Gauges eating up volts
Posted by jedigreg1984@reddit | projectcar | View on Reddit | 25 comments
Exhausted and probably missing something obvious, please help me out
Brand new gauge panel is in a 57 Ranchero that I'm preparing to break-in - voltmeter, oil temp, oil pressure, and water temp, all Stewart Warner. In the picture of the back of the panel, that's the order from left to right. When I turn on the ignition (not cranking), all gauges move a bit, and the voltmeter jumps to 11V and then settles at about 8 or 9V, when battery is at 12.6V. My Torino voltmeter reads whatever the battery is, engine on or off, but it's an Autometer part. Something is off
Info/Tests:
Gauge power and ground feeds both connect to the voltmeter first, then it chains out from there.
Removed tach ground from common gauge ground, no change.
Power feed has battery voltage (12.5V) and <0.05 Ohm resistance from the feed to the voltmeter '+' post. Ground from the voltmeter '-' post to chassis also has <0.05 Ohm resistance, so I don't think my harness/crimping work is to blame.
Voltage drop across the voltmeter posts reads 8.5V with it all connected, and isolating the voltmeter gauge from rest of the chain results in a good reading on the gauge of 12-12.5V, so the gauge is ok.
Powering the rest of the chain without the voltmeter results in the voltage drop across the other gauge posts to read about 9V - so I'm losing about 1V per gauge hooked up in parallel in the chain...
Resistance across the gauge posts are all about 380-405 Ohms all disconnected from the harness, that seems fine to me... Right?
What's going on and how can I fix this? Shouldn't I have 12.5V between each gauge post and a good known ground? Something about the way they are hooked up is affecting the voltage getting to the voltmeter, and the other gauges as well... Again, I'm hoping that the issue is obvious and I just have a blind spot for it right now.
Each gauge connected individually is pulling about 0.5-1V across the posts. It seems that I'm losing a lot of voltage between the ring terminals and the posts themselves, does that make sense? I'm using toothed washers and getting them appropriately tight for 8-32 thread. I'm going to run a totally new ground to the battery or block and see what happens
OkConsideration8605@reddit
Voltage drop is the fastest way to find the problem. With long leads on the meter. I would check the ground from the actual post of the battery to the post on the voltmeter. Then the actual post on the positive to the post on the meter. When you determine that one has the voltage loss then split it in half as close as possible and recheck, if you check from battery pussy to ignition switch feed and it is good, then check from ignition switch feed, to post on meter. That will tell you exactly which direction to go, keep splitting the bad side. Till you find the joint where the loss is.
BlasphemousBunny@reddit
Do you measure that same voltage drop happen at the voltmeter terminals with your multimeter? How about measuring at the battery with all of the gauges connected?
My best guess would be you just have a bad ground or power connection somewhere. All those gauges at 400R each in parallel is only 100R equivalent so in the ballpark of 120mA which would only take an additional 25 Ohms of contact resistance to drop 3V. But you should be able to use your multimeter to figure out where that voltage is going. Measure voltage between voltmeter + terminal and battery +, and measure voltmeter - to the actual negative terminal of the battery. You’d expect those to read 0V, but I suspect that might not be the case.
jedigreg1984@reddit (OP)
Yes, same reading on gauge and multimeter. With the panel connected and everything powered, battery still reads ~12.5V, but I have not checked battery+ to voltmeter+, and both negatives
Contact resistance is very very low which is the thing driving me crazy, more or less. I suspect my known good ground is only good in regards to the dash metal
BlasphemousBunny@reddit
Okay thats helpful to know. If the battery isn’t drooping, then your battery is good. My guess then is that there is a bit of resistance between your dash metal and the battery ground. Could be bolts, paint, corroded battery cables, all sorts of things. 25 ohms is low enough that everything can visually look perfect, but it’s still high enough to cause weird issues. Multimeter is your friend. Follow the path between your gauges and the battery and you eventually find where your losses are coming from.
jedigreg1984@reddit (OP)
Voltage drop on the multimeter from the NEG battery post to the dash ground is only 0.003V. Drop the from the POS battery post to the gauge feed is only 0.01V. Resistance for both paths is about 4 Ohms.
