Robostix pin D.0 (I2C SCL) doesn't drive high (using i2c-test)

[VaibhavGhadiok]

hi everyone

I am referring to pin .I2C SCL (Port D.0) labelled RX .

I am using the i2c-test program to cycle between 3 states - making this pin as input, making this as output and driving it low, making this as output and driving it high.

The problem is when I drive it high it only goes to 2.5 V. On driving it low I get 0 V which is fine. On turning it to input I get 0.42 V even though it is pulled high.

On repeating the same procedure on I2C SDA (Port D.1) labelled TX, it works fine. When I turn this pin to input it gives me 5V, because it is pulled up. On driving it low I get 0 V. And on driving it high I get 5 V. I should be getting the same behaviour from the other pin as well.

I removed the pull-up to VCC from the SCL pin and even now if I drive it high I only get 2.5 V. On turning it to input or driving it low I get  0 V.

Any help would be greatly appreciated.

Vaibhav
--
http://guidance.aero.iisc.ernet.in/vaibhav

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[Dave Hylands]

Hi Vaibhav,

On Wed, Oct 28, 2009 at 5:38 PM, Vaibhav Ghadiok
<vaibhavghadiok@...> wrote:

> hi everyone
>
> I am referring to pin .I2C SCL (Port D.0) labelled RX .
>
> I am using the i2c-test program to cycle between 3 states - making this pin
> as input, making this as output and driving it low, making this as output
> and driving it high.
>
> The problem is when I drive it high it only goes to 2.5 V. On driving it low
> I get 0 V which is fine. On turning it to input I get 0.42 V even though it
> is pulled high.
>
> On repeating the same procedure on I2C SDA (Port D.1) labelled TX, it works
> fine. When I turn this pin to input it gives me 5V, because it is pulled up.
> On driving it low I get 0 V. And on driving it high I get 5 V. I should be
> getting the same behaviour from the other pin as well.
>
> I removed the pull-up to VCC from the SCL pin and even now if I drive it
> high I only get 2.5 V. On turning it to input or driving it low I get  0 V.
>
> Any help would be greatly appreciated.

The only thing to do now is to try disconnecting it from the gumstix
and seeing if it still behaves the same or not. That will at least
isolate if the problem is with the gumstix or robostix.

You'll probably need to write a simple program that flashes a LED and
at the same time toggles the D.0 pin. Once that's been programmed into
the robostix, you should be able to power it off, unplug the gumstix
and then power on the robostix and it should run the program.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

I was actually doing the same. I just wrote a simple program to drive the do pin high and it gives me 5V when the robostix is disconnected from the
gumstix. As soon as I attach the gumstix the voltage falls to 2.5 V.

The modifications I had made to the robostix are as follows.

1. I have shorted the RX, TX for the gumstix in order to be able to program the atmega through the gumstix.
2. I have pulled FFUART RxD (GPIO 34) high.
3. The power connection on he robostix is attached to VCC.
4. And the SCL and SDA pins are pulled high.
5. Also I have a capacitor across VCC-GND on the side where the GPIO pins are. This is just to filter noise.

I am really not sure what Gumstix is doing, that is making the pin drop  to 2.5 V.

Dave Hylands wrote:

Hi Vaibhav,

On Wed, Oct 28, 2009 at 5:38 PM, Vaibhav Ghadiok
<[e-mail]> wrote:
> hi everyone
>
> I am referring to pin .I2C SCL (Port D.0) labelled RX .
>
> I am using the i2c-test program to cycle between 3 states - making this pin
> as input, making this as output and driving it low, making this as output
> and driving it high.
>
> The problem is when I drive it high it only goes to 2.5 V. On driving it low
> I get 0 V which is fine. On turning it to input I get 0.42 V even though it
> is pulled high.
>
> On repeating the same procedure on I2C SDA (Port D.1) labelled TX, it works
> fine. When I turn this pin to input it gives me 5V, because it is pulled up.
> On driving it low I get 0 V. And on driving it high I get 5 V. I should be
> getting the same behaviour from the other pin as well.
>
> I removed the pull-up to VCC from the SCL pin and even now if I drive it
> high I only get 2.5 V. On turning it to input or driving it low I get  0 V.
>
> Any help would be greatly appreciated.

