unknown voltage source from 2 layers of GND

[avib]

Hi

its my first post on this forum, so i hope i would make everything smooth and clear as possible , im working at pcb debug , and i got a unexplained voltage between 2 GND : 24V_GND and DGND, i cant post the scheme cause of company rules.but maybee directions or ideas will help .

on a working pcb the normal voltage between the 2 layers of GND is "0V" ...on the faulty pcb i get about 10V

the 10v possibly source is from the rs323 flat cable...cause it has been measured when the pcb only connected to RS232 flat ...on a certain pin.)on that moment no 10v between thee GND'S when the pcb is connected to a tester and gets power ...than the GND probly shorted and measured by 10V.. i have some isolaters components which have the both GND (4 channels : power in 24_GND ,power in 2 and DGND. when i replaced those isolaters or remove them the short is still there. i belive it might due to short layers ..but ill be happy for new ideas ..or directions.
if you got any quastions that i can assist you in understanding better ill try .
thank you .
avi.

 

[barry]

Not really clear what you are saying. Are you saying that you've got 10V between two "GND" layers? Apparently the PCB is either defective or designed improperly. If you've got both layers connected to "GND" with vias, then there is no way they can have 10V difference. Can you measure the resistance between the two layers? It should be less than an ohm, otherwise there's something wrong. Are 24V_GND and DGND tied together? Are the two "GND" layers you're talking about supposed to be the same net?

Forget about all that rs232 stuff and whatever else you were talking about. "than the GND probly shorted and measured by 10V"; what does that even mean? What are "Isolators components"? You need to explain your problem more clearly.

 

[avib]

the resistance between the 2 layers can`t be measured , tried on both pcb one which working fine and other isnt .probly no electric connection between those two .
the isolataros are :ADuM1400 component. later ill be able to upload part of the scheme and ill mark there some problems i found .
thank you .

 

[KlausST]

Hi,

If both GND are not connected, then it is very likely you can measure some volts between then with a DVM. No need to worry.

The resistance can't be measured.....--> you mean it is high impedance?

Klaus

 

[hobbyckts]

Mesurement of 10V at which ground out of the 2 grounds available? with reference to which point?

I'm considering that you are measuring voltages between the 2 grounds available. In that case as you said in your previous pose if you are sure that there is no reistance between the 2 grounds available. Then you should not worry about the voltages measured between these 2 grounds.

 

[avib]

Hi

thanks for the reply, i tested the pcb and compared it to other pcb that passed ATP and other kind of test ..in the good one's there is no voltage between the two GND. and the problem i get its a voltage that shouldnt be on all components that contain both grounds ,those are the isolators(ADuM1400) , they get 2 kind of voltages (5 and 3.3) and have 2 kind of GND. i removed those isolators to confirm they are not shorted or faulty it didnt help. the pcb wont boot up so i belive its cause of this voltage that shouldnt be .when im saying it cant be measured the DVM not responding to this impedance probably out of range (too high).the voltage that i measure on the 2 GND ,also affecting the components that contain both of them , e.g: if i measure on pin 5(24_gnd ) and other pin i get voltage that shouldn't be too ,it should be 3.3 for example and i get 13.3 v . cause the reference point(GND) is already with 10v.

 

[barry]

If you've got very high impedance between two grounds, then they aren't both ground, and why would you expect zero volts between them? There's something else going on here, and with your inability to post a schematic it's pretty difficult to debug your problem. If you stick your DVM probes into the air, you'll measure zero volts between them; that doesn't mean the two points are connected.

 

[hobbyckts]

Atleast provide the block where you facing the problem in order to find out what went wrong in your case. If you explain like this it will tough to understand.

 

[avib]

the red marked GND are having the the 10v voltage

 

[hobbyckts]

Now I have got few questions after seeing your circuit.

There are three power supplies available in your circuit.
1. 3.3V
2. 5V_ISO
3. +24V

But there is only two grounds are available.
1. DGND
2. 24_Gnd

The DGND you are using for 3.3V supply line and 24_Gnd you are using for both 5V_ISO, +24V. I guess you have used some DC-DC converter for converting the 24V supply to 5V. In that case how can you use same ground reference for both the voltages?

Also measure the voltages at the following points and tell us the actual values that you are getting from the measurement.
1. 3.3V w.r.t DGND
2. 5V_ISO w.r.t 24_Gnd
3. +24V w.r.t 24_Gnd

 

[avib]

the 5v_iso is using the AGND as you can see on the this photo .

 

[avib]

there is also 5v_iso(in the next photo when the moderator will aprove it ) but what you meant is 5v_io and there is 5v from 5v_io to 24_GND and 3.3v from 3.3v to DGND
the basic voltages are all fine ,thats the first thing i checked ,but than i tried some measurements and checked the the signals under 3.3v(u79) to 24_gnd(not there naturally GND) and got voltage which shouldnt be ,in compared to about 5 pcb which are fine none gave me voltage from this signals under 3.3v in (u79) .than when i saw this i assumed correct that if i will measure between the 2 GND ill get a voltage and i got about 10v which is back to my first post .

 

[hobbyckts]

I guess you didn't understood what I meant to say.

Can you tell me the part number for U82?

 

[avib]

yes part number :LM78L05AC/SO VOLTAGE REG +5V 100MA

 

[avib]

Hi

i posted 2 schemes , in the scheme u can see that the DGND and 24_GND are grounds ,i have measured lot of other working PCB and compared them to the faulty one , just the faulty one has voltage from the both GND's the isolatros components are quad or dual channel voltage with voltage channels and their GND's.