Boot Loader Flash

[Vic Wintriss]

Just got my new VicsWagon boards back with PIC soldered on the robot board with all ioio circuitry on the board, too.  Where are the best instructions for flashing a boot loader and app firmware?

[Ytai Ben-Tsvi]

There's not much to it. Get a PIC programmer, or use a second IOIO as such. Flash the bootloader image that you can download from the downloads page and finally use IOIODude as usual to flash the application firmware.

Here's an example of how to do the second IOIO way:

https://www.youtube.com/watch?v=eUKAPP63QtU

On Mon, Mar 30, 2015 at 6:40 PM, Vic Wintriss <g...@san.rr.com> wrote:

Just got my new VicsWagon boards back with PIC soldered on the robot board with all ioio circuitry on the board, too.  Where are the best instructions for flashing a boot loader and app firmware?

--
You received this message because you are subscribed to the Google Groups "ioio-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ioio-users+...@googlegroups.com.
To post to this group, send email to ioio-...@googlegroups.com.
Visit this group at http://groups.google.com/group/ioio-users.
For more options, visit https://groups.google.com/d/optout.

[Vic Wintriss]

Thanks…think I got it running with the Pickit3…I’ll let you know.

You received this message because you are subscribed to a topic in the Google Groups "ioio-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ioio-users/vOzO1t92Ryg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ioio-users+...@googlegroups.com.

[Vic Wintriss]

I got the boot loader flashed and am able to run the blink app...but having trouble with the Android charging circuit.  

Is there a detailed description of the Android charging circuit including the operation of Q1, Q2 and the VBUS and HOST pins of the PIC?  My new circuit implementation does not seem to be charging the Android properly.

On Monday, March 30, 2015 at 9:26:30 PM UTC-7, Vic Wintriss wrote:

Thanks…think I got it running with the Pickit3…I’ll let you know.

On Mar 30, 2015, at 9:18 PM, Ytai Ben-Tsvi <yta...@gmail.com> wrote:

There's not much to it. Get a PIC programmer, or use a second IOIO as such. Flash the bootloader image that you can download from the downloads page and finally use IOIODude as usual to flash the application firmware.

Here's an example of how to do the second IOIO way:

https://www.youtube.com/watch?v=eUKAPP63QtU

On Mon, Mar 30, 2015 at 6:40 PM, Vic Wintriss <g...@san.rr.com> wrote:

Just got my new VicsWagon boards back with PIC soldered on the robot board with all ioio circuitry on the board, too.  Where are the best instructions for flashing a boot loader and app firmware?

--
You received this message because you are subscribed to the Google Groups "ioio-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ioio-users+unsubscribe@googlegroups.com.

To post to this group, send email to ioio-...@googlegroups.com.
Visit this group at http://groups.google.com/group/ioio-users.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "ioio-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ioio-users/vOzO1t92Ryg/unsubscribe.

To unsubscribe from this group and all its topics, send an email to ioio-users+unsubscribe@googlegroups.com.

[Ytai Ben-Tsvi]

The IOIO decides whether it should be host or device based on whether the USBID pin is pulled low (host) or left floating (device). On the standard IOIO board this pin would get pulled low by either plugging in a USB micro A connector to the A/B socket or by forcing it low using the on-board switch.

When the IOIO application firmware detects that USBID is pulled low, it will assert the HOST signal, which will cause 5V to be supplied on the VBUS rail.

To unsubscribe from this group and stop receiving emails from it, send an email to ioio-users+...@googlegroups.com.

[Vic Wintriss]

When you say assert, do you mean assert high or assert low?

On Tuesday, March 31, 2015 at 2:36:24 PM UTC-7, Ytai wrote:

The IOIO decides whether it should be host or device based on whether the USBID pin is pulled low (host) or left floating (device). On the standard IOIO board this pin would get pulled low by either plugging in a USB micro A connector to the A/B socket or by forcing it low using the on-board switch.

When the IOIO application firmware detects that USBID is pulled low, it will assert the HOST signal, which will cause 5V to be supplied on the VBUS rail.

On Tue, Mar 31, 2015 at 1:06 PM, Vic Wintriss <g...@san.rr.com> wrote:

I got the boot loader flashed and am able to run the blink app...but having trouble with the Android charging circuit.  

Is there a detailed description of the Android charging circuit including the operation of Q1, Q2 and the VBUS and HOST pins of the PIC?  My new circuit implementation does not seem to be charging the Android properly.

On Monday, March 30, 2015 at 9:26:30 PM UTC-7, Vic Wintriss wrote:

Thanks…think I got it running with the Pickit3…I’ll let you know.

On Mar 30, 2015, at 9:18 PM, Ytai Ben-Tsvi <yta...@gmail.com> wrote:

There's not much to it. Get a PIC programmer, or use a second IOIO as such. Flash the bootloader image that you can download from the downloads page and finally use IOIODude as usual to flash the application firmware.

Here's an example of how to do the second IOIO way:

https://www.youtube.com/watch?v=eUKAPP63QtU

On Mon, Mar 30, 2015 at 6:40 PM, Vic Wintriss <g...@san.rr.com> wrote:

Just got my new VicsWagon boards back with PIC soldered on the robot board with all ioio circuitry on the board, too.  Where are the best instructions for flashing a boot loader and app firmware?

--
You received this message because you are subscribed to the Google Groups "ioio-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ioio-users+...@googlegroups.com.
To post to this group, send email to ioio-...@googlegroups.com.
Visit this group at http://groups.google.com/group/ioio-users.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "ioio-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ioio-users/vOzO1t92Ryg/unsubscribe.

To unsubscribe from this group and all its topics, send an email to ioio-users+...@googlegroups.com.

To post to this group, send email to ioio-...@googlegroups.com.
Visit this group at http://groups.google.com/group/ioio-users.
For more options, visit https://groups.google.com/d/optout.

[Ytai Ben-Tsvi]

The HOST signal is active high.

[Vic Wintriss]

When I power on, the LED blinks 3 times.  Is that normal operation?  I think that I have IOIO0500 installed.

[Ytai Ben-Tsvi]

That means that you're in bootloader mode. This will happen when you power on with the boot pin held low.

[Vic Wintriss]

Actually it blinks 5 times...only when the usb cable is connected to a phone.  Any idea what that means?  The boot pin is not being held low.

[Ytai Ben-Tsvi]

What bootloader did you put on it? Was it able to correctly talk to IOIODude? What app did you install? 5 blinks doesn't mean anything... Possibly corrupt firmware.

[Vic Wintriss]

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121321 versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Platform version: IOIO0030

How do I get out of the bootloader mode without changing the port so I can check to see what it says is loaded?

[Ytai Ben-Tsvi]

./ioiodude --port=/dev/tty.usbmodem1a121321 --reset versions

./ioiodude --port=/dev/tty.usbmodem1a121321 versions

OR:

./ioiodude --port=/dev/tty.usbmodem1a121321 fingerprint

[Vic Wintriss]

I loaded the IOIO0500 app

[Ytai Ben-Tsvi]

What do you get from ioiodude versions after rebooting into non bootloader mode?

[Vic Wintriss]

VicMacMini-2:Downloads VicMini$ ls /dev/tty.usb*

/dev/tty.usbmodem1a121311

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121311 versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Platform version: IOIO0030

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121311 fingerprint

d895d272a4b98529693fcdc4f69eeb3c

VicMacMini-2:Downloads VicMini$ 

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121311 --reset versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Platform version: IOIO0030

VicMacMini-2:Downloads VicMini$ 

[Ytai Ben-Tsvi]

This is only the bootloader. I would like to see what happens after a reset. That is, to get the "IOIO Application Detected" message.

[Vic Wintriss]

VicMacMini-2:Downloads VicMini$ ls /dev/tty.usb*

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121361 versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121361 --reset write /Users/VicMini/Downloads/App-IOIO0500.ioioapp 

Comparing fingerprints...

Fingerprint match - skipping write.

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121361 versions

It hangs up at this point and I have to exit.

^CVicMacMini-2:Downloads VicMini$ 

[Vic Wintriss]

It always says "Bootloader detected"

VicMacMini-2:Downloads VicMinils /dev/tty.usb*

/dev/tty.usbmodem1a121371

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121371 versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Platform version: IOIO0030

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121371 --force write /Users/VicMini/Downloads/App-IOIO0500.ioioapp 

Writing image...

[########################################]

Writing fingerprint...

Done.

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121371 versions

IOIO Bootloader detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Platform version: IOIO0030

VicMacMini-2:Downloads VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121371 --reset versions

IOIO Bootloader detected.

[Ytai Ben-Tsvi]

In what ways is your hardware different than the standard IOIO-OTG? Is it possible that for some reason it thinks it is in host mode? Can you probe the HOST signal?

[Vic Wintriss]

It is supposed to be an exact OTG copy.  Something is keeping it in the boot mode even after reset.  Any ideas?  Can hardware keep it in boot mode?  BOOT is not being held low.  I will probe HOST.

[Vic Wintriss]

I tried another board and got it working OK.  At least now, it recognizes the App is present.

The next problem is to make the Android detect the board.  How much charging current does the Android need in order to detect that it is connected to something?

[Ytai Ben-Tsvi]

How much the voltage drops as result of different charge current limiting and how low the voltage can drop before the Android decides it is not connected is totally device dependent. This is in general a total hack (which does work great in certain cases). The formal USB specification doesn't actually allow doing that.