The only thing to do now is to try disconnecting it from the gumstix
and seeing if it still behaves the same or not. That will at least
isolate if the problem is with the gumstix or robostix.

You'll probably need to write a simple program that flashes a LED and
at the same time toggles the D.0 pin. Once that's been programmed into
the robostix, you should be able to power it off, unplug the gumstix
and then power on the robostix and it should run the program.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[Dave Hylands]

Hi Vaibhav,

On Wed, Oct 28, 2009 at 6:57 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:
>
> I was actually doing the same. I just wrote a simple program to drive the do
> pin high and it gives me 5V when the robostix is disconnected from the
> gumstix. As soon as I attach the gumstix the voltage falls to 2.5 V.

The SDA and SCL signals come from the gumstix, run through a 3.3v to
5v voltage converter and then go to the robostix.

On the gumstix side is a pullup from each of these to VCC (3.3v) and
on the robostix side is a pullup to VBATT (5v).

So you should check those pullups as well.

> The modifications I had made to the robostix are as follows.
>
> 1. I have shorted the RX, TX for the gumstix in order to be able to program
> the atmega through the gumstix.
> 2. I have pulled FFUART RxD (GPIO 34) high.
> 3. The power connection on he robostix is attached to VCC.

Hmm. Which VCC? The signal VCC is 3.3v. The signal VCC5 and VBATT are 5v.

> 4. And the SCL and SDA pins are pulled high.

Hopefully to 5v.

> 5. Also I have a capacitor across VCC-GND on the side where the GPIO pins
> are. This is just to filter noise.
>
> I am really not sure what Gumstix is doing, that is making the pin drop  to
> 2.5 V.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

Hi dave,

I am not sure where to check on the gumstix. I have a basix and I am not able to get the schematics.

I got this:
http://sketchup.google.com/3dwarehouse/details?mid=64bb986a5deac392e55eaf79ca7dc5e7&prevstart=0
But this is not helping.

The board is really tiny. I have no clue where to probe.

I couldn't figure out the gumstix pins for SDA SCL from the UARTs table given on:
http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

The pin I was talking about in my previous post is the one on the bottom right in the picture showing the bottom side of the robostix board. I have connected that pin to 5v.
http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

Dave Hylands wrote:

On the gumstix side is a pullup from each of these to VCC (3.3v) and
on the robostix side is a pullup to VBATT (5v).

Thanks.

[Dave Hylands]

Hi Vaibhav,

On Thu, Oct 29, 2009 at 12:57 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:

> I am not sure where to check on the gumstix. I have a basix and I am not
> able to get the schematics.
>
> I got this:
> http://sketchup.google.com/3dwarehouse/details?mid=64bb986a5deac392e55eaf79ca7dc5e7&prevstart=0
> But this is not helping.
>
> The board is really tiny. I have no clue where to probe.
>
> I couldn't figure out the gumstix pins for SDA SCL from the UARTs table
> given on:
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

Just to the left of the white connector is an 8-pin chip. This is the
3.3v to 5v converter for the i2c signals.
<http://www.nxp.com/#/pip/pip=[pip=PCA9515_9]|pp=[t=pip,i=PCA9515_9]>

Pins 2 & 3 are the 3.3v SCL and SDA signals from the gumstix. Pins 7 &
6 are the corresponding 5v SCL and SDA signals.

> The pin I was talking about in my previous post is the one on the bottom
> right in the picture showing the bottom side of the robostix board. I have
> connected that pin to 5v.
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

I think that's the BOARD connector. and that would be the V_ROBOBATT
signal, which is the input to the voltage regulators. So 5v on that
pin is fine.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

I see a chip as you have described. For me it says LTNX E3. It seems to be from LT.

