Even I have burnt out a few, but usually it was due to abuse as a result unorthodox testing on my part. If I were to count, maybe 3 or 4, but it was something that was directly my fault... there was one that I accidentally reversed the polarity. It killed the external EEPROM right away, but the Propeller actually survived for a little while before dying of a slow PLL death.
I do still have an original decaped Propeller that was from an initial test run that still works. :-)
To date, I can remember burning out 3 (?) Boards - But the chips were probably still in fine condition - I have burned out / bent pins on a lot of BS2's though... -shudderes at rememberance-
I have been working on trying to drive motors and relays using the Prop supplied with 3.3v and driving the other loads with like 12V - I suspect that these three times have been by accidentally supplying the Prop with the 12V by hooking up transistors or relays wrong - although I am trying to keep things isolated I think I am not succeeding
I don't have any schematics because I am just trying to hook things up by looking at data sheets and such - like now I am playing with the ULM2003 (I think that's it PN) and the last one burnt out trying to drive a 12V relay
I suppose I could just stick to following schematics but I am trying to learn the hows and whys and not just following someone else's discovery - I suppose with that attitude I'm going to blow up a bunch of stuff but I'd rather learn I think - for me that's the whole point - thankfully the chips are just like $8
I think I really need to concentrate on keeping the Prop isolated too - but I also suspect that part of the problem has been my cheap Chinese breadboard power supply too (maybe hehe)
Thanks for all the responses - I really hope to be at the point that most of you guys are soon!
Edit: oh and all these failures have been on the breadboard set up - I DO have some working solid projects completed using the Prop Proto board
just out of curiosity ... you are using reverse biased diodes across your relays and motors ... right? ... if not, the back EMF might be what's killing the Propellers. This is not unique to Propellers. Any micro controller would need a similar protection when driving inductive loads.
Well this most recent burn out was me trying to drive a relay that was driving an electric motor - my *thought* was that the relay would isolate the motor circuit from the propeller? I had the relay clicking on just fine, but when I'd hook up the motor it would reset the prop - after about the 5th time or so it stopped responding at all ("No prop on com4 kinda message") I wish I could take a picture and show y'all but I've already cleared the breadboard and am waiting on my order from Parallax for my next three chips
I also threw away the Chinese breadboard power supply - I didn't start having any trouble until I put that in place so I'm going to take it out of the equation...
Like I said I am learning (or trying to LOL)
I don't think I had a diode on the relay either this last time
I can almost bet you money that the failure was due to the lack of a diode across the relay coil....your statement "...after about the 5th time or so it stopped responding at all" is typical of a back EMF failure.
The diode needs to be there to snub or attenuate the EMF. When you create a magnetic field through a relay coil or motor the polarity is in the same direction of the applied power. When you take away the power, the magnetic filed collapses, inducing a voltage in the opposite direction. The induced voltage is usually much higher than the applied voltage, because the load resistance is usually higher in the reverse direction. However the returned power can be as high as 90% and is significant enough to destroy a transistor if applied in the wrong direction. Remember Power = Current X Voltage ... if the load characteristics are much higher during the collapsing field, then the voltage will be higher because the lower load current. For this reason, the diode should be rated for at least that same power that you are driving the coil at to begin with.
The failure due to adding the motor, could have been causing a brownout condition to the Propeller due to a surge on the electrical system. There are ways to prevent and minimize this, but in order to do so we need more details as far as what your setup looks like in terms of a schematic.
Nearly all of my 'successful sudden kills' have been either reverse polarity or an over-voltage. Solder bridges have so far managed to either short out the regulators (destroyed a few of those too) or just be in places that drive me crazy. Before the Propeller ever existed, I managed to smoke a few brand new LCDs, maybe a BasicStamp2, and quite a few servo motors.
So by the time the Propeller arrived, I had learned to think first and act after thinking yet again. So far I haven't destroyed any. Propeller Protoboards do help as the power distribution is already done for you.
In the future I certainly will be sure to hook up at Vdd and Vss with proper coupling capacitors. But I really haven't done much with the 40pin DIP chips that I do have.
About the worst thing one can do is to work on a cluttered table top with a pile of conductive material under you circuit board. Watch out for tools as they are conductive too. Sparks suddenly fly and then everything goes haywire.
I just realized I have built several Tilden H-Bridges without any diode protect for the microcontroller. It is time to consider a redesign, though it would be hard to fit in the 4 diodes that an H-bridge requires.
I just realized I have built several Tilden H-Bridges without any diode protect for the microcontroller. It is time to consider a redesign, though it would be hard to fit in the 4 diodes that an H-bridge requires.
On very low power H-Bridges such as the Tilden designs, and others, the bipolar transistors are operated in analog mode or have slow switching times. When operated this way the transient currents are absorbed by the transistors and don't get to high in voltage.
However, if switched rapidly the 4 freewheeling diodes should be used.
If MOSFET H-Bridges are used the freewheeling diodes are built into the FETs.
