Switching PSU problem

[The Master]

Hi,

I've got several 240V to 12V 150W switching PSUs running circuits I've made. Most are working fine but today one of my circuits started resetting randomly.
After a long time checking various things I noticed that there seems to be a huge potential across the ground.

The PSU case and the 0V ground rail of my PCBs are all connected to earth through the 3-pin mains plug.
If I put a multimeter probe on the PSU case and the other probe on my PCB ground it measures over 1,300V.
With the mains disconnected I can confirm that there is definitely continuity between these points and with less resistance than my multimeter will measure.
Different points on the PCB get different readings. Points that are electrically further away from where the ground wire is connected read higher voltages.

Does anyone know what might be causing it and if it can be fixed? I've got quite a few of these PSUs now and I'm worried that everything will start failing and even become dangerous.

[richard]

The PSU case and the 0V ground rail of my PCBs are all connected to earth through the 3-pin mains plug.


If I put a multimeter probe on the PSU case and the other probe on my PCB ground it measures over 1,300V.


With the mains disconnected I can confirm that there is definitely continuity between these points and with less resistance than my multimeter will measure.

one or more of those statements cannot be true

power = voltage squared divided by resisence
if the case to gnd rail resistence was 0.1 ohms then
{ridiculous amount of power} 16.9 mega watts = 1300v x 1300v / .1 ohm
where is all the heat ?

[The Master]

1 of the components on the PCB is getting slightly hotter than usual but there's not a huge amount of heat coming from it and this is likely because one of the chips draws more current every time it resets.

I know it's not impossible to get voltage that high with all the coils and capacitors in this thing but I find it hard to believe it would happen without being specifically designed to do that.

I've tested it several times and my multimeter was showing almost 2KV last night. Could there be some capacitance or something that's confusing the multimeter?
The main PCB has 4 large capacitors between GND and 12V. I've tried the multimeter in DC and AC mode. AC mode shows 0V between ground points.

In the PSU the secondary is *almost* completely isolated from the primary.
After the 240V AC is rectified, the negative goes through what looks like a very large blue ceramic disc capacitor to the negative on the secondary. Could this be causing issues as it's essentially linking half cycles from live and neutral to my PCB ground.

The problem is also strangely intermittent. It doesn't seem to happen for the first few minutes after being powered up. My PCBs work fine and the multimeter reads 0V between all ground points. After a few minutes it goes back into that state.

While in that state, if I put my multimeter between the 0V and 12V outputs on the PSU it reads 12V and my PCB starts working as normal. As soon as I disconnect it the PCBs start resetting again.

I've had the same PCBs running overnight using another PSU of the same type and it hasn't reset once.

[richard]

lets examine this one point at a time


I know it's not impossible to get voltage that high with all the coils and capacitors in this thing but I find it
hard to believe it would happen without being specifically designed to do that.
it is impossible to get a dc voltage that high between two points connected by a low impedence unless a massive current is
flowing , circuit theory 101

I've tested it several times and my multimeter was showing almost 2KV last night.qed the fg[psu frame gnd] is not connected to the 12 gnd

Could there be some capacitance or something that's confusing the multimeter?
no

The main PCB has 4 large capacitors between GND and 12V. I've tried the multimeter in DC and AC mode.
AC mode shows 0V between ground points.
as you would expect, the voltage is basically a static build up

In the PSU the secondary is *almost* completely isolated from the primary.
as you would expect

After the 240V AC is rectified, the negative goes through what looks like a very large blue ceramic disc capacitor
to the negative on the secondary. Could this be causing issues as it's essentially linking half cycles from
live and neutral to my PCB ground.
no , the cap is for rfi suppression and usually 2.5kv rated and a few hundred pf in value
you may also have similar caps on the secondary rails, its also possible to have a MOV between the secondary rail
and fg to ensure psu isolation voltage rating is not exceeded


The problem is also strangely intermittent. It doesn't seem to happen for the first few minutes after being powered
up.
as you would expect the measured voltage is basically a static build up from switching transients and takes time to
accumulate

My PCBs work fine and the multimeter reads 0V between all ground points.
fg is not connected to dc gnd ,it simply cannot be .
it is impossible to get a dc voltage that high between two points connected by a low impedence
without a massive amount of power being consumed .

While in that state, if I put my multimeter between the 0V and 12V outputs on the PSU it reads 12V
and my PCB starts working as normal. As soon as I disconnect it the PCBs start resetting again.

i would guess that the meter impedence probably bleeds off the static buildup in some way
although "the 0V" mentioned is ambiguous

[pedja089]

i would guess that the meter impedence probably bleeds off the static buildup in some way
although "the 0V" mentioned is ambiguous
Static goes on surface of probe, and to your body to ground. Add 1MegaOhm resistor or bigger, from wall GND to 12V(- or +) and use 5W or bigger resistor, or multiple smaller in series to avoid arcing over body of resistor.

[The Master]

The ground of my PCBs (0V negative from the PSU) and the PSU case are both connected to the mains earth.
They are housed in an ABS box with screws visible on the outside so the whole lot needed to be earthed.

There is another one of these blue ceramic caps inside the PSU that goes from the negative out to case ground. I'm guessing this is the MOV you mentioned.
This is essentially bypassed by the earth wire I have running between case ground and PCB ground.

The "0V" is the negative output of the PSU.

I've got some 0.25W 1MOhm resistors that I could string together as a test.

Someone I work with has suggested the blue cap between the rectified mains and negative output may be failing.
This PSU has been on 24/7 for months without a problem but I know caps will fail eventually.
I've got several other working supplies and some that have never been turned on yet so I can try swapping out small components like that.

