Main bus undervolt

Forums › Gadgets, robotics, home automation, electronics (including wearables) › Main bus undervolt

1 | 2

neb

11294 posts

Uber Geek
+1 received by user: 10018

Trusted

Lifetime subscriber

#2292307 8-Aug-2019 14:02

It's a 12V DC-DC UPS for the main router/AP for the house, one of these. I don't know the input voltage requirements, but if they're too low it'll alarm, which is what was happening in this case.

The drop is only 100mV at 150mA load, it gets much higher as higher loads, e.g. the ~2A drawn by the UPS when charging, where it was 0.9V, which was enough for the UPS to suspect an input power problem and alarm. The device being powered runs off 8V - 20V, it has no problems with it, however the UPS monitors the input level more closely because it's supposed to guarantee 12V output.

One of the vendors of the meter, DROK/Droking has this to say about similar meters they sell:

When the system load is bigger, there will be some certain voltage drop on the connecting wire. so the meter will not properly display the output voltage of the power supply system. now the voltage compensation function can solve this problem by moderate compensation of the voltage loss on connecting wire

That's not for their PZEM-031 but is mentioned several times for other meters. Since the connecting wire without the meter inline has close to 0 ohms resistance and there's no measurable drop without the meter present, I suspect the "certain voltage drop" is more likely to be due to the meter.

tripper1000

1648 posts

Uber Geek
+1 received by user: 1176

#2292347 8-Aug-2019 14:31

OK, 0.9v @ 2amps = 0.45 ohms, so there is some inconsistency there which could be indicative of a bad connection or bad shunt resistor. That UPS requires 12 +/- 5% according to the spec here, which is 11.4 to 12.6 volts. Many of my 12v power supplies actually output 12.5v when I measure them.

What is the alleged max current for that meter? You can reduce the voltage drop problem by changing to a meter that has a higher max current - but the volt-drop numbers produced by your one are a bit shady, so it may well be faulty.

Also, if you are only interested in volts, then ditch the amp/power meter and buy a dedicated volt meter (ie volts only meter) - they influence the circuit under test far far less than an amp/power meter which need to put a resistance in line with the supply, and which is what is likely causing the effect you observe.

neb

11294 posts

Uber Geek
+1 received by user: 10018

Trusted

Lifetime subscriber

#2292385 8-Aug-2019 15:33

Alleged max current is 20A using an internal shunt, which I'm not going to try out :-). I wasn't sure how literally the data sheet was to be taken, i.e. did "12V +/- 5%" mean "nominal 12V input" or "the UPS will alarm at exactly 11.399V and below", in hindsight it does appear to be the "11.399V and below" interpretation.

I was interested in both volts and amps, the 12V 4A supply is powering a bunch of devices from the same supply - one good Class VI brick rather than half a dozen random Chinese plugpacks - and I wanted to see when I was getting close to the limit of what the brick could provide. The PZEM-031 in particular (and it's related external-shunt versions) was the most convenient because it has screw terminals rather than oddball n-pin connectors accompanied by semi-indecipherable instructions on how to connect five or more pins up to measure power, this one is just "DC in" and "Load out".

The whole thing was an attempt to clean up the power management and networking for over a dozen devices, consolidate into the smallest number of power supplies, reduce wiring mess, etc.

mclean

584 posts

Ultimate Geek
+1 received by user: 182

Subscriber

#2292442 8-Aug-2019 18:04

Keep in mind that the meter is class 1.0 accuracy, so a reading of 12.05 V means 12.05 +/- 0.12 V. And 11.95 V means 11.95 +/- 0.12 V. So the voltage drop you measured is 100 +/- 240 mV.

1 | 2