Printable Version of Topic

Posted by: john77 Mar 27 2014, 02:24 AM

With the ignition on the needle on the gauge lifts to E but goes no further despite the tank being full.

I pulled the sender, wired up a new ground and hooked up a new wire from the gauge to the sender and tested the sender manually by turning it upside down and then back. The needle goes from empty to full and back to empty. Just to double check I test the resistance, 0.2 ohms right side up, 89.9 ohms upside down (the range is 0-90 ohms).

So I put the sender back in the tank with the new wiring, switch on the ignition and... the needle rises to E and goes no further.

So I think maybe, somehow, the floats not lifting. So I pull the sender, fill a bucket of water, dip it in, and the needle on the gauge rises and tells me I have half a tank - or less or more depending how much I submerge it. In other words, the damn thing is working.

So I put it back in the tank and nothing, the gauge rises to E and stays!!!

What the hell am I missing here? It works in a bucket of water but move it a foot and put it in my gas tank and nothing.

John

Posted by: 914itis Mar 27 2014, 05:23 AM

The gauge is only receiving 12 volts, nothing from the sende. Check the wire.

Posted by: john77 Mar 27 2014, 07:50 AM

Yep, checked that. It's receiving 11.8 and putting out 5.9v to the sender with the ignition switched to ACC, which is correct according to the manufacturer.

Posted by: stugray Mar 27 2014, 09:20 AM

Are there two wires coming off of the sender unit (G and GND?).
Or does the sender unit ground through the tank?

Posted by: john77 Mar 27 2014, 11:30 AM

The sender has two wires coming off it, one that grounds it (I've rewired it to the ground under the dash by the fuse panel) and the other goes to a terminal on the gauge - these are aftermarket, not standard 914.

Just learned the dialectric constant of water is way higher than gas, so my water test was pointless.. who knew

I know the sender has a good ground and the gauge to sender wire is good as I just rewired both, and I know the gauge is getting 11.8V with the ignition switched to ACC, so I guess, seeing as it's the only thing I haven't rewired, next thing is to rewire the ground to the gauge...

Posted by: 904svo Mar 27 2014, 12:42 PM

Guess a WAG but I supect the grond is missing off the gas gauge (brown wire) to
the gauge case.

Posted by: john77 Mar 27 2014, 06:52 PM

Sorry, do you mean from the gas gauge ground terminal to the gas gauge case? So an internal wire?

I've run a new wire from the gauge ground to a ground under the dash. A new wire to give power to the gauge. A new wire from the gauge to the sender. And a new ground from the sender to a ground. So all new wires. And the damn thing still won't rise above E with 4/5ths of a tank of gas!

Posted by: timothy_nd28 Mar 27 2014, 06:55 PM

aftermarket sender? Perhaps the rubber gasket is isolating the sender from chassis ground? Post a pic of this sender.

Posted by: john77 Mar 27 2014, 07:25 PM

Yeah, the sender and gauge are both from Fuel Safe, so they're definitely compatiable.

Here's a pic of the sender. If it's relevant, when I spoke to a guy at Fuel Safe he said the aluminum shaft of the sender is grounded. (Ignore the fact both wires are white, it's all I have to test with right now. Left terminal goes to gauge, right to ground.)

As I said earlier in the thread, if I remove the sender and invert/right it the gauge works fine, it just stops as soon as I put it back in the tank.

Posted by: timothy_nd28 Mar 27 2014, 07:34 PM

Great, one more pic of the back end of the gauge.

Posted by: john77 Mar 27 2014, 08:31 PM

Back of the gauge:



As far as the DAMN NEEDLE WILL GO WITH THE IGNITION SWITCHED ON (sorry, deep breath):



And the 'instructions' the Fuel Safe guy sent me:



John

Posted by: timothy_nd28 Mar 27 2014, 08:59 PM

Okay, is this how you have it wired up?

Posted by: john77 Mar 27 2014, 09:18 PM

Yeah, that's how it's wired. I'm getting 11.6V with the ignition switched to ACC.

Posted by: timothy_nd28 Mar 27 2014, 09:21 PM

Remove the sender wire off the gauge. With the ignition switch set to on, try jumping the ground terminal (back of gauge) to the sender terminal. This should simulate 0 ohms.

Posted by: john77 Mar 27 2014, 10:34 PM

Do you mean the sender terminal on the back of the gauge or on the sender?

Posted by: timothy_nd28 Mar 27 2014, 10:45 PM

Gauge

Posted by: zambezi Mar 27 2014, 11:28 PM

sounds like the sender is not grounding properly once mounted in the tank.

Posted by: john77 Mar 28 2014, 12:44 AM

Okay, just got home and tried this out, hopefully I'm connecting it right. When I switched the ignition on nothing happened, the needle didn't move at all. But when I then replaced the ground and the sender wire and switched on the ignition the needle flipped all the way over to the other side and stayed there.

Posted by: john77 Mar 28 2014, 12:53 AM

I wondered if it had something to do with the metal shaft of the sender possibly touching the edge of the hole it sits in in the tank, so I asked a guy from the company and he said "The outside aluminum shell or shaft of our standard senders is grounded so touching the side will not hurt anything as long as the ground wire is on the correct lug or terminal.'

So unless it's a problem internally with the sender I'm not sure what it could possibly be. Doesn't the fact it read 0-90 ohms when removed and inverted show the internals of the sender are fine though?

Posted by: 904svo Mar 28 2014, 07:31 AM

Thats not a Porsche gas sender (Porsche has 3 connections). Most after market
senders read 90 ohms (empty) 0 ohms (full). Porsche senders are 90 ohms empty
0 ohms full. Try removing the sender from the tank and turn it upside down and
see what happens.

Posted by: stugray Mar 28 2014, 07:56 AM

This should only matter if the fuel level sender is the capacitive sense type.
This one has a moving float right?

ANd the pic directly above is not correct.
You have the +12V hooked up and the Sense "S" hooked to the gauge GND, but the gauge is not hooked to car GND.

And the fact that it worked in the water, but not in the gas (in the tank) points to a grounding issue.
When you are in the water, the sender body is not hooked to car Chassis, but when in the tank, it is.

Posted by: john77 Mar 28 2014, 08:16 AM

One step ahead of you - see my original post . Out the tank, hooked up to the gauge, if I flip the sender the needle on the gauge moved as it should.

Posted by: timothy_nd28 Mar 28 2014, 09:11 AM

That test is almost correct. You jumped out the S to G on the back of the fuel gauge, but you need to have another wire that ties both of them to ground or a brown wire in your dash.

This test was to simulate a 0 ohm condition, and according to the spec of this gauge the needle should of went to empty. I just wanted to confirm that 0 ohms is empty and 90 ohms is full. It seems backwards to me, and that test you did would of verified that.

Another test you can do.. Pull both wires off the sender and ohm each terminal to one of the screws that holds the sender down.

Posted by: john77 Mar 28 2014, 10:17 AM

Ha, okay, done that now. Yep, the needle just reads empty.

Posted by: timothy_nd28 Mar 28 2014, 10:36 AM

K, now measure the sender unit. Measure the resistance between either terminal to ground. Then measure the resistance between both terminals.

Posted by: john77 Mar 28 2014, 04:36 PM

Between the terminals is showing 2.3 ohm.

Sender wire terminal grounded is showing 116.5

And the ground terminal to ground is showing 118.4

Posted by: timothy_nd28 Mar 28 2014, 04:41 PM

All wires off the sender while measuring?

Posted by: john77 Mar 28 2014, 04:46 PM

Haha, no. Oh boy, I'm really showing my skills here. Both failed to give me a reading and when I tested for continuity (?) neither gave me a tone.

Posted by: timothy_nd28 Mar 28 2014, 05:19 PM

Ok, so all wires are disconnected at the sender and you read infinite resistance between both terminals?

Posted by: john77 Mar 28 2014, 05:25 PM

Between both terminals I get 2.3 ohm.

From either of the terminals to a ground is infinite.

Posted by: timothy_nd28 Mar 28 2014, 05:38 PM

K, 3 ohms should reflect an empty tank. Dumb question, how much fuel do you have in that tank?

Posted by: Dave_Darling Mar 28 2014, 05:51 PM

That sounds like a real problem there. If you have more than a few ohms resistance between something that is supposed to be grounded and something that actually is grounded, you have a problem...

--DD

Posted by: stugray Mar 28 2014, 06:05 PM

I will try again:

"I pulled the sender, wired up a new ground and hooked up a new wire from the gauge to the sender and tested the sender manually by turning it upside down and then back. "

If you can turn the sender upside down and have it change resistance, then it is not a capacitive type sender.
So:
"Just learned the dialectric constant of water is way higher than gas, so my water test was pointless.. who knew "

This should not matter.
You can make it work in the water, but not in the car (gasoline).
SO when you put it in the car, you are somehow shorting out the sender unit.

