Ground Potential

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

jin
Posts: 107
Joined: Sun Aug 26, 2012 6:35 am
Location: Singapore

Ground Potential

Post by jin »

Hi Guys,

After I set up the Stackshot wheneven I touch the metal parts of the camera, stepper motor casing or the controller chassis I keep having the tingling sensation on my finger and so I decided to check the voltage between ground and the metal and found the potential different is ~85Vac :shock:

I think the AC/DC adapter must be having high leakage current (I'm using a UK two pin version, without earth pin type) so I ground the controller chassis with a 22K resistor (that is what I have on hand) to the earth and solve the issue.

Question to the Stackshot users in this forum: did you ever have the same issue and how you resolve the issue as I'm also not sure is this an isolated case with my adaptor that might be having high leakage current due to some components fault inside?

Do let me know if this is safe to continue using with my above mentioned earth wire.

thanks in advance for the advice.

Rgds,
Jin

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

You should contact Cognisys about this issue.

The high open circuit voltage is not unusual. But the fact that the leakage current is large enough to feel suggests at least 250 microamps even with you in the loop. That does seem unusually high. My own unit (two-prong US plug) gives 43 VAC open circuit, but leakage current is only 55 microamps when short-circuited through a meter, and only 12 microamps when run through my hand (dry skin, probe tip contact areas).

If the unit were mine, I would ground it with a piece of wire instead of the resistor (for improved safety in case of component failure), and then feel OK about using it while I contacted Cognisys.

--Rik

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Jin,

As you know, my setup, the Bratcam, includes a StackShot controller and a non-standard stepper motor on my own hardware, and I often feel tingling sensations in my cheek if my face brushes a conductive portion of the camera while I’m looking through the viewfinder. The current isn’t strong enough to reliably feel with my fingers, but my face is apparently more sensitive (perhaps because facial skin is thinner?) Also, sometimes when I touch the StackShot controller, it reboots, so I presume I’m discharging some current through it—probably not a good thing to be doing. Your situation is likely very different from mine. But I do not believe, in my case, that the electrical source is the StackShot controller or my stepper motor. My first guess was the AC power supply for the camera, but it isn’t that, either.

I once poked around the rig with my multimeter, checking voltage. I don’t recall the measurements, just that the instrument confirmed that I wasn't imagining the tingling on my face.

You’ll recall that the Bratcam has a lot of steel and aluminum in it. It also has number of wires attached, for transmitting data, power, or both. It sits on a wooden platform on a slate pool table, so, absent the various wires, it is electrically isolated from ground.

In an interesting test, I disconnected all wires from whatever they were attached to at the non-camera end, but left them plugged into the camera or rig—so no power should have been coming in through them. But the tingling was still there, and so was the measureable voltage.

The next thought was that the source might be inductance from some of the myriad electrical cords I have near the Bratcam. Since all these cords run to a single power strip (located about three feet beneath the Bratcam), I unplugged this power strip from the wall. That got rid of the tingle and voltage reading. These phenomena returned when I plugged the power strip back into the wall. So, I gather, the electrical source in my case is indeed inductance.

I’ve been meaning to investigate further, and perhaps ground the rig, but have not gotten to it, yet.

--Chris

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

I am very surprised to hear these reports. Check that the primary power supply has UL approval in the USA CSA in Canada and TUV in the EU

If it does not I would replace it with one that does. Setting a grounding wire can improve the safety but if it , for example causes reboots as Chris S mentions it indicates a design or other performance deficiency that ought to be remedied.

The magnitude of leakage currents mentioned and a tingle is not a terrible hazard. 85 volts is not scary especially if it sags if current starts flowing. But it says to me that the power supply is probably a switching power supply that has no isolation transformer at all separating the circuitry from the AC line.

This is one of several reasons I dislike switching power supplies even though, like LEDs they are an idea whose time has really come. But they will work when hung on an isolation transformer and I would add one under these circumstances.

http://www.mouser.com/ds/2/410/N-68X-223861.pdf
above can be wired for almost anywhere in the world.


http://www.onlinecomponents.com/triad-m ... p=12696273
Above for 120 volt part of the world. There are step up and step down transformers that also isolate

For the same reason my computer often causes tingles and I don't like it at all.

These tingles can be dangerous if you have an undiagnosed susceptibility to cardiac arrythmias which could be triggered at lower current levels than the average person can tolerate.

