Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Raspberry Pi Grounding


Recommended Posts

Straight to the point - I've a RPi 4 4GB running astroberry successfully controlling mount, camera & guiding via KStars & EKOS.
USB connection directly to newer AZEQ6, USB C to guide-cam and USB to camera.
I've noted that touching the mount, camera and guide camera feels like it's electrified. Not high voltage, but more like continuous static. Very annoying when polar aligning and... zap!
I've the topic posted in RPi forums, and general consensus is it'll never be the pi, always a differential voltage (well, yes, that is what voltage is!) between components.
If I was to earth (GND pin on the RPi, or even taking blank USB and cutting into the black wire) to the mount and guidecam, would this eliminate the issue?
It appears to be on the RPi case (Argon ONE V2 Raspberry Pi 4 Case) without any connections, and just gets worse when things are connected.
Note the issue is apparent when imaging and not (whether connected to mains at home, outside whilst imaging, or work).
Anyone any thoughts or potential solutions like the grounding everything idea?

Link to comment
Share on other sites

its most likely that its you that is building up the charge as you flex and move within your clothes. I note that you suggest it is "more like continuous static" but I am guessing that once you have touched and discharged you are ok until you touch it again after a few minutes.

If the tickle is all the time when touched then its a mains leakage issue which needs immediate investigation.

If its static build up then one of the antistatic wrist bands worn by electronics assembler's, with one end of the curly cable grounded on the scope mount should kill off the static.

Ebay

 
113989443847
 
Edited by Tomatobro
spelling
  • Like 1
Link to comment
Share on other sites

Sounds like your metalwork is not physically earthed by your mains power supply and so is floating wherever the various capacitances take it. I don’t have this with my RPi, but for instance I do have it from time to time with my iPad charger which is connected only to live and neutral with no earth connection. The Apple mains chargers here are two pin types, and they plug in either way round. One way round I feel the iPad metalwork ‘buzz’, the other way round, not.

Link to comment
Share on other sites

Must admit I have never had this and I use a Rpi and an Argon metal case.
Also never heard of anyone earthing everything, and apart from being a bit of a pain as the idea is to reduce cables not add them I am not sure it would change anything.

The explanation offered by @Tomatobro is a real probablility.

If you have narrowed it down to the Rpi is that because you do not get this static feeling with the Rpi power supply disconnected ?

A real obscure thing but the Argon One has a fan inside and wondered if this could be rubbing on the fan casing as it rotates and builds up some static - you can either disconnect the fan or turn it off with a command in the Rpi if the Argon One software is installed. Like I said a bit obscure but try turning the fan off.

Steve

 

Link to comment
Share on other sites

Are you using a switchmode mains power supply? If so, the output of those does "float" at about half mains voltage (so 120V ish for UK mains) and it can be enough for a slight tingle. I'd be tempted to earth it with a ground spike to some convenient metalwork (e.g. on the mount), using a tent-peg or similar in the ground as the ground spike. That ought to kill the tingle. Worth noting that switchmode supplies have caused the death of many a delicate input when connected up "hot". Be careful to connect everything up with it unpowered and apply power last.

  • Like 3
Link to comment
Share on other sites

I used to get tingles from my EQ6 Pro when powered via a generic mains power brick. I also put that down to some of the problems I was having with stabillity and usability of my Ras Pi and Astroberry.

I abandoned the Ras Pi a while ago and upgraded to a mini PC and a good quality power supply with 12v DC distribution. 

No tingles since dumping the original power brick. Although floaty filter voltages might have been bothering the Pi I'm certain the Pi wasn't the source. 

Edited by Paul M
Link to comment
Share on other sites

Thanks guys.
@teoria_del_big_bangthinking it was something to do with the case, I unplugged the extra bit (taking the RPi mini to full HDMI), and ended up seeing the same, and so decided to remove the pi altogether - so just power cable (genuine RPi-4 UK plug), RPi4-4GB and attach the guide cam - same thing, feeling like the constant buzzing. Unfortunately that rules out the fan which had also been thought!

@Tomatobro, 'tickle' is constant, not just once off, when touching the setup. Worst thing is during polar alignment, when nose gets a little close to the tripod or just moving things around holding the legs and zap! Is there an easy way to measure potential leakage? I've tried with multimeter connecting to GND pin on RPi and probing around, but nothing standing out to worrying voltages.

