The Dangers of Working with a High Voltage Supply

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
:) The Pyramid supply would be for the SLA's in place of the current charger. For Li-ions I will stick with my WF-139 or nano :)

Good thing I learned my lesson with a relatively small cell that didn't explode... had it been one of my 18650's the result would have been much worse I'm sure. No more charging Li-ions in anything less than a charger built for them
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
Theoretical understanding and practical application are very different.
Regardless of the exact cause, this reads a bit like a horror story. Beside the shock thing, the first 'accident' was spiking the cell above 12v. I can't ever see calculating charge times before hand to be a viable, long-term 'free lumens' plan. Maybe I'm too ignorant of this area to see the deeper implications, but i'm completely unconvinced by any of the 'safety' measures described.
Some places engineer for 4 levels of safety, each action in a sequence works to protect you during the subsequent steps. Even if a single step fails completely it is so contained by the previous actions that it is not possible for any harm to come. A person would have to fail three procedures before the next step even exposes them to personal risk. Ideally the system is designed that you may not proceed unless the previous safeguard has been successfuly initiated. In this ideology, you have to perform a complex set of actions to successfully put yourself at risk.
In your current set-up, you have to perform a complicated set of actions to remain safe.

Way too true... my setup was the result of rushed work and ignorance into how dangerous the setup was... no excuse there.

In an ideal setup (and what I will do from now on if I need to work with these voltages again (aka when I need to))


*Operate the supply with thermocouple attached to charging medium to prevent overcharge even in case operator is distracted. This would be level 1. If I had this the supply would have shut down once the cell started bleeding off the extra charge.

* wear electrician gloves... I have 1000volt ones siting here :shakehead If I wore them the whole thing minus the ruined cell would have been avoided... but then maybe I wouldn't have learned as quickly. The reason I didn't put them on (I was too embarrassed/ashamed to even say I had them) was that I looked at the voltage and just reacted, not looking behind me to grab the gloves, a pair of pliers with 2000v insulated grips...anything which would have helped). I have the equipment but I was scared I didn't want a lithium fire as I have no class D extinguisher and don't live alone. This would be the second line of defense (the gloves)

*put a quick-blow fuse inline so that even if the leads are shorted through something the power cuts instantly. Right now the supply shuts down if shorted, however, a very fast acting fuse would be safer, plus it's very easy to install such a fuse. That's level three.


*Use one hand... this is hardly a level but a great way not to risk your life :)

I found three steps I could have taken which wouldn't have required a heck of a lot of setup and would all have prevented this.

Also, I calculated the charge time (not that it really matters but interesting to look into my planning) assuming 100mah capacity left in cell, 750mah full capacity, charging efficiency of 99%, Vin of 4.00 volts average and Iin of 300ma. I guess I could have integrated the whole thing as a function of voltage (which would rise), however I was lazy and assumed 4 volts. This means the cell is receiving

1.188 watts / hr of charge

The capacity of the cell is 750ma at 3.7 nominal or 2.775 watt hours

assuming I need 650maH to fully charge that is 2.405wh needed / 1.188 watt/he = 2.02hrs of charge time. Total charge till accident was around 1hr 15 minutes... well within what I calculated

Where I went wrong is that I assumed 100maH left because I ran the cell to cutoff, however this was at a rate of 1c and thus the cell actually had around 20% capacity left (assuming 3.3 volts is dead, 3.8 is 50% charge and 4.2 is 100% SOC). Add to that the cells are 2 years old and probably don't have 750mah of capacity full and there is the problem.

... but what really did it was that as the cell warmed up the internal resistance of the cell decreased which consequently increased the charge voltage, which at constant current meant more watts which heated the cell more... positive feedback loop :-(

Anyways, that's my long explanation justifying my stupidity :D

TYDR: I'm going to use my 0-15 volt supply to only charge SLA's, the lithium batteries stick to the lithium chargers :p
 

LuxLuthor

Well-known member
Joined
Nov 5, 2005
Messages
10,409
Location
CT
:) The Pyramid supply would be for the SLA's in place of the current charger. For Li-ions I will stick with my WF-139 or nano :)

Good thing I learned my lesson with a relatively small cell that didn't explode... had it been one of my 18650's the result would have been much worse I'm sure. No more charging Li-ions in anything less than a charger built for them

That makes me happy. Just to be clear, there are those knowledgeable and experienced users who can and do use various power supplies to charge Lithium Cobalt cells, and they may feel perfectly safe and confident. Perhaps outside on a concrete slab...hoping it doesn't rain that day.

