I too was wondering what this third center pin is about. No voltage
present to it, no resistance, no capacitance; got me wondering what
kinda surprise did Dell put in the power supply that they don't want
us to know about. So curiosity got the best of me and I very carefully
pried the glued case apart.
Standard switching power supply, but then near where the wire enters
the case and is soldered to the circuit board resides a small TO-92
device. First thought was maybe a transistor that sinks a certain
current upon attachment but after further inquiry, only two leads were
connected. This led me to reverse engineering this small circuit. It
turns out theres a 131 ohm resistor in series with the center pin wire
of the DC connecter and one pin of the 'mystery device' and then there
is a reverse biased diode going to ground. The other pin that is
connected of said device is also connected to ground (by reverse
biased I mean that the diode will only conduct if a positive voltage
were applied to the ground connection of the power supply and a
negative (ground) were connected to the center pin of the DC power
connector. This lead me to deducing, because of the series resistor,
that this diode was a zener of currently unknown breakdown voltage.
The next step was to determine the true identity of the 'mystery
device.' The part number read "Dallas 2501 (then a date code)." Dallas
being dallas semiconductor (aka Maxim IC). A search yielded only a
very incomplete datasheet refering me to the DS2502 which is a 1kbit
one-wire EPROM version. The "2501" was a DS2501 of 512 bit data space.
The datasheet gives specifications to a max programming voltage after
EPROM write instruction of 12V. This means to protect the device from
overvoltage this zener diode connected to the pins must be a 12V zener
and the sereis resistor being a current limiter protecting the diode
in the event that the inner barrel and center pin were to come into
contact.
The DS2502 and 2501 (1kbit and 512 bit respectively) use Dallas Semi.
1-Wire (R) communication protocol. It gets its power from the data
line and when the data line is low a diode protected capacitor supplys
power for its logic circuits, Parasite Power. This means that to
communicate with the DS2502/1 one only needs two lines, a data line
(logic high idle state) and ground. The power to the data/power line
is supplied by the master through a 5k ohm resistor for short cable
lengths.
Hope anyone reading this that wants to make his/her own power
converter finds this information usefull (insert disclaimer here; ie.
use this information at your own risk, I am not to be held responsible
if someone else's equipment gets fried b/c of poor design, I only
described how it works and make no claims to it being my own design
giving rights of design and operation to Dell and/or LiteOn (written
on power brick) and any other engineering firm/company/manufacturer
that was involved in the design of the motherboard, power brick and
any other associated equipment, etc.)
002100: 11 17 E1 6C 02 00 00 C2 ...l....
002110: 44 45 4C 4C 30 30 41 43 DELL00AC
002118: 30 39 30 31 39 35 30 34 09019504
002120: 36 43 4E 30 44 46 32 36 6CN0DF26
002128: 36 37 31 36 31 35 36 32 67161562
002130: D0 FF FF FF FF FF FF FF ........
002138: 41 30 30 85 90 FF FF FF A00.....
002140: FF FF FF FF FF FF FF FF ........
002148: FF FF FF FF FF FF FF FF ........
002150: FF FF FF FF EA FF FF FF ........
002158: FF
at memory address 2130, D0, is a CRC (8-bit, x^8+x^5+x^4+1 polynomial)
of address locations 2110 - 212F that was calculated by the DS2501 as
the data was read in by the bus master.
I was unable to get any useful data after 213C but then it read a byte
at location 2155. The cause of this is unknown and there was no valid
CRC calculated...this is still under investigation.
As far as the memory dump....Dell (manu.), 00 I'm guessing is an
identifier, AC describing what type of adapter, 0 another identifier,
195046 being 19.5 volts and 4.6 amps. From CN to 62 this is the
product identification number found on the barcode and A00 is the
revision code 00 being the important part b/c I obtained a second 90
watt power supply and it too and an 'A' appended before the numerical
portion of the revision number.
Hope anyone reading this finds the information useful.
(Same as before insert disclaimer here....)
002100: 11 36 14 6D 02 00 00 B8 .6.m.... = ROM
002108: F3 FF FF FF FF FF FF 00 ........ = STATUS
002110: C3 FF FF FF FF FF FF FF ........ = STATUS CRC (=C3)
002118: 44 45 4C 4C 30 30 41 43 DELL00AC = Start of page 0
002120: 30 39 30 31 39 35 30 34 09019504
002128: 36 43 4E 30 43 38 30 32 6CN0C802
002130: 33 34 38 36 36 31 36 31 34866161 = End of page 0
002138: 52 32 33 48 38 41 30 33 R23H8A03 = Start of page 1
002140: 4D 7C FF FF FF FF FF FF M|......