This is to the left of the connector that is used to connect the gumstix to the Tweener/Robostix.

In order for me to power the gumstix I need to connect it to either the robostix or the tweener board. In both cases he above specified chip gets covered.

Soldering anything on it seemed a bit dangerous.

Am I talking about the right chip?

Dave Hylands wrote:

Hi Vaibhav,

On Thu, Oct 29, 2009 at 12:57 PM, VaibhavGhadiok
<[e-mail]> wrote:
> I am not sure where to check on the gumstix. I have a basix and I am not
> able to get the schematics.
>
> I got this:
> http://sketchup.google.com/3dwarehouse/details?mid=64bb986a5deac392e55eaf79ca7dc5e7&prevstart=0
> But this is not helping.
>
> The board is really tiny. I have no clue where to probe.
>
> I couldn't figure out the gumstix pins for SDA SCL from the UARTs table
> given on:
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

Just to the left of the white connector is an 8-pin chip. This is the
3.3v to 5v converter for the i2c signals.
<http://www.nxp.com/#/pip/pip=[pip=PCA9515_9]|pp=[t=pip,i=PCA9515_9]>

Pins 2 & 3 are the 3.3v SCL and SDA signals from the gumstix. Pins 7 &
6 are the corresponding 5v SCL and SDA signals.

> The pin I was talking about in my previous post is the one on the bottom
> right in the picture showing the bottom side of the robostix board. I have
> connected that pin to 5v.
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

I think that's the BOARD connector. and that would be the V_ROBOBATT
signal, which is the input to the voltage regulators. So 5v on that
pin is fine.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

I found another 8 pin chip on the other side and it seems to follow what you had stated. 3.3 on pins 2 and 3 and 5 on pins 6 and 7. I was probably looking at the wrong chip. This one is on a corner on the side of the connector that is used to connect to the Wi-Fi stix.


I see a chip as you have described. For me it says LTNX E3. It seems to be from LT.

This is to the left of the connector that is used to connect the gumstix to the Tweener/Robostix.

In order for me to power the gumstix I need to connect it to either the robostix or the tweener board. In both cases he above specified chip gets covered.

Soldering anything on it seemed a bit dangerous.

Am I talking about the right chip?

Dave Hylands wrote:

Hi Vaibhav,

On Thu, Oct 29, 2009 at 12:57 PM, VaibhavGhadiok
<[e-mail]> wrote:
> I am not sure where to check on the gumstix. I have a basix and I am not
> able to get the schematics.
>
> I got this:
> http://sketchup.google.com/3dwarehouse/details?mid=64bb986a5deac392e55eaf79ca7dc5e7&prevstart=0
> But this is not helping.
>
> The board is really tiny. I have no clue where to probe.
>
> I couldn't figure out the gumstix pins for SDA SCL from the UARTs table
> given on:
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

Just to the left of the white connector is an 8-pin chip. This is the
3.3v to 5v converter for the i2c signals.
<http://www.nxp.com/#/pip/pip=[pip=PCA9515_9]|pp=[t=pip,i=PCA9515_9]>

Pins 2 & 3 are the 3.3v SCL and SDA signals from the gumstix. Pins 7 &
6 are the corresponding 5v SCL and SDA signals.

> The pin I was talking about in my previous post is the one on the bottom
> right in the picture showing the bottom side of the robostix board. I have
> connected that pin to 5v.
> http://docwiki.gumstix.org/index.php/Robostix_I/O_pins#In_System_Programming_.28ISP.29

I think that's the BOARD connector. and that would be the V_ROBOBATT
signal, which is the input to the voltage regulators. So 5v on that
pin is fine.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[Dave Hylands]

Hi,

On Thu, Oct 29, 2009 at 4:33 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:
>
> I see a chip as you have described. For me it says LTNX E3. It seems to be
> from LT.
>
> This is to the left of the connector that is used to connect the gumstix to
> the Tweener/Robostix.

