More stoopid questions..

[chris.richard]

My fuel guage (Fiat guage, Hawk sender) reads 3/4 full, when there's less than a 1/4 tankful. How can you calibrate them?

My speedo doesn't read zero , (and the car's definitely not moving!). Which bit of it do I tamper with?

TIA

[CorseChris]

Ideally, you need to adjust the resistance at the opposite side of the 'bridge' to the sender, but that's difficult to do without pulling the thing to pieces. An alternative is easily fitted on the back of the gauge using existing terminals, so no serious 'bodgery' required. It goes between the +12v feed and the sender terminal. For the terminally geeky among you, we need to re-balance the Wheatstone bridge that is formed by the internal windings of the gauge itself and the resitance of the sender. You're likely going to need something around 75-100 ohms or so but the only way to get it spot-on is to callibrate it with a resitance decade box. The value is governed by the resistance of the sender. My VDO dip-tube needed a 75 ohm resistor fitting to the gauge to get a good reading top to bottom.

Best bet is to either brim the tank or remove the sender so you can set it to the 'full' position, then wire it all up. Add the extra resistor (1/4 W is fine) starting with a value around 470 ohm and SLOWLY work down in value until you reach a result you are happy with. Don't go below 50 ohms - 75 ohms is probably as low as you want to be, as the lower this resitor, the higher the current passing through the sender at the full posiiton. We don't ever want this current to get anywhere near being significant.

Let me know how you get on.

[chris.richard]

Presumably it's easiest to use a potentiometer rather than a handful of resistors, since I haven't a handy stock?

[CorseChris]

You can for sure, but make sure you use a series resistor as well so you can't inadvertently reduce it to zero - it will likely pop the pot if you do, but it could pop the fuel sender. Not a desirable scenario.......

[CorseChris]

....and on the subject of stoopid, I just read (and digested) what you said was actually the problem, not what I thought you wrote.......so the problem is the reverse of the one I had, so please disregard all I said. Sorry.

When does it read a full tank? About half actual content? What does it read when the tank is physically empty? Is it a dip-tube sender or a lever type, and does it actually reach all the way to the bottom of the tank? And finally, do you happen to know the resistance readings at full & empty? Lots of questions I know, but having just made an ar@e of myself as usual, I should get the facts first ;)

If you can determine these things, I can do you a calibration resistor (& info on where to fit it) that should be near the mark.

[Chris J]

Ooooh, you two!

You're talking about electricity again, like you understand it and making me feel all inadequate.

I don't like electricity, I can't see it.

[chris.richard]

With a few inches of juice in the tank, the resistance of the sender unit is 87 ohms, which sounds in the ballpark you're talking about.
It's a dip tube, (unlike its owner who's a ...I just thought I'd say it before one of you lot does)
I'll see what I can do about the other parameters you mentioned.

[CorseChris]

Chris, that's a good starting point. If you can let me have the 'full' resistance value as well we should be onto a winner.... I'd suggest taking the sender out and turning it upside down if I didn't know just how much of a pain that job can be. Probably easier to turn the car upside down. I have a VDO dip tube sender on my Corse and it needs to come out to put a thicker gasket on. Doh.

...but if this has already been resolved and anyone knows the answer, please step forward and save us the bother....

ChrisJ - nobody understands electricity really. Some of us might have a more or less vague idea about what we see happen sometimes, but until someone can agree on what is really going on, its all a big bluff anyway.

[Chris J]

You had ME fooled, anyway. 'Hope you get to the bottom of this.

[David May]

My Corse VDO tube sensor is 86 Ohms empty (below 5 litres) and 5.3 Ohms full - sounds like it.

[CorseChris]

Thanks Dave - I'll go with those figures...

[CorseChris]

I think I probably need to retract my retraction...... (Ooh errr).

I scribbled the circuit diagram down from memory and I reckon my original fix is valid. Let me double-check for values & veracity when I get home and I'll get back to you with a working fix.

[stableblock]

My dip tube is only about 12 inches long and would zero when the tank has half full. Does anyone know of a longer one that is compatible with the Veglia fuel guage?

[chris.richard]

For the reading of 87 ohms, the very end of the dip tube, less than 1 cm, was in the petrol. With the tank halfish full, I get 142 ohms. At 90% full, I get 110 ohms. Guage is at FSD when half full. Measurements made with wire to guage disconnected. :confused:

[chris.richard]

My dip tube is only about 12 inches long and would zero when the tank has half full. Does anyone know of a longer one that is compatible with the Veglia fuel guage?

The Hawk supplied one is about 18 inches long. See the rest of this thread for compatibility issues!

[CorseChris]

OK, size issues apart, I've got the figures you need to sort your calibration Chris.

