Silicon Devices - Others

[coilcap]

I have several Xilix EPROM's that I need to program. The EPROMS I have are marked as follows: 1st line: "9428A", 2nd line: "17128", 3rd line "Xilinx logo JAPAN". Xilinx document "XC1700 Family of Serial Configuration PROMs" version 4.0 dated Jan 1996 describes marking of these devices which does not match marking on my devices. I tried to program these devices as XC17128D and XC17128X and XC17128E and XC17128L and each time I got an error. So I've ruined a few IC's already. Does anyone know what XC17128 (without D,X,E or L) is and how is it different from, say, XC17128D? Thank you.

[eteam00]

Have you already searched the Xilinx website?

Another alternative is to open a webcase.

-- Bob Elkind

[coilcap]

Bob, thank you for reply. I’ve submitted webcase however Xilinx application engineer have not been able to find much information on XC17128 so far. My only hope at this point is to see if anyone on this forum has any experience with these IC’s.

[gszakacs]

It would help to know what error you're getting, and what kind of programmer you are using.

In my experience with this family of parts, I remember that there are some settings that

control the polarity of the reset and OE signals (this is from memory, so take it with a grain of salt)

and the default polarity does not match what you need for programming a Xilinx FPGA.  Furthermore,

the polarity control bits are not part of the file generated by the Xilinx tools, so you need to set it

up in the programmer.  Bottom line - if the "error" is in using the part to configure a Xilinx FPGA,

and not in programming the 17128 device itself, it may just be a matter of settings in the programmer.

-- Gabor

[coilcap]

Gabor, thank you for the reply. I am getting the error when trying to program XC17128 using Phyton ChipProg-48 set to Xilinx P/N XC17128D. Unfortunately, there is no XC17128 part number in the programmer’s device list. The programmer software reports “incorrect device type”. When I hit “ignore” the IC gets programmed, but the checksum fails. Also when I read the memory back, its content does not match the original binary file I used to program the EPROM. This is strictly EPROM related, I have not connected the EPROM to the FPGA yet.

[rcingham]

If the first line is the datecode, and these parts are 18 years old, is it possible that poor storage conditins have damaged them?

I haven't used any of this family for over 12 years, but I would expect them to be 5V operating voltage. I cannot remember if they need an elevated programming voltage. Back in the day, they "just worked" in the programmers we were using...

You really need to get hold of a set of datasheets and compare them, but perhaps you knew this already.

Addendum:

If these are pre-1996 parts, this suggests that you might have difficulty programming them with newer kit:

http://www.xilinx.com/support/documentation/customer_notices/pcn95010.pdf

[eteam00]

If the first line is the datecode, and these parts are 18 years old, is it possible that poor storage conditions have damaged them?

Interesting possibility.  Plastic DIPs are not hermetically sealed, and humidity will eventually degrade these devices.  If this was the case, I would expect inconsistent effects from one device to the next.  If they all behave the same, that's a good sign there is some other problem at hand.

Suggestion:  Since these are one-time-programmable devices, they must be replaced in the target system for updates (hence the usual DIP socket).  If these are socketed, why not replace these (ancient) parts with a socket-compatible flash-based emulator circuit which will work for self-config for the FPGA, while allowing modern tools to program (and re-program) the devices.  This will be a pain in the butt for the new development work, but finally the modules will effectively be re-programmable versions of the 17128, and might simplify field support for the end users.

Maybe such a circuit or module already exists?  I have not searched for one, but it's worth a few minutes of effort.

-- Bob Elkind

[gszakacs]

I checked our BP-Micro BP1410 programmer.  It has XC17128 listed in the available parts (no suffix)

as well as the D suffix.  It also puts up this message when you select the XC17128:

-- Gabor

[eteam00]

Found a datasheet for 17128E/EL here.  This may not be the correct or authoritative datasheet.  The E or EL parts are one time programmable, according to the linked datasheet.

Found another datasheet for 17128D devices here.  This also describes the 17128D devices as one time programmable.

Can anyone find a datasheet which suggests erasable and re-programmable?

-- Bob Elkind

[gszakacs]

The entire XC17xxx series is an EPROM process, which means that without a quartz

window for UV erasure, they are all one-time programmable.

We used to buy AT17LVxxx series parts that were EEPROM-based, but I'm not sure about

the availability of those.  I found some very pricey ones on DigiKey:

http://www.digikey.com/product-detail/en/AT17LV256-10PU/AT17LV256-10PU-ND/1008487

-- Gabor