P

ProCoach

Well Known Member
Hot Fuel Gauges??

Okay, I hate this electrical stuff... My Vans fuel gauges are both getting VERY hot. I discovered this in flight and shut off the master. After I landed I disconnected them. Everything thing else is working perfectly. I've checked the connections and installed a new main breaker. No breakers are tripped by the gauges, but you could fry an egg on 'em. In fact, the lower rear gauge casing is warped from the melt. Needless to say they are presently disconnected. They seem to work fine... just heat up after a couple of minutes.

Anybody know what's going on??
 
Last edited:
check the alt. output

check your alternator voltage output. High voltage could be the culprit. If it is, your other guages and electronics could also be in harms way.
 
Alt...

Thanks, but according to its gauge the alternator is right on. All other gauges are working cool and perfectly. The fuel gauges will heat up just off the battery (i.e. on the ground w/ engine off). :confused:
 
Boy, that makes no sense at all (no pun intended). I would check the current draw of the gauges, which are just milliamp meters measuring the current thru the resistive tank sensor. I would think that the circuit shouldn't draw more than 2 or 3 watts,
otherwise, things are going to get warm. Three watts would be about 200ma with the alternator spinning.

Sounds like something is wrong with the meters?!
 
Is this a new problem?

Both running hot with one melted is more than alternator voltage, especially with everything else OK. I could buy one geting hot with a failure but not both. How do you have them wired up and do they indicate the fuel level that you think they should? Are you using the standard SW float sensors?

John
 
Think so (I just bought it). Wired - power to the right side gauge. Jumper to the left. The right was slightly hotter than the left. They both read accurately, even still. However, in flight (after they were really hot) they went "dead" and that's when I reached up there to see if I could feel something loose. When I did, I got burned and shut down the master.

This was after a soaking rain the night before and the humidity was very high that morning. Maybe something got wet (?). I first thought that because the faces of each gauge held moisture inside. But that condensation may have been the result of the heat up taking place.

So maybe the gauges are toast? I can replace them easy enough, but then it seems like I would be treating a symptom instead of a cause. I guess there's one way to find out.
 
ProCoach said:
Okay, I hate this electrical stuff... My Vans fuel gauges are both getting VERY hot. I discovered this in flight and shut off the master. After I landed I disconnected them. Everything thing else is working perfectly. I've checked the connections and installed a new main breaker. No breakers are tripped by the gauges, but you could fry an egg on 'em. In fact, the lower rear gauge casing is warped from the melt. Needless to say they are presently disconnected. They seem to work fine... just heat up after a couple of minutes.

Anybody know what's going on??
Click to expand...
Same exact thing happened to my Vans Gauges on the way to Arlington in July. Replaced both with new ones although mine were less that a year old.
Never could figure out what happened to them to cause the overheat. I did open one up and found burned components on the circuit board.
Regards,
Tom Velvick
rv-4
 
Misery loves company...

Misery loves company... so thanks for that. I'll plan on getting some new ones.
 
ProCoach, what else is getting wet?

Fog in the guages is not good. Those guages are toast.

My main concern is what else is getting wet? Fix the leaks first. Are we talking about $39 guages or $5K radios?

Don't care about a cheap guage but somewhere in the whole guage circuit you are getting enough current to melt the case.. this is connected to the sensor which resides in the fuel tank...

Don't like that.

This whole circuit is designed to be a low current sensor. By sensor, we mean *very* low currents that are reflected on the needle... not by melts in the case.

It may be that replacing the guage fixes the issue. Personally, I'd look more into it.

John

PS If you replace the guage, send me the old one. I'd like to dissect it..
 
Wilco... be happy to. What address?

If the circuit was getting that much current, wouldn't the circuit breaker pop? or next, the main 30A breaker? Just seems weird. How best to test?

I will get new gauges.

BTW, thanks!
 
ProCoach said:
Wilco... be happy to. What address?

If the circuit was getting that much current, wouldn't the circuit breaker pop? or next, the main 30A breaker? Just seems weird. How best to test?

I will get new gauges.

BTW, thanks!
Click to expand...
I wondered the same thing about my gauges when they got hot but the CB didnt pop. Here is a picture of what it looked like on the circuit board after the gauge overheated.
vans%20fuel%20gauge.jpg
 
rvator51 said:
I wondered the same thing about my gauges when they got hot but the CB didnt pop. Here is a picture of what it looked like on the circuit board after the gauge overheated.
vans%20fuel%20gauge.jpg
Click to expand...

Wild, but educated guess.... ;)

The voltage regulator circuit they use for dropping the 12 VDC to 5 Volts for use in their circuitry failed... If this happened, the whole thing is toast.... :)

If they use a similar circuit for all of their gauges (very likely), then the same failure mode could affect any gauge type.

Is the photo of a tach. cuircuit, not a fuel level gauge?

gil in Tucson
 
Sent you an email with address..

Thanks, that will be interesting to see.

You don't necessarily need a lot of amps to get components hot enough to damage things. I've fried lots of parts with currents as low as a few 10s of milliamps. Hopefully, there isn't any fault in a fuel sensing system that would pop a 30A breaker :eek:

As for testing, not too much to do there. Check the 12V power and ground. The disconnected sensor wire should show 20-240 ohms to ground depending on the fuel level. Full = 20 ohms. Doesn't sound like sending units, though..