Try swapping the wires on the sender unit.
One of them is probably grounded to the sender case.
The other is the variable resistance.

If it is hooked up backwards, it will work in the water, but not in the car.

Posted by: john77 Mar 28 2014, 06:24 PM

The tank is 4/5ths full.

Posted by: john77 Mar 28 2014, 07:07 PM

Sorry Stu, I keep checking back on this on my phone and completely missed your first reply.

The sender is an aluminum shaft with a float inside it that appears to run up and down a circuit board. This is the unit:

http://www.fuelsafe.com/store/sending-units-gauges/sending-units-gasoline-only.html

Maybe I'm misunderstanding how the sender works, but when I put it in a bucket of water the ground terminal on top of the sender was wired to a ground on the chassis. Isn't that how it grounds?

The shaft doesn't seem to touch the tank - it's smaller than the hole, the top of the sender is plastic, and there's a gasket between it and the tank.

When I asked the guy from Fuel Safe if the sender touching the metal of the tank would be a problem he said:

"The outside aluminum shell or shaft of our standard senders is grounded so touching the side will not hurt anything as long as the ground wire is on the correct lug or terminal. The ground will have a G by it."

I tried swapping the wires into the sender as you said and now the needle has gone all the way over as far as it can and has stayed there, even with the ignition off. If I swap the wires back on the sender it makes no difference, the needle just stays there.

Posted by: john77 Mar 28 2014, 07:30 PM

Okay, now that's all the gauge will do.

If I jump the ground on the back of the gauge to the sender lug it takes the needle back to E. But when I wire it back up correctly with both the wires the correct way round it just shoots up past F as far as it can go and just stays there, even with the ignition switched off.

Posted by: timothy_nd28 Mar 28 2014, 07:56 PM

The problem was with the sender unit. You measure 2.3 ohms between both terminals on the sender (with the sender installed on the tank). 2.3 ohms on this sender means empty tank. You then measured both pins to one of the mounting scews that fastens the sender in place. You reported back that the resistance readings was infinite. This shows that the sender is electrically isolated to ground, which is what I expected.

Again, the sender is installed, reading 2.3 ohms, yet you had 4/5th worth of gas in your tank. This was the issue. The gauge was only reading what the sender said. If you sender was reading 90 ohms, I'm sure your gauge would of read full.

The float on the sender is sitting at the bottom, and not floating. I'm not sure what's going on with that. 4/5th worth of gas should had the float pretty high up, and a reading 80 plus ohms. 2.3 ohms means the float is sitting or hung up at the bottom.

I'm not sure what you just did, but I'm afraid you may have just ruined the gas gauge, which happened to be the only part that was working. Your issues lay with the sender.

Posted by: john77 Mar 28 2014, 09:02 PM

Yep, looks like the gauge is ruined, I pulled the sender and now when I invert it the needle just sits there. As an aside, the float is free moving, and was free moving in the water, so I have no idea why it's not floating in the gas.

Thanks for your time and patience on this Timothy, I appreciate it. I'll speak to the manufacture tomorrow and see what his prognosis on the gauge is.

Posted by: timothy_nd28 Mar 28 2014, 09:24 PM

Since you have the sender out, go ahead and invert it while measuring the resistance on the 2 pins. You should read 90 ohms.

Now, put the sender right side up, you should read 0 ohms. Slowly insert the sender back into the fuel tank. You should see the resistance rise. With it all the way down, you should read 70-80 ohms, if not,,try shaking the sender.

Posted by: stugray Mar 28 2014, 09:45 PM

The website agrees with timothy_nd28 that the sender should read 0 Ohms when empty.

This is exactly the same as grounding the sense wire to the car GND.
So without the sender in the picture at all, what does it read if you hook:
+12V to +12V
GND to GND
Sense to GND

It should read empty.
Unhook the sense wire (open) and it should read Full.

If the gauge does not do that it is broken or the description of the unit from the website is wrong.

Posted by: john77 Mar 28 2014, 10:18 PM

With it fully inverted 89.9, with it right way up 2.6.

Lowering it slowly did nothing, nor did shaking it around, it just stayed at 2.6.

Pushing it in fast it rose to 40 and then dropped back to 2.6.

How does this make any sense when it worked fine in water?

