Please help me identify diode that is faulty

[frenky]

Hi guys.

I'm trying to fix a board of a plotter for a friend of mine.
When you connect plotter to the pc nothing happens... (pc should show up new usb device)

So I started with thermal inspection and measuring voltages and I believe I have found faulty diode.

There are a lot of the same H7 diodes on the board. All measure 0.484V with diode test except this one that measures 0.390V.
This diode is between 3.3V and ground and is heating up really bad. It looks like it's conducting to the ground?

At first I thought it's basic SM4007 but after googling "H7 smd diode" I got some zenner diodes as result...

I'm worried that if I try to replace that diode with wrong one I'll do more damage than good.

What do you think guys?

[frenky]

Hmmm...

I took a look at the LM5276 and it needs a shotkey diode. In the data sheet it is 1N5822.

http://www.ti.com/lit/ds/symlink/lm2576.pdf
http://www.vishay.com/docs/88526/1n5820.pdf

Now I'm embarrassed for even posting my "problem" here... 

Well if that wont' fix the problem then I still need your help. 

[Armadillo]

There are a lot of the same H7 diodes on the board. All measure 0.484V with diode test except this one that measures 0.390V.
This diode is between 3.3V and ground and is heating up really bad. It looks like it's conducting to the ground?

At first I thought it's basic SM4007 but after googling "H7 smd diode" I got some zenner diodes as result...

I'm worried that if I try to replace that diode with wrong one I'll do more damage than good.

H7 marking appears to be a NXP 11v zener diode. To confirm it, I would suggest you take out this diode from the circuit, solder a 100ohm 1W resistor in series with it, apply 12 volts across the assembly [+ve should be cathode side] then measure the voltage across the zener diode. It should read about 11 volts. For confirmation, take out a good H7 and do similar test. If both read rail voltage e.g. 12 volts, then its not a zener diode.
This also confirm whether your suspected component is in fact defective.

[do remember the orientation of the diodes before you take it out from the circuit. take few photos to be safe]

[bktemp]

Is this diode connected between 3.3V and GND, or between the output of the LM2576 and GND?
The package looks like SMA or some power SOD123, so my guess would be it is the schottky diode for the stepdown regulator.
Since both LM2576 and diode get warm, both seem to be operating but conducting a high current. So my guess would be short circuit somewhere else putting a high load on the 3.3V rail.

[frenky]

Is this diode connected between 3.3V and GND, or between the output of the LM2576 and GND?
The package looks like SMA or some power SOD123, so my guess would be it is the schottky diode for the stepdown regulator.
Since both LM2576 and diode get warm, both seem to be operating but conducting a high current. So my guess would be short circuit somewhere else putting a high load on the 3.3V rail.

Diode is between the output of the LM2576 and GND.
You are right about something else going on because there is also a 1.1A resettable fuse which has 0.1V voltage drop. On one side it has 3.450V and on the other 3.350V.
I guess the current is bigger than 1.1A so the resistance of the fuse went up.

The only other things heating up are 1.8V voltage regulator (top left of the main IC), a Differential Line Receiver AM26LS32AC (top right) and the main IC (middle).

[bktemp]

AM26LS32AC runs at 5V and draws an idle current of 52mA, so it is expected to get warm.

Is the main ic some custom part? I have no experience with plotters, but I doubt it needs a really powerful processor.
Unless there is some other fault around the main ic loading its outputs, it looks like the ic is the culprit.

[frenky]

IC is SH7709S (Renesas 32-Bit RISC).

http://pdf1.alldatasheet.com/datasheet-pdf/view/132643/RENESAS/HD6417709S.html
http://www.medteh.info/_fr/3/sh7709.pdf


I sure hope it is not this IC that is faulty...

[bktemp]

That processor is much more powerful than expected.
According to the specs it draws up to 680mA, so the high current consumption on the 3.3V rail could be normal if the core is powered using the 1.8V linear regulator.

[frenky]

Thank you for helping me I really appreciate it.
When I get home from work I'll do more measurements...