BlasphemousBunny@reddit
Man that’s just not making much sense… that voltage has to be going somewhere. how hard is it to remove the gauge(s)? I’d be curious what happens if you connected the voltmeter directly to the battery just as a sanity check. You could also bring a second battery, jump pack, or a power supply into the cabin if that’s easier than removing the gauges. Just make sure you aren’t creating voltage contention between the two power sources.
jedigreg1984@reddit (OP)
Literally just got out the Noco charger/supply
Throwing ~13.5V into the battery cables, voltmeter reads higher, about ~9V, as expected. Put the power supply directly to the gauge pack and they all snapped to life correctly, voltmeter actually read ~0.7V high, which is fine, it's not a precision device.
Absolutely a high resistance issue in the ignition switch, grounds, etc. anything that I haven't already replaced. I'm not really sure how the ground path showed such low voltage on the multimeter - perhaps it just has to do with the amount of current the multimeter uses. I suspect the battery is also in need of a repair cycle. Multiple issues then.
jedigreg1984@reddit (OP)
Going to start the diagnosis from scratch tomorrow, I'm usually very good with the multimeter but I also did make some assumptions that might have been optimistic
frogsRfriends@reddit
Honestly I would just throw like 5 grounds between dash, body engine etc. never hurts to over ground
jedigreg1984@reddit (OP)
True
AcanthocephalaAny78@reddit
When wiring in a sensor if you’re unsure of the connections there should be a dedicated 12v power and the negatives should not be mounted inside the car but rather onto the firewall to a bus bar and then the bus bar is also ran directly to the battery. It eliminates all electrical noise from bodywork grounds and ensures correct connection 100% of the time. I had to do it in my 24v cummins when the grounds were all deciding to make the throttle go wonkus
turfdraagster@reddit
Run a wired ground from your panel so you don't depend on the dash connections
jedigreg1984@reddit (OP)
What about the measured voltage drop changing based on how many gauges were hooked up? Is that barking up the wrong tree or evidence of another issue?
BlasphemousBunny@reddit
No I think that is exactly the behavior I would expect from a weird “high” resistance in your path. More gauges = more current = more voltage drop.
If one gauge is 400 ohms, V = I * R => I = V/R so I = 12.8V/400R = 32mA
Resistance in parallel add such that R1 // R2 (R1 in parallel with R2) = 1 / [(1/R1) + (1/R2)]
4 gauges in parallel would be approximately equal to 100 ohms, and 12.8V / 100R = 120mA
If there is an extra 25 ohms somewhere, at 32mA, that would mean a voltage drop of 25R * 0.032A = 0.8V. But with 4 gauges connected, that’s pulling closer to 120mA and 25R * 0.120A = a 3V drop.
Does that make sense?
jedigreg1984@reddit (OP)
Absolutely, yes. Wasn't even thinking there would be on the order of 25 Ohms somewhere so it didn't occur to me immediately that it would behave like this. Thank you for the breakdown
hydrogen18@reddit
put a current meter in the red lead while the car is on and see how much current those gauges are drawing. one could easily have a short
jedigreg1984@reddit (OP)
It's possible. And likely that I have multiple issues with the old harness, battery, and this
ruddy3499@reddit
I had similar issues, more stereo than gauges and body ground fixed a lot of it
cheeseshcripes@reddit
You may want to remove those star nuts for now, I can't tell from the pic but they should be between the nut and the ring terminal, not between the ring and the gauge.
Maybe remove the device wires from the gauges and go from there. You could have a short or an open ground in the devices that is drawing down your power.
jedigreg1984@reddit (OP)
Hard to see in the pic but I have internal-tooth washers that are between the ring and the nut on all connections, making good marks on tht rings and nuts. Those external-tooth ones are on the ground side letting the ring contact the gauge body, the gauges came with them and I left them there
I hear you about the signal wires though! I'll check them.
ThanksALotBud@reddit
Is it wired through a relay?
jedigreg1984@reddit (OP)
Nope, this section is just from the ignition, but the switch is also a suspect.
Starting at the battery tomorrow and will figure it out. Thought I could get away with it! Not this time
seamus_mc@reddit
How fresh is the battery? Is there corrosion on your grounds? How about positive terminals? Old batteries may show volts but cant push any amps.
jedigreg1984@reddit (OP)
Battery is good and was on the charger for a bit before these tests. I'll confirm age. No corrosion or damage... And gauges are drawing very little amperage. But yes I'll check everything
seamus_mc@reddit
Charger doesn’t mean anything if the battery is bad. Can you check the voltage while cranking?