Circuit sparks when touched with scope probe...

MisterBill2

Joined Jan 23, 2018
18,167
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.
The TS mentioned that it was a transformer based supply in post #1, and referenced it a couple of times since then. Nobody else mentions transformerless supplies.
 

MisterBill2

Joined Jan 23, 2018
18,167
So here we are with a mystery spark from a low voltage supply, that jumps to a (supposedly) high impedance point of a scope probe. What we have not heard, I don't think, is if the supply is actually providing the voltages it was created for. And I don't think that we know if that DC supply has the output common tied to the line cord green wire safety ground connection. That could possibly be the connection completing the spark circuit. But I don't think we know if it is or not. So there is still a lot that we do not know.
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
What we have not heard, I don't think, is if the supply is actually providing the voltages it was created for. And I don't think that we know if that DC supply has the output common tied to the line cord green wire safety ground connection.
Yes I did mention previously that the voltages are correct. And both the safety ground and circuit ground measure 0.2 ohms with respect to the steel chassis.

I'm about to try to measure the leakage current and perform the same test another device that uses a similar supply. I really hope it's not the transformer, as it's $180 custom part that's only about a year old.
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
Ok, so tried another device with a similar supply and no sparks so we can rule out the scope being the problem.

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.
Just want to make sure I'm doing this test right. This would be wired similar to a ground lift (which this device does have) where the circuit ground is connected to the 1k resistor and the other end of said resistor is connected to the earthed chassis? The current ground lift I use is a 10 ohm 5W in parallel with a 100nF film cap in parallel with a bridge wired so that 2 diodes clamp positive and the other 2 clamp negative. I assume I want to remove the bridge and the cap for this test?
 

LesJones

Joined Jan 8, 2017
4,174
If you really want us to be able to understand the problem you will have to supply us with the EXACT schematic of how everything is connected together. Have you measured the resistance between the the chassis ground of your power supply (Which I assume is connected to the negative output rail of your power supply.) and the scope ground (At the BNC connector that the probe plugs into.) with the power switch off on both the scope and your power supply. Both your power supply and the scope must be plugged into the mains supply but switched off. Also with whatever the power supply is powering disconnected from the power supply.

Les.
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
Here's the schematic. Pretty simple and I put in the aforementioned ground lift even though this power supply connects directly to the chassis. And also, to keep the problem as simple as possible, I've disconnected all loads on the supply including the LEDs.
 

Attachments

Thread Starter

liquidair

Joined Oct 1, 2009
192
Have you measured the resistance between the the chassis ground of your power supply (Which I assume is connected to the negative output rail of your power supply.) and the scope ground (At the BNC connector that the probe plugs into.) with the power switch off on both the scope and your power supply. Both your power supply and the scope must be plugged into the mains supply but switched off. Also with whatever the power supply is powering disconnected from the power supply.
Just did this and got 0.8ohms.
 

MisterBill2

Joined Jan 23, 2018
18,167
Looking at the circuit I do see a problem.It appears that the 0.1 mfd bypass capacitor for the input side of the 12 volt regulator is missing. That caP NEEDS TO BE INSTALLED very close to the input, between common and the input pin. The best place is withing 0.1 inch of the device. Forgetting that cap caused me a number of problems and I did not make that mistake gain.
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
Looking at the circuit I do see a problem.It appears that the 0.1 mfd bypass capacitor for the input side of the 12 volt regulator is missing. That caP NEEDS TO BE INSTALLED very close to the input, between common and the input pin. The best place is withing 0.1 inch of the device. Forgetting that cap caused me a number of problems and I did not make that mistake gain.
Interesting. I looked and had that cap in the previous version of this circuit but I removed it because it caused the charge pulses to increase from the rectifier which seemed to increase noise in nearby areas. It wouldn't be hard to add back, although the main filter cap is right there less than 0.75" away.
 

