Every once in a while folks will bring up questions about the instrument cluster voltage regulator (ICVR), what it does, and how to test it. I figured some folks might be interested in some of the details behind this little gadget in order to help dispel some of the mysteries behind it.

What the ICVR does.

The ICVR is a small metal can on the back of the instrument cluster (picture shown below).



The purpose of the ICVR is to provide a stable and suitable voltage supply to the truck's instrumentation (temperature, oil pressure, and fuel gauges). The ammeter (if equipped) is not part of this circuit. The ICVR is necessary for two reasons:

1) The gauges cannot withstand the full 12 volts (and higher) from the alternator. The ICVR reduces the voltage applied to the gauges to approximately 5 volts, which the gauges are designed to handle.

2) The alternator output is not tightly regulated enough to provide a stable gauge reading. Even if the gauges could withstand the 12 volts used in the rest of the vehicle, this 12 volts could fluctuate enough to cause movement in the needles of the gauges. Imagine how inconvenient it would be if your fuel gauge dropped toward 'E' as you approached a stop light, only to pick back up again as you accelerated!

The ICVR is not powered directly by 12 volts. Instead, it is connected to the 12-volt accessory bus through an 8.5-ohm resistor wire, bundled in the dash harness. This serves as a current-limiting resistor to protect the ICVR if the output is shorted to ground; it also helps isolate the regulator switching current from the electrical system upstream. If you look at where this resistor wire enters the cluster connector, it actually shares a crimp terminal with a short "dummy" wire that's taped off at the other end - this is because the resistor wire conductor is too thin to be crimped on its own.

How the ICVR works.

Many people say you can check the ICVR by checking for 5 volts at the output with a multimeter. Nothing could be further from the truth! The ICVR is an electromechanical regulator, one of the first kinds of switching regulators. It works by switching its output between 12 volts and zero back and forth. Here's the output of the ICVR, as seen on an oscilloscope:



As you can see, the ICVR output is a square wave. The frequency is constant, but the ICVR adjusts the "width" of the pulses (called a "duty cycle") based on load to control the "effective" voltage of the output - also called the RMS level. The RMS level of the ICVR output is around 5 volts, but you won't see this on a multimeter! The ICVR output moves back and forth at a rate of about 1 Hz - too slow for a handheld meter to resolve into a meaningful number. In fact, the only sure-fire way to measure the ICVR output is with an oscilloscope, as I have done here - this tool is capable of calculating the RMS level of a very slow-moving waveform over a period of time. If you look closely, the RMS level of the ICVR I am measuring is 5.584 V - right on the money!

What else?

As I mentioned earlier, the ICVR controls the voltage at the output by adjusting the widths of the pulses shown in the picture above. For example, if the current demand of all the gauges downstream were to increase, the ICVR would need to make the pulses "wider" to keep up with demand. This is a concept called pulse width modulation, and is actually the same principle behind the high-power speaker amplifiers many of you guys have sitting behind your back seat. By the way, if you're curious, the little **** on the ICVR adjusts the "default" pulse width, and therefore the output voltage. It gets calibrated at the factory.

The next question is, how can I actually test the ICVR if I don't have an oscilloscope and a multimeter won't really work? For starters, you can check that the ICVR is actually switching by watching the output with a test light. Because the ICVR switches so slowly (about once per second), you can easily see the blinking action. Another method, which is recommended by some shop manuals, is to indirectly measure the ICVR's accuracy by disconnecting the fuel sending unit and simulating an "empty" and "full" tank with resistors, and checking that the fuel gauge needle lines up accordingly. The problem with this method is that you're also testing everything else in the path, so a problem elsewhere can place false blame on the ICVR.

Conclusions...

Not much else to say, but hopefully this was useful information for those interested. At the very least, hopefully it gives folks some context that will help next time they need to troubleshoot a fuel gauge problem

 

FMC, thanks for posting this. I've been meaning to ask some questions about the ICVR and you've answered most of them here.

I do have one question on troubleshooting, though. I'm having trouble with my Crew since it's repaint. I'm getting nothing on my temp or fuel gauges. It has lights for oil and alternator. I'm suspecting my ICVR isn't working. If I check the wire to either the oil sender or fuel sender, my test light should light up, right? If not, I've got a fault between the sending unit and the cluster or the ICVR isn't working, right? I'm suspecting the ICVR since neither has worked since I put everything back together, but everything else in the cluster is working.

 

Thanks for the feedback!

The test light should blink about once per second, similar to if you were to probe the output of the ICVR directly.

You got it.

 

D1AZ-10804-A .. ICVR ~ Instrument Cluster Voltage Regulator (Motorcraft GR-513)

1971/79 F100/350, Bronco & Econoline / Misc 1971/79 Passenger Cars.

 

FMC, excellent write-up, but why do you say a multimeter "won't really work", granted it will never be as accurate as an oscilloscope but there's enough needle deflection (assuming an analog VM) to determine if the ICVR is "working" or not, how is this method any less accurate, or conclusive, than using a simple test light?

 

Thanks for the write up .. Really appreciate it

 

Thanks for the feedback.

I should clarify. A multimeter will not be able to give you a reading of the effective RMS level of the ICVR's 5-volt output. However, a fast analog multimeter can at least show you the "ticks" you'd see with a test light. To answer your question directly, a fast analog multimeter is no less conclusive than a test light.

In other words, both an analog multimeter or a test light can verify that the ICVR is at least functioning. However, only an oscilloscope has enough memory depth to properly calculate the RMS level of the output to give you a meaningful number.

Digital multimeters that are not true RMS cannot properly measure waveforms that are not sinusoidal. However, even meters that are true RMS cannot measure the ICVR because the waveform switches too slowly.

 

Thanks for the clarification, that explains why I can never get a reading from an ICVR using a digital VM, but yet my 20yr old analog VM can at least register a "pulse" from a working ICVR.

 

FMC,

Great info as usual! Thanks.

Mark

 

Great write-up. Reps sent.

 

Some further reading for those who are interested. I picked up a replacement ICVR from LMC (pictured below). This unit is interesting because it is a modernized retrofit. It appears to be fully electronic (as opposed to the OEM electromechanical version).

The output of this regulator is a solid 5 volts DC. There are no "ticks," and the output will appear as a solid line across an oscilloscope. As such, this style will register 5 volts DC on any common multimeter. You'll also notice there is no trimming ****. That's because modern voltage regulator integrated circuits are capable of generating relatively precise reference voltages that are used to control the regulator output.

 

Interesting note on the aftermarket ICVR's, does it register 5v on both digital and analog meters?

Would this replacement ICVR be akin to comparing an electromechanical voltage regulator to a solid state voltage regulator? as far as the updated electronics are concerned?

 

That's correct on both questions.

 

Don't buy Dennis Carpenter parts. They are cheap crap from China. I took the time to pull my rear gas tank and install a new sending unit from Dennis Carpenter. What a pain. Then within 4 months the sending unit starting giving false readings and then just quit. I will never buy from them again. Junk.

 

You need to send your old one into a shop that restores them. You can find those in a Hemmings motor news in the vendor section.