Notifications
Clear all

Ground PC boards.  

Page 2 / 3
  RSS
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground PC boards.

Someone asked about the power supply voltage. The Voltage produced measured from the i3 MK3 Power Supply is 24.15V, at room temperature, 110V 60Hz power applied, powering standard PCB loads. The output Voltage was not measured while driving stepper motors.

The Van de Graaff High Voltage Generator analogy is interesting. Fortunately, both belts continually operate in both directions. It's not clear the body of the floating stepper motor bodies would be able to develop a high enough Positive Voltage potential sufficient to shock someone. It may not be worth the effort to ground the motor bodies unless you happen to be a perfectionist.

Posted : 31/03/2018 8:30 am
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground PC boards.

Pictures:

Posted : 31/03/2018 8:37 am
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground PC boards.

It is unclear the pictures went. How do I send images to this forum?

Posted : 31/03/2018 8:43 am
RH_Dreambox
(@rh_dreambox)
Prominent Member
Re: Ground PC boards.

Below the Submit button you will find Attachments where you will find the Add files button

Bear MK3 with Bondtech extruder

Posted : 31/03/2018 9:16 am
ronnie12342003
(@ronnie12342003)
Estimable Member
Re: Ground PC boards.

those wires really smarten up the machine now it looks home made lol

Posted : 31/03/2018 3:43 pm
devilhunter
(@devilhunter)
Reputable Member
Re: Ground PC boards.

I sent a mail to Ultimachine on how to properly ground the board.
Lets see what they say.
(should be some answers after easter)

Posted : 31/03/2018 4:50 pm
tomp
 tomp
(@tomp)
Active Member
Re: Ground PC boards.


PCB Grounds should not "float" or exhibit AC voltages with respect to Earth ground or they're DC applied voltage. Unless someone can show a valid reason why these PC board ground planes should not be at Earth ground, I will leave my system as modified.

I am sorry, but it looks you have absolutely no idea what you are talking about.

Have you ever tried to measure AC voltage between an output of a class II power supply and mains earth? It is usually about half of the mains voltage, so about 120 V in case of 240 V mains. These power supplies are used very often, for example in all USB chargers, some laptop chargers, some TVs or displays etc. and they are safe even when this high voltage is present since the current is very limited (maximum 0.25 mA, you have to use high impedance meter to measure it). This voltage usually causes no problems, apart from for example some audio stuff. In case of class I power supplies (which is used in Prusa's printers), the AC voltage to mains earth is usually very low, those 1.4 V (measured with high impedance meter) seems all right. Try to measure the current, if it is under 3.5mA, then it has no problems, it is safe and should work without any problems.

Moreover, when the board is floating, you can connect it via an USB cable to your desktop PC that is usually mains earth grounded. If the board was also mains earth grounded, you would create a ground loop and you could blow the board or your PC up.

Posted : 31/03/2018 10:20 pm
tomp
 tomp
(@tomp)
Active Member
Re: Ground PC boards.

Pictures:
Have you really connected both the display board and the Einsy board to mains earth ground through the chassis? You have probably (I cannot check since I do not have MK3 yet) just created a ground loop. Have you at least checked that the points on the two boards that you connected through the chassis are at the same potential? I guess they are, but they might not be and in that case you would probably blow up some power supply or regulator on one of the boards.

Please, if you are not absolutely sure what you are doing, do not do any modifications to the electrical connections. I have still not received the MK3 so I cannot check, but I believe that it is built to pass all the regulations, it is safe and thus do not need any modifications like that.

Posted : 31/03/2018 10:59 pm
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Grounding USB shield.

Further investigation of the PRUSA i3 MK3 USB shield grounding yields these results:

Operating condition:
The power supply is on:
Measurements:
PRUSA i3 MK3 USB ground Shield:
1) DC Resistance of the USB shield referenced to the frame of the power supply, which is Earth Ground:
Resistance = "Overload" or OPEN CIRCUIT.
A) The USB connector shield appears to be "FLOATING."
2) DC Voltage from USB shield referenced to the power supply frame (Earth Ground) = 0.0V
3) AC Voltage measured from the USB shield referenced to power supply frame, chassis (Earth Ground) = 530.2mV RMS, 60Hz (This should be the same potential as the chassis, zero Volts)
4) AC Voltage between PRUSA i3 MK3 USB ground Shield USB shield referenced to a desktop computer's USB ground shield = 781.3mV RMS, 60Hz (This should be zero Volts)

Conclusion: The ground shield and chassis of the PRUSA i3 MK3 USB ground Shield are NOT bonded together.