002148: FF FF FF FF FF FF FF FF ........
002150: FF FF FF FF FF FF FF FF ........ = End of page 1
002158: FF FF FF FF FF FF FF FF ........
So here we have the first line being the ROM. 11 is the family code,
the serial number is: 0000026D1436 and B8 is the hardcoded CRC reality
check byte
The next line is the EPROM status register. The first byte is the code
protection bits (bits 0-3) and memory usage map (bits 4-7, reserved
for TMEX) the next 4 are page redirection bytes, bytes 6 and 7 are
maintained clear/unprogrammed and byte 8 is programmed to 00 at the
factory. C3 is the CRC computed by the DS2501, this was checked by
using the program I wrote in windows for computing the CRC
(x^8+x^5+x^4+1), and was determined to be valid. The next five lines
are the actual data space information. Dell is the manufacturer, 0 is
a field identifier, AC is the type of adapter, 090 is the wattage
rating (my thought is that it allows for 99+ watt power adapters),
195046 is the nominal voltage rating (19.5 Volts) followed by the
nominal current rating adjusted to one decimal place (4.6 Amps). From
C to H8 is the barcode number of the power adapter and A03 is the
revision of the power adapter. 4D7C is an end of information
identifier as I have found this on another adapter I own. The one
thing I was unsuccessful in acquiring was the computed CRC at the end
of the data space that the memory chip is supposed to send as a 65th
byte (which is sent at the end of the data space, this was assumed to
be the same communication protocol as is used for the DS2502).
Again I hope this helps answer any inquiries anyone may have as to the
actual data stored in the memory chip.
I'll be happy to answer any technical questions anyone may have about
this. Please only technical questions, I am not offering design
services and any inquiries of that type will be immediately ignored
and any additional blocked permanetly if I so deem them to be a
nuisance.
(insert additional lenghty disclaimers here...)
Can you also describe how you were able to read/write to the 2501? What
eeprom reader did you use? It would be awesome if you can provide the
details necessary to program up one of these ICs so that I can obtain a
replacement and get my adapter working properly again.
BTW, when the Dallas 2501 chip is damaged, the laptop (in this case a
Inspiron 8500) runs OK off mains power, but the battery won't charge.
The BIOS also complains about not knowing what type of adapter it is,
with the message "The AC Power Adapter type cannot be determined. This
will prevent optimal system performance".
Thanks for your assistance
PS
No thank you to Dell for making such easily broken products. Why can't
you guys at Dell design your systems to use a normal AC adapter without
the need for silly detection circuits?
'Ejay' (http://ejay.com.au)
Thanks for effort on this, my PA10 cord got frayed and looks like the
brief short took out the DS2105. I opened up the power supply and I do
not see the Zener and resistor internal to the supply maybe they moved
that circuitry into the laptop. In my PA10 the DS2105 appears to be
wired without any supply to it atleast internal to the power supply.
The Zener that you mentioned was it internal to the power supply or
did you trace it to within you laptop?
Thanks
Mike
I'm also curious if anyone has found a way to bypass this problem.
Conceivable methods might include modifications to the motherboard, or a
hacked BIOS. Let me know if you have a solution.
'Ejay' (http://ejay.com.au)
On Jan 28, 7:49 am, ejay <ejay.33y...@no.email.invalid> wrote:
> Mine didn't appear to have the zener anywhere. Thecenterpinof the
Would be nice if Robert could explain in detail how he did it.
I'm sure there are heaps of Dell adapters out there that have suffered
the same fate. It's very rare for an incorrect adapter to 'harm' a
battery anyway - the onboard charging circuitry knows how full the
battery is, and will apply the charge accordingly. Not to mention the
proprietary plug, which basically precludes the use of any other
adapter, so it's not a scheme to protect the battery - but rather a way
for Dell to screw the customers for more $$ when the adapters break.
'Ejay' (http://ejay.com.au)
are you sure the problem is the adapter and not the computer?
it is a common and unfortunate problem with many model laptops that the
connector in the computer becomes loose and so has poor contact with the
system board. a simple test of this condition is jiggle the plug on the
computer end while it is inserted into the laptop and watch to see if it
switches between battery and ac power. this should be covered under
warranty or there are many shops that will re solder the ac jack back on
the system board for a flat fee. depending on the age of the system it
may not be worth the expense. having said that, genuine dell pa10
adapters can be had from ebay for about $20 delivered. clone "for dell"
adapters can be had on ebay for close to $10 delivered but they are junk
and will not be as durable or long lasting.
