Author Topic: DE10-Nano JTAG broken?  (Read 4379 times)

0 Members and 1 Guest are viewing this topic.

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
DE10-Nano JTAG broken?
« on: September 20, 2020, 11:06:38 pm »
Disclaimer: I'm crossposting this from Reddit since I didn't get too many replies there. Hopefully someone here encountered this issue before and can provide more input.  |O

I recently got an Altera DE10-Nano board. Everything on it (USB-OTG, remote desktop'ing into the board, serial, LEDs, accelerometer) works, except the JTAG Mini USB. I tried on 3 computers, and on both Windows and Linux.

I started by removing the SD Card as the guide suggested, connecting the Mini USB cable, then plugging the board in.

In Windows, no new USB device appears in device manager. Quartus Prime doesn't detect anything. I've tried this many times and tried to install drivers to no avail. I don't even get the Windows "device plugged in" sound.

In Linux, nothing appears under lsusb or dmesg. jtagconfig doesn't find anything. I tried running it and the service both as root to no effect. It's like nothing got plugged in at all, just like in Windows. I also made sure to follow the relevant instructions in Arch Wiki to get it to work, but that failed too.

I've tried both with MSEL set to the factory config and with all 1's. I've also tried the AS x1 mode (10010) to no success. The JTAG port appears completely dead. I've verified continunity on the port and that the port at least isn't phyiscally broken or badly soldered. Anyone know what I could check next to make sure the USB Blaster circuitry isn't dead?

Am I missing something with the configuration switches on the board, or is it just broken? I have no idea what's wrong. Please help.
 

Offline tru

  • Regular Contributor
  • *
  • Posts: 107
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #1 on: September 21, 2020, 07:23:15 am »
There is nothing special to configure, you just plug in power to the power jack and plug in a USB cable into the USB Blaster II mini USB socket (the one next to the HDMI port).

During the installation of Quartus Prime there is a prompt at the end asking permission to install the USB Blaster II drivers, but even if you didn't opt to install them you will see it listed under the yellow icon "Other devices" tree in Windows Device Manager.

I attach picture of my Device Manager when the device is plugged in and as you can see it is listed under "JTAG cables" tree.

One thing I did notice is that when the USB-OTG micro cable is left plugged in (other USB socket) the USB Blaster II device is not recognised.

It looks like you've already tried so I think your board is broken, i.e. USB Blaster circuitry section as you've mentioned.
« Last Edit: September 22, 2020, 01:04:11 pm by tru »
 

Offline Daixiwen

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: no
Re: DE10-Nano JTAG broken?
« Reply #2 on: September 22, 2020, 07:18:32 am »
If you don't even hear the connection sound or see a device with a yellow exclamation point in the device manager (in Windows) then it is a problem very near the USB connector. Either the USB connector itself or the chip connected to it (is it an FTDI chip? I don't remember). I would check the soldering on the USB connector first.
The MSEL setting will not affect the connection between the USB device and the PC host, so as long as you don't see the USB device from the PC you don't need to try and change those settings.
 

Offline tru

  • Regular Contributor
  • *
  • Posts: 107
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #3 on: September 22, 2020, 01:11:16 pm »
Btw, the schematic is downloadable:
https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=English&CategoryNo=205&No=1046&PartNo=4

The USB Blaster II section is on page 10 and 11.  There is
1. a CY7C68013A (Cypress USB 2.0 controller) which provides the USB support
2. an EPM570GF100C5N (Altera CLPD) which provides the USB Blaster II proprietary Altera JTAG protocol

Since you don't even get a device connection I guess the Cypress circuitry section is dead.

It looks like there is a JTAG header that connects directly to the CPLD (JTAG controller) and so bypasses the USB, but a 2x5 pin header is not soldered in.  This header is located next to the 5V power jack.
« Last Edit: September 22, 2020, 01:16:15 pm by tru »
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3320
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #4 on: September 22, 2020, 01:40:41 pm »
Have you tried a different USB cable?
 

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
Re: DE10-Nano JTAG broken?
« Reply #5 on: September 22, 2020, 08:01:51 pm »
Have you tried a different USB cable?

The USB cable that came with the board works perfectly with the serial port. I've also tested continuity on the cable and the port. There's no issue with the port or cable as far as I can tell.

Btw, the schematic is downloadable:
https://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=English&CategoryNo=205&No=1046&PartNo=4

The USB Blaster II section is on page 10 and 11.  There is
1. a CY7C68013A (Cypress USB 2.0 controller) which provides the USB support
2. an EPM570GF100C5N (Altera CLPD) which provides the USB Blaster II proprietary Altera JTAG protocol

Since you don't even get a device connection I guess the Cypress circuitry section is dead.

It looks like there is a JTAG header that connects directly to the CPLD (JTAG controller) and so bypasses the USB, but a 2x5 pin header is not soldered in.  This header is located next to the 5V power jack.

I figured it would be the Cypress chip that was dead, though I'm not sure why it would arrive dead randomly. I planned to try buying a standard USB Blaster next, but I'm going to try to return it first since it was DOA. For now unless I can revive it I'll just use the SD card.

If you don't even hear the connection sound or see a device with a yellow exclamation point in the device manager (in Windows) then it is a problem very near the USB connector. Either the USB connector itself or the chip connected to it (is it an FTDI chip? I don't remember). I would check the soldering on the USB connector first.
The MSEL setting will not affect the connection between the USB device and the PC host, so as long as you don't see the USB device from the PC you don't need to try and change those settings.

