PicAxe reliability.
- Thread starter Screech
- Start date
BeanieBots
Moderator
"Can I expect a Pic-Axe microcontroller to work perfectly all the time."
Only if it doesn't matter.
That is, you must never depend on it as a matter of life and death. However, they are extremely reliable and what you are seeing should not happen. I would guess that you have missed off either the reset connection to 5v or the serin to 0v.
If both of those are OK, then suspect your bread board. (sorry Stan) Bread boards often fail to make a good connection if they have had wide legged components forced into them.
Only if it doesn't matter.
That is, you must never depend on it as a matter of life and death. However, they are extremely reliable and what you are seeing should not happen. I would guess that you have missed off either the reset connection to 5v or the serin to 0v.
If both of those are OK, then suspect your bread board. (sorry Stan) Bread boards often fail to make a good connection if they have had wide legged components forced into them.
Ahem - Picaxe<b> 20X </b> ? Little wonder it's unreliable, as you must have the only one in existence- Rev.Ed sell no such beast of course. Perhaps an 18X ? Reset R OK ?
I have to rise to the <i> unreliable breadboard </i> comment, since I estimate them as at least an order of magnitude more reliable that soldered prototypes. Following their 1970s introduction & my subsequent ~30 years of heavy educational use, I've only experienced failures of a dramatic kind - stupidly wired IC melt downs & resistor fires etc, or contacts pushed out the back/ leads broken. Just checked a handy DSE 300 hole board & see it was purchased by us 1991 as well- 15 years of faithful service! That was back in the 386 /DOS era, with the web explosion still 4 years off & Picaxes another decade away! Not many electronic products have that sort of life...
Suspect this present posting relates to well known "stray voltage wanders" ( mmm- perhaps picked up by those 2 extra pins!?),due to incorrect placement of 10k/22k programming Rs. Stan
Edited by - stan. swan on 15/03/2006 09:36:39
I have to rise to the <i> unreliable breadboard </i> comment, since I estimate them as at least an order of magnitude more reliable that soldered prototypes. Following their 1970s introduction & my subsequent ~30 years of heavy educational use, I've only experienced failures of a dramatic kind - stupidly wired IC melt downs & resistor fires etc, or contacts pushed out the back/ leads broken. Just checked a handy DSE 300 hole board & see it was purchased by us 1991 as well- 15 years of faithful service! That was back in the 386 /DOS era, with the web explosion still 4 years off & Picaxes another decade away! Not many electronic products have that sort of life...
Suspect this present posting relates to well known "stray voltage wanders" ( mmm- perhaps picked up by those 2 extra pins!?),due to incorrect placement of 10k/22k programming Rs. Stan
Edited by - stan. swan on 15/03/2006 09:36:39
Michael 2727
Senior Member
Get a standard Off The Shelf 1N4001, 1N4004
diode and plug it into every hole on a
breadboard and see how reliable it remains.
diode and plug it into every hole on a
breadboard and see how reliable it remains.
It's the 28X, not 20X, that I have been using.
The rest and resitors are ok.
I'm also powering the chip with 3 volts, not 4.5v so I dont nead resistors to power leds from the outputs.
It's also over clocked using a 16Mhz ossilator.
6 months ago I encounted problems when using a 20Mhz resonator from Rev.Ed while using 3 volts. The chip would not work.
When I change to 4.5 volts it would work again.
so, I'm still mystified.
I may blaim the breadboard.
I know that at 20Mhx , 3 volts is too low.
Thanks for the replys.
The rest and resitors are ok.
I'm also powering the chip with 3 volts, not 4.5v so I dont nead resistors to power leds from the outputs.
It's also over clocked using a 16Mhz ossilator.
6 months ago I encounted problems when using a 20Mhz resonator from Rev.Ed while using 3 volts. The chip would not work.
When I change to 4.5 volts it would work again.
so, I'm still mystified.
I may blaim the breadboard.
I know that at 20Mhx , 3 volts is too low.
Thanks for the replys.
A couple things...
First...you should not be running it at 16mhz and 3 volts. Those two dont like to play together. 5 volts would be more appropriate (4.5 volts). If you dont believe me, read the datasheet.
Second...what makes you say you dont need resistors when running LED's off of 3 volts? And..how many LED's are you running without any current limiting? Each leg may be able to source up to 20ma, but the chip, as a whole, should source no more than 100ma I believe. Again, check the data sheets.
--Andy P
Edited by - andypro on 16/03/2006 00:35:22
First...you should not be running it at 16mhz and 3 volts. Those two dont like to play together. 5 volts would be more appropriate (4.5 volts). If you dont believe me, read the datasheet.
Second...what makes you say you dont need resistors when running LED's off of 3 volts? And..how many LED's are you running without any current limiting? Each leg may be able to source up to 20ma, but the chip, as a whole, should source no more than 100ma I believe. Again, check the data sheets.
--Andy P
Edited by - andypro on 16/03/2006 00:35:22
^ "Second...what makes you say you dont need resistors when running LED's off of 3 volts?"
I read in the data sheet that at 3 volts, the outputs are current limited, and cannot be damaged.
I'm running 7 leds but at less then 50% duty.
I didn't see that 16 Mhz and 3 volts aren't compatable.
Thanks
I read in the data sheet that at 3 volts, the outputs are current limited, and cannot be damaged.
I'm running 7 leds but at less then 50% duty.
I didn't see that 16 Mhz and 3 volts aren't compatable.
Thanks
Thanks for all your help guys.
looks like my project will be fantastic after all.
The project is a scrolling digital clock.
it scrolls the hours, minutes and seconds across 6 7segment displays from left to right.
It's fantastic.
I did it to learn about multi-plexing (I think thats what it's called)
Thanks again.
looks like my project will be fantastic after all.
The project is a scrolling digital clock.
it scrolls the hours, minutes and seconds across 6 7segment displays from left to right.
It's fantastic.
I did it to learn about multi-plexing (I think thats what it's called)
Thanks again.
Mycroft2152
Senior Member
How did you interface the quartz clock to the PICAXE to get the cheap 1Hz timebase?
Myc
Myc
1Hz quartz timebase? I can answer that. I recently did this with a scrapped wall clock which runs on 1 AA cell. Throw out everything but the pcb. Make a note of the battery terminals and polarity as well as the pulser coil terminals. Since the pulser coil is wired across a push pull drive. Each side pulses high alternately at 0.5 Hz. So drive an NPN transistor with a resistor (10k or more)from each output to the base, connect emmiter to ground and a 4k7 pullup on the collector to +5v. The transistor will ground the Picaxe input with 1 Hz pulses. I powered the clock pcb from the voltage across my power on LED, (a red LED driven with 1k from 5V). This creates the low supply rail and the transistor does the level shifting.