USB Blaster replace/repair

from ALtera support "FIND ANSWERS"

Problem

Are there any known errors in the Stratix and Cyclone Configuration chapters

with reference to the external pull-down resistor value recommended for the

JTAG TCK signal?

Solution

Yes. The JTAG TCK pull-down resistor value is incorrectly specified as

10k-ohms in the Stratix and Cyclone Configuration chapters. The correct

external pull-down resistor value is 1k-ohm. This stronger resistor ensures

that the TCK signal is biased at a logic low level despite the weak internal

pull-up on this pin. The internal pull-up ranges from 20k-ohms to

40k-ohms across process, voltage and temperature.

Future versions of the Stratix and Cyclone Handbooks will be updated accordingly.

I have the 1K pulldown on TCK, and 1K pullups on TMS, TDI. TDO has nothing.

Some of the pulses on TCK look OK, but many are runt pulses. There could be timing problems as well.

In contrast, the byteblasterII produces nice clean signals.

I've had several ByteBlasterII's that either failed after a few uses or never worked. Some painful (partially due the fact Altera hasn't published a schematic for this device) investigation showed the problem to be weak vias on the PCB that weren't reliably plated through. Some vias on the board had been repaired/reinforced at the factory using a small length of wire (standard practice). Some apparently were missed during assembly/test, or were ready to fail but hadn't yet. After these were tracked down and repaired the blasters work fine.

Maybe the USB Blasters are subject to the same lack of quality control. http://forum.niosforum.com/work2/style_emoticons/<#EMO_DIR#>/wink.gif

Good to know about the vias.

Altera is taking the attidude that this is a vendor problem. I can&#39;t find the PO so I don&#39;t know the vendor was. The vendor-of-the-day on Altera&#39;s site has no record of this order, and say they would not replace it anyway since it is more than a month old. Etc,etc, kiss that $400 goodbye.

Are there really no schematics or 3rd party cables?

The USB Blaster board is 4-layer, and has an EPM7064. As long as you could get/read the code for the PLD, it would not be hard to duplicate. Registration is a bit off so I can believe it might have a via problem.

The board does work for certain operations (below). It might be easier to fix if it was completely broken.

Using ByteBlasterII, no problem with any operation using any type of file: sof,pof,jic

Using USB Blaster (http://forum.niosforum.com/work2/style_emoticons/<#EMO_DIR#>/cool.gif ,

Programming pof via active serial header works every time

Programming sof via jtag header works sometimes

Programming jic via jtag header never works, but with different errors

Using ByteBlasterII:

Info: Configuring device index 1

Info: Device 1 contains JTAG ID code 0x020850DD

Info: Configuration succeeded -- 1 device(s) configured

Info: Device 1 silicon ID is 0x10

Info: Performing verification of type standard on device(s)

Info: Successfully performed operation(s)

Using the USB Blaster:

Info: Configuring device index 1

Info: Device 1 contains JTAG ID code 0x020850DD

Info: Configuration succeeded -- 1 device(s) configured

Error: Serial Flash Loader IP not loaded on device 1

Error: Operation Failed

or

Info: Configuring device index 1

Error: Can&#39;t configure device. Expected JTAG ID code 0x020850DD for device 1,

but found JTAG ID code 0x00000000

Error: Operation Failed

Finally got around to fixing the USB-Blaster.

There was an open ground connection due to a bad via. The ByteBlasterII is still more dependable when using the debugger, and isn&#39;t that much slower to download.

Hi everyone!

I have an experience ,but I&#39;m not sure1

I have the 1K pulldown on TCK, and 1K pullups on TMS, TDI.

For TDO ,in datasheet there is no connection, but When I pull up TDO.

It&#39;s morer stable than before. I think there is a error inthe datasheet!

Best Regards,

Fei Wu