Voltmeter Behavior, Verify Accuracy Options

[Superhawk996]

Starting a new thread in response to @emeryGT 350 comment in another thread as follow up

Here’s the background:

[Superhawk996]

I have two gauges, tested both at a variety of voltages

Both do have some “sticktion” in the needle due to damping that results in a minor delay between the time voltage hits the gauge and when the needle moves.

This is more pronounced the lower the voltage. At 8v it is very noticeable. At 11v it is hardly noticeable. At anything 12v or above, I’d say the needle movement is almost instantaneous with initial needle over shoot and then the needle quickly damps out to the indicated voltage.

Both gauges read a bit on the high side vs actual voltage. Approx .5v high. This diminishes as the voltage approaches full scale.

Pictures to follow

[Superhawk996]

Gauge #1

[emerygt350]

Very interesting! Could you do some readings around 13.4-13.7? I was chatting with another teener at the northeast gathering and his gauge was looking quite a bit like mine, reading low when at the battery it was fine. Not sure if his is wired direct like mine, but when we were looking at it, the turn signals and other lights were not on...

I need to pull mine and verify what I see at the battery is what it is seeing at the posts.

[Superhawk996]

Yup will do

Here’s 14v.

[Superhawk996]

[Superhawk996]

Gauge #1 at 13.7v




Gauge #2 at full scale - was basically same indicated display as gauge #1 as I ran up through the voltages.



Will post video of the damping behavior if it is needed.

[Superhawk996]

Gauge #2 12v nominal

[windforfun]

What's the problem? How exactly do you expect your meters to agree? Your meters aren't made by HP. Are they NIST calibrated? Nope.

Wiggle your cables in the sockets to see if there'e better agreement. Corrosion will result in voltage drops. You're splitting hairs here.

(IMG:style_emoticons/default/av-943.gif) (IMG:style_emoticons/default/av-943.gif) (IMG:style_emoticons/default/av-943.gif)

[Superhawk996]

What's the problem? How exactly do you expect your meters to agree? Your meters aren't made by HP. Are they NIST calibrated? Nope.

Wiggle your cables in the sockets to see if there'e better agreement. Corrosion will result in voltage drops. You're splitting hairs here.

(IMG:style_emoticons/default/av-943.gif) (IMG:style_emoticons/default/av-943.gif) (IMG:style_emoticons/default/av-943.gif)

This whole exercise is only providing a sanity check to EmeryGT’s gauge

No issues here at all on my part

[windforfun]

Copy that. The power supply reads out to one place & your DMM reads out to two places.

The DMM is more precise & only more more accurate if it's calibrated.

Precision & accuracy are not the same.

[Superhawk996]

Copy that. The power supply reads out to one place & your DMM reads out to two places.

The DMM is more precise & only more more accurate if it's calibrated.

Precision & accuracy are not the same.

Yup

Former USAF electronics tech. Fully familiar with using a $3000 (in 1980s dollars) HP bench DMM with resolution to the 4th decimal place with fully traceable calibration to measure a stupid 9v battery.

No need for that type of accuracy or precision for automotive hobby work.

Even working on complicated electronics it was very rare to ever need that level of accuracy and precision. Flukes were disposed of like toilet paper (IMG:style_emoticons/default/sad.gif)

The sad joke was that if you were going to break or lose a piece of test equipment, make sure it was something big and expensive like a $12k O-scope or $30k spectrum analyzer. They can withhold a Fluke out of your check via GPLD if they felt so inclined (Government property Lost or Damaged) but when you made $12k a year; they were never going to recoup something expensive.

[emerygt350]

I was interested in the behavior of the voltage meter as much as the variation in accuracy across units.

[mb911]

Good data. Unfortunately mine is right on.

[914Sixer]

60's-70's are what I call ball park gauges. They were designed to be cost effective. Most VDO stuff sold for $25.

[emerygt350]

I think of it more as a dummy light or perhaps a 'general suggestion' about the charging system. However, I have found mine does some disconcerting things that don't allow me to even use it that way. If it's sitting on 12 that scares me, but then I check the voltage and its 13.4. another day it will be at almost 14 and the voltage will be 13.6. That's not helpful. The original lines to the gauge now operate a relay which has a direct (fused) connection to the battery. Hmmm.... I should double check my ground shouldn't i?

[Superhawk996]

I should double check my ground shouldn't i?

Yes! Especially since you’ve previously had issues with the starter & hot starts.

A bad negative battery cable from Batt negative to chassis will cause all kinds of grief as will a bad ground strap from chassis to transmission. The most reliable way to judge the state of these grounds is to measure the voltage drop across the cable while cranking. Visual inspection and appearance is meaningless. Anything more than about 0.2v of voltage drop while cranking is too much - replace the cable / ground strap.

Required reading:
https://www.fenderbender.com/running-a-shop...ircuits-part-11

[Edit] check battery positive cable the same way! Its condition is no less important than the negative cable.


Have you painted your fan shroud or any of the alternator mounting brackets? These are the in the ground path for the alternator.

I’ll get some video of the gauge damping behavior up later tonight. There is nothing in the damping behavior that would explain the variance your mentioning between voltage and/or any sort of extended delay in the voltmeter reporting proper voltage within +/- 0.5 volt.

[GregAmy]

Unless you want to rewire it directly to the battery, the stock voltmeter is just as accurate as the instrument cluster idiot light (binary good-bad).

I removed mine.

[technicalninja]

Thanks, Superhawk!

The dash meters are FAR more accurate than I would have thought...

I once took all of the thermometers in a shop and dropped them into a cooler filled with beer/ice/water.

The old school glass mercury variety were SUPER accurate. 33-34 degrees

Everything else including a $400 Snap-On pyrometer with "special liquid probe" was NOT!

Same shit with compression tester. None showed exactly the same readings on the same car done "lab" style.

I LIKE "lab" style comparison testing!

[emerygt350]

Unless you want to rewire it directly to the battery, the stock voltmeter is just as accurate as the instrument cluster idiot light (binary good-bad).

I removed mine.

That was one of the first things I did. Thus the relay. All new ground strap and no silly paint. Resistance from ground to neg post was all good last year. Gonna poke around with the multimeter and see what I find this year.