Programmer freezes on BeMicro CV - used to work fine

[Altera_Forum]

Help! I have 4 BeMicro CV boards that all of a sudden do not program. The programmer freezes up until I unplug the USB cable. When I try to use the JTAG debugger tool, the same happens - the Quartus application hangs until I remove the USB. 

 Everything had worked well up to this point. I had programmed each of these boards several times with the same image (.jic file). Then, my last image appears to have corrupted something beyond recovery.  The BeMicro CV is a Cyclone V with a built-in Blaster II on a MAX part. I'm using Quartus 13.1.2 64-bit on Windows 7.  Here's what seems to show that it's NOT the USB driver, my computer settings, or my cable: I have a DSE0 eval board, which also has a built-in Blaster II. With what I believe is the same programming setup (different programming file of course), I can program that board without any problem.  So, here's what I've tried so far: 1) a variety of USB cables, 2) uninstalled and reinstalled the Blaster driver, 3) attempted JTAG debugger (as previously mentioned), 4) tried an old programming file that I know worked, 5) confirmed that I can program another demo board successfully, and 6) rebooted a few times.  Any suggestions greatly appreciated!

[Altera_Forum]

Hi dave321, 

 Did you manage to find out what happened to your boards? I'm having problems with my BeMicro CV board as well, but not with the programmer freezing up, but during output of printf commands through the JTAG. Wondering if your solution might be applicable to mine.  Thanks

[Altera_Forum]

I concluded that I had somehow blown up my boards (yes, all 4 at once). I had them powered together thru a USB hub and suddenly none of them would program ever again. Perhaps it was a power surge or ESD (both very unlikely, however). And it seems to have killed only USB functionality - the FPGAs all still ran fine. I didn't have time to deal with it so I ordered replacements and then was able to program the new ones with the identical setup and files. 

 Please let me know if you solve your problem and/or find out anything from Arrow/Altera. Best of luck!

[Altera_Forum]

I am also having the same problem with the CV boards. I bought two for two different project prototypes, a quick way to get a rough demo going. So far, I have wasted more time with this programmer issue than anything productive. This applies to both boards which are connected to two different computers. None of the I/O has been connected thus far, boards were to tested right out of the box first, that has been going on for three days now. One computer is Windows XP and one is Windows 7 64bit. Both boards fail the same when trying to program them. Active Serial always fails. JTAG worked occasionally but hangs every time now, even after cold reboots. Converting to a JIC file worked once, never again. Trying the active serial again after that, it seemed to erase the chip but never programmed it. I was getting some testing done until JTAG stopped working. 

 A real UB Blaster works flawlessly on the same computers with other boards.  I sent a email to the Arrow Altera FAE late Friday afternoon, I'll repost if he has any answers. But it seems that the BeMicro blaster implementation is very poorly designed. Not sure what else to do other than try to modify for a real USB blaster interface. Any help is appreciated!  I wish Altera would produce their own lo cot board that actually works.  Also, why does this forum drop so many characters while typing? Very annoying, I have to manually correct one out of every four words.  Joel

[Altera_Forum]

Hi Joel, 

 I have a couple of BeMicro-CV boards and never had any trouble with either of them. I purchased them when they first came out. When did you buy yours? I wonder if Arrow changed something on their design (though it was pretty poorly documented anyway!). I posted an example design here (Post#5);  http://www.alteraforum.com/forum/showthread.php?t=43992  Maybe you could try downloading that design ... and then erase the EPCS ... I'm not sure it'll help, but it shouldn't make things worse (since I've used the design on a couple of boards).  Cheers, Dave

[Altera_Forum]

Thank you Dave, 

 I bought mine last week from Arrow, arrived Thursday. I did try your design, same problem, there is no access via the built in USB blaster at all. None of the usual functions like erase work, all just hang indefinitely (but a real USB Blaster works fine with other boards). Even when the JTAG "worked" it would hang at 93% completed. At first, active serial would give a quick blink of the LED and indicate failure on screen almost immediately. Now everything just hangs. I've tried deleting and reinstalling the driver and also downloading Quartus 14.0  My initial test project was about as simple as it gets, a 32 bit counter from the 50Mhz clock driving two LEDs. I saw it work briefly on two occasions where the JTAG sort of worked. I've used this project for years as a simple first test with no problems (been working with Altera parts since the days of the classic parts).  I am debating if I can rip out or disable a few pins on the Max chip and wire its 10 pin blaster header to the EPCS16 instead. Too tired to try that tonight however plus I would like to see what the FAE has to say first.  Thank you for your assistance, I wish it had worked.

[Altera_Forum]

For full functionality, you'll better connect the FPGA JTAG pins than the EPCS. According to the (incomplete) schematic in the "Hardware Reference Guide", they have test points. You'll possibly need to hold the MAX V in reset.