Many, many moons ago ... I had just started work and I had to go to a house to fix their TV. The customer kept very quiet about the cause of the problem but it was found out a bit later. The mains lead was the two twisted single wires that was deemed safe back then and their cat had bitten through it, it lost a few teeth in this effort so it didn't happen again. Instant training.
I have had a couple of rabbits and a cockatiel that could slice through insulation, far cleaner than my best cutters.
We once had a Yellow Labrador puppy that was a bit out of control. Left in the house by itself for a while it would start to munch on just about anything, the carpet, the sofa, even pulling strips of wallpaper off the walls.
One day I came in to find she had chewed through the lead to the refrigerator. Curiously after that there was no more bad behaviour:)
A customer of mine in the Caribbean sent me these pictures.
His solar tracking PV system quit tracking.
He found this wiled parrot ripped the tracker out of its weather dome
and was tearing the parts off the board.
I see the parrot has an identification band, maybe a repeat offender that dislikes humans. That is quite a curious photo with him holding a circuit board in his beak.
Thanks for the info on the diodes - I had seen them in use and ignored them on the relay because it seemed to be working and it was the motor that was not I guess my next question is - why don't they come built in to relays and motors? hehe
I think I was also playing around with the ULM2003 as well maybe when I had the relay working good and then I might have switched to try and directly control a relay from the Prop when that last one burnt - I think the ULM2003 has diodes built into it to protect against back EMF?
in other news I am not getting notified of posts on the board - does everybody get notified of new posts?
why don't they come built in to relays and motors?
Some relays do have internal diodes. But you must correctly observe the polarity.
With motors you can't have internal diodes if you want to reverse it.
However, the simple answer is its better to allow the user the choice of diode type and circuit configuration.
BTW, there are snubber circuits that will turn the relay off faster than the simple diode circuit.
A couple. One new demo board (still have my working modified original from years back) and 1 DIP chip. But I do things to them that should not be done!
The ULN2003 does indeed have built-in back-EMF diodes. Just be sure to connect pin 8 to ground and pin 9 to your relay\motor supply voltage source.
The diodes don't come built in to relays, motors, etc. because not everything that drives them is vulnerable to to large voltage spikes generated by their coils. E.g. If you directly wire a 12 volt motor to your car battery it wouldn't hurt anything.
The diodes don't come built in to relays, motors, etc. because not everything that drives them is vulnerable to to large voltage spikes generated by their coils.
Many automotive and small relays have the snubber diodes built in. See an example: 9000 Spartan Series
Duane J
I thought they left out diodes of inductive stuff because it is cheaper for the manufacturers and allows motor or relay to be powered with any polarity. Correct me if I'm wrong.
Usually they don't have the diodes, as I said in the earlier post.
But when RickIn said they never have them I had to point out that some do.
Just to be complete.
Most of you guys have probably already seen this - and it might be too basic for some too - but i just spent like an hour playing (and learning) from this
in one of the circuit examples there's a "Blocking Inductive Kickback" where the circuit is simulating driving a relay and you open a switch and watch the circuit flow through the diode - pretty cool because in my head I was picturing the diode BLOCKING - and I guess it's allowing it to flow just through the source of the induction rather than the rest of the circuit
I will be studying these simulations at length to learn - what a great tool!
Hi pi;
Usually they don't have the diodes, as I said in the earlier post.
But when RickIn said they never have them I had to point out that some do.
Just to be complete.
Duane J
Thanks for the info Duane. I suppose it makes sense for some things to have EMF protection by default e.g. automotive electronics. As my designs are usually one-off or low production runs I tend to err on the side of being conservative to avoid angry clients/service calls when a 50 cent component helps to assure long-term trouble-free operation. I also us machine sockets exclusively despite the 10x cost premium.
Good point though if you are making large production runs and excess redundant circuitry adds significantly to cost over a production run.
Most of you guys have probably already seen this - and it might be too basic for some too - but i just spent like an hour playing (and learning) from this
The simulation is new to me. Thanks for sharing. Looks like the flow of current shown is electron flow which is backwards from conventional (hole) flow. Conventional flow is positive to negative in line with voltage. In a diode circuit, conventional current goes from positive to negative (anode to cathode) but is blocked the other way. Be careful with the interpretation and save your self lots of blown LEDs (assuming proper use of current limiting resistors). If you plug an LED in backwards, it may never light again.
Comments
I do still have an original decaped Propeller that was from an initial test run that still works. :-)
@GeeksGoneBad - what kind of setup were you using? .. can you describe what you were doing during the failures?
-John
I have been working on trying to drive motors and relays using the Prop supplied with 3.3v and driving the other loads with like 12V - I suspect that these three times have been by accidentally supplying the Prop with the 12V by hooking up transistors or relays wrong - although I am trying to keep things isolated I think I am not succeeding
I don't have any schematics because I am just trying to hook things up by looking at data sheets and such - like now I am playing with the ULM2003 (I think that's it PN) and the last one burnt out trying to drive a 12V relay
I suppose I could just stick to following schematics but I am trying to learn the hows and whys and not just following someone else's discovery - I suppose with that attitude I'm going to blow up a bunch of stuff
I think I really need to concentrate on keeping the Prop isolated too - but I also suspect that part of the problem has been my cheap Chinese breadboard power supply too (maybe hehe)
Thanks for all the responses - I really hope to be at the point that most of you guys are soon!
