.
.
And here it is in place (photo is somewhat pixelated as I cropped and zoomed in on a photo I found on-line)
,
,
As you noted in the SNG diagram, the trim piece is held in place with 4x rivets and a small screw.
This photo shows the front and back of my trim pieces
There has to be a piece that slides into the groove and has a hole in the center of itself to give the rivet someplace to get purchase. No such piece on the diagram – still searching my boxes, but not sure what I’m looking for??
Here’s another question for the collective.
Caveat: I am not a mechanic and my weakest suit is automotive electronics.
That said, as I understand it, a new fuel sender unit is in my future.
This 25-second Youtube video (no audio) shows that I get an OL reading in several places and it looks to me the ohm reading jumps around on the second half of the test. I ensured the unit was oriented vertically just as it would sit in the tank; I held the unit so the float was just touching the bench (as if the tank was empty) and I slowly pressed down on the entire unit to simulate filling the tank.
New sender unit with float and new gasket is ~$90
OR
Is there a repair/calibration that I can do on the original unit?
(I could not find what “the book” readings were supposed to be at the extremes of the float positioning.)
You are looking for some thin metal buttons. They have a hole in the middle and are held rivetted to the door exterior by #19. The external trim slides over the buttons the button diameter and depth is the same as the channel in the back of the trim.
The fuel sneder is well documented anywhere and everywhere, even on this site and is about 240 ohms to 30 ohms range. It forms a potential divider with the ~62 ohm gauge so the 10v is split between the two.
Hello Craig, here’s my 2 cents worth….having watched your video, to me your sender seems ok. The fluke DMM is possibly a little over sensitive as it displays the output of the sender as the wiper arm moves over the wire wound resistor. Maybe try moving slower to see if the output is more stable or try an old school analogue meter if you have one. I’ve attached a photo of the guts of a smiths sender from a Daimler SP 250 as I don’t have any from the series 3.
This afternoon, after thinking about it for a bit, I decided to spray some stuff into the fuel sender housing. There is what appears to be an adjustable pivot rod that was my target. I considered carb cleaner, brake cleaner, drops of engine oil, some others – worse case, I drop $90 to replace it. After some consideration (the sender sat unused in my Jag for 32+ years) I went with PB Blaster. I figured there was some corrosion build up inside.
I sprayed just as little as I could and then carefully swung the float arm through its entire range of motion. I figure I slowly swept it up and down 30-40 times. After a while, I could feel the sensor bumping over the individual coils of wire inside the unit.
The last thing I did before shutting down for the day was to hook up my multimeter and check the readings through three s-l-o-w up/down passes again. I didn’t note a single “OL” and the ohm read-out numbers were consistent and didn’t jump around as before. I was gonna wait until tomorrow to post this update, but an immediate response to your comments seemed warranted.
Armed with the fact that my Fluke may be too sensitive, I’ll use an old analog Snap-On multimeter I picked up last summer on eBay and test again tomorrow. More to follow.
Since I’m in the test-the-sender mode, is there a way to wire (jumper) the gauge into the system to check movement of my newly rebuilt fuel gauge?
Hello Craig. To test your gauge, a simple test circuit can be used. See attached diagram. Your gauge should have been calibrated using a stabilised 10volts (instrument supply) however to test functionality of the gauge, 12 volts will do, bearing in mind the reading will be inaccurate. Be aware that the gauge will be slow to respond due to heating of the bi-metallic movement. If it does appear to move in the wrong direction, reverse the connections on the back of the gauge. Hope this helps.
O–U-T-S-T-A-N-D-I-N-G!! diagram. Very easy to read. Thank You.
Can I safely assume I add the IVR to the above circuit by placing it between the power source and the gauge - - - my respect for electrons (bordering on abject fear) forces me to ask: is there a right way/wrong way to hook up the IVR (can I fry it it if I hook it up backwards?).
Ooops – it turns out the analog Snap-On meter I picked up last year is a Tach/Dwell meter.
I do have a second digital multimeter (Blue Point). Not as jumpy/twitchy as the Fluke, but yielded the same results: I didn’t note a single “OL” and the ohm read-out numbers were consistent and didn’t jump around as before.
Crisis averted
Dollars saved
Standing by forguidnace on placement / orientation of IVR for second phase test
Hello Craig,
If you’re testing the original VR, you will be jumpy results, because the unit works by switching on and off quickly to give an average between off and on. Sold State units are available from SNGB and others.
Hello Craig. Here’s the MKII circuit. I second Bill’s recommendation to upgrade the IVR to a solid state device using a 7810 regulator chip. Cheap to buy & can be installed in the original can. I’ll post a photo when I can find one. Send me your can & I’ll do the mod….
I have the solid state version sold by Cool Cat – don’t know if it incorporates the 7810 regulator chip??
Maybe @Michael_Frank will chime in with the specs?
No, a 7810 does just one thing…the CoolCat IVR can be tuned to produce more or less voltage, which allows you some ability to adjust the gauge. It can even be used in American dashboards, which use 5V IVRs of the same physical configuration.
That’ll make no dfference - the heating effect on a bimetallic strip is the same whether t receives +ive or -ive or ~ac voltages. What will affect is is if it is at 30ohms or 230ohms sender resistance, so swap the two variable sender connections.
The test setup ought ot to be jumpy with the original IVR as the3 gauge is also a bimetallic type, so will adjust slowly anyway. You need to connect up B to battery positive, I to instruments and earth the casing to the battery negative.
To be honest, testing it at 12v ought to be fine - it’ll just read higher overall. Just don’t leave it connected with zero ohms of load for too long as the gauge will overheat.
Those are the buttons, held into the door skin by pop rivets - and all of it is painted body colour. The diameter is exactly the same as the channel in the back of chrome piece and the button height is exactly the same as the depth of the channel.
Expect yours to be original body colour but perhaps brass colour and only half painted where the paint is scratched off.
Robert
On second look at your photos, it appears the rivets and buttons (4 each) not only are used to secure the exterior trim piece on each door
BUT
it also is used on the top inner surface of ~the door~ (?).
Can you applify the location of the button/rivet in your last photo (you know - the one with the hairy body part in it )
)