Sparkfun XS1-L1-64 Development Board

[mio_xmoslinkers]

I guess you will only be able to use one or the other :( I guess maybe you could cut the tracks to the XTAG2 UART if you don't ever want to use that?!

Yes, I think that the problem is solved if it does so perhaps.
but , If it is possible, I want to make UART pin of XTAG2 floating (=configure as input ) from USB
(via control software).
XTAG2 uses libusb? and and the material to operate this was unpublished.
I hope to open the method for the operation of XMOS chip on XTAG2 to the public :)

The direction of the UART pin might be not revokable because XTAG2 will not be able to be rewritten of OTP :?

[mio_xmoslinkers]

I can't get the debugger working properly with my boards, for some reason. If I set a breakpoint in the code it keeps running. Debugging the same program works OK with the XK-1 and XC-1. Has anyone else noticed this?

I also actually saw some strange operation with a sparkfun board, first.
At first, the JTAG debug mode and writing in the flash memory failed.
Though it is natural that both fail in the flash writing accompanying the code execution by xrun.

Strange operation of the board was lost starting with certain.
I do not understand a true reason for the improvement.

I sequentially did the following.

First of all, I turned on the power switch of "Sparkfun Board".
Next, I supplied power to the Board by USB.
Next, I inserted XTAG2 to "Sparkfun Board".
Next, I inserted XTAG2's USB to another USB port of same PC.
Next, I wrote the program (that looked normal) to the SPI flash, and normally completed.

I did not see this strange operation after this. It was similar in two boards.

[leon_heller_xmoslinkers]

It's not just me then! I've emailed SFE about the problems I'm getting.

Using your sequence, both boards are working OK! The same sequence also works with power supplied via the 5V DC connector.

Has anyone managed to get the on-board demo software to work? I found that it uses 115,200 baud, but pressing the space bar doesn't do anything. There seems to be a problem with the baud rate, as the echoed characters are occasionally corrupted.

Leon

[leon_heller_xmoslinkers]

I got this response from SFE:

We are very sorry to hear you are having problems with your recent
purchase.

I checked out the Xlinkers forum posts. It looks like a major hardware
bug was overlooked: the Tx of the JTAG and the USB are sharing a
connection. This could be creating some of the problems you are having.
All this should mean though is that you cannot use the USB and the JTAG
at the same time.

It almost sounds like the power on reset IC is on backwards on your
board. This is the first generation of these boards so it is possible
that our test procedure does not account for this particular error.
There could also be a decoupling problem that we overlooked. Try
powering the board with 9v instead of 5v and see if that fixes any of
the problems you are having.

I noticed in the forum posts that the code you are running does not
leave time in between writing to the ports to allow for blinking of the
leds. Try eliminating the redLed <: 0 statement and seeing if the red
LED simply turns on and stays on.

Make sure you are using the XC-5.xn file not the XK1.xn file. The XK1
uses the 128lqfp package where ad the XC-5 uses the 64lqfp package. Also
make sure you are using the latest release of the XMOS design tools
(v9.9), the XTAG2 and the SPI flash require the latest release of the
design tools.

If errors persist or the UART bug is a problem for you we would be happy
to have your board returned. If this is the case please email
returns@sparkfun.com. Please include a description of your problem or
this email in your message to returns.

Leon

[mio_xmoslinkers]

At any rate, SFE understood the problem, and was good.

I will wait for the improvement and do an additional order. ;)

# If SFE sells PCB with this problem cheaply though it is glad. :)

[JonathanMay_xmoslinkers]

These are back in stock and SparkFun has posted the information about the fix they have made to the board on the store page:

http://www.sparkfun.com/commerce/produc ... ts_id=9428

Hey - for that $100 free day you could get a SparkFun board and an XTAG-2!

[TonyD_xmoslinkers]

...
Hey - for that $100 free day you could get a SparkFun board and an XTAG-2!

They were one of the first things to go on my list :D

but I don't know if I can wait until the 7th Jan for SparkFun's free day

[leon_heller_xmoslinkers]

Buy one now to satisfy your craving and get another one in January?

Leon

[TonyD_xmoslinkers]

Buy one now to satisfy your craving and get another one in January?
Leon

Sparkfun have sold out again :(

[JonathanMay_xmoslinkers]

Buy one now to satisfy your craving and get another one in January?
Leon

Sparkfun have sold out again :(

60 brand new in stock...

