Tektronix Technical Forums are maintained by community involvement. Feel free to post questions or respond to questions by other members. Should you require a time-sensitive answer, please contact your local Tektronix support center here.

Endurance & Segment Stress / Measure Mode

Post Reply
stemue
Posts: 3
Joined: August 10th, 2012, 12:50 am
Country: Germany
Location: Dresden (Germany)
Contact:

Endurance & Segment Stress / Measure Mode

Post by stemue » August 20th, 2012, 3:38 am

Hello everyone,

following question for 4200-SCS system with PG2-Unit or PMU-Unit: I would like to use Segment Stress / Measure Mode in order to evaluate the endurance and disturb characteristics of flash-type memories. No switch matrix should be used because very short pulses need to reach the sample properly. Therefore I tried to make use of the Segment Stress / Measure Mode for which it should be possible to influence the states of the SMUs using the PIN field (0 = no connection (is this equivalent to GND?), 1 = high imp, -1 = high imp (is this equivalent to 1?)). (see also figure 3-71 Model 4200-SCS User's Manual).
However, no matter how they are defined, signals appearing at a shared terminal look all the same independent of the PIN state. It appears that in every case the signal does not reach the pin properly since the SMUs are not disconnected from the shared pin.

And another question for 4200-SCS system with PG2-Unit: Using single_pulse_flash.c from flashulib, the pulse appears perfectly well on an oscilloscope. However, trying to use the same pulse within a subsite cycling (segment stress / measure) the pulse is distorted. This looks similar to false definition of voltage range (20V instead of 5V), however this is not the case. I double checked and it is written properly in the SegArb file. Maybe wrong initialization during subsite cycling? (It works perfectly well on an 4200-SCS with PMU units!).

Thank you very much for your help.

Andrea C
Keithley Applications
Keithley Applications
Posts: 1612
Joined: October 15th, 2010, 10:35 am
Country: United States
Contact:

Re: Endurance & Segment Stress / Measure Mode

Post by Andrea C » August 30th, 2012, 5:19 am

Hi,

I can confirm that with KITE 8.x and the subsite cycling, the pulse may be distorted. The application is always using the high voltage range and therefore fast pulses which require the faster range will be distorted.

Do you have the 4225-PMU and KITE 8.2? If so, have you investigated the use of the NVRAM libraries? With those, you can get the endurance testing without using the subsite cycling. Here is an application note:
http://www.keithley.com/support/data?asset=56338

Andrea

stemue
Posts: 3
Joined: August 10th, 2012, 12:50 am
Country: Germany
Location: Dresden (Germany)
Contact:

Re: Endurance & Segment Stress / Measure Mode

Post by stemue » August 31st, 2012, 1:10 am

Thanks for the reply. Yes, we also possess a 4200-SCS with 4225-PMU. And strangely, the subsite cycling works on that system, i.e. pulses < 1us arrive undistorted. However, it does not work on the 4200-SCS with 4205-PG2 and hence subsite cycling can only be used for pulses >1us, that's a pity (due to the high voltage range, as you described). Furthermore, I am aware of the NVM library and this could be a workaround.
Could you tell me which KITE version to use for our 4200-SCS with 4205-PG2 in order to allow for fast subsite cycling in the ns-range?
The more severe problem, as I described, is that the SMU states, as defined within Segment Stress / Measure, do not affect the SMU relay states. Connecting SMU and PVU to the same pin is therefore not possible. Could you clarify on that?

Thank you so much.

Post Reply

Return to “Memory (Flash, Phase Change, NVM, Resistive)”

Who is online

Users browsing this forum: No registered users and 1 guest