I meant on the robostix board. In this photo:
<http://docwiki.gumstix.com/index.php/Robostix_troubleshooting>
It's to the right of the VCC that's in red, and above the V_BATT that's in red.

You should be able to test the signals without the gumstix connected.

When SCL on the robostix side is at 5v, SCL on the gumstix side should
be at 3.3v.
Similarly for SDA.

When one side is at ground, the other side should be at ground.

If this looks good, then the problem is almost certainly on the gumstix itself.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

I checked the voltages and they seem to be fine.

EN is 5V. VCC for the chip is 3.3V. And the SCL, SDA at pin 2,3 are 3.3V and Pins 6.7 are 5V.

Dave Hylands wrote:

Hi,

On Thu, Oct 29, 2009 at 4:33 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> I see a chip as you have described. For me it says LTNX E3. It seems to be
> from LT.
>
> This is to the left of the connector that is used to connect the gumstix to
> the Tweener/Robostix.

I meant on the robostix board. In this photo:
<http://docwiki.gumstix.com/index.php/Robostix_troubleshooting>
It's to the right of the VCC that's in red, and above the V_BATT that's in red.

You should be able to test the signals without the gumstix connected.

When SCL on the robostix side is at 5v, SCL on the gumstix side should
be at 3.3v.
Similarly for SDA.

When one side is at ground, the other side should be at ground.

If this looks good, then the problem is almost certainly on the gumstix itself.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[Dave Hylands]

Hi Vaibhav,

On Thu, Oct 29, 2009 at 5:59 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:
>
> I checked the voltages and they seem to be fine.
>
> EN is 5V. VCC for the chip is 3.3V. And the SCL, SDA at pin 2,3 are 3.3V and
> Pins 6.7 are 5V.

Then I think that the problem has to be on the gumstix, unless you can
see a solder bridge or something around the pins on the white
connector.

You should be able to short the 3.3v side to ground and see that the
5v side goes to ground as well.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

I grounded SCL pin on the 3.3V side and I was getting 2.5V (on the 5v side). Should have been getting 0.
I grounded the SDA pin on the 3.3 V side and was getting 0.5 V on the corresponding pin on the other side. SCL is the pin that has been giving me problems.

Dave Hylands wrote:

Hi Vaibhav,

On Thu, Oct 29, 2009 at 5:59 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> I checked the voltages and they seem to be fine.
>
> EN is 5V. VCC for the chip is 3.3V. And the SCL, SDA at pin 2,3 are 3.3V and
> Pins 6.7 are 5V.

Then I think that the problem has to be on the gumstix, unless you can
see a solder bridge or something around the pins on the white
connector.

You should be able to short the 3.3v side to ground and see that the
5v side goes to ground as well.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

Just to add. I drove the pin D.0 to low. When I try to pull this using a resistor, it doesn't get pulled up. This is with no gumstix attached.

Also on the chip, I get 0 on the SCL on the 3.3 V side and consequently a 0 on the 5V side.
For the SDA, I see 3.3 V on the 3.3 V side and 5V on the other side. This is the expected behaviour for this pin.

Both SDA and SCL pins are being driven low by me and both are pulled up but only the SDA one is getting pulled up.


I grounded SCL pin on the 3.3V side and I was getting 2.5V (on the 5v side). Should have been getting 0.
I grounded the SDA pin on the 3.3 V side and was getting 0.5 V on the corresponding pin on the other side. SCL is the pin that has been giving me problems.

Dave Hylands wrote:

Hi Vaibhav,

On Thu, Oct 29, 2009 at 5:59 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> I checked the voltages and they seem to be fine.
>
> EN is 5V. VCC for the chip is 3.3V. And the SCL, SDA at pin 2,3 are 3.3V and
> Pins 6.7 are 5V.

Then I think that the problem has to be on the gumstix, unless you can
see a solder bridge or something around the pins on the white
connector.

You should be able to short the 3.3v side to ground and see that the
5v side goes to ground as well.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

A small correction to:
Also on the chip, I get 0 on the SCL on the 3.3 V side and consequently a 0 on the 5V side.