Edit: oh and all these failures have been on the breadboard set up - I DO have some working solid projects completed using the Prop Proto board
Jamie
Advice: OPTOISOLATORS!
just out of curiosity ... you are using reverse biased diodes across your relays and motors ... right? ... if not, the back EMF might be what's killing the Propellers. This is not unique to Propellers. Any micro controller would need a similar protection when driving inductive loads.
I also threw away the Chinese breadboard power supply - I didn't start having any trouble until I put that in place so I'm going to take it out of the equation...
Like I said I am learning (or trying to LOL)
I don't think I had a diode on the relay either this last time
I can almost bet you money that the failure was due to the lack of a diode across the relay coil....your statement "...after about the 5th time or so it stopped responding at all" is typical of a back EMF failure.
The diode needs to be there to snub or attenuate the EMF. When you create a magnetic field through a relay coil or motor the polarity is in the same direction of the applied power. When you take away the power, the magnetic filed collapses, inducing a voltage in the opposite direction. The induced voltage is usually much higher than the applied voltage, because the load resistance is usually higher in the reverse direction. However the returned power can be as high as 90% and is significant enough to destroy a transistor if applied in the wrong direction. Remember Power = Current X Voltage ... if the load characteristics are much higher during the collapsing field, then the voltage will be higher because the lower load current. For this reason, the diode should be rated for at least that same power that you are driving the coil at to begin with.
The failure due to adding the motor, could have been causing a brownout condition to the Propeller due to a surge on the electrical system. There are ways to prevent and minimize this, but in order to do so we need more details as far as what your setup looks like in terms of a schematic.
So by the time the Propeller arrived, I had learned to think first and act after thinking yet again. So far I haven't destroyed any. Propeller Protoboards do help as the power distribution is already done for you.
In the future I certainly will be sure to hook up at Vdd and Vss with proper coupling capacitors. But I really haven't done much with the 40pin DIP chips that I do have.
About the worst thing one can do is to work on a cluttered table top with a pile of conductive material under you circuit board. Watch out for tools as they are conductive too. Sparks suddenly fly and then everything goes haywire.
However, if switched rapidly the 4 freewheeling diodes should be used.
If MOSFET H-Bridges are used the freewheeling diodes are built into the FETs.
Duane J
solushun iz simple. give kat shrimpz an he wont chew wirez. no brainer.
-browz
I have had a couple of rabbits and a cockatiel that could slice through insulation, far cleaner than my best cutters.
We once had a Yellow Labrador puppy that was a bit out of control. Left in the house by itself for a while it would start to munch on just about anything, the carpet, the sofa, even pulling strips of wallpaper off the walls.
One day I came in to find she had chewed through the lead to the refrigerator. Curiously after that there was no more bad behaviour:)
A customer of mine in the Caribbean sent me these pictures.
His solar tracking PV system quit tracking.
He found this wiled parrot ripped the tracker out of its weather dome
and was tearing the parts off the board.
Unbelievable!!
Duane J
I think I was also playing around with the ULM2003 as well maybe when I had the relay working good and then I might have switched to try and directly control a relay from the Prop when that last one burnt - I think the ULM2003 has diodes built into it to protect against back EMF?
in other news I am not getting notified of posts on the board - does everybody get notified of new posts?
With motors you can't have internal diodes if you want to reverse it.
However, the simple answer is its better to allow the user the choice of diode type and circuit configuration.
BTW, there are snubber circuits that will turn the relay off faster than the simple diode circuit.
Duane J
The diodes don't come built in to relays, motors, etc. because not everything that drives them is vulnerable to to large voltage spikes generated by their coils. E.g. If you directly wire a 12 volt motor to your car battery it wouldn't hurt anything.
9000 Spartan Series
Duane J
Usually they don't have the diodes, as I said in the earlier post.
But when RickIn said they never have them I had to point out that some do.
Just to be complete.
Duane J
http://www.falstad.com/circuit/
in one of the circuit examples there's a "Blocking Inductive Kickback" where the circuit is simulating driving a relay and you open a switch and watch the circuit flow through the diode - pretty cool because in my head I was picturing the diode BLOCKING - and I guess it's allowing it to flow just through the source of the induction rather than the rest of the circuit
I will be studying these simulations at length to learn - what a great tool!
Thanks for the info Duane. I suppose it makes sense for some things to have EMF protection by default e.g. automotive electronics. As my designs are usually one-off or low production runs I tend to err on the side of being conservative to avoid angry clients/service calls when a 50 cent component helps to assure long-term trouble-free operation. I also us machine sockets exclusively despite the 10x cost premium.
Good point though if you are making large production runs and excess redundant circuitry adds significantly to cost over a production run.