Edit: 59

[leon_heller_xmoslinkers]

I've just ordered another two. They've cut some tracks to get round the problem that some people have encountered.

Leon

[phampson_xmoslinkers]

1)It failed in the UART receipt from FTDI when XTAG2 was connecting.

Yeah, I thought that might be a problem (as we discussed above when someone spotted it in the schematic). The FTDI & XTAG2 have their serial lines connected together. The issue there is that they are probably interfering with each other in some way. I guess you will only be able to use one or the other :( I guess maybe you could cut the tracks to the XTAG2 UART if you don't ever want to use that?!

:? - I'll be writing an invoice for my consultancy fee shortly :mrgreen:

[mio_xmoslinkers]

1)It failed in the UART receipt from FTDI when XTAG2 was connecting.

Yeah, I thought that might be a problem (as we discussed above when someone spotted it in the schematic). The FTDI & XTAG2 have their serial lines connected together. The issue there is that they are probably interfering with each other in some way. I guess you will only be able to use one or the other :( I guess maybe you could cut the tracks to the XTAG2 UART if you don't ever want to use that?!

:? - I'll be writing an invoice for my consultancy fee shortly :mrgreen:

I want to ask for the error report discount when being order next time :mrgreen:

[leon_heller_xmoslinkers]

I asked Nate at SFE about the mod details, so that I could fix my boards myself:

I'm sorry we haven't gotten to yourself and the 14 people that had
purchased the version with the bug in it. I will send out an email to
the rest of the Sparkfun customers tomorrow explaining how they can fix
the bug.

Here is a summary of the issue people have encountered and how it can be
remedied:
With the XTAG-2 plugged in, the XS1-Dev-Board becomes partially powered
by the programmer when the switch is not turned on. As a result the XS1
core supply does not rise monotonically when the switch is flipped. This
is what triggers the connection issue and power cycling has to be done
so that the IC can be programmed.

The non-monotonic rise in core supply is due to the XTAG-2 driving the
UART signal, RX-I, high. This signal is connected to the ESD diodes of
the FTDI chip. As a result, the IO supply of the chip on the board is
driven to about 2v and the core supply is powered as well.

The long of the short of it is, the RX-I and TX-O pins of the XSYS
header have to be cut. The connection between the FTDI and the XS1
should remain intact however the connections between the FTDI and the
XSYS header should be cut. These two traces are the traces on the
unpopulated side of the board. They correspond to pins 17 and 19 of the
XSYS header. They are the traces connected from two vias to two header
holes.

Again, we here at Sparkfun are very sorry for the inconvenience and
assure you this is not the norm around here.

Regards,
Nate Bernstein

He had previously offered to fix them if I returned them, but it isn't worth the hassle when a couple of seconds work with a scalpel will cure the problem. I've attached a pic showing where the tracks are cut. The surgery cured the power up problem, BTW.

Leon

[mio_xmoslinkers]

The letter came from Sparkfun also to me :?

> The issue is that the board has to be powered and programmed in a very
> specific order. The order is
> 1) supply power to the board
> 2) make sure the power switch is on
> 3) insert the XTAG-2
> 4) program the board

The method that had been done without understanding the reason by me was almost a correct answer. ;)

By the way, I cut the corner of the pin of the connector directly with nippers.

[leon_heller_xmoslinkers]

What is the circuit on your board?

I'm just about to interface a miniSD card to mine with a little breakout board I made. I've tested it with a Parallax Propeller, so I know it works.

Is your avatar a cat, also?

Leon

[mio_xmoslinkers]

Hi, Leon

>What is the circuit on your board?

This circuit is ADC on pin change circuit board (SSOP->DIP).
This ADC , AD9283BRS-80 which was sold with $5 in the "famous" Japanese parts shop "AKIZUKI-DENSHI".
I will make 20MHz data logger with this.

>I'm just about to interface a miniSD card to mine with a little breakout board I made. I've tested it with a Parallax Propeller, so I know it works.

I am very interested in "Parallax Propeller",but I have never used it.
Is "SPIN language" easy?

>Is your avatar a cat, also?

Yes, It is the black stray cat (like me, I thought) which was in the train station :D