Author Topic: Wayward DC currents  (Read 1109 times)

0 Members and 1 Guest are viewing this topic.

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Wayward DC currents
« on: May 02, 2023, 10:26:57 am »
I keep doing these things, getting bizarre readings and then failing to explain it properly.  I have another problem.

I have a single source of 24V DC which is running 2 DC/DC bench PSUs.

One is set to 13V and is floating a lead acid.  The lead acid is powering the network switch and the internet/wifi router (and 2 USB hubs).

The other is set to 19.50V and is running the work laptop docking station.

The trouble is, the currents are very suspicious.  The lead acid's loads have been pulling pretty much exactly 1.3 Amps for the past 2 weeks.  This morning the current out of the battery fell to 1.0 Amps.

The PSU powering the laptop started out at 4.5 Amps, topping up the laptop battery and fallen to around 1 Amp.   Which is "about right", 20 Watts or so.

So where did the other 300mA go?

As a "finger in the air" I connected the multimeter to the "-" pole of the lead acid and via the "Amp" range to the copper radiator pipe beside it.  Hmmm.  0.5 Amps, falling over a few seconds to.... 300mA.

At the weekend I tried to power a monitor from the PSU which is powering the work laptop and I ended up disconnecting it pretty quickly.

The monitor came on fine.  Worked perfectly, but when I look at the PSU running it it said "0.00A" and the other PSU was showing 2 Amps more.  I could even unplug the "-" from the PSU and the monitor continued to run.

So 2 amps which was not returning to the PSU that made it.  The only other path from the monitor to anywhere was via the HDMI and DP cables, from there t the USB hub and back down the "-" rail via the battery and back untimately to the 24VDC that provided.  I didn't like the sounds of 2 Amps on the lowside of 19.50V "leaking" that way, so I shut it down.

This mornings oddity is different.  Those 300mA appear to being "added" to the circuit, not re-routed.
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Wayward DC currents
« Reply #1 on: May 02, 2023, 11:36:29 am »
Could you please draw and share a wiring diagram first? You have described your setup pretty clearly -- but it is still easy for readers to make implicit assumptions and maybe misinterpret your description. And drawing all sources, loads and connections might also help you understand what is going on.
 

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Re: Wayward DC currents
« Reply #2 on: May 02, 2023, 12:21:05 pm »
This is not a schematic, I just used a scematic editor to diagram it.  I added the "para-paths" as un-connected lines, they are just suspicions.  (The DCDC converter outputs are swapped, please ignore that :))

The path out of the laptop docking station I have confirmed is the HDMI cable.  When removed the 300mA appears coming out of the battery again.

In the case of the monitor, the only change was replacng the laptop dockign station with the monitor and it showed no current and everything coming from (or returning via the 13V PSU)


EDIT:  This might show how I can lose 300mA, sort off via the other PSU, which will be making both PSU's current readings completely b*ll*cks. 

It's still very curious as to how can I also produce a return path for that current from the local earth!
« Last Edit: May 02, 2023, 12:24:34 pm by paulca »
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Re: Wayward DC currents
« Reply #3 on: May 02, 2023, 01:24:09 pm »
I remembered I have a current clamp meter.  So I was able to bunch all 4 wires from the PSUs into the clamp and there was a stray 20mA (neglibible, noise or error margin).

The lower voltage circuit appears to be providing/sourcing 0.5A to the higher voltage circuit.  One shows -0.5A total current and the other shows +0.5A total current.

I don't think I need to be that concerned with 0.5A over HDMI etc, all of the cables/interfaces involved support about that current or more for power transfer.

I suppose the only way to stop this is use isolated DC/DC converters?
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 8276
  • Country: de
  • A qualified hobbyist ;)
Re: Wayward DC currents
« Reply #4 on: May 02, 2023, 01:40:46 pm »
Many possibilities: shielded ethernet cables, PCs (grounded via mains PE), audio cables and all the other stuff.