130 Ohm resistor, easiest to fit it on the back of the gauge itself, between the Yellow/Black (+12v) and the Salmon (sender)

So, to collate what we know;

60 Ohm VDO sender needs a 75 Ohm correction resistor
90 Ohm VDO sender needs a 130 Ohm correction resistor
Standard sender is around 330 Ohm.

Sender figures are for 'empty'. They should go to zero or thereabouts at full.


....and just maybe, I'll get around to making an adjustable warning light driver that can be used to, well, light the low fuel warning light at a point of your chosing. Might make an article for the newsletter....

[chris.richard]

Lots of Boffin points being won here Chris! ;)

[chris.richard]

I put a 130 ohm resistor in as you said - With fuel 1/2 way up the dip tube, the gauge now reads bang on 4/4, instead of FSD as it did previously. Can your calculator tell me what reistance I'll need to bring it down to 1/2?
Muchas Gracias.

[CorseChris]

Mmmm, that doesn't sound right. How accurate is your resistance meter??? From what you say, I'd suspect you have a 60 Ohm sender like mine.

Try a 75 Ohm resistor instead of the 130 Ohm


...edit 'cos I missed your post with the extra values....


....and they don't make any sense. If you are happy that all the connections are sound & clean, then I think the sender is SNAFUed. I'm always open to correction, but I think the dip tubes come in 2 flavours, 90 Ohm & 60 Ohm when empty, down to zero when full, smooth, linear transition. Yours is behaving like a very strange thing indeed....... It might be time to pull it out and see what's going on. Could be a bad earth, either internally to the sender itself, or sender to loom, could be the sender 'track' itself is worn/damaged.

Worst case, pull it out and send it me, along with the fuel gauge, so I can have a gander at it and get it calibrated.

[chris.richard]

Now, why did I sell my spare sender unit? :mad:

I've had a look at it again, Chris, and I agree it doesn't add up.
Sender unit - my ohmeter says 130 ohms (fuel half way up dip tube), which agrees roughly with the 12V, 100mA readings I get off the wiring.
Gauge - Reads 4/4 whether there is 130 ohm, 75 ohm or no resistor across the pink/black-yellow wires.
I've a spare gauge, so I can try that tomorrow and see if it reads differently. I can see I'll have to pull the sender out of the tank soon. :(

[Sando]

......don't forget the Earth to the tank?
I only have a single wire connected to the sender so it needs an earth from somewhere. If you've rubber mounted the tank this may not help? ............guess how I found out!
Rob
PS I haven't had to add a resister? seems to work fine, Which type of VDO tube sender would this be?

[CorseChris]

All I can say for certain, is what I have measured and seen myself. The VDO dip tube sender I bought from Merlin reads 60 Ohm when empty and 0 Ohm when full (well, to be fair, when turned upside down, so possibly over-full). The Veglia fuel gauges I have tested (I have 3), all expect about 330 Ohm empty and about 10 Ohm full. To calibrate my own sender to the gauge I am using, I had to put a 75 Ohm resistor from +12v to the sender terminal. This has the effect of increasing the current passing through the sender itself, thereby increasing the voltage across it to a point matching what it would have been, had the sender been 330 Ohm....

The gauge is really quite an elegant design IMWO. The windings of the field and moving coils are arranged so that it reads pretty constant from about 8 volts supply to over 16 volts (at which point I decided enough was enough...). As long as the appropriate resistance ratio can be established by adding trim resistors, this stability can be maintained with any sender. Worst case would be a sender of more than 330 Ohm. This would have to be trimmed with a parallel resitor, so while it would read OK at full & empty, points in between would not be linear....but then it's not likely to be a linear reading of actual fuel content to gauge depth anyway, unless the tank shape & sender were designed for one another.

But I doubt anyone much cares about all this anyway......

[chris.richard]

I care! :)
I removed the sender, upside down is <2ohms empty is about the 90 ohms. So far so good.
I've tried my spare gauge and it acts in the same way - 4/4 with no resistor, and moves a millimeter down the scale with the 75 ohm resistor on. But, if I earth the sensor wire, it still reads 4/4. Methinks there might be something amiss with the wiring. Any ideas what?

[CorseChris]

OK - sounds like you have a proper, working, 90 Ohm sender then. You'll need the 130 Ohm trim resistor between Yellow/Black & Salmon coloured wires on the back of the gauge......assuming all else is as Veglia & Fiat intended.

From the other evidence, and what you just say, I'd concur with your diagnosis Doctor.... summat is clearly up with the wiring. I'm not familiar with the Hawk wiring colours so the best thing is to describe what you need to achieve, then you can do it with whatever colours suit your fancy.

Looking at the back of the gauge, bulbs toward the bottom. +12 (Yellow/Black) goes to the gauge in two places - upper left LUCAR terminal and the adjacent middle LUCAR terminal. GND (White/Black) goes to the lower middle LUCAR terminal that is riveted through the plastic low fuel bulb isolator. Sender (Salmon) goes to upper right LUCAR connector. Lower left lamp is illumination (white wire) lower right lamp is low fuel warning (Red wire).

