EZoFlash+4v4 serial memory support modify

[Ezo]

Find here changes to support serial memory on EZoFlash+4v4 -
schematic changes in red, generic serial memory chip pinouts (DIP-8,SO-8 ), layout changes and connections.
Actions required
1) cut line between IC7/pin9 and V1/k
2) place resistor 1k between IC7/pin1 and IC6/pin7 (D0)
3) place wire between IC7/pin2 and IC7/pin9
4) apply wires to external adapter or DIP socket
- DIN to V1/k,
- DOUT/DATA to IC7/pin1,
- Vcc,D1,GND,WE to responding BU3 pins.
-----------------------------------------------------------
Common comments:
- software versions 0.98dx can be used for serial memory;
- no hardware changes to pcb45 version required;
- latest SW 0.98d9 support SPI Flash memory, voltage Vcc=3.6 required.

[Ezo]

My tests results find in latest chip_testfile.
Comments:

- I2C series 24xxx
1) No problems found on generic devices.
2) Special devices CSI64LC10, 24LC21A, X24645 fail/not supported

- Microwire 93xxx
1) Use always chip datasheet to find pinout, supported mode (8bit, 8/16bit, 16 bit), pin ORG.
2) For 8/16bit chips keep pin ORG open in 16bit mode, pin ORG ground in 8bit mode.
3) Found SW bug. Most 8/16 bit chips can be programmed OK in 8bit and 16bit mode, but compared read in both modes different - bytes in word are swapped. From verify results on other programmer - byte swap exist in 16bit mode.
8bit mode looks OK. Probably this is reason why practically all willem HW versions support 93Cxxx devices in 8bit only by default, pin ORG grounded permanently. Software allow 16bit mode, software has pin ORG usage instructions, but no public comments on HW change/line cut and related jumper implementation.
4) 16 bit mode can be used. Load file, Action/Swap BYTE, Program and/or Read chip, Action/Swap BYTE, Save as. Making copy from one chip to other one you can skip Swap.
5) Notes from tests.
Atmel chip read is equal in 8 and in 16bit mode.
CSI93xxx read in both modes have more differencies (seems swapped half memory area, swapped bytes; chips failed on other programmer too).Only one mode can be used for CSI93xxx programming, more tests and verify required to confirm reliability in 8bit mode.
Microchip C version chips 93AA66C, 93LC76C fail, probably version compatibility issue.
Specific chips 93CS56, M93S66 fail/not supported.
6) Waiting your comments, experience, confirmation or denial of listed ideas and results.

- SPI EEPROM
1) initial tests done on SW 0.98d9. Most chips are programmed OK, but read always return clear buffer (all 0xFF).
2) main tests on SW 0.98d5 with SW bug found. Most chips are programmed OK, but read always return 0xFF on first byte (except Microchip 25LC320), separate verify fail on first byte. Verified chip on other programmer - chip has record in first byte.
Equal first byte problem exist with programmer pcb45c, with older SW 0.97ja
3) Looking to have chip copy in file (backup), read and check always first byte in buffer, edit before Save.
Actions step by step - read chip; check buffer; if first byte FF run Verify; from Verify error result get chip content in first byte; open willem SW Edit Hex function in buffer; edit first byte; run verify; save in file.
4) Another bug - no success with 4kb eeproms (AT25040, CSI25C05, M95040) , programming fail.
Fail Microchip 25LC080 , probably due to 16byte page (selected device 25x080 with 32byte).

- SPI Flash.
1) supported by SW 0.98d9
2) apply voltage 3.6V , jumper Jp2 only.
3) erase chip before programming
4) AMIC chips verified OK.
5) ST Micro M25P040 and Winbond W25X40 fail.
Reason unknown, not found. Mentioned chips programmed succesfully on modified (Vcc) programmer pcb45c.
6) serial part of modified ezoflash+ is different from pcb45version . What components can be critical on timing in my version for SPI Flash chip range? Experiments take a time .

Serial memory modify support PIC microcontrollers. No tests done.

Waiting your feedback.

[md_bluelily]

Dear EZO,

If I make this changes on my working board (EzoFlash+ v4.4), then is it able to program the chips that it used to program ? Or it is only able to program SPI chips then ?
I want to program these two chips: M95080 and M95160. Will it be able to program these two or not after making new changes to it ?
Is there any adaptor available for DIP-8 and SO-8 packages ? If no, then how am I supposed to connect my chips to the board ?
Please help me.

Thanks in advance for your attention and support.
Best Regards.