I also can't imagine that they are good for the camera , and associated equipment such as flashes, the stackshot controller and whatever else might be there.

Probably a case where the power supply vendor fibbed or was evasive to Cognisys.

jin
Posts: 107
Joined: Sun Aug 26, 2012 6:35 am
Location: Singapore

Post by jin »

Hi,

Rik - thanks for the advice, I wrote to Cognisys last night, pending for reply and meanwhile I now ground the controller chassis to earth without a resistor.

Chris - in my case this voltage is the same even with the adapter stand alone. I actually saw your system with a 'ground' wire assembly?

Image


g4lab - in my country we are using UK plug with earth pin, so no issue with earth-to-chassis potential different as they are all tied to earth and use of isolation X'fmr is not needed. The adapter is basically isolated from primary otherwise it will trip my home ELCB - Electrical Leakage current breaker, 50mA sensitivity.

I checked the adapter it has TUV/GS mark (for Europe?) on the name plate so should be safety type approved already, I assumed.

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Gene, I can't speak for Jin's situation (he posted while I was writing, and his scenario is very different from mine), but let me emphasize that I'm fairly certain that my StackShot items are not the source of the electrical discharge I feel from my camera. The tingling occurs even when the Stackshot's power cord is unplugged from an AC source, so Cognisys is innocent on all charges ;) . Also, I just checked, and their power supply is labeled as UL listed, and, unsurprisingly, a switching adapter.

The rebooting that sometimes occurs when I touch the StackShot controller is probably caused by electrostatic discharge (ESD) from my hand into the controller. I think that StackShot controller is not the source of this zap, but the target of it.

After my earlier post, Rik Littlefield kindly sent me information about a non-serious issue that some early StackShot controllers had with ESD discharges temporarily confusing the LCD display. Later controllers included a design change to avoid this, and earlier ones can be fixed with a bit of metalized tape and about 10 minutes time. Cognisys will happily do the modification for free, but if my controller needs this (it's a fairly early specimen), I'll likely just take care of it myself. One comforting thing I learned is that Cognisys tested their controllers with large amounts of ESD, and never caused permanent damage.

I'm quite sure, if any problems emerge with Cognisys products, they will take good care of us. When it comes to customer care, they are superb.

There are usually four separate power supplies powering elements of my rig, including the Cognisys one. Since the rig still discharges "tingling current" when all of these are disconnected from AC, I don't think we can blame any of them for being the source of the "tingle current."

Jin, I'll have to look back at pictures of the Bratcam, to see what wire might have appeared to be a ground wire.

--Chris

--edited typos
Last edited by Chris S. on Sun Dec 08, 2013 1:50 am, edited 1 time in total.

jin
Posts: 107
Joined: Sun Aug 26, 2012 6:35 am
Location: Singapore

Post by jin »

Hi Chris,

Thanks for sharing. Mine has no functional issue except for the tingling sensation. Also, my entire system are battery operated ie, camera and flash so the adapter is the sole DC power source.

Will keep guys posted once I hearing anything from Cognisys.

Rgds,
jin

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

jin wrote:...and use of isolation X'fmr is not needed. The adapter is basically isolated from primary otherwise it will trip my home ELCB - Electrical Leakage current breaker, 50mA sensitivity.
With respect, I think you have too much faith in your ELCB. A device could leak 40 ma without tripping the ELCB. But as described at http://en.wikipedia.org/wiki/Electric_shock, 10 mA can prevent letting go and 30 mA can cause ventricular fibrillation. The ELCB specifications are a compromise that prevents most life-threatening leaks while not infuriating consumers with too many spurious triggers. It is not intended to prevent merely painful shocks.

g4lab's advice about the isolation transformer is well worth considering.

As it happens, I have one similar to what he describes, packaged up in a PowerVar 0.5 Power Conditioner / Toroid Isolation Transformer like the one advertised HERE.

The supply I mentioned earlier that measures 43 VAC open circuit and 55 microamps short circuit leakage when plugged into the wall, suddenly measures 0.7 VAC and 0.0 microamps to ground when plugged into the isolation transformer.

