ATF1504AS: Questions about this CPLD

[maded2]

plasmo wrote:

maded2 wrote:

plasmo wrote:

Mine USB blaster is just a cheap $5 clone.

Maybe 7128S is different than 7064S regarding the JTAG enabling. Or perhaps the 7128S you have are something else. I'm interested to try the JTAG enabling method on 7128S. Did you buy yours from eBay?

Got them from Aliexpress. PM me with your address and I send some over after the holidays for you to try. I've 12 of them.

thanks,
eddie

Eddie,
I received 4 pcs EPM7128SLC84. I verify their JTAG ports are locked. Three of the 4 parts draw fairly high current, 200-300mA; one draws about 100mA. I have a proven programming fixture for EPM7128SLC84. Only signals hooked up in the programming fixture are power/ground and 4 JTAG signals. I added a 1K resistor to OE1 (pin84) where 12.5V can be applied through the 1K resistor. I apply 5V first and 12.5V next. I was able to erase the JTAG lock for the part with low current. Once JTAG is re-enabled, it programmed normally. However, with the three high-current parts, I was not successful in erasing the JTAG lock. I've tried many times, but failed every time.

Here is what I've observed with the successful part: I programmed with a working EPM7128SLC84 POF file. The programmer reported unsuccessful JTAG connection and programming appeared to have failed, but once I removed 12V and recycled 5V, I'm able to establish JTAG connection and program the part normally.

So there is a small success, but more works are needed to reliably erase the JTAG lock.
Bill

Thank Bill,
It's a step forward at least. I think those chips with high current draw maybe dud onces as I've no way to test it. Is the 12.5V the minimum?

anyway many thanks for your work.

eddie

[plasmo]

I really don't know what's the right JTAG-enable voltage. I've read 12V being used, so I tried slightly higher voltage to overcome potential voltage drop across the 1K resistor.

I have had a batch of EPM7128S from UTSource that were bad and drew high current as well. I returned them and received the replacement last night. They have low initial current of about 70mA and programmed & verified OK.
Bill

[okwatts]

Hi I came across this website which talks of programming these with an apparently inexpensive solution.
See https://www.hackup.net/2020/01/erasing- ... 1504-cpld/ for more details.
Talks about the re-enabling JTAG also.
Hope it is of interest and I may give it a go at least copying others CPLD equations for now.

[okwatts]

I successfully got this to work using this technique with the FTDI 232H board and a 44 pin PLCC to 40 pin DIP adaptor that came with my eprom programmer. I struggled at first to get ATMISP to convert the JED file to SVF until I did the following. Start the program without the FTDI plugged in and it will complain that there is no adaptor, ignore that and add a device after the default 0 so it becomes device 1 then you can load the JED file and tick the create SVF box and run. This should produce the file required. I copied the command line given in the article and in Windows 10 power shell when I tried openocd it complained it couldn't find a device with the parameters for the FTDI and quit. I then went to linux (a Ubuntu version) installed openocd, changed the path to the config file and plugged in the apparatus and it proceeded to program correctly. I'm not sure why Windows didn't detect the FTDI properly even though it showed up in Device Manager with the proper id. Perhaps if I had rebooted a few times it might have got it right, there is still mystery in Windows!

[plasmo]

So you don't need the expensive ($60) Atmel programmer? By converting JED to SVR, an inexpensive FT232H like this can program ATF1504 CPLD? That's cool. To bad I've already bought the Atmel programmer.

[cbscpe]

Believe me, that was a good investment. The Atmel Programmer is very reliable and you don't have to convert the JED to SVF files. Having less manual steps saves you a lot of time and avoids problems due forgotten steps or incorrectly selected files. It also comes in a nice Case. I even consider buying a spare one just to be safe in the future, in order to be able to easily and properly upgrade or modify my projects that use Atmel CPLDs. In general I always buy the official programmers for all my programmable stuff just. Debugging a project that goes wrong only because the programming went wrong as very annoying, been there done that, never again. Many times there are special offers where they sell the programmers with a discount, especially Atmel/Microchip tends to make such sales with a 50% discount. So I always wait for such an occasion. That's how I got my Atmel-ICE and I would never give it back.

[okwatts]

I agree that if you have the expertise and numbers and use a lot of CPLD devices, especially other than just the one 1504AS then a proper programmer makes sense. For those like me who have only one or two devices that came in designs from someone like Bill, the inexpensive solution is adequate to update the programming for new features or updating the code for the Boot Rom portion of the devices. I'm unlikely to do more than that, so the investment is minimal to accomplish what I need it for.

[Zapotec]

cbscpe wrote:

Believe me, that was a good investment. The Atmel Programmer is very reliable and you don't have to convert the JED to SVF files. Having less manual steps saves you a lot of time and avoids problems due forgotten steps or incorrectly selected files. It also comes in a nice Case. I even consider buying a spare one just to be safe in the future, in order to be able to easily and properly upgrade or modify my projects that use Atmel CPLDs. In general I always buy the official programmers for all my programmable stuff just. Debugging a project that goes wrong only because the programming went wrong as very annoying, been there done that, never again. Many times there are special offers where they sell the programmers with a discount, especially Atmel/Microchip tends to make such sales with a 50% discount. So I always wait for such an occasion. That's how I got my Atmel-ICE and I would never give it back.

Hi
You Mean that with Atmel_ICE it si possible do program ATF1502AS devices ? wich software than I Can use ? (Microchip studio I think don't support CPLD devices).

I ask this because I have some Altera EPM7032S new (not from China, but from old stock). So Jtag is enabled, Quertus , via USB Blaster, see the device witout any problem.
I can program & verify sucessuflly the CPLD with a simple program that leave JTAG enabled.
BUT!
I need to use also JTAG pin. Here the problems begin.

Preface : In the test condition as before (no programming problem) if I put pin 44 (OE1) at >9.6V, the JTAG become unavailable (also on new device). I try also with a JTAG scanner : with OE float (or <9.6V) it found the device, with OE @ Vpp, the device disappears from JTAG chain.

If I try to program device with .JAM (If I try with .POF, quartus warn me that will be disabled JTAG, and abort file selection), Quartus ends program, and fail verify. The device programmed work fine in circuit, expect JTAG pin that are still enabled (I can reprogram the device without problem)

The funny situation is that I have a Elenec Programme (labProg 48LV). It fails to program the device !
- fail ID, but it is already reported as issue by elnec, must be disabled the check
- then erase & blank ok, program OK
- verify fail... device is still blank
After this if I put the CPLD back under USBBlaster, Quartus still find the device in JTAG scan, but can't program again the device

I try also with an Elnec BeeHive 204... same tour as before...

So I recover an old EPM7032 (no JTAG). My programmer work fine with it. But .POF file is non compatibile, and I don't know how to convert from 7032S project (I made the project under quartus, so I have sources, but Quartus don't still support old 7032 (not S) device

So now I'm thinking to buy a Atmel device and pray that my Elnec program it (or a good ansewer to my first question in this post )

Thanks !
Paolo