Even if I thought I had 100% Triple Dog Dare safety procedures in place, there is still that Murphy's Law that can completely distract you away from following your procedures. Like you see a dog attacking your daughter. Turning off the power switch would be the last thing on your mind.
 

Linger

Well-known member
Joined
Feb 17, 2009
Messages
1,437
Location
Kingston ON
Li-ions I will stick with my WF-139 or nano
:thumbsup:That is a very intelligent decision, demonstrating theoretical knowledge and practical sensibility. Check voltages before your charge, use protected cells in a safe charger, know when to expect termination and monitor tempurature. An appropriate use for the proper device, brilliant decision on your part, because out of the whole range of possible options, from thowing a bare wire over a high-voltage tower to connecting it to a running alternator, you know enough to pick the safest most reliable method. That, I think, is how to be a show-off.
 

ctom

Member
Joined
Mar 26, 2009
Messages
10
Location
San Mateo, CA
D-Dog,

First, I'm glad to hear your injuries were relatively minor - you are one lucky dog! Secondly, I'm dismayed the scenario that I described in my second post came to pass...

I suppose it's all academic now since you are no longer using the power supply but just to close the loop on the constant power output question, I reviewed the schematic and came to the following conclusion on the operation of the power supply:

The supply will sample the voltage and current at the output and multiply the signals via the MC1494L. This provides the actual power (V*I) being delivered by the power supply and this signal is fed to an opamp. The opamp "calculates" the difference between the actual power being delivered to the load and the desired power level setpoint. If there is a difference, it will turn on (or less on) the series pass transistor (Q15) to modulate the load current to minimize the difference between the actual and desired setpoint.

Practically speaking, the supply will output the highest voltage (500 or 1000 volts, depending on the user) and control the current until the target power level has been achieved. In terms of charging batteries, I don't know if batteries have sufficient creepages and clearances to withstand these voltages without arcing.

As an aside, I noticed you mentioned you had high voltage gloves you could have used to protect yourself. Don't be so sure! Back when I was designing 5 kV power supplies, we were issued 20 kV gloves for lab work. We were required to test them by fitting them onto metal hands and hi-potting them. I was always surprised at how often they failed due to cuts, punctures and nicks that were not always visible to the naked eye.
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
Thanks for closing the original question, it is very appreciated even though I won't be using the supply anymore :)

Wow, I thought I could trust the gloves although what you state makes perfect sense, one crack in the rubber and you have a problem. Perhaps the leather exterior with rubber interior gloves are safer? I understand neither are failsafe although something is sure better than nothing, right? Just another reason why redundancy is the only true answer.

As I continue to look at what happened (I can remember the moment like it was a few seconds ago still), the reason I only got surface wounds with no scarring of the deep tissue/ why it didn't hurt worse was that I disconnected the lead connected via conducting magnet to the anode of the battery. This brought the output voltage to 537 volts, however, when I shorted through my hand to the anode of the battery, the voltage immediately dropped back to 12 volts (actually a little less than this because the protection circuit was bleeding off charge). Thus, I was only exposed to whatever the caps has stored in them at the time, not continuous 500 volts from the supply.
 

PeAK

Well-known member
Joined
Jan 30, 2009
Messages
238
.
.
... I also have a Pyramid PSU which is 0-15 volts (for the SLA's) so from now on I'm using that, getting rid of this one on Monday.
.
.
.
Does your Pyramid supply have a current limit ? If so it is possible to configure it to charge your Li-ion cell safely.

PeAK
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
Nope, only a pot to control voltage which is why, especially in lui of what happened I'm not even considering it. In order to charge lithium cells I would need CC for bulk and then CV with a thermocouple and be present as the minimum... even then being present with a charger built for Li-ions is the best bet :)

The Pyramid is a great supply, however, it's only CV variable current from 0-25 amps... For SLA's I match the output voltage to the current voltage of the battery and then adjust upward until I'm charging at where I want to be. Quick and easy way to charge the cells plus as long as you have the no load calibrated correctly no chance of blowing anything up. Of course, being present is good too. I really hope I can obtain a CC/ CV charger for the sla's soon... that would be even safer :) Most of the chargers I come by, including the latest are surplus so i'ts only a matter of time before I find a CC/CV one...
 

Alan B

Well-known member
Joined
Nov 19, 2007
Messages
1,963
Location
San Francisco Bay Area
Glad the injuries are minor.