@teoria_del_big_bangAgreed about minimising cables, whole intention of having the RPi mounted on a dovetail and only 'power to it and the mount' cables from the ground!

@AvocetteI've seen same thing with MBP when using in the magnetic-easy-port, although can't say I noted it disappearing if connector was turned around.

 

  • Thanks 1
Link to comment
Share on other sites

@wulfrun not sure about what that is. Issue is reproducable in work when connecting RPi straight to ring main or at home using several different sockets, main observing done via outside plug.

@Paul MPower supply to RPi is the genuine one, and when using mount, the nevada power supply available from FLO, although again the effect is noticable on ONLY the RPi.

Grrr!

Link to comment
Share on other sites

14 minutes ago, pete_81 said:

@wulfrun not sure about what that is. Issue is reproducable in work when connecting RPi straight to ring main or at home using several different sockets, main observing done via outside plug.

@Paul MPower supply to RPi is the genuine one, and when using mount, the nevada power supply available from FLO, although again the effect is noticable on ONLY the RPi.

Grrr!

could well be a faulty RPI power supply albeit a genuine one try changing it for a phone charger and try

Link to comment
Share on other sites

Just now, pete_81 said:

@fozzybear, unfortunately got 2,one for this and another for pi400, both same issue so doubtful it's that🤔

Ok worth a try, I am racking my brain as there was something mentioned on RPI forum a couple of years ago about device's attached to an RPI with separate power supplies but can't find the link and "feedback" into the RPI i'll keep looking 

Link to comment
Share on other sites

4 minutes ago, fozzybear said:

Ok worth a try, I am racking my brain as there was something mentioned on RPI forum a couple of years ago about device's attached to an RPI with separate power supplies but can't find the link and "feedback" into the RPI i'll keep looking 

Ok so searched and not applicable to RPI4 so something else as rectified in this version

Link to comment
Share on other sites

Hi

Not sure if this is as @wulfrundescribes above, but isn't this just a case of hamming a length of copper pipe into the ground and attaching a wire between it and a tripod leg using (Spanish: abrazaderas) round clips? Wet the ground around the copper with salty water. 

Cheers.

Link to comment
Share on other sites

This is common with most kit these days (should be banned in my view) the problem originates with most of the PSUs currently in use having 70-110V on the earth terminal, grounding is best done through a 100 ohm or similar 5 Watt resistor with each item in use, this will stop the tingle but keep some isolation of signal/data between each unit..

Alan

Edited by Alien 13
Link to comment
Share on other sites

Thanks again guys. @Alien 13 Alan, how to incorporate this into my setup to test? Assume I have access to all pins on the RPi, am happy dismantling USB cables and confident in soldering stuff, just the electrical layout I'm unsure of here.

I also note through all this I'm seeking a solution rather than potentially addressing the issue. Don't want to just rectify the issue, but get to the bottom of why it's happening!
How to investigate mains leak?

Link to comment
Share on other sites

Switchmode mains supplies have a capacitor from each mains terminal to an output terminal. They have to be fitted so that RF interference generated by the switchmode circuitry is bypassed back to the mains and not radiated as interference to radio/TV systems. The side-effect is that the outputs float at about half mains voltage, the two capacitors act as a potential divider. The capacitors are meant to be sized so there's no electrocution risk AND they must be fail-safe types (don't bank on that if it's a cheapo supply though).

You can't do anything about the cause, it's in-built by the supply design but you could earth the mount to a ground spike, possibly through a resistor as described above. Whack a tent-peg/piece of copper pipe etc into the ground and make a lead with a crocodile clip each end and a resistor in-line (the above suggested 100-ohm is as good as any). Clip one end on the mount metalwork (bare metal) somewhere and clip the other end on your peg/pipe.

If that fails to solve things, your only real alternative is to use a linear power-supply, which won't suffer from the issue in the first place.

EDIT: you can check if the supply "leaks" by using a decent test-meter (DVM), connect one test lead to a mains earth and the other to each of your supply power terminals in turn. Set the meter to AC volts. If it reads some tens of volts on both terminals, that's pretty convincing evidence of the issue. If the supply is a switchmode, it's pretty much guaranteed.