Measurement and comparison made showing typical USB shield ground conditions of a conventional desktop computer:
Results should compare similarly to measurements above on the PRUSA i3 MK3 USB Shield ground. THEY DO NOT COMPARE!
5) DC resistance of USB shield ground referenced to the frame of the computer (Earth Ground) = 0.2 Ohms (VERY DIFFERENT FROM 1) ABOVE)
6) DC Voltage from USB shield referenced to the frame of the computer (Earth Ground) = 0.0V (SAME AS 2) ABOVE)
7) AC Voltage from USB shield referenced to the frame of the computer (Earth Ground) = 0.0mV (VERY DIFFERENT FROM 3) ABOVE)

Conclusion: The ground shield and chassis bond together.

ground shield measurements between two desktop computers:
These measurements were made to show the compatibility of grounding systems between both ends of a typical USB communication cable.

Typical measurements between two USB ground shield systems of one desktop computer measured against another USB ground shield system on a different desktop computer:

8) DC resistance of USB ground shield from one computer to USB ground shield to the other computer's USB ground shield (not a laptop computer) approximately = 3 +- 1 Ohms.
A) Both Earth ground systems connect through Earth grounds of their respective power cords, thus linking both USB ground shields to each other as well as attaching both to Earth ground.
B) These resistance measurements DO NOT COMPARE with the PRUSA i3 MK3 USB ground Shield measurement.

9) DC Voltage from USB ground shield of one desktop computer compared to USB ground shield Voltage on another desktop computer = 0.0V.

10) AC Voltage difference between USB ground shield of one desktop computer to USB ground shield on another desktop computer, = 0.0mV RMS. (VERY DIFFERENT FROM 4) ABOVE)

Conclusion:
USB ground shields on both computers bond to their respective chassis, following the Universal Serial Bus Specification.
The PRUSA i3 MK3 USB ground shield does not follow the Universal Serial Bus Specification. It does not bond to the chassis.

Summary:
Common sense would say the Voltage difference of ground shields on each end of a USB cable should be at the same potential, zero Volts. Additionally, there should be no AC Voltage differences between two USB ground shields.

Universal Serial Bus Specification, Revision 2.0, April 27, 2000:
6.8 USB Grounding, page 114:
The shield must be terminated to the connector plug for completed assemblies, the shield and chassis are bonded together. The user selected grounding scheme for USB devices, and cables must be consistent with accepted industry practices and regulatory agency standards for safety and EMI/ESD/RFI.

6.9 PCB Reference Drawings
The drawings in Figure 6-12, Figure 6-13, and Figure 6-14 describe typical receptacle PCB interfaces. The drawings are included for informational purposes only.

Posted : 01/04/2018 11:03 am
tomp
 tomp
(@tomp)
Active Member
Re: Grounding USB shield.


The shield must be terminated to the connector plug for completed assemblies, the shield and chassis are bonded together.

This is for the cables, not for the devices. And of course, if you buy a USB cable, the shield around the wires is bonded to the chassis of the connectors at both sides.

Your measurements have no point, look at the schematic of the Einsy board (it is open source) and you will see that the USB section is isolated from the rest of the board, therefore it is floating unless you connect it to a mains earth grounded PC. This is the safe way to do it.

Posted : 01/04/2018 12:07 pm
Nikolai
(@nikolai)
Noble Member
Re: Grounding USB shield.


Further investigation of the PRUSA i3 MK3 USB shield grounding yields these results: ...

Your measurements look pretty wild mixed to me. It was already pointed out different power supply approaches. Everything you've measured is perfectly fine and within specifications.
In hardware design there are always trade offs. A simple thinking "Ground need to be 0V AC/DC with 0 Ohm resistance" does not apply.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Posted : 01/04/2018 6:39 pm
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground PC boards.

[attachment=1]IMG_0090.jpeg[ https://shop.prusa3d.com/forum/download/file.php?mode=view&id=13862/attachment ]

Posted : 02/04/2018 4:53 am
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground PC boards.

