Dead Picaxe? Can't program 40X1

teraflop122

New Member
Well, I've been troubleshooting my current robot project for a little while now, and for the life of me I can't find anything wrong.

I'll start from the top, I guess.

Yesterday afternoon, I was trying to get a Compass module from Parallax working. I never did figure out the code, but that's aside the point. About midway through my experimentations, the software stopped connecting to my 40X1 as if it was off... but it was on, and the program was still running. I know this because the USB to Serial adapter's activity light was still going. I unplugged the USB from my computer, plugged it back in, and for the oddest reason Vista thought it was input from my USB mouse. I unplugged and plugged back in a second and third time. Finally, the editor software froze for a minute, and then proceeded to work correctly. I found that I could once again download programs to the chip.

About thirty minutes later I selected "clear hardware memory" and a few minutes later, when I tried loading a new program, the software once again couldn't connect. I had assumed "clear hardware memory" just meant it would erase the current program. Did I make a bad mistake, there? This time the issue seemed terminal: it still doesn't work today. Reseting the chip per instructions did nothing. I changed out batteries and took a few readings with a multimeter. I'm getting 5.6 volts from four new AA's, and then almost exactly 4 volts from an L7805 regulator. Aside from the 40X1, the logic supply from the L7805 is also running six IR rangefinders, a compass module, and the logic portion of a dual micro serial motor controller from Pololu.

I'll draw up a wiring diagram if need be.

Any help at all would be appreciated.
 

tiscando

Senior Member
the 'clear hardware memory' only just downloads an empty program on the picaxe, so it shouldn't be a problem.

I'm getting 5.6 volts from four new AA's, and then almost exactly 4 volts from an L7805 regulator.
If you get 4 volts from the regulator's output, then either more than 0.9A is drawn, or your multimeter might be reading lower voltages than actual.
If the regulator is really supplying 5V, then your new batteries may have been about 6.4V (New cells (batteries) tend to be very little over 1.5V), so when they were connected to the picaxe, the voltage may have been too high.

If the 4AA batteries are actually connected to the regulator input, and the 40x1 has never been connected directly to the batteries, then the above wouldn't be true (apart from about the multimeter), but the 'new' batteries you got may already have been used.

Things to try:
Disconnect the download cable from the picaxe, then power up the circuit.
Measure the voltage on the Rx (serial in) pin. Is the voltage above 1V?

If yes then check the download circuit is connected properly. If connected properly, then try touching a 1K resistor between the Rx pin and Gnd. If the multimeter still reads >1V, then you may have got the same problem as i had.

If no, then reconnect the download cable, and try and download a program. while the download progress bar is showing, read the voltage on the Rx pin. If it reads <3V, then it's the cable's or port's problem, otherwise connect the multimeter probe to the Tx (serial out), and does the voltage flashes briefly above 1V every 3 seconds?
 

eclectic

Moderator
@teraflop

Can you build a minimum download circuit?

Manual 1, page 35 and 31.

Then try to program the Picaxe.

e
 

teraflop122

New Member
I have some multimeter measurements here.

With a second, different model multimeter, I still get about 5.5 volts from four "new" AA's, and just Under 4 volts from the L7805. Disconnecting most of the IR rangefinders bumps it up to maybe 4.25 volts. I used an analog multimeter instead of the digital this time around.

I disconnected the download cable and powered the circuit. The voltage on Rx was zero.

I connected the download cable and the voltage on Rx went negative by about half a volt. When I try to download a program, the voltage jump to ALMOST 3 volts. Even with the IR rangefinders removed, it still only reached just Under 3 volts. I get 0 volts on Tx when trying to download. EDIT: Tx stays at 0 all the time, or at least the multimeter doesn't read anything.

@eclectic

After the multimeter tests above, I cut power to everything except the Picaxe and the download circuit, which, by the way, is one of these: http://www.sparkfun.com/commerce/product_info.php?products_id=8331 The problem persists.

I read through the "testing the system" section starting on 35 as you recommended. The first part wasn't useful, because I cannot transfer the test program. The only thing I found in the check list which applies was the logic voltage recommendation of 4.5 to 5.5. Even after disconnecting everything from the L7805 except for a small tantalum and a 10 uf 24v electrolytic capacitor, I still get less than 4.5 volts, about 4.3.

So...what, I guess my L7805 is bad? Do 7805's just spontaneously go bad? The 0 volts on Tx would imply the Picaxe isn't checking for downloads like it's supposed to, right?
 
Last edited:

eclectic

Moderator
Just a quick check.

Are you running the Picaxe from

A. 4 x AA cells, or

B. 4xAA cells, then through a L7805 regulator?

e
 

teraflop122

New Member
@ eclectic

The second one. Only the motors were powered directly by the batteries, when I had them connected.

@ SilentScreamer

I looked up L7805CV, and it has a dropout of 2 volts. I must be missing something though, because it had always worked fine on 4 AA's in the past.

I'll try hooking up another battery or two and see what happens.

EDIT:

I added two AA's, and the 7805 is putting out just about 5 volts now. Yet, I still can't transfer programs. I'm still getting slightly less than 3v on Rx when I try to connect, if that's relevant, and Tx is still zero with no signs of activity at any time.
 
Last edited:

tiscando

Senior Member
If the circuit is on breadboard, try unplugging the 10K pulldown resistor from the ground and plug it in to the supply voltage, so it pulls the Rx pin high. What is the voltage on the Rx pin? And is there any activity on the Tx line? Connect it back to gnd after.


Try leaving the download cable unplugged, and try and download a program. What is the voltage on the cable's Rx line? (connect the black multimeter probe onto the tip (gnd), and carefully touch the red probe onto the middle ring (Rx) of the downloading jack. What voltage do you get?)
 
Last edited:

hippy

Ex-Staff (retired)
There doesn't seem to be anything wrong with the voltages you are seeing when doing the serial tests and Clear Hardware Memory does nothing but download blank source code.

I would suggest trying Hard Reset. Hold the reset button or keep powered off until the "Connecting to hardware..." message is shown in the download pop-up window.

If that doesn't work, then try again with enough input voltage to get the L7805 delivering 5V.

Do you have another PICAXE chip you can try downloading into ? You could also try downloading from another PC in case Vista has got itself confused after seeing the connection as a USB mouse.
 

teraflop122

New Member
Ok, the hard reset with enough volts to get 5v from the regulator worked!

Thanks for your help, everyone!

It always ends up being something simple...
 
Top