Edited by wulfrun
  • Like 2
Link to comment
Share on other sites

Is your Nevada supply the PS-08 linear power supply and the Raspberry Pi 5V supply the only power supplies in the system? If so, then connect a wire from the Nevada -ve output terminal to the indicated Ground screw terminal on the rear of the Nevada PSU and it should solve your problem. I assume your Rasp Pi 0V gets connected to the Nevada -ve terminal via your normal cabling.

Before doing this though, to measure the tingle voltage, with your multimeter set to AC volts, measure between the ground terminal on the rear of the Nevada PSU and the Pi 0V. Also measure between the same Nevada ground terminal and the Nevada -ve output. You need your meter on AC volts and not DC volts to measure the tingles as it's an AC leakage voltage superimposed on the DC output voltage causing the tingling. You'll probably measure a higher AC volts on the Pi 0V than the Nevada -ve as your connections between the two are via fairly lengthy relatively thin cables and the Pi is electrically closer to its power supply.

The Pi power supply is only a two pin mains connector and is likely causing the problem as wulfrun says above. Low power output switch mode supplies are normally 2-pin mains plugs as they're cheaper to make and the capacitors to reduce EMI (electro magnetic interference) can be a small enough value that the leakage tingle effect is not usually enough to notice. Larger power switch mode supplies use 3-pin mains plugs and the output -ve is either connected to ground, or the EMI capacitors can be connected from input to ground and output to ground avoiding AC leakage on the output. 

If your supplies are powered from a long mains extension in the damp air this tingling effect can be more noticeable.

The Nevada linear power supply DC output is floating, so your whole system has no Ground reference, and the AC leakage volts have nothing to limit them. Connecting the Nevada -ve to the rear Ground terminal will clamp any AC leakage to Ground. This is assuming your extension cable if used has a good Earth continuity, which it needs to be safe. The linear supply has floating outputs for versatility as it has no need to eliminate EMI the same way as switch mode supplies.

Nevada, or any other reputable high current output switch mode power supplies have the -ve output connected to mains Earth internally so you shouldn't have this problem if using one of them..

Alan

 

Link to comment
Share on other sites

Thanks Alan @symmetal... So details of setup... from outside mains socket, I connect masterplug (stating here for good quality) heavy duty extension lead to patio area (around 3m from external socket, although flex is 10m). This powers the Nevada PS-08 which only powers the AZEQ6 via the cigatette plug-GX12 connection cable. I also plug in the RPi (PSU in white, in white) to the extension.

On occasion during setup, as the RPi has no screen, I use my MBP running on battery to do alignment, start EKOS, and all the bits that require the close proximity to the setup, before moving indoors and starting imaging. So the MBP can require power sometimes, meaning this is also gets plugged in, but only if battery starts running low.

I also plan to use a power supply to DSLR (Camera Supply), which will enable longer sessions without requiring watching the camera for it to run out of battery charge!

There is no connection from the RPi to the Nevada (Alan, you mentioned this as an assumption), but this is what I was thinking with spare USB socket (4 on RPi => CCD, GuideCam & EQMod + 1 spare) - tap into the black on it and ground somehow, reading this, ground to PS-08? Or open the Pi up (access to the GPIO pins) and use one of the Gnd's from this to the PS08?


Using the Linear Power Supply I'm also guessing is a potential option... If I used the Nevada to power the RPi via a 12V-5V reducer (example?) and higher current USB-C cable (thinking min of 3A as original PSU) - would get around the issue altogether?

Link to comment
Share on other sites

Pete,

From your description the Rasp Pi EQMod cable 0V wire should be connected to the mount 0V which should connect to the Nevada -ve terminal via the mount power cable. You can check this continuity with your multimeter on ohms, with all the power off.

However it's always best if using separate power supplies to have a permanent solid connection connecting their negative terminals together, if all your voltages are positive relative to ground as in your case, to avoid the voltages floating apart by unplugging a cable to the mount. A permanent connection from the Pi GPIO 0V pin to the Nevada negative would achieve this.

As the Nevada linear supply outputs are also floating this wouldn't cure the AC leakage problem from the Pi power supply, so connecting the Nevada -ve to the rear Ground terminal as mentioned previously should fix this issue. These bonding cables between power supplies and Ground should be as thick as is practical to use, and not using very thin wires, especially if the supplies are supplying higher currents of a few amps or more to avoid currents in these cables causing slight voltage variations between supplies due to any current flowing in the bonding cables.