I have a somewhat similiar setup with a central DC power supply for powering small devices and have grounded the negative output to PE to minimize ground currents (which create noise) through the tons of cable between all devices.
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: Wayward DC currents
« Reply #5 on: May 02, 2023, 03:20:25 pm »
I suppose the only way to stop this is use isolated DC/DC converters?

Hmm, I would have thought that you want a common ground after the DC/DC converters. Are their respective negative outputs connected?

(The polarity of the left DC/DC output seems wrong, but that's most likely just a glitch in the drawing, right?)
 

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Re: Wayward DC currents
« Reply #6 on: May 02, 2023, 03:36:11 pm »
Yes, drawing glitch.

The voltages at all points check normal.  13.00V out of the PSU with the battery, 12.99V across it, 19.50V going out to the laptop, just un-matched but mostly symmetrical current.

Voltage between them checks out too.  0V between their negative posts.  6.50V between their positive.  Checks.

I suppose it is just alt paths.  The only concern is the route they end up taking involving monitor cables and USB hubs.

Is it any wonder my audio sounds like I'm in a room with a bunch of dot matrix printers in the background?

The only concern of course is that neither current limit function is fully aware of what is really going on.  I also have concerns about one of the PSUs suddenly disconnecting and the 19.50V managing to also migrate back up stream.

I have a diode across one of the PSUs, the other fell off ages ago, not sure they would help with the later concern though.
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Re: Wayward DC currents
« Reply #7 on: May 02, 2023, 03:39:52 pm »
Also, it points a finger at one set of cables as having a high enough "-" impedence that the current has felt it necessary and easier to travel so far.

As I also got a resistance tester today I might be able to check that.
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline paulcaTopic starter

  • Super Contributor
  • ***
  • Posts: 4364
  • Country: gb
Re: Wayward DC currents
« Reply #8 on: May 04, 2023, 09:36:29 am »
So I am re-doing the whole lot to use an inverter to bring 240V up to the office where I can convert to DC and power the same DC loads at about half the efficiency.  Granted it means I can far more easily power those devices which aren't easy to surrogate a power supply for.

In the process of doing this, the only direct DC loads on the system where reduced to the "Admin" rail on the solar power system itself.  A Raspberry PI, two ESP32s and a 2.4Ghz travel wifi router.  4 watts.

What does the charge controller who load output I am using see?  0W.  0.0A.  Why?  The DC current is returning via the RaspberryPI, the USB interface to the inverter and back to the battery and not via the "Load shunt" in the charge controller.

Accepting this is just part of the parcel with DC it does raise concerns with making "DC power buses" in the first place.

Consider a use case I was considering.

Battery -> Inverter -> DC Brick -> 48VDC -> Fan-out DC power with appropriate boost/buck to power 5V, 12V, 20V DC power outlets around the office.

The 48VDC 10A power brick won't even be impressed with the 100W load or so.  It will barely notice the 2A of current and it's fan might hit 20% duty.

Let's say I know turn on the DC/DC bench PSU to work on a micro-controller project.  When I connect the MCU to the USB stuff a DC current path is formed and some much larger current decides to source itself through that USB connector and via my little weedy 100W bench supply.  It might appear my MCU is pulling 2 Amp!  Not least will the current limiter go nuts and be completely unable to stop the 2.0A, but in some cases even switching the PSU off by it's single pole switch results in it remaining on and displaying voltage AND current.

It would suggest that putting the AC/DC conversions close to the devices might have advantages as those DC sources are usually "decoupled" or at least partially isolated.  Device to device currents probably can still exist, but there is a much more limited set of pathways it can take.

Have I just stepped down a rabbit hole?  Is it worth anything other than frustration going further?

EDIT:
"Load shunts".  Low side "load shunts".  Basically this comes down to who has the smallest load shunt R value or better yet, as a DC current I would be looking for the device which has NO shunts at all!
« Last Edit: May 04, 2023, 09:38:17 am by paulca »
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29812
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Wayward DC currents
« Reply #9 on: May 06, 2023, 02:43:09 am »
I suppose it is just alt paths. 
Certainly but at any time did you measure with a LoZ DMM ?
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf