Instrument (a/k/a Constant) Voltage Regulator

[TonyC]

Hi, all, and thank you to all those have morally supported me during the rebuild!

With that ordeal behind me for the most part, I have decided to do some smaller dash work. Currently the HVAC surgery is on hold pending some needed parts, but there was something else that prompted my attention. The fuel gauge is acting really wonky; I mean, it has for a long time, but it seems to be acting up more now. When I changed the fuel sender in the tank, the needle changed its reading habits, which I adjusted to over time. But now, with the needle pointing to the near-empty mark, where the old sender would begin to starve off, there are still 8 gallons of fuel left in the tank. How do I know? That was my experience today: A fill-up stopped at the top with just over 16 gallons. So...enter diagnosis of the title of this topic.

Now, I don't really know for sure if the ICVR is to blame for the wonky gauge readings, as it's also supposed to regulate the idiot lights for the oil and temp circuits, and those lights seem to work as they should. But, I felt it would probably be a good idea to explore the possibility of a replacement, as this one is factory-original, never touched. With several designs out there just for Fords, I couldn't be sure if the one I found listed to fit Frankenstein was exactly right, so I went into the dash to yank it and verify its design. Where that is concerned, I have to say that the shop manual seriously oversimplifies the removal and installation. I'll elaborate on what I needed to do to get to it; but first, I wanted to say that I did verify its design and put in an order for a new one. From the look of it, maybe it is due for changing...or maybe not, IDK. Look at the picture below, and tell me this: Is that shiny wet-looking spot in the corner normal for OE ICVRs, or is that a sign of possible trouble?

That question established, allow me to get into the more-involved details of accessing that thing. It is not an easy-peasy, three-step process like the shop manual makes it out to be. What I had to do, after dropping the lower housing, was remove the knobs and retainers to the radio, HVAC control panel, and wiper control, then remove four small screws with 1/4" hex heads that are deep inside recesses under the dash (a long socket was necessary to access them. This freed up the sloping dash panel, which I then removed. Next step was to unscrew and unplug the fuel gauge, which is held in place by two more 1/4" hex-head screws. That allowed a clear view of the ICVR, which was screwed into a metal strap that was in turn screwed into the dash under the speed gauge, by more 1/4" hex-head screws. Another thing I learned during this disassembly is that the radio needs to be dropped in order to improve access to the screws and parts involved, so I did that as well. I discovered that the strap was coming loose, so I hit that screw with a wrench; because of the angle and clearance, getting a socket and ratchet in there is impractical; I used a box-end to tighten it down. I then used the same box-end to undo the screw that held the regulator and its companion suppression condenser from the strap, so I could finally confirm what it looked like so I'd know what to order (no, the shop manual does not give a clear picture of the ICVR, except for a cutaway view of its innards). After confirming how it looks and how it's hooked up, I put everything back together, reversing the steps I took to get it out. Altogether, the job from start to finish took me two hours; maybe now that I know what to do to get to it, the actual replacement job will take less time, an hour and a half, maybe. I'm sharing this for anyone who may want or need to replace their ICVRs on their '66—9 Suicides. Now, keep in mind that this is specifically for the '66 and '67 models, and the '68/9 models may have some difference to them. But I hope this is at least a good reference to help you get an idea of what is involved, which is a bit more than the shop manual will suggest. I'm also including a few more pictures of what it looks like, both in my hand and in its place, so you know what to look for.

---Tony

[action]

IVR only controls gauges not lights. It is a voltage limiter. The key is the letter "I". Instrument.
Fuel, engine temperature and oil pressure gauges as equipped

The "shiny" spot is likely varnish or some sealant. Under that is likely an adjustment. Removing the sealant will tell anyone looking at it someone made an adjustment. Most IVRs are not adjustable. Mostly because most people making an adjustment do not know what to do other than adjust. In addition, they are not that expensive to just replace.

For a fuel gauge, the tank sender modifies the signal to ground by changing resistance further.
Just like other gauges, the sender modifies the electrical path to ground.

Which may be the issue you are having that the ground at the tank sender is not solidly connected.

Action

[TonyC]

So, essentially, the thing is inconsequential. I was actually wondering how it could have any influence on idiot lights.

From what I saw just now, looking at the picture I took a bit closer, maybe there is a provision for an adjustment to the ICVR; I didn't see it as such at first. Or maybe it's a ground issue, IDK...but with the ground wire being part of the three-wire plug that goes into the sender, and which is fully secured in its seating, I consider that rather unlikely, unless the wire needs attention at its other end, wherever that may be. But, if it were a grounding problem, I would think the gauge would stop working, at least intermittently. That has not been the case, only showing a lower reading than the tank actually has, except at all-the-way full. Is it possible that an ICVR can just wear down after 55 years, which included a massive shorting of the whole electrical system?

---Tony

[Dan Szwarc]

The red stuff is likely glyptal, which is an insulating paint. It also helps secure what is likely a very small wire underneath.

[tomo]

Your shop manual should give you a simple test to check out the ICVR. It probably is similar to the 1964 test, where you check the voltage at the input to the fuel and temperature gauges for voltage that varies between 10 and 0 volts. Use an analog meter as it will be more likely to show the fluctuation. You should also see the same voltage at the sending unit yellow wire with a full tank, it will be less with an empty tank.