As said above it's a Cypress chip and I already tested the USB port's and cable's continuity. The FTDI chip was at the serial port iirc.

I've also tested both USB 2.0 and 3.0, which obviously made no difference.

I think it's really just dead. If anyone has suggestions to potentially revive/debug the cypress chip that would be great.
 

Offline tru

  • Regular Contributor
  • *
  • Posts: 107
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #6 on: September 22, 2020, 09:34:00 pm »
Had a relook at the schematics and the Cypress is using 3.3V.  There is a blue LED that should light up as soon as you apply power to the power jack, yours is lit?
 

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
Re: DE10-Nano JTAG broken?
« Reply #7 on: September 22, 2020, 09:36:50 pm »
Had a relook at the schematics and the Cypress is using 3.3V.  There is a blue LED that should light up as soon as you apply power to the power jack, yours is lit?

Yup. And the other green LEDs are also dimly lit as in your picture. Literally everything on the board (HDMI, Serial, Power, USB-OTG, accelerometer, GPIO, etc, etc) works except the JTAG which is completely dead.
 

Offline Dmeads

  • Regular Contributor
  • *
  • Posts: 164
  • Country: us
  • who needs deep learning when you have 555 timers
Re: DE10-Nano JTAG broken?
« Reply #8 on: September 26, 2020, 02:39:29 am »
I had a similar issue with a DE10-lite.

All my classmates had the same board and mine was the only one bugging out. unlike yours, mine would show up, but disappear when I tried to program it. There was nothing going through the JTAG chain when I tried to test continuity and it said the JTAG ID didnt match up.

Even though I knew the cable was fine, I swapped it out and I was able to program the board like 5 times before the same issue started happening, then it wouldn't program any more. If youre really in a pinch, I would just try a bunch of cables, even if they might only work once or twice. (Also a complete reinstall of quartus worked for a very short time). Basically no long term fix  |O

Don't even try asking the question on the intel forums, total waste of time. Im pretty sure my issue was with the USB-blaster chip.
 

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
Re: DE10-Nano JTAG broken?
« Reply #9 on: November 30, 2020, 04:44:39 am »
I got back to messing with this board. JTAG's dead regardless of the USB cables I try. I tried a Chinese USB Blaster next (IC is sanded. PCB is blue and has FYDK on it.) which also failed on the JTAG header I just populated. jtagconfig -d returned
Code: [Select]
Unable to read device chain - JTAG chain brokenQuartus similarly can't detect any devices and JTAG chain debugger gives an unhelpful
Code: [Select]
Error: JTAG chain problem detected
Error: No device detected
The chinese blaster appears in device manager and accepted the quartus drivers just fine. I've also verified continuity on the JTAG cable & header.

I heard some other Terasic boards had something that needed to be removed in order to use the onboard JTAG header? I'm connecting the USB blaster and powering it on before powering the board on if that makes a difference. (Reversing the order makes no difference)

Does anyone have any suggestions for where to go from there? I'm starting to think the FPGA BGA IC might not be soldered correctly.
 

Offline tru

  • Regular Contributor
  • *
  • Posts: 107
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #10 on: December 03, 2020, 05:54:31 am »
I see JTAG chain error message.  For the JTAG chain the HPS and FPGA needs to be in a closed loop (see page 2 of the attached pdf).  It is a pity the PCB doesn't have silkscreens for the resistors, but anyhow I've discovered most of the JTAG related ones, see the attached pictures.

Are any of those resistors on my board missing on yours?
Also they must measure 0 ohms.
 

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
Re: DE10-Nano JTAG broken?
« Reply #11 on: December 05, 2020, 12:15:37 am »
Hello, my board is identical to the board you've shown in the pictures. No missing resistors except those marked DNI. I'm under the impression that I should bridge RN9 if I want to use the onboard JTAG header, is that correct?
 

Offline tru

  • Regular Contributor
  • *
  • Posts: 107
  • Country: gb
Re: DE10-Nano JTAG broken?
« Reply #12 on: December 05, 2020, 03:03:58 pm »
Hello, my board is identical to the board you've shown in the pictures. No missing resistors except those marked DNI. I'm under the impression that I should bridge RN9 if I want to use the onboard JTAG header, is that correct?
Oh yeah, DNI = not populated.  No, as I see it RN9 is for the JTAG socket to connect to the USB chip, and this isn't needed for external JTAG. In the pdf, the JTAG connector connects through RN10 which directly goes to the same traces as the EPM570GF100C5N CPLD (onboard blaster).
JTAG_Blaser_TDI, JTAG_Blaser_TDO, JTAG_Blaser_TMS, JTAG_Blaser_TCK.

The only components left for inspection are those two pullup resistors R26 and R27, but I cannot identify them on the PCB.  >:(
« Last Edit: December 05, 2020, 03:07:51 pm by tru »
 

Offline IndependenceDueTopic starter

  • Newbie
  • Posts: 6
  • Country: 00
Re: DE10-Nano JTAG broken?
« Reply #13 on: December 19, 2020, 05:08:58 am »
Hello, I've found diagrams of the board's layout and have located R26 and R27 (both were 1k ohms as expected...)

I've attached it below.
« Last Edit: December 19, 2020, 05:16:08 am by IndependenceDue »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf