Please be gentle!

magnustullock

New Member
ok total newb question, I used to have a serial cable (intil my work mates chucked it!) so purchased a USB cable last week. I also purchased the breadboard adaper. Do I still need to supply a separate 5V to the chip for download? I have plugged the cable in to the PC and to the adapter and then put a multimeter on it. So far I see no voltage reading. Seems a but unnecessary to need a 5V additional connection when USB produces that anyway? Before I go and check all my solder points to see if I have a bad joint, I thought I would ask here.

cheers
 

hippy

Technical Support
Staff member
The PICAXE does need a separate 5V supply.

Using the PC's USB supply is generally not recommended as any issue with the PICAXE circuit wiring could potentially damage the PC.
 

Technoman

Senior Member

AllyCat

Senior Member
Hi,

+1 for the 2x or 3x AA batteries. ;)

However, USB ports should (always) be internally current-limited, so to simply power a PICaxe and a few indicator LEDs, then using the USB supply should be fairly "safe". Connecting via a cheap "sacrificial" USB Hub (e.g. from a Pound / Dollar store) might be a reasonable compromise.

BUT, PICaxes are often used with inductive loads (relays, motors, solenoids and even ignition coils) or with an external 12 volt power supply, etc.. In those cases a simple "mistake" could easily feed a higher voltage back INTO the PC / USB socket and damage or even destroy the PC (even via a hub). Better to be safe than sorry :(

Cheers, Alan.
 

hippy

Technical Support
Staff member
While PC USB ports should be safe to use, should have over-current and other protections, it can be the case that they don't. Sometimes the 5V on a USB port will come straight from the PSU which can deliver tens of amps which can burn out cables and tracking without the PSU even flinching. It's the reason we don't recommend using a PC PSU as a 5V power supply.

There are, as noted, potential faults which protection devices may not be designed to handle and those could destroy the protection devices themselves, render USB ports unusable, and cause cascade failures.

It's bad enough losing a PICAXE in a puff of smoke, far worse to lose a PC and possibly everything on its disk.

PICAXE users are welcome to take a 5V feed from a USB port or PC PSU when they accept full responsibility for the risks and consequences of doing that, but we cannot recommend it, cannot accept any responsibility for whatever adversity may befall anyone if doing that.
 

hippy

Technical Support
Staff member
there should be a device somewhere informing of the fault like that one
Again should be, but that's no guarantee that there will be.

Also, if one looks at the datasheet the maximum voltage on the output is allowed to be 5V3. A situation where more than that is placed on the PICAXE power rails could destroy that chip.

Most of these protection devices are designed to protect from over-current and shorted cables, short to 0V situations. They are not designed to cope with other faults which could more easily occur with a PICAXE or other electronic circuit.

As AllyCat notes, forgetting to use a back EMF diode on a relay, or that diode not making contact on a breadboard, could lead to more voltage on the power rail than expected, than is allowed.
 

mortifyu

New Member
Hello...

My 2 cents worth: The PC USB 5V could technically be used to power your PICAXE. As stated earlier in this thread, there are a number of factors that must be considered. To use a USB port for power, I'd strongly recommend it only under the supervision of a teacher or educated mentor.

Here, to understand what/why/how regarding back EMF from inductors:
https://en.wikipedia.org/wiki/Counter-electromotive_force

To limit current draw from the PC USB port, you could use a LM317 setup as a CONSTANT CURRENT SOURCE. Read through this datasheet to see an example on page 14:
http://www.ti.com/lit/ds/symlink/lm317.pdf

Using a diode to supply the power to the PICAXE will ensure current cannot flow back into the USB port under ANY condition.

As an absolute protection mechanism, a 5V1 zener diode could be placed across the USB power supplying rail.

Here is a diagram to demonstrate the above mentioned points:

USB Powered PICAXE.png


Hope this helps.

Regards,
Mort.
 

Pongo

Senior Member
At max current that will lose 1.2 volts across the resistor, something(?) across the LM317, and 0.7 volts across D1, taking the supply voltage to the 08M2 close to 3 volts. My preference would be to use a low drop out 3.3 volt regulator to buffer the USB port from the external circuitry, something like the LP2951 which has a current limit. (But I would only do this in very rare and tightly controlled circumstances, not as a general source of experiment power.)
 

mortifyu

New Member
Right you are Pongo, perhaps I should have also made note that D1 should be a SCHOTTKY type diode. This would reduce the voltage drop across D1 to around 0.1 volts. The suggestion of LM317/2R7 (462mA constant current) simply provides current limitation for the supplying USB port. As a general rule a USB port can be expected to safely provide at least 500mA. Although LP2951 is only rated to 100mA, it is indeed another possibility for this application. However, I too would only do any of this under absolutely necessary circumstances.

Regards,
Mort.
 

depeet

New Member
Since a few years, I use a mosfet with a very low on-resistance instead of D1. I took the datasheet of a random Schottky diode and the voltage drop was around 0.5Volts @ 500mA. The mosfets I use have an on-resistance of 0.3 ohm. At 500mA this results in a voltage drop of only 0.15 Volts. If you work with "high voltages" (12 or 24 volts) the drop voltage might be a minor issue. But with lower voltages like 5 volts, this could be a major problem.
 

Technoman

Senior Member
Look at this chip specially designed for that purpose :
  • LTC4362 (about 2$) Comes in a very small package (8-Lead DFN 2mm × 3mm) , could eventually fit within a plug.
 

depeet

New Member
Look at this chip specially designed for that purpose :
  • LTC4362 (about 2$) Comes in a very small package (8-Lead DFN 2mm × 3mm) , could eventually fit within a plug.
Funny, yesterday I looked at the LTC4361 which does about the same thing, but needs a few external components and monitors voltages up to 80V
 

AllyCat

Senior Member
Hi,

Current-limiting may indeed give some useful protection to the overall system, but note that the LTC4362 (and most others) is intended to protect the downstream electronics (e.g. the PICaxe) from overvoltage/current, whilst the issue raised by the OP and particularly in post #4 is the protection of the upstream electronics, i.e. the PC/Laptop/USB ports.

Unfortunately series semiconductor switches or transistors may not, (and sometimes cannot) isolate the reverse flow of current, back into the source and thus may not protect against damage. For example many FETs and ic transistors have an integral reverse diode connected between their "output" and "input" terminals (either intentionally or as a "parasitic") which might, or might not, be shown on their schematic symbol.

Bench power supplies, and other high reliability systems often include "shunt" protection in the form of a triggered SCR/Thyristor (sometimes called a "Crowbar") or a Zener diode to ground, to "dump" any excess voltage (from wherever it might originate). Generally, under severe overload conditions SCRs and Zeners are considered to "fail safe" (i.e. to a short circuit) whilst the bond wires in integrated circuits can blow open circuit, (like a fuse) and cease to protect the system.

Cheers, Alan..
 
Top