Your gauge should (may??) have a reverse polarity protection diode on it. It's soldered between the GND & +12 terminals. It's not that significant.


If you start at the gauge, make sure it has a good solid 12 volts across the GND & +12 terminals, remove the sender wire, then short the sender terminal to GND - it should go to just over 4/4. With the terminal open, it should read less than zero.

Not knowing the car or it's loom, my first guess would be that it is wired to allow for a low fuel warning light switch and the wrong wire is connected, but there are many, many ways to go wrong so it's far better to assume nothing and start from scratch. All you need to find is the one wire that ends up going between sender terminal on the gauge to the sender terminal on the sender itself.

[CorseChris]

Oh, forgot to say, I made up a wee circuit last night to drive the low fuel warning lamp with the sender voltage. Small adjustment range centred around the area where the needle points at the lamp. Put some damping and hysterisis in it so it shouldn't flash on & off too much when driving along with low-ish fuel (unlike my Alfa....)

I can make up a batch if anyone is interested.

[chris.richard]

The description of the connections is useful. I remember being puzzled that the markings on the case didn't seem to tally with the wires. I've a feeling the ground and +12v wires might be reversed, although there is a diode in there. Hopefully this weekend will sort it.

[CorseChris]

The markings on the one I'm looking at are almost unreadable, but it sure doesn't help that there is a 12v mark right next to the sender terminal, while it's correct lable S (for Salmon) is somewhat removed....

The colour markings are GN (Giallo/Nero) for +12, S (Salmon) for the sender, BN (Bianco/Nero) for the GND. Can't make out any others apart from a very faint 'Veglia'

If the 12v & GND have been reversed, it will almost certainly have seen off the diode - it's only a small one. Not even sure why it's there to be honest, unless the 12v supply normally has a very small fuse in it.

[David May]

It's holiday month here in Italy, so there's nobody at the clock factory to ask the reason for the diode, although I rather doubt that there is anybody left from the 1970s project groups anyway! The last 'Veglia' crossed-coil movements to have a PCB (still in use in the mid 1990s) only mounted the occasional trimming resistor. (The current version - still in use on at least one performance car from this area - has a single resistor spot-welded to lugs.)
I'm not sure if all Fiat 124 variants used the same instrument calibration - there were dozens of combinations of coil and trimming resistance in production over the years.

[chris.richard]

.......a faulty earth to the gauge. Probably the commonest auto electrical fault. The 127 ohm resistor is about right too. Thanks Chris, another triumph for the Stratossupersite :)

[CorseChris]

Well done Chris. My motto applies then - look for the simple stuff first.

Dave. Interesting background info. A later vintage oil pressure gauge I found kicking around yesterday also has the diode, but it's directly in the line for the warning lamp. Still can't think why it's there other than maybe to reduce the brightness a wee bit.

[David May]

Chris,
I can't think of any other reason to have diodes around, especially when makers are notorius for not spending on anything not 100% essential, and especially not relatively unreliable semiconductors.

[stableblock]

Hi
I don't have any first hand knowledge of this but a reverse connected diode will prevent high voltages should the supply to the coil in the instrument be interupted briefly e.g. by a poor sliding contact. You can try this for your self using a battery and you may well get a nasty shock with no diode.

The purpose might then be to prevent a spark at the point where the circuit is diconnected - if this was due to poor sliding contact on the fuel level sender this spark would gradually make this worse. Sparks in fuel tanks aren't a good idea anyway! although in practice there may not be enough air to cause an explosion! Depends on what the Upper Explosive Limit for petrol is!!

Having a reverse connected Diode across the meter sounds a good idea to me.

Hot wire instruments don't suffer from the same problems but don't have the same linearity in their scales / calibration either.

Peter

[David May]

I don't think that a small instrument (which has very little in the way of an iron core) poses much of a spark risk, although now you mention it, the diodes may well have been there from the days when the instruments had their own supply. The 'regulator' was nothing more than a hot bimetallic strip, which crudely opened and closed a contact to give an average voltage around 8V. This continuous on/off switching may well have caused interference and probably contact arcing, which would have been suppressed with diodes.
The lack of regulation on more modern designs may well have made them redundant.

[stableblock]

The bimetal regulators were used with hot wire instruments rather than moving coil ones. I'm pretty sure that the veglia instruments are moving coil - no warm up time!
I've certainly had a nasty belt in the past from a moving coil fuel guage - I guess the contact set up would determine whether it would be enough to spark?

[CorseChris]

The matching VDO gauge to the dip tube senders is a moving coil type, but doesn't have a clamping diode. I'd guess the 'Q' & inductance of the coils in a meter would be so low that a spark isn't likely, but if you've had a belt, then clearly enough for a few more than 12volts.... Current is low enough so that arcing isn't an issue.

Interesting question though...