Protection from incorrect power supply?

[RobA]

The Zener should behave like a normal diode for forward voltages below the breakdown voltage, so it should work for reducing the voltage of the leakage current from the 1N5817. I tried it with a +/- 15V supply on a 12V Zener (using a CLR in front of the Zener). When I put the +15V on the CLR, I get about +12V at the Zener. When I stick the -15V on there, I get around -0.7V. So, I think that part is a working OK.

When I do the proposed setup of PTC (about 10Ω resistance) into 1N5817 followed by a crowbar Zener or 1N4001, with the -15V I get -0.55V with the Zener and -0.41V with the 1N4001 at the power output.

Caveats:
This was all done with my now suspect DMM.
The power supply I used can only source 30mA, so this didn't test the PTC part at all, but I really wanted to see what the voltage would be before the fuse could blow.

[electrosonic]

I haven't considered using PTCs before. Looking at the data sheet on the Tayda site, I see a max trip time of 2.5 seconds. I think that is too long to be considered reasonable protection - a lot of components can be fried in less than that time.

Andrew.

[RobA]

I haven't considered using PTCs before. Looking at the data sheet on the Tayda site, I see a max trip time of 2.5 seconds. I think that is too long to be considered reasonable protection - a lot of components can be fried in less than that time.

Andrew.

There are others available with different characteristics. But even if you could get one that is really fast, I don't think I'd use it as the primary protection. In the proposed setting, it's really only protecting the diodes on the input. There have been bunches of reports of Zeners blowing from overheating. The PTC could be a good way to protect them without having to put a big CLR in front of them in the cases where you can't take that much voltage drop if the correct voltage is being used. Even with that, there are going to be lots of cases where you wouldn't want to use the PTC because the 10Ω or so resistance is too high. It's all going to depend on what the circuit is that you're trying to protect.

[electrosonic]

This thread reminded me of the Plimsoul schematic that is floating around.

Like this for example:
http://1.bp.blogspot.com/-nKJVWJcLM_k/T1X4UruQ5_I/AAAAAAAABDM/vhqoDtuIUHI/s1600/plimsoul.PNG

I couldn't figure out what D3 was for. Now I think it's purpose is to shunt the leakage current that will pass through D4 if the wrong polarity power is applied.

Andrew.

[GrindCustoms]

The Zener should behave like a normal diode for forward voltages below the breakdown voltage, so it should work for reducing the voltage of the leakage current from the 1N5817. I tried it with a +/- 15V supply on a 12V Zener (using a CLR in front of the Zener). When I put the +15V on the CLR, I get about +12V at the Zener. When I stick the -15V on there, I get around -0.7V. So, I think that part is a working OK.

When I do the proposed setup of PTC (about 10Ω resistance) into 1N5817 followed by a crowbar Zener or 1N4001, with the -15V I get -0.55V with the Zener and -0.41V with the 1N4001 at the power output.

Caveats:
This was all done with my now suspect DMM.
The power supply I used can only source 30mA, so this didn't test the PTC part at all, but I really wanted to see what the voltage would be before the fuse could blow.

That makes lots of V and mA drop in the end, where you able to take the mA reading at source and after the whole array?

I've got some UF4004 diodes handy, might worth to see what the differences could be using them... but those will let the AC pass thru with no restriction at all.