Well that is a tip-off for sure. But if it doesn't, it doesn't mean it
isn't the jack itself or the solder connections to the board. As
sometimes the jack can break (and it is still can be solid as a rock) or
a solder connection or something... and no amount of wiggling will make
it work for some of them.
--
Bill
Gateway MX6124 ('06 era) 2 of 3 - Windows XP SP3
Can you elaborate. Lately, my Inspiron does switch from battery to AC when I
jiggle the chord a bit.
What does that indicate?
Is it easy to fix?
Mel
"Christopher Muto" <mu...@worldnet.att.net> wrote in message
news:gfqdnSs0IpA7_t_W...@speakeasy.net...
to repair a jack problem means entirely removing the system board from
the laptop, removing the old jack, soldering in a new one, and then
reassembling the laptop. not particularly hard but not a job for a
complete novice either. bad soldering skills can ruin the board... but
a board with this problem is pretty much useless anyway. if you can
remove the board (visit support.dell.com and under documentation look
for the service manual for you model for details) then you can send off
the board for repair. look on ebay for some people that do the job for
a flat rate, and that might even include removal of the board (ship them
your whole computer less memory, hard disk, optical drive, etc).
Of course there is also a chance that the problem is less dire than
this... well traveled and roughly handeled ac adapters can have a issue
with the wire where it enters the laptop plug, or the wire could be cut
or frayed, or the removable power cord may simply not be fully inserted
into the adapter... but those conditions are no so common, but worth
considering. in other words, make sure you have a good ac adapter
before you proceed with a motherboard power jack repair.
here is a example of the work involved...
http://www.youtube.com/watch?v=KjOVOLOW1Hc
I always wondered why some genius hasn't made fake batteries with a jack
on them. For those that use their laptops on AC alone, I figure this
would be a gold mine for somebody. Even adding a power regulator inside
(since the AC adapter voltage is higher than the battery) shouldn't be
too costly. After all, you can buy AC adapters for 10 bucks or less on
eBay.
Another solution is if your laptop can handle a docking station. Well
there you go. No power jack required. <grin>
--
Bill
Gateway M465e ('06 era) - Windows XP SP2
So far, it has not been a problem. But I suspect it will get worse over
time.
I don't think it is the chord.
This Inspiron 2200 (much maligned by many on the newsgroup) has worked
flawlessly for me for about 7 years. When it gets worse, I'll buy a new
laptop.
I also have an Acer netbook which I like a lot.
Mel
"Christopher Muto" <mu...@worldnet.att.net> wrote in message
news:pcidnbw9TfFILt_W...@speakeasy.net...
Actually I've found that the power supply's cord/wire to the computer
is the *most common* problem I've seen. Not just at the connection to
the computer, but especially at the junction where the wire enters the
power supply itself.
If the problem is at the end connector or the jack, the customers
usually had issues with putting the computer away (in a laptop bag,
etc.) with the power supply still connected. (I know, Duh!)
If the problem is at the junction of the wire and the PS, it's usually
from a "road warrior" that travels a lot and pulls the cord tight when
wrapping the cord around the PS when packing it up. I always tell folks
to leave a 1-2 inch loose loop in the cord so that you don't bend the
cord at the junction to the PS. If any of the 3 wires in the cord get
damaged from repetitive bending, you've got to repair or replace the
PS.
Luckily, it's pretty easy to pop open a Dell Laptop Power Supply and
unsolder the three wires and cut off about 6 inches of the "damaged"
section, use heat shrink tubing to insulate the 3 wires properly,
re-solder the wires and glue the PS back together. But you've got to
remember to test ALL the wires in the cord for shorts to each other or
open connections to the plug. Sometimes I've had to cut down some cords
to 1/2 their length or less. Some were totally trashed.
I have more PS boxes without wires than I expected. Found a place that
sells these wires for $15 each including shipping
(http://dcplug.net/dell.html), but it's higher than I'd like to pay.
They should be about $5 or less each plus shipping ($3-$5).
I don't have a video showing you how to do it, but it's not that hard
for someone experienced with a soldering iron. That $15 cord (above)
would make it very easy for most self repair folks to do.
Good luck!
(snip)
>Luckily, it's pretty easy to pop open a Dell Laptop Power Supply
It is? Do you have a magic method, or is it something we mere mortals
could hope to emulate?
Mortal? Hmm.
I started to put this info here, but decieded it would be off topic for
this thread and created a new thread. Here's the link:
I hope it helps!
Someone to check up this?
emails at corfu.t35.com jan kajander