Circuit testing handheld switch

[DaveC426913]

Help! I've got this device that I feel responsible for getting working and I want to test its circuits. It is currently dead.

It's this http://www.canadiantire.ca/browse/p...lder_id=1408474396672427&bmUID=1230269280757".

The attached pic is the handheld controller. Power comes in from an AC adapter or a 12V cigarette lighter adapter. The other wires go back to the device and control various heaters, fans and vibrators. One set of wires goes to a dead-man switch that deactivates all electronics if there's no one in the seat.


First step: using a multimeter, I've verified that the power supply is indeed reaching the board.

Second step: check the fuse. Um... could someone verify how I test a fuse with the multimeter? I set the dial to Ohms and, if it's blown, I should read infinite resistance across it, right? i.e. the needle should not move.

But it does give me a reading. It gives me the same reading as when I short the leads together. So that means fuse is OK, right?

Now I'd like to verify that this dead-man switch is working. I cannot find the switch itself (I've looked extensively). But it seems to me that I should find some combination of two of those off-board wires that read an infinite resistance when I'm off the seat and less-than-infinite resistance when I sit down. So far, all pairs read less than infinite resistance, but I'll have to do an exhaustive check.

While I do that, anyone have any suggestions (I mean other than 'take it back to the store' and ' don't buy that junk') as to how to proceed?

 

[MATLABdude]

Help! I've got this device that I feel responsible for getting working and I want to test its circuits. It is currently dead.

It's this http://www.canadiantire.ca/browse/p...lder_id=1408474396672427&bmUID=1230269280757".

The attached pic is the handheld controller. Power comes in from an AC adapter or a 12V cigarette lighter adapter. The other wires go back to the device and control various heaters, fans and vibrators. One set of wires goes to a dead-man switch that deactivates all electronics if there's no one in the seat.


First step: using a multimeter, I've verified that the power supply is indeed reaching the board.

That's good. If you're getting 12V at the point that's labelled 'DC' on the circuit board, you're getting power to the system. Alternately, it looks like there's some LEDs that should work be on, or at least blinking, or doing something.

Second step: check the fuse. Um... could someone verify how I test a fuse with the multimeter? I set the dial to Ohms and, if it's blown, I should read infinite resistance across it, right? i.e. the needle should not move.

But it does give me a reading. It gives me the same reading as when I short the leads together. So that means fuse is OK, right?

Yes. Good fuse = (near) short circuit

Now I'd like to verify that this dead-man switch is working. I cannot find the switch itself (I've looked extensively). But it seems to me that I should find some combination of two of those off-board wires that read an infinite resistance when I'm off the seat and less-than-infinite resistance when I sit down. So far, all pairs read less than infinite resistance, but I'll have to do an exhaustive check.

While I do that, anyone have any suggestions (I mean other than 'take it back to the store' and ' don't buy that junk') as to how to proceed?

Has this unit ever worked? Usually, if something comes DOA out of the box, I just take it back for an exchange, unless it's trivially easy to fix, or can't be returned (i.e. came as a "no guarantees, no returns" and heavily discounted / free deal).

First off, does the car socket work (outputs 12ish volts)? You can also use that AC adapter (if it comes with one). If it does work, did you try checking the resistance between the incoming 12V (red) and GND (black) lines (without it being plugged in, of course). A short circuit would indicate a blown power regulator or protection diode (which might be the root cause, or just another indicator).

If that's okay (and since there doesn't seem to be any obvious sign of catastrophic failure), I'd check to see if the switches are okay (and especially the power one). You can use the ohm meter or continuity check to see if the terminals switch properly (they look like SPDT switches, so there should be a common terminal, and two switched ones).

Following that, I'd check the active components (again, with the seat unplugged). Check to see if the diode in the corner is acting as a short circuit, regardless of which way you're measuring the diode. If your DMM has a diode check mode, you should also be able to confirm that the diode forward voltage is correct (and around 0.2 to 0.5 volts), and that there is no backwards diode voltage. You can also perform the same tests on the BJTs(?)--those D-shaped black things with the three leads. While sometimes two of the legs are shorted together (you can tell if there's a trace joining the two terminals, or a small resistor) this is usually not the case, and the middle terminal should not be shorted out to either of the side ones.

Outside of that, I'd guess the IC, or perhaps the actual heating and vibrating units. Unless your dead man's switch really *isn't* working. You might want to check that right off the bat if the control unit is otherwise lighting up, and doing it's thing (there's only 9 or so connections--and chances are that you'll see one of the lines going to ground when you sit on it, or set something heavy on top).

That's all I've got. Good luck!

 

[DaveC426913]

Unless your dead man's switch really *isn't* working. You might want to check that right off the bat if the control unit is otherwise lighting up, and doing it's thing (there's only 9 or so connections--and chances are that you'll see one of the lines going to ground when you sit on it, or set something heavy on top).

This is where I am so far. No point in testing the rest of it until I determine that the dead-man is functioning. So far, I haven't found a combo that changes resistance when I sit on the seat. That would indicate that the dead-man switch is the culprit. At least, so far.

Then again, I haven't found the switch itself either, so I can't even be sure I'm activating it. (Presumably, it's under the seat.)

 

[DaveC426913]

Update: I tested all 8 wires against black. All had something less than infinite resistance. None changed when I sat on the seat.

Either I'm testing incorrectly, or the dead-man switch is permanently closed and there's some other fault in the system.

 

[Integral]

Dave,
The yellow and purple wires seem to be in series with the switch. If that is a power switch then those 2 wires need to be connected for it to function. Try shorting between them see if it comes on.

 

[turbo]

Dave, this is a cheap wave-soldered board, and looking at the solder joints on the underside of the board, some of them look funky. This can happen if manufacturer is pushing production and the solder in the pool is contaminated or the temperature control is not good. I would suggest looking at the joints on adjacent traces, and if some of the solder is bridged from one joint to another on adjacent traces, carefully trim it away with an Xacto knife. Sloppy soldering can cause shorts that are very hard to diagnose. It is for this very reason that I refused to service guitar amps with PCBs. You can waste a LOT of time troubleshooting only to do an inspection and find a factory defect.

 

[DaveC426913]

Dave,
The yellow and purple wires seem to be in series with the switch. If that is a power switch then those 2 wires need to be connected for it to function. Try shorting between them see if it comes on.

I'll look at that but if you're talking about the black switch on the right it is not a power switch. That is actually a 3-setting knob for temperature. But the first white switch on the right has settings for Heat / Off / Fan, so I guess that's the power switch.

Dave, this is a cheap wave-soldered board, and looking at the solder joints on the underside of the board, some of them look funky.

The board looks fairly clean up close but it's as if someone sprinkled cake dough or Plaster of Paris on it - which then dried..

 

[DaveC426913]

Verdict is in:

'twas the dead-man switch and 'twasn't the dead-man switch.

The switch was the culprit - but it wasn't failing, it's just very poorly designed.

If I locate the switch exactly (required someone who already owned one to show me, because it's so obscure), and press it hard, the thing works great. Anything less is not enough.

Now to decide whether to bolster the switch's performance, or simply short it out...

 

[berkeman]

Glad you figured it out, Dave.