I've been arguing with a colleague whether it is a sensible solution to clean dusty PCBs with compressed air. While I think this is the best solution (nothing touches the PCB), he claimed that spraying off the dust can cause ESD, because the charges that may have accumulated in the dust win't bleed off in the process (and it's better to leave everything as it is). I'm not really convinced by this argument and my (limited) knowledge of ESD production doesn't help getting peace of mind in this matter. What is your input?
IPC-A-610: Acceptability of Electronic Assemblies Section 3.1.2 mentions compressed air as a possible source of electrostatic discharge.
This forum post which claims to lay out the "Truths, myths, and flat out lies" about ESD notes that compressed air is a source of ESD due to the air rubbing against the air. But it claims that most of the charge is dissipated from the air particles before they hit the surface.
However, if you read the IPC spec, it talks about building up charge due to the air moving over insulating surfaces. So the charge will build up as it leaves the nozzle from air to air friction. Some (but not all) of that charge will dissipate as it travels to the board. But more charge will build up as the air moves across the laminate of the board itself.
The compressed air will defiantly have some charge as it leaves the can and it will build up more charge as is blows across the board. Whether or not the amount of charge is enough to damage your parts depends on a lot of factors. But it is very possible. When in doubt, test it.
You'll indeed risk building up high static voltages. Blowing air can do this easily. Think of thunderstorms when a cold front slides under the warmer air pushing it high upwards: the air-against-air movement can build up millions of volts.
In the case of the compressor your "lightning strikes" will be restricted to a few cm maximum, with energies of a few mJ, but that will be enough to destroy your CMOS parts.
I really don't think that clean, dry air can carry a charge or cause ESD. After all, ionized air is often used to reduce ESD risks.
However, if there are any particles or droplets at all in rapidly moving air, these can easily transport charge from one place or another, acting just like the belt in a Van de Graaff generator. It's the raindrops in a thunderstorm that create lightning, and I have gotten some surprising jolts from the plastic hose of my shop-vac when cleaning up large quantities of sawdust.
That said, I still think that "cleanliness is next to Godliness" with electronics, and the consequences of leaving dust in equipment are far worse than the risks of removing it. Dust inhibits the transfer of heat, and dust+humidity can develop into conductive surface contamination that can be very hard to remove later. I regularly clean dust out of gear using both a vacuum cleaner and/or compressed air.
While I'm not a technical guru like the likes of you all, I believe the key element of this part of the debate hinges on "conductive." When the air is dry, your body (or the ESDS devices) are the best conductor around, so the static suddenly discharges into you or the tech, as the best path of least resistance. When humidity is high, the air is relatively better as conductor (as Dave Tweed said), so the static generated/accumulated by your movements, clothing, carpeting, etc. dissipate and discharge more readily, in a diffuse fashion. (And yes, it is the water and ice droplets passing up and down, shearing off electrons, that negatively charges up thunderclouds, as Dave also mentioned.) Thus the reason electronic assembly areas keep humidity above approximately 40% to reduce worker's charge build up.
Ionization of course, can be the problem or a cure: neutralizing static through active air ionization becomes an option in cleanrooms and the like (ex. discharging injection mold plastics for use in devices), in addition to or as an alternative to insulating and grounding. "Active air ionization employs high-voltage ac or pulsed dc to produce ionized air to neutralize surface charges." A little plus works a treat with the negative, and discourages particles otherwise drawn to charged surfaces too. So, to my pal Dave T, the key there is that the ionized air used to reduce ESD levels is positively ionized, to balance the negatively charged (electron) static. (Much like, in lightning strikes, the flow of negatively charged energy downward is met by a jolt of positive flowing upward to the highest point of a tree or rod--the opposites attract. This is how lightning chooses its path.)
Ultimately, the concern is less about preventing the creation of static, but where the charge is going to go when it is. As embedded.kyle pointed out far more technically, compressed air will in fact create charge particles in the air and on the boards. Take preventative steps, consider risk vs. need, test and good luck.
Yes, fast moving air can cause ionisations and therefore present a risk of ESD damage to your pcb.
Second, why do you wish to clean the pcb? If the board is working, then leave it alone as much as possible- otherwise you risk introducing defects or damage in addition to the ESD issue.
If the board is not working and you wish to repair it...find the fault and do as little as possible to the board to fix it.