I keep blowing VR1
I have a BOE-Bot based robot that I have been having trouble with...
The bot is designed around a BS2, with a rev B BOE.
I have 4 drive servos - the 2 left and the 2 right servos are connected with a y-cable and plugged into P12 and P13.
I have 2 SSIR sensors (P14 and P15) as well as a Ping sensor and a servo for it (P0 and P1).
I would like to add a Memsic 2125 for tilt, a speaker and maybe a few other gadgets once I have my power situation resolved.
I wanted an extended battery life -
After some research I found a 5V 1A Switching voltage regulator from Dimension Engineering (http://www.dimensionengineering.com/DE-SW050.htm). I was using (2) 7.2v 2000mA Ni-Cd batteries in series (see attachment). The switching voltage regulator made a huge difference - almost doubled my battery life, but at a cost. I smoked VR1 on the BOE.
My first thought was too much current so I removed one battery, and replaced it the a shunt (so I did not have to re-wire everything). The bot ran this time for about 6 hours until it took out VR1 again.
Can anyone tell me why this keeps happening and what I can do to prevent future failures?
The bot is designed around a BS2, with a rev B BOE.
I have 4 drive servos - the 2 left and the 2 right servos are connected with a y-cable and plugged into P12 and P13.
I have 2 SSIR sensors (P14 and P15) as well as a Ping sensor and a servo for it (P0 and P1).
I would like to add a Memsic 2125 for tilt, a speaker and maybe a few other gadgets once I have my power situation resolved.
I wanted an extended battery life -
After some research I found a 5V 1A Switching voltage regulator from Dimension Engineering (http://www.dimensionengineering.com/DE-SW050.htm). I was using (2) 7.2v 2000mA Ni-Cd batteries in series (see attachment). The switching voltage regulator made a huge difference - almost doubled my battery life, but at a cost. I smoked VR1 on the BOE.
My first thought was too much current so I removed one battery, and replaced it the a shunt (so I did not have to re-wire everything). The bot ran this time for about 6 hours until it took out VR1 again.
Can anyone tell me why this keeps happening and what I can do to prevent future failures?
Comments
The switching regulator doesn't have enough current capacity for the servos. It's rated at only 1A and one servo at a near stall can draw 1A all by itself.
A voltage regulator is very well protected against current overload and thermal overload. It will shut down until the current drawn drops below its operating threshold and an overheated regulator will shut down until its temperature drops adequately. Some regulators will instantly be destroyed if they're connected backwards with ground positive and the Vin input negative. Some regulators, like the LM2940 used in the Rev C BOE are protected against this and can be connected backwards without becoming damaged. I think the Rev B BOE uses the same regulator.
On the Rev C BOE, the Stamp runs directly off Vin and, by using an external 5V regulator to supply Vin, the Stamp's own regulator won't have enough "headroom" to properly regulate its own 5V supply. You really need to supply at least 6V to Vin for the Stamp to operate properly. Dimension Engineering has an adjustable voltage 3A regulator that should work nicely in your application when its output is set to 6V.
None of this really answers your question. The setup you described should not "smoke" the voltage regulator on the BOE under any circumstances, particularly not after running successfully for a time. There must be something either that you haven't described or your actual wiring doesn't match what you think it should be.
I do have a rev C BOE sitting around that I was using on another project - I could switch over to that.
So to recap - I should replace the switching voltage regulator with the 3A variable model set to 6V, use the Rev C BOE, and be sure the servos are using Vin - correct?
With this configuration could I safely reintroduce my second battery back into the circuit?
"...None of this really answers your question. The setup you described should not "smoke" the voltage regulator on the BOE under any circumstances, particularly not after running successfully for a time. There must be something either that you haven't described or your actual wiring doesn't match what you think it should be. ..."
- I have checked and re-checked everything and aside from the labeling mistake on my diagram (thank you) the circuit is as described. Could the y-cables be part of the problem (i.e. 2 servos on P12 and 2 more on P13)? All I know is I replace VR1, it runs like a champ for several hours then I get a failure. This has happened 3 times now.
I am not opposed to a complete re-work of my battery system. I have searched the forms but was unable to find any relevant posts... Do you have any suggestions as to how I could safely incorporate larger batteries so I can extend my run-time?
Connecting two servos in parallel will double the current drain. Maybe, when the voltage regulator is not operated at its proper voltage, the current limiting circuitry doesn't work properly and the current overload is what's damaging it. I don't know, but I can't think of anything else.
Frankly, the best power system you could have would involve using a 6V battery pack for the servos only using either a 6V Gel-Cell battery or 5 NiMH size-D or larger cells in series (unregulated) and a separate 6V battery pack for the Stamp and other logic circuitry using the normal 5V regulators on the BOE and Stamp. Alternatively, you could use one 6V source for everything, but use a series inductor and large (1000uF or so) electrolytic capacitors on the BOE side of the inductor to isolate the logic circuitry from the noise and instantaneous demand of the servos. You'll need a common ground for everything and a single tiepoint for the grounds to avoid ground loops.