PICAxe 28X1 Firmware Issues

[harvey_twyman]

We have 1 older PICAxe 28X1 device with Firmware Version A3

This programs at 5V under Vista correctly.

The newer 28X1 devices I'm having problems with have Firmware Version 5

The devices initially reported their firmware using the
"Options/Mode/Firmware?" command correctly but wouldn't program
with the error message:

"Error - Memory Verification Failed (Byte 255)"

After repeated trials with supply voltages down to 4.5V
I managed to get them to program successfully.

However after several programming sessions at these lower voltages
the devices DIE! The firmware command then returns a message:

"Error - Hardware not found on COM"

All your datasheets mention using 5V suppply so why do
Technical Support always suggest using 4.5V to power the device?

We've used these chips powered by a regulated 5V supply for
several years now without any problems.

So what's changed? - the Firmware!

I've managed to KILL 5 devices so far!

Harvey Twyman
Department of Electronics
University of Kent

 

[hippy]

To my knowledge there has been no change to the internal 28X1 downloader between versions A.3 and A.5 so the firmware change should not be the cause of any download problems.

The reason we frequently recommend a 4V5 supply is for students who use these in class where 3xAA batteries is more convenient and can be less problematic than using other power supplies. The current PICAXE range will work entirely satisfactory with a properly regulated 5V supply. We also recommend trying to download using 4V5 when a laptop is used as this can alleviate some problems caused by laptop serial ports which do not produce a fully compliant RS232 serial output signal.

The PICAXE range is quite robust though not totally immune from failure caused by external effects, over voltage, shorted outputs and so forth.

It can be that a PICAXE fails to acknowledge a download and may appear to be 'dead' when it is not; download can be prevented by a program which has been downloaded which does not yield control to the download initiation. In those circumstances it is recommended to perform a "hard reset" - power off the PICAXE, initiate a download, when the download pop-up shows "Connecting to hardware..." power the PICAXE back up. It should also be possible to hold the PICAXE in reset and release the reset button at that point as well if one is fitted.

We would recommend trying the hard reset procedure and also try downloading from another PC and using a desktop if a laptop is currently being used to help identify where the problem may be.

 

[inglewoodpete]

I can understand the frustration when you only see the symptoms and not the cause of a download problem.

I am currently working on a project with over 40 28x1s connected in an i2c network. Before the new year these were all A.3 firmware and we had to return them to Rev-Ed for reflashing to a later version (A.5) due the i2c/PWM interaction problems.

Each PICAXE borad has its own 7805 regulator with associated close-mounted capacitors as per the datasheets for both devices. I have not had any problems with the chips failing to program with either A.3 or A.5 firmware. The PICAXEs are working in a relatively 'noisey' electrical environment with 2 12v motors associated with each PICAXE, althought the motors are not running when downloads occur. We are using 2 laptops and a desktop (All Windows XP) to do the downloads via (aftermarket) USB to serial leads with Prolific chipsets).

The only problems I have encountered have been occasional lockups of the USB interface to the cable. Generally, this is rectified by unplugging the USB plug at the computer, waiting 5 seconds and reconnecting. One laptop, a Toshiba, sometimes needs a reboot to reset the USB port, but this is rare (several weeks since this has been required).

(For those interested, the hardware is for a Semi Living Art project which leaves Australia today for its New York exhibition. I hope to post a url soon when it becomes available.)

 

[Brietech]

I live in NYC! Where is it going?

 

[inglewoodpete]

Rather than hijack this thread, I'll post details in a separate thread in a day or two.

 

[harvey_twyman]

VISTA Issues on Downloading to PICAXE 28X1

Reply to "Inglewoodpete"

"(All Windows XP) to do the downloads via (aftermarket) USB to serial leads with Prolific chipsets)."

I don't have a problem with Windows XP either, it all works fine with my Prolific Chipsets - it's VISTA that doesn't work!

I'm using the "Latest" 2005 Prolific Drivers which would have been created before Vista.

The PICAxe-08M manages to download successfully under VISTA but NOT the PICAxe-28X1

I've tried all the suggestions from Technical Support regarding the "Power Up Sequencing" and the use of a 4.5V supply etc.
None of these make any difference!

I manage a Lab of 36 PCs running VISTA so I have to get this problem sorted out soon as I have students waiting to use it.

Harvey Twyman
Department of Electronics
University of Kent

 

[harvey_twyman]

AXE027 Has a VISTA Driver?

My VISTA problem continued - The AXE027 mentions a VISTA Driver.

The AXEO27 uses the "FTDI Chipset"

The "PICAXE Serial Download Cable" mentions "XP Only" Prolific drivers. Could this be my problem?

PICAXE does NOT support VISTA on this cable?

 

[hippy]

The AXE027 "USB Driver for XP / 2000 / 2003 / Vista" is available for download if you follow the Software/Drivers link at the top right of the page. For the USB010 there is no specific Vista driver available but the XP driver should work and others have reported success ( Posts #4 and #5 ) -

http://www.picaxeforum.co.uk/showthread.php?t=8748&highlight=USB010+vista

If the PICAXE-08M is downloading under Vista it would appear that you have got a usable driver already installed and working and that should also work with the 28X1's - as it appears to have done before you started to encounter problems.

Do the 28X1 Version A.5 PICAXE's still download correctly using XP ?

Are you able to try with a physical 9-way serial download cable rather than via USB ?

 

[Technical]

To confirm there has been no change whatsover in the download/programming section of the 28X1 PICAXE boostrap in any version.

A 'Memory Verification Error' means that when trying to reprogram its own EEPROM/program memory, the data byte sent from the computer is
not the same as the data byte returned from the chip back to the computer. The chip is however communicationg, as it would not get this far to start with.

The ONLY main reason for this error message on an undamaged chip is a fault with the voltages - either:

1) power supply, too low, so the chip cannot operate correctly
2) power supply, too high, so the oscillator runs too fast and causes issues
with communication (often seen at 6V or 9V, which is above the maximum rated
supply of the chip).
3) RS232 voltages not correct, so the data byte returned is not detected
correctly by the computer. This can be caused by an incorrectly assembled
circuit (e.g wrong values of 10k/22k resistors) or faulty cable or faulty
USB adapter (or incorrect USB driver installed.) On rare occasions it can also be other bits of software (e.g. interactive whiteboards) polling the serial port in the background.

If a voltage is on a threshold point unexpected behaviour may result e.g. if the 10k is incorrectly placed on the wrong (PICAXE) side of the 22k, download will still work on some computers but not on others, due to subtle difference in RS232 voltage levels between different computers.

Due to different internal pin setups in the silicon, at an (incorrect) threshold point it is also possible for one type of chip (e.g. 8 pin) to work when another (e.g. 18 pin) does not.

However at 4.5V from 3xAA alkaline cells (or 5V) and a correctly functioning cable/usb adapter all chips will work reliably. It is useful to use the 'port test' function and measure the actual voltage across the PICAXE serial input pin/0V when the LED is clicked high and low.