| Author |
 Topic  |
|
|
cheebster
Canada
12 Posts |
 Posted - 02/21/2006 : 16:50:44
|
Hi there. I have a tsop-48 adapter here, modded by a guy at MCUMall directly to unprotect 29LVXXXX tsops. It has a jumper, and sends VPP 12V to the reset pin12, when i select Temporary Sector Unprotect in Willem EEPROM 0.97ja, i see 12V on that line too.
I try to erase a 29DL323GT, so i use the settings for 29LV320 in the menu, and have just Temporary Sector Unprotect checked.
The erasing process takes more than 40 seconds, it goes up to 100%, then i get an error at the bottom 'Error erasing chip'
If i do a blank check, the chip it says Device Not empty.
The device ID gives me the correct infos, other than detecting the chip as a -DT instead of a -GT. What i get, is
Manufacture : 0x01 AMD
Device Code : 0x50 Am29DL323DT
Is there anything i'm missing i need to check, or any settings that i need to use ? I'm using the enhanced willem from MCUMall too.
If anyone could give me some tips, i would really appreciate it. Thanks !
cheebster.
|
|
| Reply #1
ZLM
2945 Posts |
Posted - 02/21/2006 : 21:20:14
|
The ID is correct. 50H is the ID of AM29DL323G.
It seems the chip not been erased.
Turn on the temporary unprotect, see if it is can be erased correctly. If not, your chip may be a bad chip.
Also, check your address line A19,A20,A21 are connected correctly. |
Edited by - ZLM on 02/21/2006 21:24:24 |
 |
|
| Reply #2
cheebster
Canada
12 Posts |
Posted - 02/22/2006 : 06:10:41
|
Hi ZLM, EZoSat. Thanks for your replies. The chip comes from a Dish PVR 510 Receiver. I know the secsi isn't factory locked, as i'm usually able to erase those chips using a procedure called 'vidmod' on the receiver itself - just sending 12V on the reset line to erase and program fully.
It's about the same that the programmer is doing i think. If i just check the temporary unprotect, that should be all i need to do to erase the chip fully ?
It usually takes 45 seconds to erase them in the receiver, but via jtag. Is it normal that it takes more on the programmer ?
Thanks a lot. |
|
|
| Reply #3
cheebster
Canada
12 Posts |
Posted - 02/22/2006 : 06:53:17
|
Um.. can i trust the 'Error erasing chip' in the software ? I just did a read on the 29DL323GT, and all i get is FF's and 00's. Does that mean the chip has been totally erased ? What's 00's, same as FF's ?
Thanks. |
|
|
| Reply #4
ZLM
2945 Posts |
Posted - 02/22/2006 : 13:20:56
|
We should trust the software erase. I have used this 0.97ja for a while on 29LV320, for erasing and the writing. On my experience, erasing the 29LVXXX normally take about 40-60 seconds. But some chips will take time longer if the chip is not a "good" chip. Not "good" means it can be erased but takes longer time, and write the data may have difficulties on some addresses.
The erase chip action is sending the erase commands to the chip, same as the read ID command. So, I think if the software can get ID correctly, then the software should have a correct communication connection to the chip, the hardware connection should be correct. The software uses same method to send the read ID commands and erase chip commands. |
Edited by - ZLM on 02/22/2006 13:24:25 |
|
|
| Reply #5
cheebster
Canada
12 Posts |
Posted - 02/22/2006 : 15:26:32
|
Hi there. I have tried with 3 different 29LV323GT's and i got the same results everytime. I triple checked them, they are all cleaned and the pins are perfect. They all get the id's too. I just programmed some 29LV160's, all went fine. I seriously don't know what to check now :| I will try erasing them a few other times, to see if i get all FF's in the buffer..
EZoSat, i get records starting from offset 1A0000 i get some pattern like this;
Offset 0 1 2 3 4 5 6 7 8 9 A B C D E F
002EEA50 00 00 FF FF 00 00 FF FF 00 00 FF FF 00 00 FF FF ..ÿÿ..ÿÿ..ÿÿ..ÿÿ
002EEA60 00 00 00 FF 00 00 00 FF 00 00 00 FF 00 00 00 FF ...ÿ...ÿ...ÿ...ÿ
002EEA70 00 00 00 FF 00 00 00 FF 00 00 00 FF 00 00 00 FF ...ÿ...ÿ...ÿ...ÿ
002EEA80 00 00 FF FF 00 00 FF FF 00 00 FF FF 00 00 FF FF ..ÿÿ..ÿÿ..ÿÿ..ÿÿ
Until offset 2F000, then it's all 00's until the end.
Thanks, cheebster.
|
Edited by - cheebster on 02/22/2006 17:01:33 |
|
|
| Reply #6
ZLM
2945 Posts |
Posted - 02/22/2006 : 20:58:43
|
All commands on 29LV160 and 29LV320 are the same. The different points are size, number of sector and the address line used. 29LV160 only used A19, and A20. But the 29LV320 used A19, A20 and A21.
|
Edited by - ZLM on 02/22/2006 21:00:23 |
|
|
| Reply #7
cheebster
Canada
12 Posts |
Posted - 02/23/2006 : 07:57:32
|
| Hey. Yeah well on the back of the tsop-48 adapter, i got a jumper that puts 3.3V on the chip instead of 5V. I will try erasing them another time. Thanks :) |
|
|
| Reply #8
cheebster
Canada
12 Posts |
Posted - 02/23/2006 : 11:06:41
|
Um. I've been trying to erase using various settings in the software. Now i did 3 reads after trying to erase, and i find out that, the chip is all 00's until offset 0x30000. The last 1 Megabyte of the chip is FF'ed out. :|
Anyone got a clue ? I'm confused, it's not the same results that i gave the first time.. i don't figure out why...