I get 0 on the 5V side and 0.5 V on the 3.3 V side.


Just to add. I drove the pin D.0 to low. When I try to pull this using a resistor, it doesn't get pulled up. This is with no gumstix attached.

Also on the chip, I get 0 on the SCL on the 3.3 V side and consequently a 0 on the 5V side.
For the SDA, I see 3.3 V on the 3.3 V side and 5V on the other side. This is the expected behaviour for this pin.

Both SDA and SCL pins are being driven low by me and both are pulled up but only the SDA one is getting pulled up.

VaibhavGhadiok wrote:

I grounded SCL pin on the 3.3V side and I was getting 2.5V (on the 5v side). Should have been getting 0.
I grounded the SDA pin on the 3.3 V side and was getting 0.5 V on the corresponding pin on the other side. SCL is the pin that has been giving me problems.

Dave Hylands wrote:

Hi Vaibhav,

On Thu, Oct 29, 2009 at 5:59 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> I checked the voltages and they seem to be fine.
>
> EN is 5V. VCC for the chip is 3.3V. And the SCL, SDA at pin 2,3 are 3.3V and
> Pins 6.7 are 5V.

Then I think that the problem has to be on the gumstix, unless you can
see a solder bridge or something around the pins on the white
connector.

You should be able to short the 3.3v side to ground and see that the
5v side goes to ground as well.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[Dave Hylands]

Hi Vaibhav,

On Thu, Oct 29, 2009 at 8:16 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:
>
> A small correction to:
> Also on the chip, I get 0 on the SCL on the 3.3 V side and consequently a 0
> on the 5V side.
>
> I get 0 on the 5V side and 0.5 V on the 3.3 V side.

So the pullup should have no effect if the line if being driven low or
driven high. If the 5v SCL is configured as an input, and you ground
the 3.3v side and measure 2.5v on the 5v side, then that suggests that
there is a problem with something on the 5v side, most likely a solder
bridge or something around the pullup resistors.

If the pin is configured as an input, which means that the line is
floating, then that's when the pullup kicks in.


--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

hi Dave

My think worked for a while and then stopped. I could cycle through the various states of the SCL pin on the robostix using the i2c-test program on the gumstix. and get the right results. I tried using i2c-load during this time but got the same error that it could not locate 0x0b. Now I am back to square one. Things have stopped working.

However, I noticed something interesting. When I run the i2c-test program on the gumstix, my SCL pin does get pulled high. Or for that matter even before starting the program. Essentially the robostix i2c-test program in running on the controller and the first thing it does is to set the robostix SCL, SDA ports as inputs. And then the robostix i2c-test program waits in a while loop, waiting for a character input from the gumstix i2c-test program. And in this state the SCL,SDA should and they do get pulled high to 5V.

The problem is when I run the program and start changing things. The robostix i2c-test program essentially changes the states of the robostix pins. It changes between driving high, low and changing to input. When I cycle through these states and again make the SCL pin as input, then it stops being pulled up. In fact even when I change the state to high, the SCL pin goes to 2.5V instead of 5V. Is it something the gumstix side of i2c-test program does that changes things?

And yes, if I remove the gumstix and then change the state of the SCL pin to either low, high or input, I get the right results ie 0, 5 and 5 (since it is being pulled up) resp.

Everything on the NXP chip is fine, when I ground the 3.3 V side I get a  0.5 V on the 5V side, which I think is fine. This is when I have configured the SCL pin on the Robostix as an input.

Thanks

Vaibhav

[VaibhavGhadiok]

On the other hand if I change the i2c-test program on the robostix to drive the pin high by default (initially), before it accepts any input., then I notice that the pin only gets driven to 2.5 V.

Maybe the gumstix pins are in some state that cause the voltage to drop. Since I don't do anything to the gumstix side of things. This is without running the gumstix i2c-test program.

