Am29F040B content degradation

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
fdfd_4525841
Level 1
Level 1
First like received First like given

I have used AM29F040B for years and still in production. Last experience is that sometimes the device can degrade the code loaded. Is there any historic about this issue? Is there any situation that the memories can degrade the data loaded without any intentional write cycle during use?

0 Likes
1 Solution

Hello

Sagitron is not recognized as an authorized distributor of AMD, Spansion or Cypress FLASH memory products,

and Cypress is unfamiliar with Sagitron's environmental-controlled storage protocol.

Therefore, it may certainly be possible that the Am29F040B-90FI may have developed oxidation on the leads.

However, I believe you may be experiencing 'soft-errors' that may be due to excessive x-ray or radiation exposure.

'Soft-errors are correctable through 'erase' and 're-programming.'

Best regards,

Albert

Cypress Semiconductors

View solution in original post

12 Replies
AlbertB_56
Moderator
Moderator
Moderator
500 replies posted 50 likes received 250 replies posted

Hello,

Thank you for contacting Cypress Semiconductor.

In what kind of environment is the target application (using the Am29F040B) operating?

What is the Operating Temperature Range = ?

What is the Operating Altitude = ?

What is the Operating Depth = ?

What is the frequency of erase and programs operations that occur = ?

Use in Aeronautics/Avionics/Space = ?

Any Excess EMI present nearby = ?

Any Radiation Field present nearby = ?

What is the complete Ordering Part Number (OPN) of the Am29F040B?

Please respond at your earliest convenience...

Best regards,

Albert

Cypress Semiconductor.

0 Likes

Hi Albert, sorry by delay

The memory is working in an electronic board with CPU Motorola 68332.

The failure is at ambient, laboratory conditions in diferent places.

The failure occurs in the same module but only in some units, not in all of them, and when happens is after many hours of use (more than 50h) or many startups sequences (more than 500). then is difficult to reproduce.

The issue is that if we program the memory, it works well during time and sometimes the units come back to repair because this fail, but if we reload SW the fail dissapear.

We tested a lot to know if a cassual SW write, but never happens.

The most frequent fail is that un one position one or two bits with value "1" goes to "0", and the address use to be the first position or the middle of the memory, but sometimes is in other positions.

Part Number: AM29F040B-90FI

0 Likes
lock attach
Attachments are accessible only for community members.

Hello,

No worries on your delayed response.  It is understood that we are all !tasked with various projects and issues, simultaneously.

I believe you meant to indicate the Am29F040B-90EI, as the Am29F040B-90FI is an invalid Ordering Part Number (OPN),

which does not exist.

It certainly appears the Am29F040B is experiencing "Soft Errors", as soft errors are not damaging to the cells, but correctable

through erasing and re-programming of the FLASH device, itself.  Please refer to the attached legacy (Spansion-2011) presentation

in regards to X-Ray physics, as there may be the possibility that your specific assembly inspection process may be exposed to

these similar conditions.

Best regards,

Albert

Cypress Semiconductor

0 Likes

Thanks Albert by this information, it is very interesting, we have not ambient with expected X-ray, but we will try to excite to see if reproduced.

It could be the case of the end user, having a high interference environment that produces the faillure.

Other way we work is the case that power cycles have any type of influence, let me know if you have something on this.

If we finnally find the root source I will let you know.

Best regards.

0 Likes

Hello Albert,

There is another question from my side.

Is there any issue if we are using old parts from AMD?

We are using memories with date code 0427 and 0421 that are possibly manufactured at 2004.

0 Likes

Hello,

Yes, I believe the date codes (0421 & 0427) are from the 2004 production year.

NOTE :   The Am29F040B-90FI did, at one time, exist as a vaild Ordering Part Number (OPN),

but it was officially discontinued in January, 2006   (F = 32-pin TSOP, reverse pinout)

Moisture Barrier Bags : FLASH memory products are sealed in moisture barrier bags, and should be stored

in an environment with the temperature less than 40°C and the relative humidity (RH) less than 90 percent.

The moisture barrier bag meets the requirements specified in MIL-PRF-81705D, Type I, Class 1 for flexibility,

ESD protection, mechanical strength, and puncture resistance.  Furthermore, the bag complies with the drop

test requirements defined in ASTDM 4169-94 and the water vapor transmission rate defined in ASTM F-1249.

All bags are coated with antistatic material to provide ESD protection on all surfaces and to comply with the

requirements set for in EIA 625.

Humidity Indicator Card : The Humidity Indicator Card is included in each moisture barrier bag.  The Humidity

Card has humidity-sensitive elements that turn from blue to pink whenever the specific Relative Humidity (RH)

level is exceeded.  The Humidity Indicator Card can be reused if all the humidity-sensitive elements are blue.

If the Humidity Card has turned pink, bake the card at 125°C ± 5°C for 15 minutes. The humidity indicator cards

used with Cypress FLASH memory products meet the standards specified in MIL-I-18835.

Desiccant : Each sealed moisture barrier bag has Desiccant.  The desiccant pouches placed in the bag  greatly

reduce the presence of moisture. The desiccant pouches help to maintain the environment in the bag at no greater

than 10 percent RH, thereby protecting the devices during shipment and storage for at least 12 months. 