I'd love to see this schematic. What is the internet address for the "schematic of the Einsy board (it is open source)"?

Posted : 02/04/2018 5:08 am
NiHaoMike
(@nihaomike)
Eminent Member
Re: Ground PC boards.


I'd love to see this schematic. What is the internet address for the "schematic of the Einsy board (it is open source)"?

https://github.com/ultimachine/Einsy-Rambo/blob/1.1a/board/Project%20Outputs/Schematic%20Prints_Einsy%20Rambo_1.1a.PDF
That's apparently for the 12V version. The 24V version probably just has a few different components.

Inspired to get into 3D printing by Micah Elizabeth Scott, Naomi Wu, and an anonymous Bitcoin girl I met in college.

Posted : 02/04/2018 6:16 am
gregnier
(@gregnier)
Eminent Member
Topic starter answered:
Re: Ground the USB shield.

Why would anyone not ground the USB shield, but instead let it "float"? Computer manufacturers do not "float" they're USB ground shields.

The purpose of USB Ground shield is to connect the outer conductive layer of the USB cable. This conductive layer serves a couple of critical functions: It helps as a shield to prevent coupling of external signals. And it forms part of a transmission line system to accommodate high-frequency signals passing through the USB cable.

Letting the ground shield "float" negates both of these purposes. Not to mention what it will do to foul up the Group Delay properties of the transmission line. If Group delay is allowed to be out of specification, the signal integrity of the system will be compromised. Data, in the presence of excessive Group Delay, will be corrupted. If this happens, the signals arriving at the other end of the transmission line will not be correct.

If letting the ground shield "float" is OK, why do ALL computer manufacturers connect the ground shields to Chasis Ground? Ideally, The USB cable should be able to transmit signals from DC to Daylight with perfect signal integrity. In practice, realistically, it is impossible to construct such a transmission line device which will do that. All we can do is work with the Physics and technology of our time allow us. We need to try and not screw things up by doing something strange.

Posted : 02/04/2018 6:32 am
Korsakoff
(@korsakoff)
New Member
Re: Ground PC boards.

Dear all,

Assuming the rev. 1.1 schematics of the EINSY RAMBO board on the Ultimachine website is similar (or equal) to the Prusa Mk3 EINSY board,
the ground of the USB interface is floating w.r.t. the rest of the board. As can be seen on sheet 4, the USB ground is called "GND1" and the ground of the remaining part of the board is called "GND". An inductive coupler (ADUM7441) is used to separate this USB ground (plus +5V and data) in order to prevent any ground-loop or interference (galvanic decoupling). So the USB side will be grounded by the PC or laptop.
The 4 mounting holes (FGND) are connected to the board GND as can been seen on sheet 2. The board ground is also connected to power input J1 pin 2. I don't have my Mk3 kit yet, so I'm not sure if the PSU frame is connected to the 24V ground terminal (minus). If so, then the EINSY board ground will not float. If not, the EINSY board mounting holes can/should be connected to the frame and will be grounded (not floating).
Hopefully this technical explanation makes sense. I'm not a native English speaker.

Posted : 02/04/2018 7:27 am
Korsakoff
(@korsakoff)
New Member
Re: Ground PC boards.

I was looking for the schematics of the Prusa PSU but couldn't find anything on the web yet.
However when I look at the Meanwell PSU (I found the schematics of a 12V 350W type), the Frame ground is NOT connected to the output voltage minus connection. I see some capacitors that connect the -12V to Frame ground.
I'm not a power supply specialist, but in that case it seems not wise to connect the frame ground to the minus voltage output.
Can anyone explain what is the reason for this design?
So the EINSY board is connected to the minus voltage output terminal and so the board ground is not floating but still not connected to the frame ground directly (only through capacitors inside the PSU)!
The LCD board has the same potential as the EINSY board by the 2 flatcables runing between both boards.

Posted : 02/04/2018 8:07 am
NiHaoMike
(@nihaomike)
Eminent Member
Re: Ground PC boards.

The PSU doesn't ground V- internally because there are applications - e.g. audio - where it would be undesirable to do so.

For the printer, I speculate that the reason is so if one of the heaters fail partial short to chassis, it will not go into thermal runaway. Using a 47 ohm fusible resistor to connect V- to ground would limit worst case ground fault current to about 500mA during the few seconds it would take for the resistor to burn open. (And that's my suggested fix for those having problems with the floating ground.)