VaibhavGhadiok wrote:

hi Dave

My think worked for a while and then stopped. I could cycle through the various states of the SCL pin on the robostix using the i2c-test program on the gumstix. and get the right results. I tried using i2c-load during this time but got the same error that it could not locate 0x0b. Now I am back to square one. Things have stopped working.

However, I noticed something interesting. When I run the i2c-test program on the gumstix, my SCL pin does get pulled high. Or for that matter even before starting the program. Essentially the robostix i2c-test program in running on the controller and the first thing it does is to set the robostix SCL, SDA ports as inputs. And then the robostix i2c-test program waits in a while loop, waiting for a character input from the gumstix i2c-test program. And in this state the SCL,SDA should and they do get pulled high to 5V.

The problem is when I run the program and start changing things. The robostix i2c-test program essentially changes the states of the robostix pins. It changes between driving high, low and changing to input. When I cycle through these states and again make the SCL pin as input, then it stops being pulled up. In fact even when I change the state to high, the SCL pin goes to 2.5V instead of 5V. Is it something the gumstix side of i2c-test program does that changes things?

And yes, if I remove the gumstix and then change the state of the SCL pin to either low, high or input, I get the right results ie 0, 5 and 5 (since it is being pulled up) resp.

Everything on the NXP chip is fine, when I ground the 3.3 V side I get a  0.5 V on the 5V side, which I think is fine. This is when I have configured the SCL pin on the Robostix as an input.

Thanks

Vaibhav

[Dave Hylands]

Hi Vaibhav,

On Fri, Oct 30, 2009 at 7:14 PM, VaibhavGhadiok
<vaibhavghadiok@...> wrote:
>
> On the other hand if I change the i2c-test program on the robostix to drive
> the pin high by default (initially), before it accepts any input., then I
> notice that the pin only gets driven to 2.5 V.
>
> Maybe the gumstix pins are in some state that cause the voltage to drop.
> Since I don't do anything to the gumstix side of things. This is without
> running the gumstix i2c-test program.

Just to clarify - you need to make sure that the i2c-pxa and i2c-dev
drivers are NOT loaded on the gumstix when using the i2c-test.
Otherwise they will see the activity on the SCL and SDA lines and may
get confused and may try to drive the lines.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
gumstix-users@...
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

They are not loaded.

Dave Hylands wrote:

Hi Vaibhav,

On Fri, Oct 30, 2009 at 7:14 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> On the other hand if I change the i2c-test program on the robostix to drive
> the pin high by default (initially), before it accepts any input., then I
> notice that the pin only gets driven to 2.5 V.
>
> Maybe the gumstix pins are in some state that cause the voltage to drop.
> Since I don't do anything to the gumstix side of things. This is without
> running the gumstix i2c-test program.

Just to clarify - you need to make sure that the i2c-pxa and i2c-dev
drivers are NOT loaded on the gumstix when using the i2c-test.
Otherwise they will see the activity on the SCL and SDA lines and may
get confused and may try to drive the lines.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

[VaibhavGhadiok]

Right now, it has again started to work magically but the i2c-load still gives me the same error.



They are not loaded.

Dave Hylands wrote:

Hi Vaibhav,

On Fri, Oct 30, 2009 at 7:14 PM, VaibhavGhadiok
<[e-mail]> wrote:
>
> On the other hand if I change the i2c-test program on the robostix to drive
> the pin high by default (initially), before it accepts any input., then I
> notice that the pin only gets driven to 2.5 V.
>
> Maybe the gumstix pins are in some state that cause the voltage to drop.
> Since I don't do anything to the gumstix side of things. This is without
> running the gumstix i2c-test program.

Just to clarify - you need to make sure that the i2c-pxa and i2c-dev
drivers are NOT loaded on the gumstix when using the i2c-test.
Otherwise they will see the activity on the SCL and SDA lines and may
get confused and may try to drive the lines.

--
Dave Hylands
Shuswap, BC, Canada
http://www.DaveHylands.com/

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
gumstix-users mailing list
[e-mail]
https://lists.sourceforge.net/lists/listinfo/gumstix-users