Preventing ESD damage in DIY electronics

[DcibeL]

Hello,
Where I live it is usually very cold and dry...If I walk across the living room to change the volume or something on my stereo, it's very likely for me to get a shock from it due to ESD. It doesn't help that I have a corduroy couch!

All my equipment is DIY apart from the source which is either my DVD player or my PS3. The DIY equipment is a DAC, pre-amp, and power-amp. When I get a shock from touching my stereo there is usually a pop in my speakers, especially if I touch the DAC/pre-amp (these are assembled within the same chassis).

The other day I zapped my stereo and blew up the output op-amp on the pre-amp. This is the last device before the signal is sent to the power amp. I replaced the part and my system is alive again, but would not like to see this happen again.

The DIY equipment has been all wired the same, and I admit it's not really "proper". The electronics itself is isolated from the chassis entirely, and the chassis is connected to earth ground. At no point does my circuit ground connect to the earth ground on the chassis, but there are some points which are close that a large electrostatic spark could possibly jump across.

Does anyone have any recommendations of what could be done to better the prevention of ESD damage to my electronics? What can be done to improve my grounding arrangement to improve the situation without compromising safety (I would like to keep the chassis grounded for safety).

Thanks!

A quick internet seach showed this.

My thoughts at this point would be to wire a 1MOhm resistor to ground, so I can ground myself (slowly dissipate through big resistor) before I touch my stereo, but this is not the optimal solution.

A remote control is another solution, but that is asking for a lot of work...

I am planning to purchase some anti-static spray and spray the couch...but how long until that wears off? I would like something more permanent.

[leadbelly]

I've gone through this and your best bet is to leave your electronics alone and focus on the room environment. My ultimate solution was to put a good furnace humidifier in.

[DcibeL]

The place has a boiler system for heat, but you're probably right a good quality humidifier is a good option. My girlfriend has a cheap humidifier that doesn't work for crap. It just makes things wet, did not improve the humidity situation at all, at least not as far as ESD is concerned.

[wakibaki]

I used to wear a fabric jacket to work, every time I got out of the car I would get a belt, sometimes painful, and you could hear the crack of the spark.

Since I changed to a suede jacket the shocks have gone away.

Friction with modern artificial fibres often causes static build up.

You might be able to reduce the problem by changing your shoes, clothes, carpet or furniture. Natural materials are often less of a problem, although this is not always the case.

w

[Conrad Hoffman]

Humidity is key, but whatever you do, don't get an ultrasonic humidifier. They deposit white dust everywhere. Some are filtered, but sooner or later you'll give up on replacing filters. Get one with a paper cartridge evaporator, a rotating belt, or a furnace type.

I don't understand how you avoid hum if your chassis' aren't connected to circuit ground somehow, somewhere. Unless you have double wall construction. Most of my equipment uses three wire grounds and the circuit grounds are tied to chassis ground. I can shuffle across the carpet and zap any chassis, with only a very minor snap from the system. When other appliances switch on or off, I get no audible noise at all.

[len_scanlan]

Static charge can be a killer when it finds its way into integrated circuits.

At my work, the products we design are tested for immunity to ESD (electrostatic discharge). The typical model of the human body is a 150pF capacitor in series with a 2K ohm resistance.

The ESD 'gun' used in the test contains the capacitor and resistor. The capacitor is charged up to the test voltage and then applied to the electronics under test. We test at 8KV, 15KV, and for automotive products accessible from outside like a keyfob, 25KV (!).

In very dry conditions, these kind of voltages are easily generated. Especially if you have a wool carpet and are wearing sneakers with plastic soles. Remember the experiment where you rub a balloon across your hair and the hair stands on end.

You mentioned the inputs of your preamp are not connected to chassis ground. So any ESD strike (from your hand connecting a cable to a socket) has to go somewhere - and it finds the path of least impedance to ground, which seems to be right thru your opamp!

The best point to place any ESD protection component is directly between the input pin and the closest possible point on the chassis. Remember to keep the path as short as possible as any inductance from device leads leads to a 'surge' on the input pin as the static charge can't instantly drain away to chassis.

There are components called 'transient suppressor diodes' which consist of back-to-back zener diodes. They are specified with a maximum voltage to which they will clamp the input. For example a 30V rated device will start to conduct at 30V and clamp the input surge to something like 35-40 volts maximum. You want to choose a device with a high enough standoff voltage that it won't conduct under any circumstances, yet will clamp any static 'zaps' to a safe level. As a rule of thumb, most opamps will be OK with a 30-40V glitch on their input. It helps if there is some resistance between the connector and the input pin of the opamp, to limit the fault current.