Posted by: john77 Mar 28 2014, 10:31 PM

It does nothing in either of those scenarios. So yes, the gauge is now also broken.

Posted by: timothy_nd28 Mar 29 2014, 04:48 PM

Yah, something isn't right with that sender. If it wasn't so new and possibly under warranty, I would say take it apart. Get on the phone with them, maybe they will take all of it back.

Why are you using an aftermarket sender and gas gauge? Does your old combo gauge not work?

Posted by: john77 Mar 29 2014, 05:08 PM

It's a 6 conversion with a 2.7RS. The guy that built it swapped out the gas gauge for a 911 oil temp/oil pressure gauge.

The sender and gauge aren't new, the set up just wasn't working when I bought the car. So I could take it apart, I'd just need to find a plate to cover the hole in the tank... And probably increase my skill level...

I'll give the guy a call Monday and if my only option is to replace them I'll consider taking them apart.

Posted by: timothy_nd28 Mar 29 2014, 05:21 PM

Consider purchasing a triple or quad gauge, it would look better IMHO. IF the sender is old, there is a possibility that the float somehow got saturated full of gas. This would explain why it doesn't float in gas, but only floats in denser fluids like water. I'm pretty sure the whole car is ruined now, please pull the engine and send it to me

Posted by: bandjoey Mar 29 2014, 05:37 PM

I'm no expert here but. You can probably get a working original float from someone here. Cheap. Why not get one and use it? At least it'll eliminate one part of the problem. Just a thought. Good luck.

Posted by: john77 Mar 29 2014, 05:48 PM

Haha, I have a guy crating the engine as I type.

There's a chance the sender was sitting in gas for a long time - that sounds like a very plausible explanation. Is that something I could possibly take apart and reseal?

Never knew you could get triple/quad gauges, any idea who makes them? It would certainly look cleaner. The only bitch is that the builder screwed the holder for the new gas gauge straight into the front of an otherwise perfect dash. Well, otherwise apart from the hole they then cut in it to run the wires into the back of the gauge cluster. So if I do clean it up with a triple gauge I get to stare at two screw holes...


Edit: I just used this new thing called google and found a couple of places: North Hollywood Speedometer and New Vintage gauges. Not sure how much $$$$ they are, so f anyone has any other recos feel free to chime.

Posted by: john77 Mar 29 2014, 05:50 PM

I'm not sure if they're compatible, I'd have to look into it. Good suggestion if it is though, thanks.

Posted by: timothy_nd28 Mar 29 2014, 08:20 PM

Not compatible. Porsche senders are 0 ohms full and 90 ohms empty

Your aftermarket gauge/sender set is the inverse. 90 ohms full and 0 ohms empty.

Unless you are talking about stealing a float out of the 914 sender to fix this after market sender?

Posted by: timothy_nd28 Mar 30 2014, 05:43 PM

These triple gauges or quad gauges do pop up time from time in the classified section. Price seems to be around 250-400 dollars. You could post a WTB ad in the classified section

or

You could consider buying this kit:
http://www.914world.com/bbs2/index.php?showtopic=205600&st=80

Sounds like you already have the 911 gauge for this project

Posted by: john77 Apr 15 2014, 12:18 AM

Thanks for the link, that triple gauge looks like a great option.

I finally got the correct sized torx bit today and was able take the sender apart. I can't see any bad connections but the circuit board has a layer of crusty dried gunk over it. Any idea what I could use to clean it off without damaging the board? For starters I was thinking of trying 99% alcohol with q-tips.

Posted by: timothy_nd28 Apr 15 2014, 08:36 PM

Cool sender, I never seen one like this. It needs to be cleaned, but I would feel terrible if I gave you bad advice which could destroy it.

If you are serious about the triple gauge, you would need a stock sender.

Posted by: john77 Apr 16 2014, 10:45 PM

Ha, yeah, that occurred to me this afternoon - I'll actually need to find a late 914 gauge and sender.

Probably for the best, this sender is driving me insane. I scraped off some of the crud gently with a razor blade and then soaked it over night. Put it back in the car today and what do you know it worked... kind of. I placed it back in the tank hooked up to my multimeter and it rose to 13.6 ohm and stayed there. Then I hooked it up to the gauge and turned on the ignition and the needle rose up past E for the first time and just stayed there.

That is, until I rebolted the gauge into the tank. And then it drop back past E, and read 2.6 ohm again on my multimeter...