| T O P I C R E V I E W |
| rochoz |
Posted - 11/26/2011 : 21:39:25
Hi all,
I am having problems when continually verifying Dallas DS1250Y NVRAM modules using the GQ-4X. The problem is if I continually verify one of these DS1250Y devices, usually after 3 or 4 verifies eventually some random location no longer verifies. The byte that gets corrupted stays corrupted, until you verify enough times that a different location gets clobbered. I initially suspected a bad DS1250 part, but I have been able to recreate this problem on 4 different devices, 2 of which are brand new. I do not have this problem if I bring any of these devices to work and verify them on an expensive DataIO programmer. I suspect the GQ4X is doing something at power off.
I connected my digital scope and found that there are numerous glitches on the WE signal during power down, as well as undershoot of around 900mV, at around when VCC is 3V. Neither of these are a desirable feature. In fact, the datasheet has a very bold warning that absolutely no undershoot is allowed when the part is in battery backup mode (ie, when VCC is less than 4.5V).
Is this a known problem? Can it be fixed?
Thanks,
Mike |
| 2 L A T E S T R E P L I E S (Newest First) |
| rochoz |
Posted - 11/28/2011 : 18:31:57
Thanks for the reply.
Actually I'm re-evaluating my problem. It appears that this problem is happening only to ONE of my DS1250Y parts, and the problem is also occuring on that device on an expensive BP Micro programmer at work (we have both DataIO and BP at my work). At any rate, this problem may not be happening on the other devices. I am going to look at this a little closer to see if I really have a problem with my programmer or not. |
| ZLM |
Posted - 11/28/2011 : 11:53:16
undershoot is common on a low cost universal programmer. It may be caused by mismatched impedance between chip and the programmer circuit.
One way to reduce the undershoot noise is to make a adapter between chip and programmer. But that may not easy.
Try to use slow speed and see if it makes different.
|
|
|
