Go Down

Topic: 74HC595 and L293D overheating .  (Read 400 times) previous topic - next topic

amine2

Mar 03, 2019, 06:51 pm Last Edit: Mar 03, 2019, 06:52 pm by amine2
Hello guys , i had a PCB made for this schematic


i assembled it and soldered the parts , and everything went fine i aploaded the bootloader on the Arduino (SMD atmega328P) and managed to upload code onto it without problems .

the problems began when i started testing . first , and even before connecting the 12v line , one of the 74HC595 started heating up , only on of them , as the second one worked correctly without any problems . which was very confusing for me . with that , when i measure the supply voltage (the 5v supply voltage) with a multimeter i get 5V , the thing is after connecting it to the circuit , when i measure the voltage between the 5v and gnd lines of the 595 i get 3.8v , which is odd . also when i measure the voltage on the output pins of the 595 i also get 3.8v when it's supposed to be HIGH 5v .
the 595 that was overheating ended up being fried and stopped working all together after a full day of testing . the other one though , continued to work normally without overheating , the only problem with it is the weird 3.8v voltage .

after that i connected the 12v rail to test the L293D ICs , two of them worked correctly without overheating  , but the third (IC5) overheated so much , that after a couple of seconds from me wiring the 12v rail in , it got so hot that it burned my finger . it also ended up getting fried while the other two continued to work normally .
All this happened without me connecting the motors . all of those experiments happened without any load what so ever being demanded from the ICs .

i can't figure out what's wrong .
thank you .
it's all about the melons .

larryd

No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

amine2

Hello, thank you for your Attention LarryD , it's very well etched , i got it from JLC PCB in china . it's a very good pcb .
it's all about the melons .

Paul_KD7HB

Hello, thank you for your Attention LarryD , it's very well etched , i got it from JLC PCB in china . it's a very good pcb .
I think Larry meant, did you pay t have the board tested for shorts?

Paul

larryd

Sometimes two traces can have microscopic filaments between them.

Check out all traces on these I.C.s with an ohm meter.


No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

amine2

I think Larry meant, did you pay t have the board tested for shorts?

Paul
i really don't know , but the board looked good to the eye .
it's all about the melons .

amine2

Sometimes two traces can have microscopic filaments between them.

Check out all traces on these I.C.s with an ohm meter.


i did that already , i found no shorts ! i will do it again now just in case .
could that also explain the 3.8 voltage if it were true ?
it's all about the melons .

larryd

Shorts can be microscopic that no eye can see.

Show us good images of the 'actual' PCB.



No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

amine2

it's all about the melons .

WattsThat

You should have stopped before applying 12 volts, you already knew there was a problem. You can only make things worse by ignoring the obvious.

Unfortunately, cannot do much with the limited size photos, cannot see traces well enough to troubleshoot.

Possible plan of action:

Using another board, install the '328 with crystal and caps, the two '595's in sockets* and all bypass caps. Apply 5 volts. If you have that odd 3.8 volt level on IC1 or it gets warm, you have a board design error.

Install program in the '328. Scope the outputs of the 595's.

Is it working? If yes, board #1 has a shorted trace on IC1 somewhere. Got back and fix it, replace IC1 and finish testing. Finish board #2 and move on.

* - sockets are always a very good idea on first article builds.

Vacuum tube guy in a solid state world

amine2

thank you Watts , i solved the problem by following those steps .

i soldered a second board , and it worked quite well without overheating , the 3.8v problem was gone .

i decided to re-check the first board with a multimeter , i ended up finding that there was connection between VCC(5v) and GND . i unsoldered the 595 that was overheating , and when i placed the electrodes between the VCC and Ground lines of that chip , it also showed a connection . for some reason , that 595 chip had the 5v and GND pins connected with a resistance of 30ohms , now that i know that , i don't understand how it even worked with 3.8 in the first place.


Thank you very much guys .
it's all about the melons .

Go Up