Perplexing "FUEL" light malfunction

[Dave WM]

Just finished testing. Results are in. works the same with one or two lights (in series). That is it stays off in gas (warmed to about 118f) but will come on within seconds of removal of the sensor from the gasoline. I had to cut the on time short for the single bulb as I did not want to overload the filament. Based on my test I can definitively say that it works with either one or two GE47 bulbs. Are they the same as the stock unit, no, but I feel the cover a broad enough range that its safe to say the circuit should work with a low amp 12v bulb. the GE47 is approx. 1 watt at 6v from what I can find.

The only other test I can think of is to try the setup in the car (I will set it up on the floor board and directly connect to the factory light). I will set that up later today and report back I have a video coming on the test I just reported on.

below is a few hours later. What I am not clear on is I tested it for well over 1 hr with 2 bulbs and then well over 1hr with a single bulb. The overall current was pretty much the same, I suppose due to the ratio of the bulb resistance to the thermistor resistance. The only diff was if I had allowed it to go on much longer on the single 6v bulb I suspect it would have burned out.

Another observation, when testing with water vs gasoline, the thermistor can dumped MUCH faster with gas vs water. I suspect a surface tension issue.

Las thing, I forgot the thermistor was in warm gas when I checked the resistance, about 1k I took it out sprayed out the warm gas cooled with the solvent it was over 2k, I suspect when it settles down to room temp (aprox 78f inside the house AC) it will return to the 1.7kish resistance.

 

Edited August 10, 2020 by Dave WM

[Captain Obvious]

So basically your summary is that the Toyota thermistor works great, but the ZCD part does not. And not only that, but the Toyota thermistor seems to work great* with two wildly different bulbs.

Did I get that right?  

 

 

* Assuming you don't leave it on so long that you fry your 6V bulb by applying 12V to it.

[Dave WM]

1 minute ago, Captain Obvious said:

So basically your summary is that the Toyota thermistor works great, but the ZCD part does not. And not only that, but the Toyota thermistor seems to work great* with two wildly different bulbs.

Did I get that right?  

 

 

* Assuming you don't leave it on so long that you fry your 6V bulb by applying 12V to it.

in a nut shell..

:)

Tomorrows test will involve using the light bulb that is installed in the Z, which I presume is nothing special, same as all the other OE dash bulbs.

Edited August 11, 2020 by Dave WM

[Captain Obvious]

Got it.

Dr. Dave, Did you drill out the holes in the ZCD version yet to see if getting more flow in there helps? I'm still not yet convinced this is a completely electrical issue.

If you're reluctant to make modifications to a unit that you still might want to return, you could possibly simulate larger holes by encouraging flow? Read... "Wave it around in a cup of liquid (to try to force fluid through the little holes) while monitoring the state of the lamp."

[Captain Obvious]

25 minutes ago, Dave WM said:

 the light bulb that is installed in the Z, which I presume is nothing special, same as all the other OE dash bulbs.

Yup. Most of them should all be the same. The glove box might be the easiest to get to if you just want to pull a bulb. That or maybe the ashtray bulb?

Thought of something though... After all the years and PO's though, you should make sure you've got the right bulb. Original is Toshiba A12V3.4. There are other numbers that have been used over the years as replacements, but if you find that Toshiba bulb, you know you've got an original.

Edited August 11, 2020 by Captain Obvious

[Zed Head]

That plastic piece just has to be for locating and protecting.  It's just slipped in after the thermistor is fed in to the can and the wire is soldered to the bottom.  Give it a tug.  It will probably slide right out and expose the thermistor. 

[dmorales-bello]

Got it.
Dr. Dave, Did you drill out the holes in the ZCD version yet to see if getting more flow in there helps? I'm still not yet convinced this is a completely electrical issue.
If you're reluctant to make modifications to a unit that you still might want to return, you could possibly simulate larger holes by encouraging flow? Read... "Wave it around in a cup of liquid (to try to force fluid through the little holes) while monitoring the state of the lamp."

Yes, I did. I carefully enlarged the hole in the middle of the white plastic cap at the top with a fine Dremel burr, and enlarged the three holes in the bottom of the can with a 1/16 drill bit on the Dremel. I performed the "dunk" test in water before and after the mods and flow into and out of the can definitely improved.
That modded unit is in my gas tank now and unfortunately produced the same malfunction as before enlarging the holes. I'm testing the different bulbs on it as reported in my previous post.
I think the "trapped air bubble" theory for the malfunction can be discarded. At this point, with all the tests@Dave WM and I have run, the most probable cause is that the ZCD thermistors in both units I received are not adequate for this application.
Attached are pics of the enlarged holes.


Sent from my SM-N950U using Tapatalk

[Dave WM]

what is interesting to note is the thermistor did exactly what Dr Dave had when I used water (just in case that got lost in all my post). I attribute this to poor fluid flow perhaps due to surface tension. If Dr. Dave wants to try it I will be happy to send him this one, I would like to test a bit more, but at this point the best test would be to install in a car that has a top loader (mine is a side loader). If it does not work in Dr Daves  car would assume he is driving around with 1/2 tank of water.

 If she weighs the same as a duck then she is a witch (a man of science) .

 

Edited August 11, 2020 by Dave WM

[dmorales-bello]

"Witch! Witch! Witch!"
Sure, Dave. I'll stick it in my tank. .
Seriously now, of course I'll try it.

Sent from my SM-N950U using Tapatalk

[Captain Obvious]

1 hour ago, dmorales-bello said:

I carefully enlarged the hole in the middle of the white plastic cap at the top with a fine Dremel burr, and enlarged the three holes in the bottom of the can with a 1/16 drill bit on the Dremel.

I think the "trapped air bubble" theory for the malfunction can be discarded.

Oh. Well that's too bad. That would have been easier to deal with. So we're back to needing a different thermistor, or a workaround for the current one.

My read on the thermistor is that (within reason), the cold resistance doesn't really matter. As long as it's on the order of KOhm(s), it should be fine. And as for the hot resistance, as long as it's on the order of a couple Ohms, it should be fine too. But the absolute numbers have wiggle room.

What's really important is that it's small enough that it'll heat up when you push current through it, but not so small that it'll get so hot that it can't be cooled by a gas bath. In other words, what's most important is it's ability to generate and reject internally generated heat.

If it's huge and has big winged heat sinks on it that do a fantastic job of getting heat out of it's core, it'll never heat up enough to get into thermal runaway.

Conversely, if the core is surrounded by an insulating layer, you won't be able to get the heat out of the center fast enough and once it's in thermal runaway, you won't be able to recover. This sounds, at first glance, what is happening here.

[Zed Head]

3 hours ago, dmorales-bello said:

Yes, I did. I carefully enlarged the hole in the middle of the white plastic cap at the top with a fine Dremel burr, and enlarged the three holes in the bottom of the can with a 1/16 drill bit 

I think the "trapped air bubble" theory for the malfunction can be discarded. 

But the insulation effect can't be discarded.  You don't know where the thermistor itself is sitting.  It might be jammed in to a hole in the bottom of the plastic piece.  Insulated.  The extra flow could be solely from the larger holes in the can.

Sorry.  But assumptions are the bane of drawing firm conclusions (that's not from a movie, as far as I know.  But I tried to make it sound like is.).

Edited August 11, 2020 by Zed Head
no know

[Dave WM]

I liked the title object of the movie, search for the holy grail, not too far off topic.

btw, what's your favorite color? (from the bridge keeper)

Edited August 11, 2020 by Dave WM