MSP430-JTAG-TINY problem

I’ve been using my JTAG-TINY for a few months now and I’ve found that it is really nice most of the time, but very bad some of the time. Here’s what I’m seeing happen:

JTAG-TINY is plugged into my USB port

Device is self powered with 2AA batteries (I’ve used it with 7 different devices, all self-powered)

When the JTAG cable is plugged in, my PC shuts off instantly

Out of around 1000 uses, this has happened four times now. One of the times this happened, it took out my PC’s internal USB hub permanently. Since there is not any documentation saying otherwise and it always works fine with the parallel JTAG programmer, I assumed this procedure was ok.

Does anyone have any comments about this?

I think you should check the connection between the Vcc on your board and the JTAG connector. Your Vcc should be connected to either JTAG pin 2 or pin 4.

(1) Since your board has its own Vcc power, you should connect your Vcc to JTAG pin 4. In this case, the JTAG Tool will adjust the voltage of its I/O signals to match your Vcc.

(2) If you decide to turn off your own Vcc power supply and let JTAG Tool supply Vcc power to your board, you should connect your Vcc to JTAG pin 2 instead. But do not do this when your board is self-powered. The two power supplies (one on your own board and one in the JTAG Tool) will fight; and one (or both of them) might get hurt.

OldCow:
I think you should check the connection between the Vcc on your board and the JTAG connector. Your Vcc should be connected to either JTAG pin 2 or pin 4.

(1) Since your board has its own Vcc power, you should connect your Vcc to JTAG pin 4. In this case, the JTAG Tool will adjust the voltage of its I/O signals to match your Vcc.

This is my how my boards are configured. I’ve witnessed it take down my PC when I plugged it into three different designs. One was powered by two AA batteries, one was powered by a 3.3V, 1.5A, isolated supply (this is the one that permanently damaged my PC’s USB hub), and one was powered by a 50mA limited 3.0V lab supply.

The 2xAA-powered design reliably breaks my USB hub about 20% of the time now (so far it comes back after a reboot), but always seems to work fine if I plug in the JTAG header first, then the USB. When I’m working with ~50 boards at a time, this is fairly annoying.

Are there schematics posted for this programmer? It seems like there’s something wrong with the level shifting circuitry.

Here’s what my JTAG interface looks like. The circuit is the same for all our recent designs, but we’ve used the MSP430F1611, 1612, and 135 processors (all three have had the same problem).

http://drew.drawingimpressions.com/images/JTAG_MSP.png

try shorter USB cable

Tsvetan

OLIMEX:
try shorter USB cable

Tsvetan

I’m using the standard 6’ USB A to B cable that Spark Fun lists at the bottom of the page for the programmer. Is there a reason why you think this cable would cause such a spectacular failure as blowing out a USB hub?

our experience with FTDI chips show that not all PC USB hubs are same (different vendors different implementations) and sometimes shorter cable helps

Tsvetan