Circuit sparks when touched with scope probe...

Thread Starter

liquidair

Joined Oct 1, 2009
192
Hey all,
So I'm working on a project using a power supply design I've used many, many times. The supply in question is a simple linear cap input supply regulated to 12V via an LM1085, then split via 2 diodes where one part stays 12V (11.65V), the other is regulated down to 5V. At first power on something didn't feel right and I thought I heard little zaps, but I continued. Voltages measured correct so I then fired up the scope to check for any ripple/oscillations. After hooking the scope probe's ground lead to the supply's ground, I checked the 12V diode's cathode and it would spark from cathode to probe tip.

I immediately shut down and measured the ground connections from supply ground to chassis and scope ground, rechecked the supply for shorts, checked my scope probes for shorts...everything appears OK. I tried testing the circuit again and thought it was resolved but again got sparks after a few times touching that part of the circuit. It seemed as if switching to the 10X setting on the probe made the sparking worse. A second probe did the same thing.

Any clue what could be happening? It seems very odd to me that and arc could occur to a high-impedance scope probe, especially an arc that doesn't seem to damage anything. It's got me a little scared to proceed.
 

KeithWalker

Joined Jul 10, 2017
3,063
Can you explain what a "simple linear cap input supply" is? From your description, it sounds as though it is not isolated from the mains supply. If that is so, It is potentially dangerous and is on the banned list of subjects for discussion on this forum.
Regards,
Keith
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
Could you take a picture of the power supply,a schematic would be great.Are you in the United States or Europe?
Hi! I'm in the States, I'll get a schematic in the morning!

What is the make and model of the oscilloscope? It sounds like your scope is not properly grounded.
It's a B&K 2190B. Ya, that's what I assumed but I measure 0.2 ohms between probe ground clip and the mains earth of my safety socket (light bulb wired in series with the outlet) which the DUT is connected to.

Can you explain what a "simple linear cap input supply" is? From your description, it sounds as though it is not isolated from the mains supply.
No way would I attempt to connect direct to mains when I'm scared of a little spark in what should be a benign 12V/5V circuit, lol! It's textbook, like this:

13VAC secondary -> 100nF X-Class (across winding) -> Schottky bridge rectifier -> 3,900uF cap -> LM1085 -> 100uF cap -> 2 diodes to split supply
D1.) (12V node) 1000uF cap + 100nF ceramic
D2.) 100uF cap +100nF ceramic -> KA7805 -> (5V node) 1000uF cap + 100nF ceramic

The 12V node is where I was seeing the spark. I tried again earlier tonight without any loads and I shut it down as I was beginning to smell a little something but nothing was hot or even warm.
 

Ian0

Joined Aug 7, 2020
9,680
From your description, it sounds as though it is not isolated from the mains supply. If that is so, It is potentially dangerous and is on the banned list of subjects for discussion on this forum.
Regards,
Keith
Since when has discussing a non-isolated supply been banned? I see people discussing triac circuits, motor drive circuits, three-phase wiring etc. all the time.
 

Ian0

Joined Aug 7, 2020
9,680
The 12V node is where I was seeing the spark. I tried again earlier tonight without any loads and I shut it down as I was beginning to smell a little something but nothing was hot or even warm.
Are you sure it isn’t just bad soldering? When you prod it, it makes the circuit and you see a spark.
 

LesJones

Joined Jan 8, 2017
4,174
Verify that the 13 volt secondary winding is totally floating. I.E Infinite resistance reading between the secondary winding and mains earth, mains live, and mains neutral on the mains input to your power supply. (With it NOT plugged into the mains socket.)

Les.
 

MisterBill2

Joined Jan 23, 2018
18,179
If something is sparking to the scope probe, and not the probe ground, then there is a problem with the probe, OR it is set to provide a 50 ohm termination.
 