[Altera_Forum]

--- Quote Start ---   I bought mine last week from Arrow, arrived Thursday.   --- Quote End ---   Ok, so at least the Arrow FAE has a reason to respond, given that these are new devices.  If they had not worked at all, I'd suspect that the FTDI EEPROM was programmed (causing it to be recognized as a USB-Blaster), but the MAX V USB-Blaster IP was not (so it would not function as a USB-Blaster).  My next guess would be there is a problem with the USB-Blaster, eg., power supply or clock. Perhaps you could probe around on the boards you have and see if voltages are incorrect, or that the clocks look ok. I'll upload a copy of the schematic that I found at some point (its better than the copy in the user guide).  Cheers, Dave

[Altera_Forum]

Agreed but even if I can just program the EPCS16, I'll be happy. I've found a few errors in the reference materials, one being where it refers to the Blaster as a " USB Blaster II" in some places and not others. Win 7 would not recognize it as a Blaster II, I had to install he Blaster drivers. Win XP did detect it as a Blaster II, which I also find very odd. I didn't get to look at the Blaster part of the schematic yet but good idea on the reset.

[Altera_Forum]

The Arrow FAE has been VERY good at helping so I do expect a reply. Basically I am just not clear on who actually designed and supports the BeMicro stuff. My feeling is that the problem is somewhere in the Max V IP but indeed you are correct about checking everything thoroughly first. My hope was to do a lot of work at home this weekend to get a head start on it so I don't have all of the equipment here but will check it out at work tomorrow. The fact that Windows always recognizes a connect/disconnect event tells me the FTDI is probably operating okay. 

 I appreciate the better copy of the schematic, the one in the user manual has a few issues.  Again, everyone's help is appreciated a lot!  Joel

[Altera_Forum]

Joel or anyone else here have you found any resolution to this problem? 

 I got a brand new BeMicro CV today as part of the BeScope kit. I programmed it first shot, zero trouble. The FPGA came up and is running great. The system console would not connect (just hangs; never detects the device) and I discovered I can no longer access the board with JTAG nor program it with any method. The symptoms are exactly as described here; the USB enumerates but nothing else.  I supplied the board with external 5V with no change.  I routinely develop with other Altera devices and have no trouble.  I am not even sure where to request support or report the issue. It took two months to get a SoCKit board RMA'd through Arrow when I had to do that before. These devices are supposed to promote designing with Altera parts, right? The only thing they are telling me is that you will have to spend more time fussing around on the test bench than you will be able to spend working!

[Altera_Forum]

Hello, 

 I found this solution on another forum and it fixed the issue. Please rotate jumper J11 clockwise from the default 1-3 position(Vccio=2.5V) to 1-2 (Vccio=3.3V).

[Altera_Forum]

--- Quote Start ---   I found this solution on another forum and it fixed the issue. Please rotate jumper J11 clockwise from the default 1-3 position(Vccio=2.5V) to 1-2 (Vccio=3.3V).  --- Quote End ---   I just went through the schematic I posted and looked at the devices connected to VCCIO, and they all should operate fine at 2.5V and 3.3V. I do not see an issue in the board design that would cause JTAG not to work with VCCIO set to 2.5V.  I'll test the board I have at work tomorrow and see if it fails ...  Can you post a link to the "other forum" discussion?  Cheers, Dave

[Altera_Forum]

--- Quote Start ---  I just went through the schematic I posted and looked at the devices connected to VCCIO, and they all should operate fine at 2.5V and 3.3V. I do not see an issue in the board design that would cause JTAG not to work with VCCIO set to 2.5V.  I'll test the board I have at work tomorrow and see if it fails ...  Can you post a link to the "other forum" discussion?  Cheers, Dave  --- Quote End ---    Hi Dave, The discussion on this forum discusses the BeScope design which uses the BeMicroCV. Post# 67 mentions the jumper to fix the programmer. http://forums.parallax.com/showthread.php/156989-bescopebundle-tip-for-those-who-want-to-buy-an-fpga-development-board/page4  I was having the same issue and changing Vccio=3.3V fixes the communication issue. I haven't researched enough to figure out why this is.  Thanks Carl

[Altera_Forum]

--- Quote Start ---   I'll test the board I have at work tomorrow and see if it fails ...   --- Quote End ---   The board I have works fine with the VCCIO jumper in either position.  The programmer hangs if the jumper is missing, since the FT245R VIO and the JTAG interface is powered via the jumper. The jumper is a 2x2 header, with only two valid settings, but four ways to insert the 2-pin jumper, so perhaps the people experiencing a problem did not jumper to pin 1 (which has a square shaped pin when you turn the board over).  Cheers, Dave