Not all supplies will have their leakage drop so much. Measuring the output of the isolation transformer directly, each of its power connectors measures 47 VAC and 12 microamps to ground, probably due to capacitive coupling. It's possible to imagine a supply that could continue to leak that much even when connected to isolation. In truth, I was quite surprised to see that the supply I tested dropped nearly as much as it did.

The explanation is that the two power connectors coming out of the isolation transformer are symmetric with respect to ground: when one goes positive the other goes negative. As a result, a switching supply that sort of "sits between" its two power inputs also ends up sitting very close to ground, so its leakage drops way down as in the case I measured.

--Rik

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

I am pleased that Rik got ahead of me and said a bunch of the same things that I would have said.

And I would add this. In addition to the fact that 40 or 50 ma is plenty of current to give a jolt or prevent letting go most of the rest of the world other than the North America and parts of Japan (my apologies to anyone I forgot or didn't know about), use 220-240 volts AC so it is even easier to have these leakage currents flowing.

I really don't like it when I discover that a piece of gear has no transformer isolation from the AC line. Even here in the USA at 120 volts AC. There is just too much hash on the line. Transformers stop a lot of it. But they cost money and use copper and iron so they are being largely eliminated.

Why should there be an AC voltage riding on what should be a DC output??? The input impedance of the Fluke meter in the picture is 10 megohms. You should also measure the other DC lead and you will probably find a similar voltage. This is probably common mode leakage.
It is probably coming through some leakage capacitance but if it makes a tingle that is too much. Set the meter to measure AC current on a milliamp scale, and read the leakage current from the tingling surface to ground.

TUV certification should mean its a good product but tingles and leakage voltages that high don't seem right to me. I admit to being old fashioned but I keep an assortment of isolation transformers and isolated variacs around to test things. Especially things that have switching power supplies. (Hiss , spit!)

Chris if your unit is still tingling with all the power supplies unplugged it sounds like a flash capacitor might be holding a charge for a while. Does it keep tingling for a long time. Usually these things are AC leakage currents. Usually they have discharge resistors to discharge the capacitors after a period of time for safety reasons. But they make them discharge the capacitors slowly so that they don't slow the recycle performance.

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

g4lab wrote:Chris if your unit is still tingling with all the power supplies unplugged it sounds like a flash capacitor might be holding a charge for a while. Does it keep tingling for a long time.
Gene, thanks for sharing your insight. The potential for a flash capacitor to be the culprit is good to have on the radar screen. In my particular case, though, I don't think flash capacitors are involved. First, if you recall the Bratcam's lighting stage, it physically and electrically isolates the light sources from the camera and subject stages. The lighting stage and camera/subject stages are not touching, and each sits on a wooden platform. Not visible in the linked image, there is also a sheet of non-slip rubber between the lighting stage and the wooden platform. Flashes are set off by radio triggers, and have no contact with the camera. Further, when I use small flashes, only their plastic mounts contact the Noga support arms, so they should be electrically isolated even from the lighting stage. Also, I have moved more and more toward the use of continuous light, and now rarely use flash. The halogen illuminators are suspended from the ceiling, on a wooden shelf hung by nylon ropes from wooden joists, with the fiber optic light guides lamped on the Noga arms attached to the lighting stage. So even these are well isolated from the camera, where I feel the tingling.

No doubt you know far more than I about the behavior of electrons. Is there something that makes you dubious that inductance could be the cause of the tingling? The Bratcam has a lot of metal, and it is situated above--though rather far above--a bunch of electrical cords carrying full wall power. Perhaps more importantly, the Bratcam has a number of data and low-voltage cords running to it, which, even when disconnected at their distal ends, may be crossing or lying near these wall-power cords. When the wall-power cords are disconnected from wall power, the tingle goes away. When they are reconnected to wall power (not reconnected to the Bratcam), the tingle comes back. I'll repeat and emphasize that in both the wall-power connected and wall-power disconnected cases, the cords running into the Bratcam are unconnected at their distal ends. So they should not be getting power directly, but could well be getting it through inductance and carrying this induced charge to the Bratcam. I only barely comprehend electricity, but this sounds like induction to me.

If you see holes in my thinking, please point them out! I love to learn, and value being corrected when wrong.

