Do brake light switches need power to be bench tested?

[sneezy]

Howdy TheSamba.com! It's story time!

The Setup:

I just bought a '63 bug with a bone dry brake system. It included one old corroded probably-original master cylinder + brake light switch installed on the car, plus a newer Brazilian master cylinder + brake light switch in a parts box.

So I pulled the old MC, set up a bench bleeding rig, and got it bench bled with not too much ado (though it had leaky seals, which I assume is why it was so corroded, and also why the system had been drained) – then I ran a volt/ohm meter across the two terminals on the brake light switch to test that, which is the first time I've tried to do that. Figured it should work, the wiring schematic shows it as just a plain switch with power in and ground out.

But! No matter how hard I pushed on the piston, the meter showed infinite ohms – no circuit.

So I replaced that switch with the switch from the Brazilian MC – but I couldn't get a circuit on that one either.

So I swapped out the old MC for the newer MC in my bench bleeding rig, and tested both switches on that – no circuit.

I wasn't happy with the action on the Brazilian MC anyway so I bought a new $25 MC and switch, and bench tested that one too – no change. (At least the action is nice on it. For now.) I tested all three switches with all three MCs – no circuit.

Out of desperation I threw all three switches in a vise and blew 40 psi into each of them – no circuit.

The Question:

How can an old VW two-terminal brake light switch be bench tested? Do I need to run some volts through it first before applying pressure will do anything?

Maybe I really have three bad brake light switches, or maybe I'm just missing something silly. Regardless, the plan is to put the best-functioning master cylinder I have in the car and attach the shiniest brake light switch I have to it, then fill up the system, bleed it until I get pressure in the master, and see if the lights work… if not then it's back to the bench. Either way I'm still curious about the answer to this question.

Thanks!

[herbie1200]

Brake light switch are pressure switches, so to turn them on you need a lot of pressure in the MC.

Which pressure?

A rapid math:

Pedal is pushed with 1kg; the pedal lever multiplies this force by about 10: 10kg on the MC pushrod.

This pushrod acts on a 19mm cylinder, whose area is ~282mmq, 0.00028 Mq.

Pressure (in Pascal) = F/A = 10kg*9.81 /0.00028=345643 Pascal =~ 3.4 Bar.

To simulate 1kg on the pedal you should push your MC with 10kg... are you sure you did this?

Try to inflate the switch with a foot pump, you should listen a "click" between 2 and 4 bars.

[bluebus86]

to test the switch electrically, wire up two brake bulbs and power it, then when activated measure the drop across the switch. excessive drop is a problem. however if you test first with an ohm meter, if it fails, the drop test wont pass also.

Bug On
_________________
Help Prevent VW Engine Fires, see this link.....Engine safety wire information

Stop introducing dirt into your oil when adjusting valves ... https://www.thesamba.com/vw/forum/viewtopic.php?t=683022

[Cusser]

With brake fluid pressure applied (pedal pushed), an electrically disconnected brake light switch should show continuity/small resistance, and infinite resistance with the pedal not pushed (no pressure). No need to have electrically connected to test.

For later 3-terminal switches, same holds but use the two parallel terminals.
_________________
1970 VW (owned since 1972) and 1971 VW Convertible (owned since 1976), second owner of each. The '71 now has the 1835 engine, swapped from the '70. Second owner of each. 1988 Mazda B2200 truck, 1998 Frontier, 2014 Yukon, 2004 Frontier King Cab. All manual transmission except for the Yukon. http://www.thesamba.com/vw/forum/album_page.php?pic_id=335294 http://www.thesamba.com/vw/forum/album_page.php?pic_id=335297

[bluebus86]

[sneezy]

Thanks for the replies – so to summarize:

1. "Try pushing harder, and try a foot pump."

I just tried blowing 100 psi into the switches, but there was no visible change in the results on the multimeter. I don't have a foot pump, and the shop air is noisy – but shouldn't I see a different result on the meter regardless?

3. "Run power across it and look for a voltage drop – but if the ohmmeter test fails, this won't help."

So the meter test should work, even unpowered, no?

4. "An electrically-disconnected switch should show some continuity under pressure."

A +1 for my theory and test… and does suggest all three switches are bad.

5. "The ohms test won't tell you if the switch is good, just that it's responding to pressure."

Noted, and thanks for the tip!

So it sounds like my tests should have shown some resistance change, and the switches are likely bad. I have a new one on the way, I'll try again with that one and update here.

Thanks all! 👍

[Sharp64]

Why not just screw them in and test them in car? Takes 5 minutes. The cheap switches suck. Even some of the less cheap switches suck. I only buy NAPA anymore as I’ve seen others fail after a very short time.
_________________
1960 Indigo Blue Ragtop (Blue Barry)
1967 Ghia
1985 Westy Money Pit
“...some men aren't looking for anything logical, like money. They can't be bought, bullied, reasoned, or negotiated with. Some men just want to watch the world burn.” - Michael Caine

[mukluk]

[sneezy]

Hey all! Thanks for the comments and suggestions.

"Why not just test it in the car?" A few reasons: when I started trying to test the switches, the system was bone dry, the switch and MC were already out and on my bench, and I was already bench testing the MC. Additionally, I didn't want to just test a switch, I also wanted to learn how to bench test a switch – not least because I couldn't seem to find any information about how to do this online, so I wanted to fix that problem too.

Since my last post, I installed the best master cylinder into the vehicle and tested the shiniest brake switch – and it worked! …Which means using shop air and a multimeter did not give me a useful result.

So for now it looks like the answer to my original question is "yes", but I'll test that hypothesis too.