[Vic Wintriss]

I don't understand.  Which voltage are you talking?  The ioio is not measuring its 5 volt supply voltage. I would not expect the ioio 5 volt supply to drop a measurable amount in any case, regardless of load. The variable resistor is setting a charge current to the Android, therefore I would expect the Android supply voltage to raise...not drop.  Is the Android monitoring it's own supply voltage?

[Ytai Ben-Tsvi]

The way USB works is that the device detects the presence of a host by sensing the VBUS rail. If it is about 5V, a host is present and the USB specifies what is the minimum the host must provide in order to be detected correctly. Under this specified voltage, the device may or may not detect the presence of a host.

Now, what we're doing on the IOIO is taking a 5V signal and passing it through (effectively) a resistor to the VBUS rail. This means that the voltage on VBUS will drop linearly in the amount of current drawn (5V - I*R). So the more you try to limit the current by increasing the resistance, you'd also be lowering the VBUS voltage until eventually causing the Android to stop detecting the presence of a host.

[Vic Wintriss]

I got the Android to detect the presence of the ioio board.  By measuring the Vbus rail current, it looks like the Android that I am using (an older HTC) takes about 450 ma.  When you say device, what are you referring to?  When you say host, what are you referring to...assuming that I have the ioio board connected to an Android.

[Ytai Ben-Tsvi]

Device and host are standard USB terminology. Loosely speaking, the host is the one providing power and the one requiring to have intimate knowledge of the device's interface.

In the case of the IOIO/Android, either way works (i.e. the IOIO can act either as a host or device). In your case, you're probably more interested in the IOIO being host scenario, which is the case in which that IOIO charges the Android.

[Vic Wintriss]

I have a new problem.  Suddenly, I cannot load a test program from Android Studio into the processor.  I can load the Blink program with ioioDude and it works OK.  I can load the Android Studio test program into other IOIO boards and it works OK.  The test program just blinks the led.  Any ideas?

[Vic Wintriss]

When I run versions with IOIOdude when not in the bootloader mode, I get:

VicMacMini-2:IOIODude VicMini$ ./ioiodude --port=/dev/tty.usbmodem1a121351 versions

IOIO Application detected.

Hardware version: SPRK0020

Bootloader version: IOIO0401

Application version: IOIO0500

VicMacMini-2:IOIODude VicMini$ 

[Ytai Ben-Tsvi]

What do you mean when you say "cannot load a test program from Android Studio into the processor"?

Which processor?

[Vic Wintriss]

The PIC.  I have laid the ioio circuit down on the new VicsWagon board.  Everything was working fine.  I even was getting ultrasonic readings from the sensors on the VicsWagon and then all of  a sudden I couldn't load from Android Studio any more.

[Ytai Ben-Tsvi]

You don't download code from Android Studio to the PIC.

You can either:

• Download code from MPLABX to the PIC (firmware)
  OR:
• Download code from Android Studio to Android (software)

I don't understand what you're saying really.

[Vic Wintriss]

Of course...I guess what I was trying to say is that my Android test program works fine with an ioio board on an earlier version of the VicsWagon.  All of a sudden the Android has stopped talking to the ioio circuitry on the VicsWagon.

[Ytai Ben-Tsvi]

So, if I understand correctly, you are holding two different IOIO variants in your hand and an Android with your custom app. One of the variants doesn't work as expected.

A few questions come to mind:

1. What's the (known) difference between them as far as hardware?
2. Have you verified that the firmware (both bootloader and app) are installed correctly? Hint: you can force-reinstall with ioiodude just in case.
3. What's not working with the "bad" variant? Does it charge the Android when connected? Does it pop the OpenAccessory dialog?
4. Does the "bad" board work with a PC?

[Vic Wintriss]

I found the problem.  The ioio board is not detecting the presence of the Android...again.  I'll work on it and see if I can figure out why it suddenly stopped seeing the Android.

[Vic Wintriss]

In trouble shooting the new VicsWagon board with the ioio circuitry on it, I find that even though USBID is solidly low (ground), HOST does not go high after startup.  Do you know any reason why that might be?

...

[Ytai Ben-Tsvi]

Is that with the application firmware running? The behavior of raising HOST in response to USBID pulled low is software-defined. It is not an inherent property of the hardware. For example, the bootloader doesn't do that.

You can verify that the application firmware is properly installed and running by running HelloIOIOSwing / HelloIOIOConsole on a PC connected to the IOIO.

[Vic Wintriss]

Are there instructions for running helloIOIOSwing?  I couldn't figure out how to run it.

