1

I have some prototyping boards like this:-

Prototyping board

It's a standard 0.1 inch pitch and I like them. Is this suitable for a medium voltage of about 200 V DC? I'm concerned that the voltage might just repetitively creep /jump between pads right across the board.

Paul Uszak
  • 7,327
  • 5
  • 37
  • 69
  • not really an answer but I've done projects (220V AC) on prototyping boards by using extra spacing, scraping unused pads between "HV" and conformal coating.. might have been overkill but it can be done – Wesley Lee May 03 '16 at 21:03
  • @WesleyLee Were your scrapings on the basis of some published science or feel good instinct? – Paul Uszak May 03 '16 at 21:24
  • I scraped the unused tracks to make the minimum (+ some margin) recommended spacing based on some Clearance Vs Voltage tables on the internet – Wesley Lee May 03 '16 at 21:35

2 Answers2

6

No, don't put 200 V between adjacent pads.

When nothing is going wrong, it should work. It varies with pressure, temperature, and humidity, but air can roughly hold off 1 kV per mm. Using that as a guide, 200 V would be 200 µm or 8 mils. Most likely you have 10-20 mils between pads.

However, since failure can be dangerous and cause all kinds of undesirable things, you want to be conservative. First, the 8 mil rough guide is the expect-to-fail point. Obviously you don't want to be there. It doesn't take much dirt or accidental solder ball or something to turn a 20 mil gap into a 8 mil gap or less.

Usually you want ¼ inch or 5 mm between line voltage and anything a human can touch. That's deliberately very conservative, but again, that's because the cost of failure is so high.

Don't use a breadboard like that for anything you're not willing to touch directly, which pretty much means stay below 48 V.

Olin Lathrop
  • 310,974
  • 36
  • 428
  • 915
  • What are adjacent pads? My thought was that if you jump one inter pad gap, then you can just jump the next gap too. Are two gaps any different to a single gap of twice the width? – Paul Uszak May 04 '16 at 13:04
  • @Paul: Multiple gaps do increase the overall insulation capability most of the time, but it's difficult to count on. You have to know what voltage the nodes in the middle are floating at. If there are two gaps and the middle node is the same voltage as one of the ends, then one gap can arc over, which makes it conductive, which then puts more stress on the next gap, which arcs, etc. – Olin Lathrop May 04 '16 at 13:45
0

TLDR: treat higher DC voltages with great respect.

DC power at high voltage is a more ferocious creature than low voltage DC or even high voltage AC. AC reverses polarity every 9ms, dropping to 0 volts and 0 amps, which creates an opportunity to snuff an arc. That is also true for directly rectified DC - but not at all, once it is smoothed out to continuous DC.

It's a good thing to think about in design.

8 mils may hold off an arc from starting under ideal conditions, but once that arc does strike, it'll take a lot more gap to extinguish it. It will indeed vaporize conductor material until it has widened enough gap. Here's a video of 600VDC doing that to a streetcar.

DC contactors have blowouts designed to snuff the arc (or more accurately, throw the plasma path into an arc chute which cuts it up). They use permanent magnets, residual magnetism from the collapse of the coil field, or a blast of air. This page shows a photo of a contactor rated to 300VDC, only 50% more than your design, and you see the not inconsiderable arc suppression gear.

Of course, semiconductors take a different approach to circuit interruption, but if your semiconductor is interrupting anything inductive, you may have a voltage spike. That could find an insulation weakness and strike an arc which a lower voltage could maintain.

To prevent these arcs from starting, you want insulation that might be appropriate to a much higher AC voltage.

You might also consider arc-fault detection on the DC power supply.

  • 1
    Fine anecdote, but how does it answer the question? – pipe May 04 '16 at 00:10
  • @olinlathrop The question is "is this suitable yes/no" and the answer is "no and here's why". Why does that seem like an anecdote? Honestly it sounds like you're quoting theory, and out of that have come up with this 8 mil stuff. I've worked with the stuff live, and it's a different breed of animal. It's not 20 volts x 10. I treat it with healthy fear and respect. That seems absent from your answer which troubles me, but I'm glad your answer was popular. – Harper - Reinstate Monica May 06 '16 at 03:16
  • You seem a bit confused. Olin never commented your answer, and I never answered. The word "no" is not even in your answer. You talk about how DC is more dangerous than something you never mention, presumably AC. The question was not about the difference between AC and DC. With this answer it sounds like 10 volt DC is equally dangerous. – pipe May 06 '16 at 04:51
  • 1
    Oh. That's actually a pretty good point. Sorry for getting my back up, clearly I am thinking all wrong on this one. i'll either clarify or delete since this is a dup anyway. – Harper - Reinstate Monica May 06 '16 at 04:52
  • Wouldn't the breaking capacity depend on the current capability of the source? If the 200VDC source is only capable of, say, 200mA (such as you can find inside a tube amplifier) or 10mA, its arc would be much weaker and be possible to extinguish even in a small 250VAC/10A rated relay. – Pentium100 May 06 '16 at 13:53
  • You'd think... Worth testing... wouldn't count on it. I had a situation where a 30A, 250VAC rated relay was used to interrupt a 1A generator field. It wasn't snubbed, and when the relay tried to interrupt, sometimes the inductive kick struck an arc, and 1A sustained it. This generator made about 50V, so field voltage had to be less than that. And the relay wasn't welded shut, it worked next cycle. Teardown showed 1mm+ contact gap was intact with mild arcing, which I attribute to the current being naturally limited to 1A. – Harper - Reinstate Monica May 06 '16 at 20:48