MisterBill2

Joined Jan 23, 2018
18,167
Interesting. I looked and had that cap in the previous version of this circuit but I removed it because it caused the charge pulses to increase from the rectifier which seemed to increase noise in nearby areas. It wouldn't be hard to add back, although the main filter cap is right there less than 0.75" away.
That capacitor is needed to prevent oscillation in the several megahertz range. Unfortunately the larger caps farther away do not have the same effect, both because of capacitor issues and because of interconnection led inductance. Perhps the previous cap was the wrong type.
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
I might have found something. I moved the circuit ground wire to the ground lift circuit and I removed all of the ground lift components in preparation to solder the 1k resistor in to test leakage. Oddly, when I measure the resistance between circuit ground and the chassis I get 2 different sub ohm values while actuating the switch. With those components removed, we should be getting sub ohm and infinite, so something is clearly fishy!
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
That capacitor is needed to prevent oscillation in the several megahertz range. Unfortunately the larger caps farther away do not have the same effect, both because of capacitor issues and because of interconnection led inductance. Perhps the previous cap was the wrong type.
It was a ceramic, and actually I had it positioned immediately after the rectifier. My intention when I first started doing that was to quickly clean up any high frequency diode hash giving it a small loop back through the X-class cap across the winding. But, as it was explained to me, and it may be wrong, the smaller cap value has a much faster time constant and therefore needs brief but high current charging pulses. So maybe a 10uF ceramic might be a good compromise?
 

BobTPH

Joined Jun 5, 2013
8,804
Do what the datasheet says. They normally recommend specific capacitors on the input and output side.

Why do you think you need a cap across the secondary? Never seen that before.

And why do you think you need diodes between the bridge/ filter and the regulators? Diodes can be used when sharing two power sources of different voltages to select the higher one and isolate them. The same is not needed when two loads share a single supply.

Bob
 

Thread Starter

liquidair

Joined Oct 1, 2009
192
Why do you think you need a cap across the secondary? Never seen that before.
Hi Bob, that cap lowers the winding's resonant frequency so any diode switch ringing is slower can no longer couple as easily to other windings and nearby circuits. It needs to be an X-class cap according to regulations since it's directly across the winding. I'm not 100% if that requirement is still necessary because I put it after the fuse, but I do it anyway to be safe.

And why do you think you need diodes between the bridge/ filter and the regulators? Diodes can be used when sharing two power sources of different voltages to select the higher one and isolate them. The same is not needed when two loads share a single supply.
That's a really good question, lol. I started doing that a long time ago and I'm pretty sure my intention was that since the 5V supply has a microcontroller on it, and the 12V supply goes near audio, that splitting the supplies with diodes makes it a little harder for digital noise to find a way to the 12V ground (the 5V and 12V grounds join at C85) with a very slight dissipation advantage. That could not even work but I tend to be happy about the noise performance I get so I still do it.
 

Ian0

Joined Aug 7, 2020
9,667
It needs to be an X-class cap according to regulations since it's directly across the winding
It only needs to be a class-X cap if it is directly across the MAINS.
Its effectiveness as a filter it enhanced if the transformer is of split-bobbin construction because the leakage inductance is higher.
 

MisterBill2

Joined Jan 23, 2018
18,167
Coupling of noise back through a correctly installed IC regulator is not a problem usually. An incorrect ground connection is far more likely to be the source of any noise coupling. The goal is to have none of the noisy current flowing in any of the audio ground conductors. and here "ground" slipped in when the word should have been "common". Ground is a mythical zero impedance plane that is seldom found in reality.
 

LesJones

Joined Jan 8, 2017
4,174
Thank you for posting the schematic. The 0.8 ohm reading you got in post #27 confirms that there is no break in the earth path on the mains wiring. The only other thing I can think of on the mains earth wiring that could cause a problem is if there was something else in the building with the neutral wire on something else being swapped with the earth in error so it was drawing current via the earth wiring causing a small AC voltage between different points on the earth wire. You could check for this reote possibility by measuring the AC voltage instead of the resistance between the points I suggested in post #25. I think what you found in post #31 could lead to solving the problem.
Now just a few comments about the off topic comments about the schematic. My understanding of the datasheets on the LM78xx series regulators is that the 100 nF capacitor close to the input is only required if it more than a few inches away from the main reservoir capacitor.
I install a diode on the input to the 5 volt regulator (With a large value decoupling capacitor after the diode.) when it is feeding a microcontroller if the raw DC supply is also feeding a motor (Or similar load) that draws high current pulses which causes the raw input voltage to drop for a short time. In that case the capacitor after the diode supplies the 5 volt regulator but can not try to hold up the raw supply.
To reduce rectifier noise in audio circuits I install small capacitors (Say 10 nF) directly across each diode of the bridge.
The off topic discussion about the schematic has also given us a better idea of your level of knowledge.

Les.
 

DickCappels

Joined Aug 21, 2008
10,153
I may have missed this if it was mentioned above, but what is the voltage between earth and the chassis? Also, between the earth (assuming the scope is earthed -is it?) and the point that drew sparks? I suggest measuring with a high impedance scope probe.
 
Top