[Ytai Ben-Tsvi]

At the bottom of this page:
https://github.com/ytai/ioio/wiki/Using-IOIO-With-a-PC

[Vic Wintriss]

I saw some warnings about using the new intelliJ AndroidStudio...which I am using...since Eclipse no longer supports Android.  Will there be any problems getting helloIOIOSwing running with AndroidStudio?

[Vic Wintriss]

What does the code look like that checks USBID and raises HOST?  I don't see that code in any of the apps that I have been using.  Is it in one of the libraries?

On Wednesday, April 15, 2015 at 5:11:58 PM UTC-7, Ytai wrote:

[Ytai Ben-Tsvi]

1. A precompiled version, HelloIOIOSwing.jar is included in the software bundle. You don't need to build it yourself for testing.
2. It is not true that Eclipse no longer supports Android: it supports Android as well as it used to, only in the future won't be developed further. I'm working on Android Studio porting, but it takes time.
3. The code you're talking about is in firmware/microchip.

[Vic Wintriss]

Still having problems with the Android detecting the VicsWagon version of the ioio board.  Since my ioio board will only be communicating with an Android phone, do I need the automatic sensing?  Would it work if I just went back to the way you do it on the V1 board with just a resistor?

[Ytai Ben-Tsvi]

This circuit is needed if you want to be able to work in device mode, e.g. for using the device bootloader with IOIODude. It is possible that it would work even if you forced 5V on the VBUS output, but it wouldn't be a great idea.

Also, I see no reason to assume from what you wrote that there is any particular problem with this (VBUS switching) part of the circuit, as it is behaving as expected considering HOST is low.

So the real question is why is HOST low. A couple of options I can thing of:

1. An electrical problem causing the IOIO to not detect the USBID is pulled low. This can be verified by probing the pin itself - maybe it is not connected to the circuit as result of a manufacturing defect? Maybe that PIC pin got electrically damaged as result of your testing?
2. Similar to (1), but for the HOST pin. Or maybe something is hard-pulling that net to GND, like a short across the pull-down resistor? Easy to check with a resistance meter while the board is un-powered.
3. Something got messed up with the firmware: either you're not running the stock firmware and something is wrong with your specific build or flashing doesn't work right. Also, are you sure you're using the right PIC? Running an app on a similar but not identical PIC might result in a the firmware kinda working but not perfectly.

[Vic Wintriss]

Good ideas...I'll check them all.  The PIC is PIC24FJ256GB206 -I/PT  Does the -I/PT mean anything important to us?

...

[Ytai Ben-Tsvi]

Is the right chip.

[Vic Wintriss]

Any idea why I can't flash the bootloader wth the Pickit3 on my VicsWagon ioio?  It erases OK and verifies empty.

...

[Ytai Ben-Tsvi]

It should work. Care to add some details about how you're doing that? Also, have you been able to flash a standard IOIO using the same procedure?

[Vic Wintriss]

At first I could flash the bootloader without any problems using the same procedure and even had the app working and getting ultrasonic readings.  Then all of a sudden it stopped working.  I now get.....

2015-04-21T17:15:02-0700- Completed loading IPE.

*****************************************************

Connecting to MPLAB PICkit 3...

Currently loaded firmware on PICkit 3

Firmware Suite Version.....01.36.10

Firmware type..............dsPIC33F/24F/24H

Target voltage detected

Target Device ID (0x0) does not match expected Device ID (0x41040000).

from the Pickit3.  I've double checked connections.  It is the right chip. Very discouraging!

...

[Ytai Ben-Tsvi]

This is usually an indication of either the PGC or PGD connection being broken.

[Vic Wintriss]

It was a broken wire.  Thanks so much for the hint.

How do I load the apps with the Pickit3?  It wants hex files.  Are they available someplace?

...

[Ytai Ben-Tsvi]

The recommended procedure is to flash a bootloader, then flash the app using ioiodude. If you insist on doing both in one step, the downloads page has combined bootloader+app images, which are also available as a zip of hex files. Or you can merge yourself using tools/merge-hex.

[vic.wi...@jointheleague.org]

Where can I find the blink hex file for the V1 board?

...

[Ytai Ben-Tsvi]

I didn't provide this one as a merged hex file. If you don't want to install it using the bootloader (as it was intended for), you'd need to build it yourself from source and merge it with the bootloader hex.

Visit this group at https://groups.google.com/group/ioio-users.

[vic.wi...@jointheleague.org]

I can't use the bootloader with ioiodude, because ioiodude is broken for the Mac.

...

[Ytai Ben-Tsvi]

No need to repeat this message on multiple threads. IOIODude works fine on Mac in most cases. There must be something different about your setup. Try different versions of IOIODude or different machines.

Visit this group at https://groups.google.com/group/ioio-users.