Try something like this soldered between the input pin and chassis. If you can't solder to the chassis, crimp a lug on the end of the component lead and bolt to chassis.

http://littelfuse.com/data/en/Data_S...KE_revised.pdf

Try P6KE27CA available from Digikey (p/n P6KE33CALFCT-ND)

This part is rated at 27V, clamps at 37V max, and has a self-capacitance of about 1000pF. This last point is important as the self-capacitance of the device is connected between your input and GND. If large enough or if the source device has some appreciable output resistance, then it can act as a low-pass filter and roll off your treble somewhat. Assuming your CD player has an output impedance of say 200 ohms, this works out as a cutoff frequency of about 318KHz, well above the audio band, and nicely positioned to remove any RF noise from local transmitters.

Remember you need a Bidirectional suppressor diode, as a unidirectional part will clamp any negative going voltages on your input (audio!) to GND.

regards,

Len.

[DcibeL]

Originally posted by Conrad Hoffman
Humidity is key, but whatever you do, don't get an ultrasonic humidifier. They deposit white dust everywhere. Some are filtered, but sooner or later you'll give up on replacing filters. Get one with a paper cartridge evaporator, a rotating belt, or a furnace type.

I don't understand how you avoid hum if your chassis' aren't connected to circuit ground somehow, somewhere. Unless you have double wall construction. Most of my equipment uses three wire grounds and the circuit grounds are tied to chassis ground. I can shuffle across the carpet and zap any chassis, with only a very minor snap from the system. When other appliances switch on or off, I get no audible noise at all.

Thanks for the humidifier tips. No hum whatsoever in my system, the amp will hum if there's nothing connected to it though. I've been doing some research on grounding schemes today and I think that I will connect the circuit ground to earth ground and see how that affects things, not just for ESD but for the general operation of the electronics.

I'll use a small resistor, or ferrite bead between circuit ground and earth ground, from the star point on the circuit (right after rectification), to the same point that ground connects to the chassis.

Originally posted by len_scanlan
At my work, the products we design are tested for immunity to ESD (electrostatic discharge). The typical model of the human body is a 150pF capacitor in series with a 2K ohm resistance.

The ESD 'gun' used in the test contains the capacitor and resistor. The capacitor is charged up to the test voltage and then applied to the electronics under test. We test at 8KV, 15KV, and for automotive products accessible from outside like a keyfob, 25KV (!).

We do the same thing at my work, as our customers have expectations of the ESD susceptibility of our products.

Originally posted by len_scanlan
You mentioned the inputs of your preamp are not connected to chassis ground. So any ESD strike (from your hand connecting a cable to a socket) has to go somewhere - and it finds the path of least impedance to ground, which seems to be right thru your opamp!

I'm not sure why this part in particular was the one to die. I'm going to have a close look at the PCB and see what I can find.

Originally posted by len_scanlan
The best point to place any ESD protection component is directly between the input pin and the closest possible point on the chassis. Remember to keep the path as short as possible as any inductance from device leads leads to a 'surge' on the input pin as the static charge can't instantly drain away to chassis.

There are components called 'transient suppressor diodes' which consist of back-to-back zener diodes. They are specified with a maximum voltage to which they will clamp the input. For example a 30V rated device will start to conduct at 30V and clamp the input surge to something like 35-40 volts maximum. You want to choose a device with a high enough standoff voltage that it won't conduct under any circumstances, yet will clamp any static 'zaps' to a safe level. As a rule of thumb, most opamps will be OK with a 30-40V glitch on their input. It helps if there is some resistance between the connector and the input pin of the opamp, to limit the fault current.

Try something like this soldered between the input pin and chassis. If you can't solder to the chassis, crimp a lug on the end of the component lead and bolt to chassis.

http://littelfuse.com/data/en/Data_S...KE_revised.pdf

Try P6KE27CA available from Digikey (p/n P6KE33CALFCT-ND)

This part is rated at 27V, clamps at 37V max, and has a self-capacitance of about 1000pF. This last point is important as the self-capacitance of the device is connected between your input and GND. If large enough or if the source device has some appreciable output resistance, then it can act as a low-pass filter and roll off your treble somewhat. Assuming your CD player has an output impedance of say 200 ohms, this works out as a cutoff frequency of about 318KHz, well above the audio band, and nicely positioned to remove any RF noise from local transmitters.

Remember you need a Bidirectional suppressor diode, as a unidirectional part will clamp any negative going voltages on your input (audio!) to GND.

regards,

Len.

Thanks for the recommendation of voltage suppression diodes. I would be curious to know how an ESD spike would harm one of these diodes, ie, how much ESD can they take before they need replacement?

I'll be strongly considering a proper humidifier in my place, but unfortunately at the moment I don't have the funds to invest in anything for a while. Need car repairs, and have to deal with some disaster in my home that occurred over the weekend.

Thanks everyone for this great information and advice.