That schematic of the Meanwell PSU looks to be a very old design and a modern PSU will likely use a substantially different one, most likely two switch forward for 200-500W or so. (Above that resonant full bridge is common and below that one switch forward or flyback is common.)

Inspired to get into 3D printing by Micah Elizabeth Scott, Naomi Wu, and an anonymous Bitcoin girl I met in college.

Posted : 02/04/2018 8:51 am
tomp
 tomp
(@tomp)
Active Member
Re: Ground the USB shield.


Computer manufacturers do not "float" they're USB ground shields.

The USB connector shield is usually connected to the PCB ground, but not all USB port shields are connected to the mains ground. Again, please, check the difference between IEC class I and class II (double insulated) power supplies. USB connector shields in appliances powered by class II power supplies (some computers, some TVs, most CD and DVD players etc.) are not connected to the mains ground (they cannot be, because class II power supply does not have mains earth connection, it event must not have it).

This conductive layer serves a couple of critical functions: It helps as a shield to prevent coupling of external signals.
The data line in USB cable is a differential pair. Not shielding it at all will cause a lot of common mode noise, but only a small amount of differential noise. This is very similar to UTP cable used for ethernet.

If letting the ground shield "float" is OK, why do ALL computer manufacturers connect the ground shields to Chasis Ground?
For EMI reasons. But not all computers have conductive chassis and not computers are powered from class I power supply, thus not all USB connector shields are connected to mains ground (class II power supplies do not even have connection to mains ground).

As robert-jan.g said, according to the schematic, the USB connector shield on Einsy board is connected to the local ground of the board that is responsible for USB communication. This part of the board is then insulated from the rest of the board (and it is powered by the USB, by the way). There is nothing wrong about it. This configuration is used in many measurement instruments if you want to avoid ground loops or have your circuit floating.

By the way, not all mains earth grounds are at exactly the same potential. Especially in big buildings. And of course, the potential can be very different between different buildings. A few years ago, we had to run a coax from one side of our business building to the other. One side of the coax shield was connected to the mains ground. At the other side of the building, the voltage between the shield and the mains ground (i.e. the voltage difference between the mains ground at different parts of the building) was 40 V. If someone would have connected the shield also on the other side, we would blow up our expensive measurement system. That is why ethernet uses transformers, thus the whole ethernet cable is completely floating.

If you want to do some measurements that make sense and that can told us at least something about the safety of the printer, you could measure the power supply leakage current. Just switch your meter to AC mA, connect one probe to power supply minus (or plus) and the other to mains ground. However, you would have to use a different meter, Fluke 115 is not appropriate for this measurement since it do not have any mA or uA range.

Why are you even trying to solve this "issue"? Do you have any problems with USB communication on your printer? If not, you should probably stop doing modifications like this as you seem to not understand what are you doing. As I mentioned earlier, by connecting random points at your printer, you can damage it.

Please check this video, it explains what can happen if you connect ground without actually knowing what you are doing:

I am not en electrical engineer, I am just a physicist working with a lot of electronics measurement stuff and also an electronics hobbyist. However I am confident that the USB circuit on Einsy PCB (according to the schematic, I guess it is exactly the same what is in the printer) is designed correctly it meets all the standards and it is safe and should work without any problems.

Posted : 02/04/2018 9:08 am
Korsakoff
(@korsakoff)
New Member
Re: Ground PC boards.

Thanks NiHaoMike for your explanation.
If I read trough all the previous posts on this subject is it correct to conclude that:
1) the USB is floating for a reason (isolate lpatop/PC from printer), do not ground it.
2) it is okay to connect the Prusa frame to PSU frame (ground it), as it is floating now.
3) grounding the stepper motors is okay, not sure if this will help (than you should also consider to shield the motor cables).
What is your opinion to increase your proposed "fusable resistor" from V- to PSU frameground to a much higher value (e.g. 100 Kilo Ohm), it will prevent the EINSY board to completely float from Frame ground and therefore might solve the reported electrostatic sensitivity?
I'm afraid I can not be of much help as I have just basic knowledge of electronics (so it is better to leave this to the experts before I will create confusion or make wrong conclusions 😀 ).

Posted : 02/04/2018 9:20 am
Page 2 / 3
Share: