Adding protection from ESD/EMI to basic 8051 programmer

[mik3ca]

The above is identical to the circuit in **broken link removed** except that I added a few diodes. I coloured each one differently because I feel as if at least one is mandatory from microcontroller burnout due to sudden reverse voltage spikes.

I read somewhere that if power is suddenly removed from a circuit then that circuit is supposed to produce a large reverse voltage and a "flyback" diode is supposed to prevent that from happening?

The power is from a regulated power supply.

So my question then is which diodes (based on the ones I added to the circuit) should I use at minimum to prevent ESD/EMI and are the 1N400x series acceptable here?

 

[Nigel Goodwin]

I read somewhere that if power is suddenly removed from a circuit then that circuit is supposed to produce a large reverse voltage and a "flyback" diode is supposed to prevent that from happening?

An inductor does that, a 'circuit' doesn't - there seems no need (or use) for such extra diodes.

 

[mik3ca]

Ok, but I'm just trying to figure it out because when I use this programmer, I use a power supply that I plug in when I start the programmer and when I'm done, I remove the power supply. The sad truth is I might be able to program the micro a few times before it no longer works even though the datasheet states it can be program 1000+

Doesn't reverse high voltage spikes happen when the power is gone?
and do microcontrollers function as inductors at some point during power disconnection?

 

[dknguyen]

Ok, but I'm just trying to figure it out because when I use this programmer, I use a power supply that I plug in when I start the programmer and when I'm done, I remove the power supply. The sad truth is I might be able to program the micro a few times before it no longer works even though the datasheet states it can be program 1000+

Doesn't reverse high voltage spikes happen when the power is gone?
and do microcontrollers function as inductors at some point during power disconnection?

All conductors are inductors at some level so there's always a voltage spike of some level when current is interrupted unless there is a low impedance path for the flyback current to flow. The slower and smaller the current that is interrupted, the smaller the spike. Nothing says this voltage spike has to be sufficiently high enough to damage your devices either and this is the case with your scenario.

In practice, it's not practical to worry about this unless the inductive spikes are of sufficient energy and frequency to cause damage. Things like motor drivers interrupt much larger currents through much larger inductances at tens of kHz so they have protective devices added.

CMOS chips are already delicate enough that they already have built-in ESD diodes to give marginal protection so they don't die every time you touch them and these do something similar for flyback at low energy levels. There are also parasitic capacitances and decoupling capacitors that provide such a flyback current path sufficient for low energy levels.

Not to mention that using a 14N00x diode won't do anything for inductive flyback or ESD since they are too slow to turn on in time to clamp either inductive spikes or ESD events. You need something like schottky diodes for flyback and TVS diodes for ESD.

The only diode I see that is even remotely useful is the red one in that it sort of gives a grave tiem if you connect 5V up backward. But why would you ever connect it up backwards if it's going through a polarized connector?

I don't know what you are trying to achieve with the others. They seem pointless and ineffective at transient, ESD, and reverse polarity protection regardless of the type of diode being used.

 

[Nigel Goodwin]

Doesn't reverse high voltage spikes happen when the power is gone?
and do microcontrollers function as inductors at some point during power disconnection?

No and no.

 

[mik3ca]

I don't know what you are trying to achieve with the others. They seem pointless and ineffective at transient, ESD, and reverse polarity protection regardless of the type of diode being used.

Just like the question. I'm trying to add protection to my circuit so that I get maximum use out of the microcontrollers. When one turns on a circuit, isn't there a "bouncing" effect that could produce oscillations? I mean with buttons, that is true because I know from microcontroller experience when a button is pressed, it initially returns alternating values then after it returns a pressed value.

I guess than if I use a diode, I should go with 1N914? That one is smaller and red.

 

[dknguyen]

Just like the question. I'm trying to add protection to my circuit so that I get maximum use out of the microcontrollers. When one turns on a circuit, isn't there a "bouncing" effect that could produce oscillations? I mean with buttons, that is true because I know from microcontroller experience when a button is pressed, it initially returns alternating values then after it returns a pressed value.

I guess than if I use a diode, I should go with 1N914? That one is smaller and red.

Diodes don't debounce. And even if there are oscillations when turning on such as from a mechanical switch, that's what the MCU's brownout and reset circuitry are for.