cancel
Showing results for 
Search instead for 
Did you mean: 

PSoC 5, 3 & 1

GuNo_288966
Contributor

Hi,

   

I have a serious problem with a CY8C5868AXI-LP032 and SPI.

   

When I send data over SPI, after a few seconds this toasts the controller. The resistance between VCC and VSS goes down to 1.1 Ohms, the controller is no longer found with the miniprog.

   

Details:

   

I use the SPI Master V2.4, bidirectional, 1000kbps, internal clock, no interrupts. SDAT is connected to a bidirectional pin, strong drive. SCLK and SS are connected to a digital output pin, strong drive. There is nothing externally connected to these pins yet.

   

First I sent some bytes by SPIM_PutArray(). On the scope I saw the SCLK switching, but SDAT and SS were permanently low. After a few seconds the system broke down, 1.1 Ohms between VSS and VCC, controller became hot. Controller was no longer accessible by the miniprog.

   

Exchanged the CY8C5868AXI. Resistance was then in the MOhms range. Removed the SPI master user module, connected the pins to a control register and switched them manually. Then the output pins behaved as expected..

   

Added the SPI master again, connected the pins as described above. started the program. After a few seconds breakdown, again 1.1 Ohms between VCC and VSS.

   

Any ideas what I am doing wrong?

   

Best regards,

   

Guenter

0 Likes
9 Replies
Bob_Marlowe
Expert II

This smells a bit like wrong power sources.

   

Check

   

-Vdda

   

-Vddd

   

-SPI master GND and PSoC5 GND

   

and check the SPI Master's voltages

   

 

   

Bob

0 Likes
GuNo_288966
Contributor

Checked. All connections are as in figure 2-5 in the data sheet.

   

VDDD=VDDA, VSSD=VSSA, VCCD pins connected.

   

I will now exchange the controller again.

0 Likes
GuNo_288966
Contributor

Controller replaced. All voltages checked, VDDD=VDDA=5..0V, VCCD=1.8V, VSSD=VSSA=0V.

   

Removed the SPI from the design, switched the lines manually => it works.

   

I don´t want to test the SPI again before I have an idea what went wrong. Something inside the controller makes a massive short between VDD and VSS. The controllers do not reply to the miniprog because there is not enough voltage because of the short. Is there anything I can do to "reset" the controller?

0 Likes
ETRO_SSN583
Esteemed Contributor

Generally speaking this indicates the parasitic SCR in CMOS has been triggered,

   

which shorts internally Vdd to Vss trails, causing large current, chip heat, and in

   

many case the internal die bond wires blow open for either Vdd or Vss.

   

 

   

The cause is a pin being driven outside its supply rails by roughly a diode drop,

   

this in turn triggers the SCR mechanism.  This can be simple transients, like due to

   

L and C, and the trigger can be as low as a few tens of nS.

   

 

   

So use DSO, set two triggers, Vtrig < Vss - .5 OR Vtrig > Vdd + .5, trigger on

   

Normal mode, and see if you can find root cause.

   

 

   

    

   

          http://en.wikipedia.org/wiki/Latchup

   

 

   

Regards, Dana.

0 Likes
Anonymous
Not applicable
        Do you apply power to the VCCD pin from external regulator? And use externally regulated mode?   
0 Likes
GuNo_288966
Contributor

VCCD is the internal 1.8V.

   

Now I got it up and running.

   

I think that I might have caused the problem myself when measuring with the osci probe. There is +/-15V on the board... Thanks for the help,

   

Guenter

0 Likes
Bob_Marlowe
Expert II

Fine that you got it solved...

   

BTW: I'm located near Bremen, did I ask you where you are from?

   

 

   

Bob

0 Likes
GuNo_288966
Contributor

Bob,

   

I am located in Elmshorn.

   

My PM is ...removed

   

Cheers,

   

Guenter

0 Likes
EvPa_264126
Valued Contributor II

I had a similar problem.
For CY8c5868AXI-LP035 + USBFS + 5V power - from USB.
Several months ago, chip overheated and broke down.
Yesterday, the situation with the fast heating repeated, but I had time to turn off the power.
 Burned lip hurts so far))).
A little experiment for the project USBUART:
1. If a board is connected to the programmer () 3.3 or 5V) and then add USB - no problems, the device works.
2. In the several USB connectors device works in other - not or not always.
3. The situation does not change when i use a different project and PC.
4. The heating occurs even if only Vdda = 5V and Vddiox = Vddd = 3.3V.
I temporarily solved the problem by installing Vdda = Vddiox = Vddd = 3.3V
 

0 Likes