I thought I was going nuts!

[Roff]

I built this little set-reset latch out of some 2N7000's. I turned it on, and the LED it was driving came on, but instead of 10mA as designed, it was glowing dimly with about 1mA through it. I looked for solder splashes - none. I started disconnecting wires. After disconnecting both the drains that were connected to the LED, it still glowed. I disconnected the gate of the cross-coupled device, and the LED went off. I checked the gate-source resistance. It was not high, like it should be, so I replaced that FET. Same problem. I checked the wiring again carefully, and it was fine. I checked the gate-source resistance of the FET I had just installed. Same problem. I checked the gate-source resistance of an unused part. It was fine.
I had an ah-hah moment. My soldering iron must be blowing gates. I checked the tip, and it had 24V rms on it! The iron is an old Weller TC202 with the Curie point pencil. The tip is grounded by design, but I discovered that the receptacles in the 3-wire connector where the pencil plugs into the power unit were stretched to the point where they were intermittent. I closed them up a little bit with an icepick, which solved the problem, at least temporarily.
I wound up having to replace all 3 MOSFETs. The little flip-flop works now.

 

[dknguyen]

THat's annoying. So I take it if the soldering stand was grounded it would have helped? or no?

 

[Roff]

THat's annoying. So I take it if the soldering stand was grounded it would have helped? or no?

Well, I don't know what you mean exactly. What I meant was, the tip, by design, was grounded to earth through the pencil connector (where the pencil cord plugs into the main power unit) to the power cord ground lead. The pencil connector ground lead was intermittent, so the tip was picking up AC (coupled capacitively, presumably) from the power to the heating element.

 

[dknguyen]

Oh, so it wouldnt have helped since the the pencil would just get charged up again.

 

[Roff]

Oh, so it wouldnt have helped since the the pencil would just get charged up again.

It wasn't static charge. It was an AC voltage, coupled from the secondary of the transformer that powers the tip.

 

[Hero999]

You're lucky it wasn't a really expensive IC you were soldering.

 

[Roff]

You're lucky it wasn't a really expensive IC you were soldering.

I know. I'm surprised I didn't zap something long ago. It's probably been intermittent for a long time. I haven't been soldering many MOSFETs, I guess. I did some PIC breadboarding recently, but I used sockets (thank God!).

 

[Boncuk]

Hi Roff,

imagine your "plumber iron" was connected to mains. That way you could have tracked it easier.

 

[Roff]

Hi Roff,

imagine your "plumber iron" was connected to mains. That way you could have tracked it easier.

It is connected to the mains, but fortunately, the tip wasn't shorted to the high side.

 

[Pommie]

Would I be right in assuming that "you" supplied the earth return path?

Mike.

 

[audioguru]

Hi Ron,
My Weller soldering iron is also about 41 years old but it hasn't gone intermittent yet.

Get one of those lousy battery operated ones that heat and cool quickly.
Or get a butane powered blow torch soldering iron.

 

[Roff]

Would I be right in assuming that "you" supplied the earth return path?

Mike.

That's probably true. I didn't have my scope probe connected while I was soldering, and that was the only grounded instrument I was using.

Since there seems to be some confusion (not on your part, Mike) on what the problem was, I created this crappy drawing. I solved the problem by closing down the sprung receptacles with an ice pick.

 

[Roff]

Hi Ron,
My Weller soldering iron is also about 41 years old but it hasn't gone intermittent yet.

Get one of those lousy battery operated ones that heat and cool quickly.
Or get a butane powered blow torch soldering iron.

I have a butane-powered iron, but I think the only time I used it was when I was installing reed switches in my door and window frames as part of my security system. It worked OK, but I've never tried it on other electronic components.

 

[Bob Scott]

My Weller soldering iron is also about 41 years old but it hasn't gone intermittent yet.

My 40 watt gold tipped Ungar Imperial is only 39. I ripped the ground prong right of of the plug 39 years ago when grounded receptacles were not in common use. It has never "blown" any CMOS or FET (and it has soldered a LOT of those. Maybe I'm lucky.

Is Curie temperature the temperature threshold where iron loses its ferromagnetism? How hot is your iron Ron? Wattage?

 

[Roff]

My 40 watt gold tipped Ungar Imperial is only 39. I ripped the ground prong right of of the plug 39 years ago when grounded receptacles were not in common use. It has never "blown" any CMOS or FET (and it has soldered a LOT of those. Maybe I'm lucky.

Is Curie temperature the temperature threshold where iron loses its ferromagnetism? How hot is your iron Ron? Wattage?

Yep, the Curie point is the temperature where a ferromagnetic material loses its ferromagnetic ability. The magnet is in the barrel. I had always assumed it was in the tip, but I never really thought about it, so I checked it out just now. The slug in the tip apparently loses its ability to be magnetized at the Curie temperature, which allows a switch in the barrel to open.
My iron draws 60 watts. Right now, I have three 700 degree tips and a boss hog 800 degree tip.
I actually ordered a replacement heating element a couple of years ago, after the old one died. I think it was about 25 bucks - a heckuva lot cheaper than a new station.

 

[Hero999]

Do you think earth bonding one side of the mains transformer's secondary might have helped?

The secondary side was probably floating at 60VAC or more and bonding it would reduce the maximum voltage relative to earth to just 24VAC. This means that the tip voltage will be even lower if this happens again.

Even better, if it's a centre tapped transformer, earth the middle so the maximum voltage with respect to earth is just 12VAC RMS or 17V peak which is low enough not to fry a MOSFET gate.