[Ezo]

Changes affect to all serial devices.
Chips M95080 and M95160 are programmable. SW has bug, that read can fail on first byte (0xFF), but chip programmed OK.
Here is no adapter for serial chips, you can use wired connection or create adapter yourself.
Topic will be copied in EZoFlash forum english section.

[md_bluelily]

Dear EZO,
Thanks for your reply.
So, do you mean that I can still program other not serial chips too without any problem ? I mean chips that are not serial like PLCC32 and etc.

Thanks in advance.

Best Regards.

[Ezo]

Programmer will keep functionality on other adapters/chips with 0.97ja
Report if my utilities fail, because changes in serial part makes my programmer Willem compatible.

[md_bluelily]

Programmer will keep functionality on other adapters/chips with 0.97ja
Report if my utilities fail, because changes in serial part makes my programmer Willem compatible.

Thanks again.
Sure I'll put the test result here.
Best Wishes.

[md_bluelily]

Dear EZO,

I made the modification to the existing working EzoFlash+ 4v4. I connected the M95080 to the programmer using the 25xxx connection model as you have described in the modification schema which is as follows:

Chip pin 1: WE
Chip pin 2: D Out
Chip pin 4: GND
Chip pin 5: D In
Chip pin 6: D1
Chip pin 3: ---- Chip pin 7------ 2k2 Resistor --------|
Chip pin 8: -----------------------------------------------|----- Vcc

I hope this is the right connection.

I put the jumpers as follows: 1st and 3rd.
Now when I try to read the chip (which is totally empty) using 0.98D10, it returns a buffer full of 0xFF.
When I try to write back the read buffer, it prompts that it has written it successfully and it even verifies it and reports "Device Programmed OK".
But when I try to write another file onto the chip, it fails with the following error:
Error at 0x000000 Chip=0xff buffer=0x46

I noticed that the address (0x000001) is the address of the character which can not be written on the chip and the buffer(0x46) is the code of that character. I also noticed that the address (0x000000) is 0xFF.

I should also mention that I'm able to use 0.97ja to read and write PLCC32 chips.

Thanks for your help and support.
Best Wishes.

[md_bluelily]

Dear Ezo,
Do you have any idea ?

Best Wishes.

[Ezo]

But when I try to write another file onto the chip, it fails with the following error:
Error at 0x000000 Chip=0xff buffer=0x46

Read chip, what is in buffer, comparing to source ?

[md_bluelily]

But when I try to write another file onto the chip, it fails with the following error:
Error at 0x000000 Chip=0xff buffer=0x46

Read chip, what is in buffer, comparing to source ?

Dear Ezo,
I tried to read the chip after I write it with error. The first time that I read it, it returns a buffer filled with 0xFF and exactly after that when I read it it returns a buffer filled like the following:

Hex Address: -----00--01--02--03--04--05--06--07--08--09--A--B--C---D--E--F
Read Data: -------C0--C0--00--00--00--00--00--00--0C--0C-00-00-00-00-00-00
Then
Hex Address: -----00--01--02--03--04--05--06--07--08--09--A--B--C---D--E--F
Read Data: -------80--C0--00--00--00--00--00--00--00--0C-00-00-00-00-00-00

But the important thing is I think I can not write on the chip at all. The process of writing finishes too fast with a file loaded in buffer. This process takes longer with a clear buffer and also it verfies it too.

What's your idea dear Ezo ?
Thanks in advance.

Best Regards.

[Ezo]

Check 95080 pin layout.

[md_bluelily]

Dear Ezo,

The problem is fixed now. I tested another two chips and found out that my chips were corrupted. So I got two new M95080 and M95160.

I programmed M95160 successfully. Even it reads the chip without any problem. The first two bytes are ok too. I should perform the read command twice. The first time it returns 0xFF bytes and the second time it returns the right data.
But I'm not able to read the M95080 chip. Verify command reports that the content on chip and on buffer are identical and works correctly but whenever I read the chip it returns a buffer filled with 0xFF no matter how much I try.
Is there any solution to fix this problem too ?

Anyway, thanks for your help and support.
Best Wishes.

[Ezo]

I do not have experience on serial chips. M95080 was verified OK*(first byte read error), you can read from my chip test file. Another 8KB chip Microchip 25LC080 fail due to different page size.

[md_bluelily]

I do not have experience on serial chips. M95080 was verified OK*(first byte read error), you can read from my chip test file. Another 8KB chip Microchip 25LC080 fail due to different page size.

OK. No problem.
Thanks dear Ezo for your help.
I really appreciate it.