Your symptoms point more towards the sending unit than the ICVR. The sending unit is a variable resister that could have contact problems between the wiper and the resistance coil or some corrosion on the coil mounting rivets.

[TonyC]

Well, the sending unit is relatively new, replaced about two years ago, so the corrosion idea is not applicable. On the other hand, that is not to say it may not be defective; I have not seen the low-fuel warning light come on at all except during cranking in all that time, and that was why I bought the thing from M2E. So, if it is in fact the sending unit, I'll just live with it for now; it's better to think my supply of fuel is lower than it actually is, anyway, as opposed to vice-versa.

---Tony

[action]

Well, the sending unit is relatively new, replaced about two years ago, so the corrosion idea is not applicable.

---Tony

My comment a few posts up was, "..... the tank sender modifies the signal to ground by changing resistance further."

In a two wire sending unit that has NO low fuel light, one wire to the sending unit is from the gauge on the instrument panel.
The other wire goes to ground at a location near the tank
Corrosion at the point the ground wire connects to the unibody would be an issue. Or any ground path back to the battery.

That could include the sender, and many times it is NOT the sender.

In addition, if the connection to ground is poor, that does not mean the connection point is always poor. Depending on relative humidity the signal path to ground may change. If so the reading of the fuel gauge will chnage.

Action

[TonyC]

Okay, I follow that, makes sense. That is something worth hunting for, I think; I should find that by studying the wiring schematic and poking around the underside.

---Tony

[tomo]

I had to rethink my reply and try to explain it better.

The sending unit is a resistance type that changes the resistance in the circuit to modify the amount of current in the circuit available to move the gauge needle. When the tank is full the sending unit resistance is lowest and the gauge needle moves to the full mark. When the tank is empty the sending unit resistance is at it's highest and the current going to the gauge is at it's lowest causing the needle to move toward empty.

If you have another source of resistance in the circuit it will make the gauge read empty before the float is at the bottom of its travel. This could be corrosion in the sending unit or on the grounding wire.

You said that your sending unit was acting "wonky" ever since you installed it. That leads me to believe that the sending unit is your problem. The reproduction sending units for the Fords and Thunderbirds are not the same quality that Ford used. It could be the arm adjustment or the sending unit resister element. You would have to pull the unit and make sure that you have enough arm travel to go from full to empty and that the resistance measurements are linear.

To adjust the arm you should measure the tank depth and make sure that the float would be near the bottom of the tank at the empty position and at the top of the tank for the full position

[TonyC]

Thanks, Tom. Who knows, maybe the new sending unit does have design or construction issues...but at least it isn't fully inoperative, which would be a major problem. Right now I'm going to go with Action's suggestion, track down the other end of the ground wire and clean it up, as I know it hasn't been touched since the factory. Then, when the new ICVR comes in, I'll swap it in; can't hurt anything, may help. Even if it doesn't change anything, at least it's better to show less fuel than is actually there. I can live with that.

---Tony

[Solid]

I got a bad rebuilt sending unit at one point. Check the resistance on it by slowly actuating the arm with the sender connected to a multimeter set to measure resistance. If you see non-linearity (IE it skips vs being a continuous change) then the sender is bad. Don't assume it is bad if the min and max values aren't exactly what the manual says - there is some wiggle room on either end of the range so that is generally harmless.

If you still have the original wiring harness and voltage regulator then you can take a bit of a hit as the harness in particular has almost certainly increased in resistance since it was built. Depending on the tolerances of the rebuild, "full" or empty or both from the sender might not quite indicate as expected on the gauge. A poor ground at the sender end can also cause all kinds of weirdness. You can add another ground wire as a daisy chain from where the original one is chassis-grounded to another existing screw location in the trunk area to resolve this. Make sure the original eyelet has not been painted over or covered in random gunk so that it makes a good connection to the eyelet for your daisy chain. I ended up pulling new 4 gauge to the trunk while the interior was out and putting in a power and ground distribution block.

I would never replace the original regulator with the same type. Get one of the cheap solid state ones you can find on eBay. That original design is totally obsolete and very definitely does not produce clean or consistent output. There are write-ups from me and others about swapping it in.If you buy the right one you don't even need to change the connectors on the harness. They're cheap as dirt.

[hanko mechanico]

1957 Lincoln Capri. So if the constant voltage regulator seems to be working correctly( volts vary between 2 and 8 at gauge) and I ground the disconnected sender wire which has continuity to fuel gauge and gauge does not move off empty does that definitely determine my gauge is faulty?

[Lee]

Probably. But check and see if there is any resistance across the meter. If that reading is infinite, then the heating wire is broken somewhere inside (and is repairable, I’ve done it).

[hanko mechanico]

Thanks Lee but it will be such a pain removing the fuel gauge I'll probably try to get a new one in case I'm not able to repair my original. It is so uncommon for the gauge to be faulty instead of the sending unit so I want to be certain it is the gauge

[Dan Szwarc]

I always thought that 90 ohms would drive a gage to full and an open would drive it to empty.