I bought 20 Beagleboards for Mouser and 20 from Digikey recently. 5 boards from Mouser and 2 boards from Digikey are defective. Today, I asked my purchaser to ask refund from Digikey and the Customer Service Rep of Digikey asked her to get RMA from Beagleboard RMA team. The two boards from Digikey have never been reworked. So, I think my refund request is reasonable. These two boards have NAND ECC error. They failed to boot.
I think Beagleboard.org should let the CM be aware of the poor QC.
Here is the reply from RMA team:
Beagleboard.org does not receive any money from the sale of the BeagleBoards. So, how do you expect beagelboard.org to refund you the money? All refunds are handled by the compay from which you purchased the boards. This is the way it has always been and will always be.
Gerald
What's this NAND ECC error look like?
Regards,
I just spoke to the RMA team and the described issues was “Defective Board”. So they have not had the opportunity to even try to figure out what the issue was. The NAND ECC issue has not even been reported to the RMA team. The CM is in the process of contacting Mr. Fung direct.
Gerald
I asked my purchaser to return the 2 BB to Digikey and get refunded. But, Digikey ask her to get RMA instead. I think it is reasonable to get refunded by Digikey if the purchased good is “sick on arrival”. So, Digikey should refund me and return the sick Beagleboards to the CM.
-Eric
They will once the RMA team agrees that the board is bad and has received enough information from you do determine the issue. Just saying the board is defective is not enough information to adequately determine the issue and get it resolved. We are interested in making sure all issues are resolved, not just getting boards returned and not being able to find the issue.
Your help in getting issues resolved for everyone else is greatly appreciated. You do benefit from the community, so just look at this as your way of giving back to the community to the benefit for everyonene in the community. It only takes a few minutes to answer a few questions.
Gerald
Below is the screen capture of the terminal. The first section is the message after power on. The second section is the message after I pressed the reset button. My second board is not as bad as the first one on power-on reset. But, it will come to this point after a few boot failures and "button reset". On both boards, I cannot use my keyboard to break it.
Texas Instruments X-Loader 1.4.2 (Feb 19 2009 - 12:01:24)
Loading u-boot.bin from nand
ECC Failed, page 0x00080000
12 0 0 ea 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e
5 14 f0 1f e5 40 1 68 80 a0 1 68 80 0 2 68 80 60 2 68 80 c0 2 68 80 20 3 68 80 8
0 3 68 80 78 56 34 12 0 0 68 80 0 0 68 80 74 c9 6a 80 9c dc 6d 80 0 0 f e1 1f 0
40 e3 d3 0 0 e3 0 f0 29 e1 68 0 4f e2 4 0 0 e2 40 20 20 e3 2 20 0 e0 1 11 20 e3
2 36 20 e3 3 10 1 e0 3e 3b 20 e3 3 10 1 e0 f8 7 30 e8 f8 7 21 e8 2 0 50 e1 fb ff
7f 1a d7 5 0 eb 19 0 0 eb a4 0 4f e2 68 10 1f e5 1 0 50 e1 7 0 0 a 70 20 1f e5
70 30 1f e5 2 20 43 e0 2 20 0 e0 f8 7 30 e8 f8 7 21 e8 2 0 50 e1 fb ff 7f da 94
0 1f e5 1 7 40 e2 80 0 40 e2 c d0 40 e2 7 d0 4d e3 a0 0 1f e5 a0 10 1f e5 0 20 2
0 e3 0 20 0 e5 1 0 50 e1 4 0 0 e2 fb ff 7f 1a 4 f0 1f e5
Texas Instruments X-Loader 1.4.2 (Feb 19 2009 - 12:01:24)
Loading u-boot.bin from nand
ECC Failed, page 0x00080000
12 0 0 ea 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e5 14 f0 1f e
5 14 f0 1f e5 40 1 68 80 a0 1 68 80 0 2 68 80 60 2 68 80 c0 2 68 80 20 3 68 80 8
0 3 68 80 78 56 34 12 0 0 68 80 0 0 68 80 74 c9 6a 80 9c dc 6d 80 0 0 f e1 1f 0
40 e3 d3 0 0 e3 0 f0 29 e1 68 0 4f e2 4 0 0 e2 40 20 20 e3 2 20 0 e0 1 11 20 e3
2 36 20 e3 3 10 1 e0 3e 3b 20 e3 3 10 1 e0 f8 7 30 e8 f8 7 21 e8 2 0 50 e1 fb ff
7f 1a d7 5 0 eb 19 0 0 eb a4 0 4f e2 68 10 1f e5 1 0 50 e1 7 0 0 a 70 20 1f e5
70 30 1f e5 2 20 43 e0 2 20 0 e0 f8 7 30 e8 f8 7 21 e8 2 0 50 e1 fb ff 7f da 94
0 1f e5 1 7 40 e2 80 0 40 e2 c d0 40 e2 7 d0 4d e3 a0 0 1f e5 a0 10 1f e5 0 20 2
0 e3 0 20 0 e5 1 0 50 e1 4 0 0 e2 fb ff 7f 1a 4 f0 1f e5
Below is the boot message from the second BB. It waits forever when count down to 7.
Texas Instruments X-Loader 1.4.2 (Feb 19 2009 - 12:01:24)
Reading boot sector
Loading u-boot.bin from nand
U-Boot 2009.11-rc1-00601-g3aa4b51 (Jan 05 2010 - 20:56:38)
OMAP3530-GP ES3.1, CPU-OPP2 L3-165MHz
OMAP3 Beagle board + LPDDR/NAND
I2C: ready
DRAM: 256 MB
NAND: 256 MiB
*** Warning - bad CRC or NAND, using default environment
In: serial
Out: serial
Err: serial
Board revision C4
Die ID #1e6600040000000004037f680c02101a
Hit any key to stop autoboot: 7
-Eric
Hi Gerald,
I did not know you have such special RMA process in place. I appreciate your effort to make the Beagleboard project successful. And, I am aware of that beagleboard.org is non-profit. However, the manufacturer that make the BB is for-profit. So, the manufacturer should ensure good quality on the product it sold. All my Beagleboards are used in my manufacturer testers. And, I deal with CM weekly if not daily. So, I am familiar with their process and operation. All of my “sick on arrival” BB is due to manufacturing issue rather than design issue. If I did not raise the flag, I do not think anyone will care.
I also learned that the Beagleboard clinic is part of the manufacturer’s operation. Let me tell you my experience with Beagleboard clinic. In February, I send a Beagleboard to the clinic that has a GPIO pin on the expansion header stay LOW even if it was configured to have weak pull-up by my custom patch code in omap2-beagleboard.c. Since I used this pin for my MMC slot’s “card detect”, the MMC driver continuously interrupted once it was initialized. Ironically, the technician asked me to prove that the pin was stuck to LOW. Otherwise, the board would not be repaired. Then, I put the faulty board aside. Last month, when I did not have enough Beagleboard to use, I decided to put an header on the expansion bus and let my expansion board’s pull-up resistor to fix it. Fortunately, the problem was fixed.
Best Regards,
Eric
I find that you would think that I do not care extremely offensive. I care, probably too much. I deal with this stuff 7 days a week. I run the manufacturing operation and talk to these guys 5-8 times a day. I designed all the testers. I wrote all the test SW. As with any manufacturing operation, things can happen, new component batches, new technicians, new PCB batches, new assembly people, new batches of chips etc. Things change from one run to the next.
So, now you are educated. Please use your new information in the future before you come out blasting beagleboard.org and thinking that I do not care.
Gerald
Gerald,
I am not going to respond to this subject further. All I can say is that you and me have different perspective on this matter. While you see this as a community project, I see this as producer/consumer relationship when I buy Beagleboards from either Digikey or Mouser. If beagleboard.org sell directly to its customers, it is a different story.
Regarding how you manage the Beagleboard manufacturing activity, I do not want to make any comment. But, if I was the one who manage it, I or the CM would pay a qualify person to manage the day-to-day manufacturing. This is part of cost to do business. We need to create more jobs in America.
Best Regards,
Eric
I agree with your perspective as you are a user of the boards. While we have people trying to buy a single board for their use, you are buying them in larger quantities to use in your business. That is where the difference is in our perspectives and intent.
Gerald
Why are you using such ancient xloaders? Have you tried with a more recent one? E.g. http://www.angstrom-distribution.org/demo/beagleboard/MLO
Hi Koen,
I did not put the "ancient xloader" in the NAND. The problem always exists in this board whether the SD card is inserted or not. It behaves just like that when it is brand new. I did not and do not have chance to change any environment variable in the NAND of this Beagleboard. In fact, my SD Card works on all good Beagleboards. I have my shell script to "make" the SD cards for my Beagleboards. My SD card's only have u-boot.bin and uImage in the DOS partition, and the roofs in the Linux partition. After power-on reset, the X-Loader that resides in the NAND will load u-boot.bin and uImage from the SD card sequentially.
Why are you using such ancient xloaders?
So, I do not know how to answer your question. The xloader came with the Beagleboard. I think my other Beagleboards show the same version of xloader.
-Eric