When you replace the guages, you might want to measure the current that they both draw and put a small fuse in line with the source that will pop if a guage gets zapped again.

Have fun.
John
 
az_gila said:
Wild, but educated guess.... ;)

The voltage regulator circuit they use for dropping the 12 VDC to 5 Volts for use in their circuitry failed... If this happened, the whole thing is toast.... :)
If they use a similar circuit for all of their gauges (very likely), then the same failure mode could affect any gauge type.
Is the photo of a tach. cuircuit, not a fuel level gauge?
gil in Tucson
Click to expand...
Hi Gil,

The photo is of the internal circuitry of the Vans float-type fuel level gauge that failed on my wifes rv-4.
 
What did Van's say about your wife's fuel gauge? (You're a lucky man, btw) Did they get a bad batch of instruments or were they thinking this just just an isolated anomaly?
 
Chips.... and not the type you eat...

rvator51 said:
Hi Gil,

The photo is of the internal circuitry of the Vans float-type fuel level gauge that failed on my wifes rv-4.
Click to expand...

Thanks Tom....

An update, all the parts are 15V parts, so the power circuitry is probably a step up circuit....

The bigger chip (a Cherry Semiconductor CS289 - check the logo) is a tach. generator chip, which converts frequency to a 20 mA drive for a meter. However, it does also have a DC input option, so I guess they are just using the same chip in all of their gauges in this DC mode...
If any other EEs are interested, the other chip is a Dual Op-Amp...

gil in Tucson
 
Last edited:
Photo Update

Here's an update... I replaced the fuel gauges and took a photo. There was a small fire inside the gauge. That's what melted the casing. I knew it was very hot, but this is kind of scary. Luckily I caught it in time and shut down the master. The new ones seem fine and are not heating up (so far :cool: ).

Looks very similar to yours, Tom.

DSC05282.JPG


John, I'm sending you the other one.
 
John/Tom

... it looks like a component completely vaporised itself. Absolutely no trace left...

Can you see the connections into the circuit board by the lower mounting hole in John's picture this component used to have?

Tom's picture has the same evidence.

I wonder what that component was...... definitely the weak link in the system.... :)

gil in Tucson
 
Has the gauge manufacturer (UMA?) and vendor (i.e. Van's) been informed of these failures? I would hope that they would be interested in improving the design to solve this issue. But the problem doesn't exist from their point of view unless they are informed about these failures.
 
Vans was made aware

Vans knows, as this is where I started my inquiry into what was going on. It (the problem) evidently stumped them. The best information, by far, was given here.

I also informed Vans of this thread. Don't know what, if anything, that means to them... I think they were just content to sell me new gauges.
 
Some test results

ProCoach sent me the scorched guage and I was able to generate a schematic from it. My best guess as to what went wrong is that D3 failed shorted. Its a zener diode across the input of the 5 Volt regulated power supply and when it shorts, the big resistor (the smoked one) carries enough current to burn out.

The resisistor-zener network appears to be a pre-regulator for the little 5Volt regulator IC3 (78L05) to keep its power dissapation down or (less likely) a transient protection deal (zeners are louzy at that).

At any rate, I removed the shorted D3 and piped in the power past the smoked resistor and the guage actually works! I'll use it in MY plane :D

I did not see any evidence of water damage.

So why did D3 fail? Hard to say since it and the resistor were too smoked to get any values off of it but it could be that its undersized and runs hot. It could also be getting hit with transients (maybe from a master contactor without a suppression diode on it or other source). At any rate, looking at the other pix here it looks like its a common failure point.

Other interesting thingies:

The apparently 'missing component' was just not stuffed. Its a jumper to bypass some resistance in the amplifier section. There were no residual component leads in the holes.

With a 22 ohm sense resistor (full fuel) the guage pulls about 130 milliamps at 14 volts supply. I think these guages should be fuzed and that will give a starting place to pick one. Other guages in the lineup may have similar issues, but I don't know that for sure.

The CS289 tach chip is really there as a meter-driver. The air-core meter (necessary for 270 deg swing and a good choice for high vibration environments) works by applying voltage to two coils set at 90 degrees to each other. The CS289 converts an input current to sine-cosine signals to set the angular displacement of the needle. The chip also has circuitry to convert a pulsed input from a tach to the same input current but its not used in the fuel guage.

As Gil A. indicated, the other chip is a 358 OpAmp. One section is used to scale the varying sensor resistance and apply an offset voltage as required by the CS289. The blue pot calibrates the guage.

The 5V regulator is used to get a (sort of) regulated voltage source for the resistive fuel sensor. It still works fine after D3's noble sacrifice.

As far as what to do about it, if the zener is undersized and running hot, not much can be done from the outside other than fuzing. If its transient sensitive, a MOV across the power terminals may help.

Thanks again PC, for the guage, it was fun messing with it. If someone has an old working one it might be fun to see what the levels on that power circuit are and what happens under transient conditions.

And then I'd have TWO :p

John
 
Last edited:
You must log in or register to reply here.