[richard]

The ground of my PCBs (0V negative from the PSU) and the PSU case are both connected to the mains earth.

yet you claim a 2kv potential difference .

one more try

it is impossible to get a dc voltage that high between two points connected by a low impedance unless a massive current is flowing , circuit theory 101

[The Master]

My multimeter says 2KV DC. I'm not sure I can trust it.

I understand that Ohm's law doesn't allow this condition. 2KV across even 1Ohm would give 2,000A of current flowing through the PSU case and some wires that only handle around 4A.

I'm sure all of the ground points I've checked are connected by low impedance.
There's no excess heat or melted wires etc.
My components haven't fried despite apparently having hundreds of volts difference between their grounds.

The only conclusion can be that there isn't really 2KV. There has to be something though as my circuits keep resetting and the multimeter is showing such high voltages.

Is there anything I can do to help narrow it down? I will try the 1MOhm resistor idea to see if it makes any difference.

[Dave]

First thing I would have done when I saw such LARGE potential is to, Get another meter and double check it...

[richard]

My multimeter says 2KV DC. I'm not sure I can trust it.

I don't doubt the possibility that the voltage is real, as a tv serviceman I have experienced those sorts of voltages between supposedly earthed TV's and properly grounded antennas many times.

maybe a photo of the psu indicating where the measurement is being made could shed some light

ps its not a meter auto_scaling measurement error

[The Master]

I've attached some pictures which I hope will help. The earth pin on the IEC socket has 2 green wires coming from it. (I've attached an old picture so you can see what's going on behind the PCB. There was only 1 wire back then.)
One wire goes to the PSU and then the front panel. The other goes to the screws holding the IEC socket then splits to the 10 metal sockets on the back and the ethernet module (little PCB top-left).
All 3 screws on the ethernet PCB and the 2 screws on the front panel are connected to PCB ground.

The 12V supply goes into the PCB at the back. The negative side connects to this PCBs ground plane. That ground plane is then connected to the ground plane on the ethernet module through the short wires between them (Red: 7V out, Black: ground, Green/Blue: data).

The ribbon cable to the front panel links the ground planes on the ethernet module and the front panel. It also has 5V/3.3V supplies and some data lines.

So all 3 PCBs have their ground planes connected but the ethernet module and front panel also have their ground planes connected to mains earth (and the PSU case). The PCB at the back is the only one without a separate earth connection as it's only contact with the outside world are the 4 pins on each mini-XLR socket. It doesn't even make contact with the metal cases of these sockets.

While the PCB was resetting I connected the multimeter to the 12V screw terminals on the PSU. This is what stops the problem until I take the probs off.
The high voltage was measured with 1 probe touching the top of the PSU case and the other touching the 3 screws on the ethernet module. The screw where the earth wire connects has the lowest voltage. I also put the probe on the back PCB where the ground wire connects to it from the ethernet module and this had the highest voltage.

There were 4 of these that were all turned on several months ago. The other 3 are still running fine now. This one was sat on top of 1 of the others so both would have experienced the same ambient temperatures and electrical noise etc.
There are another 4 of these which are still running after a year.

I opened up every PSU before powering up for the first time to check for solder balls, bad joints, lose components etc. Can never be too careful with cheap Chinese electronics ;)

I'm trying to source another multimeter to double check the reading.

[pedja089]

PCB in top right corner with metal port isn't grounded. Look at ground plane. And solder mask...
Also secondary of PSU isn't grounded. Remove blue capacitor looking thing, and place short. Then measure static build up.
PIC probably resets when spark jumps from screw to GND.
Multiple path to for ground can act as antenna. Move green wire from all PCB's to - terminal of PSU. Then try to test it.

[richard]

I agree with pedja
relying on the psu cover to be screwed on for earthing continuity is not a good look.
ps has meter reading been verified yet ?

[The Master]

I know it might not be clear from the images but every single screw hole in both PCBs (and the metal connector) are all definitely connected to PCB ground. The ground plane is on the bottom of the PCB and top is 3.3V which is why the screws holes aren't connected to anything on the top of the PCB. They are plated through to the bottom though.

The PSU secondary isn't grounded internally (though I will try that as you suggested). It is still grounded though because it connects to the PCB ground which has the green wire connected to it. It's a longer path but it's still connected to earth.

Is removing the green wire from the PCBs just to see what happens or are you suggesting this as a permanent fix? If I replace that capacitor looking thing with a short then I guess everything would still be grounded but I'd feel better knowing I have an actual earth wire attached in case one of the connectors comes loose.


relying on the psu cover to be screwed on for earthing continuity is not a good look.
Is this referencing the blue capacitor like thing that connects to the case rather than directly to the earth terminal?
I don't think that would be so bad as the case is always supposed to be on and I have separate grounding wires for the PCB grounds.
I could add a wire between the earth terminal and the negative secondary and also an earth wire directly to the case just to be sure.

I'll test everything you've suggested as soon as possible and report back. I know someone who has a better multimeter so I will see if I can try with that.

[pedja089]

If - is bottom, then metal connector should go at bottom.
Screw doesn't make good contact. Also on vertical board with LCD, I don't see exposed copper or tin around metal connector.

[The Master]

I can move the connector to the bottom. It shouldn't be causing this problem though. The PCB has about 6mm of tin around the hole on both sides of the board and the hole is plated through.
The LCD board is exactly the same (6mm tinned both sides and plated through). On that board the ground plane is on the same side as the green wire.

[pedja089]

Can you measure resistance between socket and GND on board and post picture while measuring.
I really can't understand that there is static build up, and all symptoms associated with it, and such low impedance path to ground.
Can you test your main's ground? I don't know from where you are, but try to measure voltage between neutral and GND.