Do i need, to enable 'Fast Mode Programming' (For AMD, Fujitsu, ST Chip) or Fast Programming mode in the programmer settings ?
I did not have any problems writing 29LV160's for some reason. I have checked the VCC is at 3.29V during the whole erase process, and VPP is at 12.28V on pin 12 stable too. |
|
|
| Reply #9
ZLM
2945 Posts |
Posted - 02/23/2006 : 11:26:15
|
I'll suspect the chip is not in a good condition. If I'm not wrong you are not be able to get erase completed before the erase time out. That means your chip not been erased correctly. In this case, the content of your chip is unpredictable due to the unsuccessful erase. This is why every time the reading is different after erased the chip.
You may need find a way to approve the chip is a good chip for erasing,writing.
Playing the 29LV323 chip is time consuming. One round may take at least an half hour. You must be patient. |
Edited by - ZLM on 02/23/2006 11:29:20 |
|
|
| Reply #10
cheebster
Canada
12 Posts |
Posted - 02/23/2006 : 12:09:35
|
Um. Alright ZLM. Reading your post, i've did some other tests with Another 29LV323 now. I did an erase, it took the whole 100% to go through, and it said error erasing chip. I checked the buffer then, all FF'ed out... If i do a blank check it says the chip isn't blank ? I don't get it. I loaded the chip in winhex, it seemed all FF'ed out..
Maybe the first chip was just bad ? Um. I will try to write a bin to that one now. |
|
|
| Reply #11
cheebster
Canada
12 Posts |
Posted - 02/23/2006 : 12:36:59
|
I'm now trying to write to the other chip. I get an error at 20% Each time, Error at 0x0D000, Chip = 0x84, Buffer = 0x25 ( the first place where there is actually data in the file i load, before this offset it's just FF'ed out.
Selecting other options get the programming to hang at 20% instead of giving an error out.
I'm thinking about giving up hehe. |
Edited by - cheebster on 02/23/2006 12:41:11 |
|
|
| Reply #12
ZLM
2945 Posts |
Posted - 02/23/2006 : 15:20:29
|
The erase failed. Because you got whole 100% and "error erasing chip" message.
Some where must have value other than 0xFF. The blank check will tell you the location on the status bar.
The reason for taking whole 100% timeout is because the software does not check the DQ5 when erasing. The software only check the DQ7 for the completion of erasing. The DQ5 will be set if the chip internal erase failed on timeout. So, to confirm this, you can measure the D5 before reach the 100% erase, you will get voltage on this pin. This pin is corresponding to ZIF pin 19. Which is the last 3rd pin on right-bottom of ZIF socket.
I do not understand why it is fail on timeout. Maybe the chip needs more accurate timming for the commands. I do not know. |
Edited by - ZLM on 02/24/2006 16:49:38 |
|
|
| Reply #13
cheebster
Canada
12 Posts |
Posted - 02/24/2006 : 06:54:13
|
| Hi. I just checked activating A21 and it does gets voltage on chip pin 10 when i check it in check H/W. Odd thing is, that the voltage is 5V, not 3.3V ?!? Is that normal ? Thanks for all the help. |
|
|
| Reply #14
ZLM
2945 Posts |
Posted - 02/24/2006 : 17:08:37
|
| I got one 29LV320, first two times erase failed on timeout. Progress bar went to 100%. But third time erase succeed at 90%. |
Edited by - ZLM on 02/24/2006 17:10:04 |
|
|
| Reply #15
cheebster
Canada
12 Posts |
Posted - 02/26/2006 : 16:24:08
|
| Sorry, but what is WP/ACC ? :| I'm quite noob with programing flashes. I need to apply a voltage on pin 14 ? What's a pull-up resistor ? Thanks in advance, cheeb. |
|
|
| Reply #16
ZLM
2945 Posts |
Posted - 02/26/2006 : 21:27:38
|
| WP/ACC is the pin for accelerated writing when it is applied a 12V voltage. Here we only connect it to Vcc through a resistor. The resister is called Pull-up resister. |
Edited by - ZLM on 02/26/2006 21:28:25 |
|
|
| Reply #17
cheebster
Canada
12 Posts |
Posted - 02/27/2006 : 14:06:04
|
| So, i need to apply 3.3V (VCC used for that chip) on Pin14. Not 12V, right ? |
|
|
| Reply #18
cheebster
Canada
12 Posts |
Posted - 03/08/2006 : 12:15:01
|
| Hi there. I checked, and pin14 HAS 3.3V dc (3.17V in fact).. it's already tied to VCC. Guess i all bought this for nothing, i really needed to program 23LV323's :| Thanks to all for the replies and great help. |
|
|
| |
Topic |
|
AM29DL323GT Read/Write Problems
Forum Locked