Last edited:

KeithWalker

Joined Jul 10, 2017
3,063
Since when has discussing a non-isolated supply been banned? I see people discussing triac circuits, motor drive circuits, three-phase wiring etc. all the time.
My understanding is that discussing transformerless non isolated AC mains circuits is banned unless the circuit is totally enclosed and has no external connections. Moderators, please correct me if I misunderstand this policy.
As it turns out, in this case it is not an issue because liquidair mentioned in posting #6 that he is using an isolated supply.
 
Last edited:

Thread Starter

liquidair

Joined Oct 1, 2009
192
Verify that the 13 volt secondary winding is totally floating. I.E Infinite resistance reading between the secondary winding and mains earth, mains live, and mains neutral on the mains input to your power supply. (With it NOT plugged into the mains socket.)
Hi Les, I just checked and I get no reading from line and neutral to the secondary, but mains earth I do get something like 121k (which I assume is the diode resistance through the rectifiers).
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
If something is sparking to the scope probe, and not the probe grouns, then there is a problem with the probe, OR it is set to provide a 50 ohm termination.
This is what I thought initially but seeing it happen with the 2nd probe too made me doubt a fault with the probes. I'm measuring infinity between the probe tip and the ground clip/the BNC shell, 0.2 ohms from ground clip to BNC shell, and 300 ohms from probe tip to the BNC center pin. Both probes show the same measurements.
 

MisterBill2

Joined Jan 23, 2018
18,179
So far nobody has mentioned the fact that a scope probe is supposed to be a very high impedance input, and you would not be able to draw a spark to the probe tip even touching the mains directly.

So for starters, you need to use an ohm meter to check the resistance from the probe tip to the scope common. That should show at least a few megohms on most ohm meters, or else open.

Have you verified the voltage with your multimeter? It should show either 5 volts or 12 volts, depending on which output you check. Then check the supply voltage, and common, relative to the scope common terminal.

Based on the information given it seems like the only reasonable guess is that somehow the scope probe tip has become grounded. Many good scopes do have a switch to ground the input, used for setting the zero point of the trace. Or the scope probe may have a ground button. I have seen that also.
 

MisterBill2

Joined Jan 23, 2018
18,179
OK, an additional response arrived while I was keying in my response.
So one more suggestion is to use your meter to check for a voltage between the scope probe tip and the point being probed. That reading will verify the nature of the spark causing power. And I just realized that the resistance measurement was made with the probe disconnected from the scope. So the next thing would be to check the resistance between the BNC center and shell.
 
Last edited:

Thread Starter

liquidair

Joined Oct 1, 2009
192
So the next thing would be to check the resistance between the BNC center and shell.
Ok, so I checked this and it's infinite.

So one more suggestion is to use your meter to check for a voltage between the scope probe tip and the point being probed. That reading will verify the nature of the spark causing power.
For this, I noticed a brief moment where the meter measured 4V then fell to 0. After this, I removed the 5V regulator (just to see) and probed the spark point. I thought I noticed this last night but I've confirmed...it won't spark right away. I was able to touch this point 2-3 times before it began sparking and was consistent after that. Not sure if that's a worthwhile clue or not.
 

DickCappels

Joined Aug 21, 2008
10,152
Discussion of transformerless power supplies were not allowed until (from memory) less than a year ago at which time the subject was removed from the list of forbidden topics. Whether a project being discussed is likely to lead to damage, injury, or death will be considered on a case-by-case basis.

You have leakage from the Line to your secondary. If you are relying on a transformer for isolation you might consider changing your transformer. If there is no transformer, then please post the schematic so people can make suggestions.

You can measure the leakage from the circuit to earth by measuring the AC voltage across a 1k resistor as the leakage current passes through the resistor on the way to earth. Anything I can feel is enough to make me either add an earth ground connection or stop using the offending appliance. I think 15 ma is one of the standard maximum leakage currents allowed on consumer products.
 

Ian0

Joined Aug 7, 2020
9,680
If the secondary is totally floating, then it will be connected mains via the inter-winding capacitance, and will tend to float to half the mains voltage.
There may be a spark when it gets connected to earth. It depends on whether the transformer was concentrically wound, or a split-bobbin, or whether it had in inter winding screen.
 
Top