'm running v24 RC6.2 on
>> perhaps a dozen local wireless hot spots and business customers, with
>> zero issues. *I haven't tried RC7 yet.
>
>I mean the newer versions of the router. I've got a v5, and v6. They
>seem to need reboots more often than the v1. I'm not sure which
>version of the dd-wrt they are running, maybe a firmware update is in
>order. Those two were harder to get dd-wrt on, so once I had it
>running I didn't try updating.

kc - who just found a V2 with SpeedBooster for $8 :)

[suckithard@gmail.com]

Well I ordered the parts, and I've managed to buy another wrt54g v1
for $20. So in a week its time to solder.

I just picked up a wrt54gs v4. Do you guys know the voltage range of
this particular model? I opened it and looked, an half the chips
around the plug are covered in super glue.