Out-of-Bag Time :   Out-of-bag time varies depending upon the moisture-sensitivity rating for the product and is

based upon assumptions made about the factory environment to which the product is exposed.

There should be no issue to use older date code FLASH memory product, as long as strict adherence to the above

conditions are fully satisfied.

However, baking of the FLASH memory product is required if :

1).   The out-of-bag time from initial exposure to board mounting is exceeded;

2).   The out-of-bag product has been stored at an RH of 10% or greater,  or

3).   The humidity indicator card is greater than 10% when read at 23°C ± 5°C.

Please provide specific details as to whom the third-party vendor or distributor is,

from where the Am29F040B-90FI were purchased.  The identity of the  third-party

vendor or distributor may assist with determining the actual environment of how

the Am29F040B-xx were stored.

Best regards,

Albert

Cypress Applications Support

0 Likes

Hi Albert,

We had already thought about the possibility of bad storage. Following the typical recommendations the components are stored in bags in small packages with their desiccant and humidity indicator. In addition, in the manufacturing processes, one is added that includes the stoving of this and other components.
However, there is always the doubt that in a small batch of manufacturing there has been a lack of procedure or an error, that is why we investigate whether the fault we have responds to something known.

By the way, speaking of known things, are there previous cases of flash memories that have failed similarly to our case? And if there are what has been the cause of these failures?

Best regards,
Felix
0 Likes

I forgot the other question.

We know that it is obsolete and we purchaded devices time ago from Sagitron. At this moment Sagitron is not alive any more, I don't know if another company replaced it.

0 Likes

Hello

Sagitron is not recognized as an authorized distributor of AMD, Spansion or Cypress FLASH memory products,

and Cypress is unfamiliar with Sagitron's environmental-controlled storage protocol.

Therefore, it may certainly be possible that the Am29F040B-90FI may have developed oxidation on the leads.

However, I believe you may be experiencing 'soft-errors' that may be due to excessive x-ray or radiation exposure.

'Soft-errors are correctable through 'erase' and 're-programming.'

Best regards,

Albert

Cypress Semiconductors

Following the case that we could have suspect in the devices procured, I have two photos of devices in stock. One is lot 0427 and the other is lot 0421. They have very different marking.

Is it possible that we received non official devices?

sn2_1_7.5x.jpgsn4_1_7.5x.jpg

Best regards,

Felix

Hello Felix,

As I had already mentioned in my previous response, Sagitron is not recognized as an authorized distributor of AMD.

Therefore AMD, Spansion or Cypress, are not familiar with Sagitron, as well as with its environmentally-controlled

storage protocol.  Sagitron is not a recognized authorized distributor or Sales Representative.

Cypress does not extend any portion of its product warranty and/or guarantee to any Cypress (AMD, Spansion) FLASH

memory products that were purchased through Third Party Vendors and/or Non-Authorized Distributors.

There may be the possibility that the Am29F040B FLASH memory devices may have been Illegally re-produced by counterfeiters.

Therefore, you will need to contact Sagitron directly for a Return-Materials Authorization (RMA).

Best regards,

Albert

Cypress Semiconductor

0 Likes

I received the list of procurements for this component:

Fecha

Cantidad

Pedido-Línea

Borrado

Nom Proveedor

14/02/2001

500

16100265-1

SAGITRON,S.A.

14/02/2001

500

16100265-2

SAGITRON,S.A.

14/02/2001

500

16100265-3

SAGITRON,S.A.

14/02/2001

500

16100265-4

SAGITRON,S.A.

14/02/2001

800

16100265-5

SAGITRON,S.A.

06/07/2001

40

16101558-3

SAGITRON,S.A.

09/10/2001

1600

16102185-7

SAGITRON,S.A.

08/07/2003

208

26305165-5

ARROW IBERIA ELECTRONICA SLU

09/07/2003

1200

26305165-13

ARROW IBERIA ELECTRONICA SLU

11/10/2004

780

26410066-3

FUTURE ELECTRONICS LTD.

18/10/2005

195

26509500-4

ARROW IBERIA ELECTRONICA SLU

18/10/2005

2005

26509500-5

ARROW IBERIA ELECTRONICA SLU

25/07/2006

138

26607828-3

ARROW IBERIA ELECTRONICA SLU

23/01/2008

2400

26800808-1

OMNIAELECTRONICS, S.L.

15/07/2008

180

26809936-5

GIZA TECHNOLOGIES, INC

30/01/2009

390

26901188-6

GIZA TECHNOLOGIES, INC

24/11/2009

900

26913484-5

GIZA TECHNOLOGIES, INC

20/03/2012

74

4200073269-60

GIZA TECHNOLOGIES, INC

04/08/2017

350

4200273208-10

MILEXIA IBÉRICA, S.A.U.

As you can see SAGITRON was an old representative, after we procured to Spain local representatives ARROW, FUTURE, OMNIAELECTRONICS, GIZA, and MILEXIA. They use to have the representation of main distributors on Europe or USA.

In any case we always procured AMD as the most reliable manufacturer for this component. Then I sent the photos to know if it is possible identify if the manufacturer is AMD official as probably there was several production lines. Let me know if you have this information.

Best regards,

Felix

0 Likes