You asked if the tingling lasts a long time. If my memory is correct--and it might not be, as I didn't take notes--it lasted for a few seconds to a minute or two after disconnecting wall power from the nearby electrical cords. And it seemed to take a few minutes for the tingling to recur after reconnecting wall power to the nearby cords. My thought was that the rig was exhibiting both inductance and capacitance. But without good notes, this thought should be taken with a large grain of salt. As my scientist friends tell me, if an experiment isn't documented in a lab notebook, it didn't happen. A couple of solid tests suggest themselves, and perhaps one of these days, I'll conduct them. In the meantime, the issue has not interfered very much with my photography.

--Chris

jin
Posts: 107
Joined: Sun Aug 26, 2012 6:35 am
Location: Singapore

Post by jin »

Just gotten an email reply from Cognisys, sharing for your info if you experience the same issue :-)

Code: Select all

Paul: Some leakage is normal, however the leakage from your adapter seems excessive.  I would recommend grounding the StackShot with a wire (not a resistor), this will prevent any issues and keep everything at zero volts (ground).   This can also happen if you have several different electrical components plugged into different outlets and there is a difference in potential between them.  Either way grounding the system is definitely the safest way to go.
g4lab: my understanding is that ac adapter employed high frequency switching technique or PWM to eliminate the traditional iron lamination core transformer to reduce the adapter size a lot. Instead, high frequency ferrite core transformer is used and this is also acted as an isolation transformer between the AC primary to the secondary circuit. so the leakage current I believe is due to the parasitic capacitance residing between the transformer pri-sec coil/windings - look like something unavoidable. In fact I checked all two pins ac adaptors I have at home, they all have this ground voltage, but the voltage level are all different :-)

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

Yes sometimes there are ferrite core transformers doing some isolation and sometimes also blocking capacitors. And sometimes, apparently, not.

Yes they are smaller and lighter and cheaper. Yes they can more easily eat any input voltage and frequency. (Not that THAT was a problem)
But they are not repairable usually. And they don't keep lightning ( a big consideration here in Tornado Alley) out as well as a power transformer.(which won't always either but often will)
The only guy that can fix them is the guy that designed them and he never works for the company anymore and also half the semiconductors in them are frequently customs. All the leakage considerations it old fashioned power supplies were worked out shortly after WWII. Switchers also dump high frequency noise back into the AC line which can bother other equipment. (especially sensitive gear in a science building) I dislike them and don't think anything will change my mind about that.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Both jin's and g4lab's descriptions sound correct to me.

For the particular model of power supply I tested, I searched the web and found some more specifications. Mine is a model FJ-SW1202000U, input 100-240 VAC 50/60 Hz 0.6A, output 12 V DC 2000 mA. There is a data sheet HERE that says

DIELECTRIC STRENGTH Hi-Pot
Primary to secondary, 2000Vac/5mA/5s

Insulation resistance:
Primary to secondary :10MΩ Min at 500V DC

I don't have a circuit diagram for that particular power supply. Details definitely vary between designs. The PC power supply diagrammed in Figure 5 HERE appears to have nominally complete isolation between input and output. The adapter HERE has not quite complete isolation because if the external ground connection is removed, there are a couple of small high voltage capacitors explicitly connecting the output back to the input. That circuit is nominally balanced between two opposite phase inputs, but due to component variations the balance would not be perfect.

The DC supply inside the LED lamp that I disassembled is completely different. (See HERE.) In that case the connections from input to output are much more direct and the unit is clearly designed to run inside a well sealed insulting enclosure. When plugged direct into the wall, one output connector measures solid at 230 VDC above ground and there's a big filter cap on it so I would not like to touch that beastie by accident.

--Rik

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

One of the design principles of switchers seems to be to buck the voltage up often considerably higher than you need it and regulate it downwards to what is required using a variety of filtration and regulation schemes.

This is why it is recommended that the average person never try to repair a PC power supply. Even though the voltages supplied to the computer are 3.3volt 5 volt and 12 volt in plus and minus polarities, there can be capacitors charged to several hundred volts and the circuitry seems to proscribe a bleeder resistor as is done in flash circuits. I have seen repeated safety advisories recommending that they not be opened. Exercise caution.

soldevilla
Posts: 684
Joined: Thu Dec 16, 2010 2:49 pm
Location: Barcelona, more or less

Post by soldevilla »

I had similar problems some time ago, even my Stackshot stopped working and I had to reset it. I put a wire from the mass of the power plug to the grounding pin of the plug from the wall and have not had any more problems.

Post Reply Previous topicNext topic