Avoiding using the Pi power supply altogether, and using a 12-5V buck converter like this or similar, from the Nevada output would also fix the issue and be more elegant. The buck converter I indicated can be used via the screw terminals, or the USB power output can be used to plug into the Pi power input with a standard USB cable.

Alan

Edited by symmetal
Link to comment
Share on other sites

20 hours ago, pete_81 said:

Issue is reproducable in work when connecting RPi straight to ring main or at home using several different sockets, main observing done via outside plug.

This should not happen. And did you say its a 3 pin power supply?

Link to comment
Share on other sites

I have a single 12v supply which powers the mount, heat straps and also a powered USB hub (which then in turn powers the PI and all USB components).

If you have a different power supply for the PI and mount it could could cause an issue unless you bond ground for each together well (as @symmetal points out above). Letting ground link between PSUs through the components being powered is not a good idea.

Also don't connect the neutral of the 12v supply output to ground in a 3pin mains plug (unless the instructions say to do this), as you risk creating a local earth point outside for your house (and maybe for the whole neighborhood too).

Edited by 7170
Link to comment
Share on other sites

Thanks @symmetal Alan, & @7170James... Think this is the best approach for me too... I'm going to do both supplies from the Nevada as this should obviously eliminate worrying about lots of cables and grounds etc. I've the USB attachment rather than the EQMod cable (newer skywatcher mounts have the USB connections), hence all the talk about hacking apart a USB and grounding things. But in summary, it looks like the ideal solution is to use a voltage converter and USB cable to power the RPi via a fast-charge capable USB-C cable (to cope with the RPi requirements).

Edited by pete_81
  • Like 1
Link to comment
Share on other sites

@pete_81 - it is all in pieces at the moment as I've got the HEQ5 which forms my portable rig apart for re-greasing but I can describe in detail:

"JnDee 12V 10A (120W) AC / DC POWER Supply ADAPTER Transformer, for LCD Monitors, LCD TVs, LED Strip and CCTV System" 12v power supply. The description is taken from Amazon where I bought it, and it is still for sale now.

It gives 10A max at 12v using a standard 2.1/5.5mm DC plug.

This is then plugged into a standard CCTV power splitter cable (designed to feed x4 cctv cameras from one power supply). A generic product you can get for a couple of pounds.

I then use a short 12v 2.1/5.5mm extension type cable plugged into the power splitter cable to goto the mount, then another one to goto my custom made dew heater, then another goes to the USB powered hub.

The powered USB hub is a AUKEY USB3 hub (x4 USB 3 sockets, and x3 power only ports). The key thing is it is a USB hub designed to be powered off a 12v supply. There are lots of them out there, but they tend to be a but more expensive.

One of the power only USB ports from the hub is plugged into the Pi3 power supply in socket to power the Pi. Then the camera, wheel and focuser are plugged into the USB hub, and GPS dongle/EQDir cable go into the PI;s USB sockets as I found that more reliable.

The PI, USB HUB and PSU are all in a waterproof box fixed to the tripod legs. The only thing with the AUKEY hub is you have to press a button on the hub to turn it on (it doesn't automatically go on when it starts getting 12v power), so I can't flip a switch remotely by the house and go back inside, I need to actually goto the rig, but its not an issue as I need to take the cover off anyway.

Everything is running off the same 12v10A power supply, so common neutral etc. I've never had an issue with the setup.

Hope the above makes sense, key thing is of course is the powered USB hub designed to work off a 12v supply.

Edited by 7170
  • Thanks 1
Link to comment
Share on other sites

Thanks @7170 James, really helpful and yes, I can visualise your setup from your description. The powered USB hub sounds good but without having all the power available from one socket (the Nevada splits between banana-type plugs and cigatette plug), I decided to go with the buck converter idea for now. This should be arriving with me today and tonight is meant to be clear from 9 so might actually have it working then too for my setup without any buzz!
@fozzybear you'd mentioned trying phone charger. I have one of these arriving also to test running the RPi if not in the astro-setup (don't want to cart the Nevada to just run such a small computer!) so I shall let you know how that test worked out. I'll be a bit disappointed if both of our RPi supplies are causing the same fault (one for my astro setup and other for Pi400)

Thanks to all for your help on this. Follow up to come later, including hopeful measurement of mains leakage.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.