Power supplies are not designed to charge batteries and so don't automatically have a diode to protect them against back current. Switcher power supplies often have a diode in them and may be compatible with batteries, but this is not always the case.

Constant power mode with this supply is most likely not going to generate significantly more current than the current limited mode. The constant power feature will only work in a limited range of the supply's output range.

Anyone working over 50 volts professionally (in the US) is required to follow NFPA70E work procedures, though a lot of folks haven't learned that yet.

There must be barriers between the body parts and the high voltages, tools must be insulated, etc. If there are no barriers then voltage rated gloves that are periodically tested and certified must be worn. In no case can body parts be allowed to come into contact with the circuits. There must always be a barrier of some type (gloves, plastic, closed panel, etc) between body parts and energized circuit parts.

We all used to work without these safeties, but it is now generally a violation of workplace safety rules to do so. At home there are no such requirements, but not following good work practices can lead to injury or death and periodically does.

This high voltage power supply is completely inappropriate for charging these batteries.

Thanks for sharing your story, hopefully it will save others from potentially serious or fatal accidents.
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
Still thinking about the stupidity and decided to figure out what actually went through me.

The supply was charged to 534 volts at the time I made contact. Current the supply provides I "believe" is irrelevant because by I = V/ R the body will only allow a certain amount of current to pass through. Thus the current kills you but you need a potential difference to "guide" the current.

Resistance between my two fingers very generously was 125000ohm (more like 300k to 400k) after taking a shower (recreated everything except actually touching the supply/ hooking up a battery obviously) To get the voltage measurement I connected my 2000 volt test probes through 1000 volt gloves and turned the supply to 500 v CC mode after I made sure the leads were fixed in the supply (aka my hands weren't even in contact... almost 100% safe :) ) Voltage spiked at 534 volts and stayed there until I shut down the supply.

Using Ohms law rearranged we get I in amps =534/125000 = 0.0042 or 4.272ma. The resistance must have been lower than this though because I can't imagine that small of a current even making a blister. I did have muscle control in my hand at the time so it must have been below 60ma but above 10ma. Perhaps after the current penetrated it lowered the resistance of the skin? Interesting to note is that the resistance between the blisters is now 25 million ohms :) My body put up its own defense mechanism it seems :)


Edit: taking another shower now... perhaps I have to take a longer one to allow my ody to retain more water. A current of 35 ma would be possible initially if my resistance was only 15000 ohms

Not to decrease the severity of what happened, however, I'm just curious where/if I'm going wrong with the math. What I did was stupid no doubt, however, did I really put my life in danger (forget about the 1 hand rule for a sec because that will save you in most cases (minus the hand of course :p ))

Please note this changes nothing... supply going back tomorrow and no more playing around with it for me :). I just want to know what went through me... because it hurt :D (But I was 100% fine the second the voltage dropped (aka right after the initial capacitor discharge). I just want to learn as much as I can from the experience ;)
 
Last edited:

LuxLuthor

Well-known member
Joined
Nov 5, 2005
Messages
10,409
Location
CT
There must be barriers between the body parts and the high voltages, tools must be insulated, etc. If there are no barriers then voltage rated gloves that are periodically tested and certified must be worn. In no case can body parts be allowed to come into contact with the circuits. There must always be a barrier of some type (gloves, plastic, closed panel, etc) between body parts and energized circuit parts.

What Alan is delicately trying to say is make sure you are also wearing a condom. He doesn't like to talk about his "accident." LOL!
 

PeAK

Well-known member
Joined
Jan 30, 2009
Messages
238
Still thinking about the stupidity and decided to figure out what actually went through me.

The supply was charged to 534 volts at the time I made contact. Current the supply provides I "believe" is irrelevant because by I = V/ R the body will only allow a certain amount of current to pass through. Thus the current kills you but you need a potential difference to "guide" the current.
Current is the flow of charge but the voltage is a measure of the energy of each charge. The current still kills but is able to surmount barriers (i.e. resistance) when there is higher voltage/energy-per-electron.

Static from a balloon can have high voltage but the reservoir of electrons (capacity/capacitance/reservoir) is low. The key is the energy exchanged given by

E = q * V

The voltage can be high but if the number of electrons is low, then the Energy (E) will be low...as in your case.

Now take a capacitor from a flash than can take 500V.
If you charge it up, the charge will be

q=C*V where is "C" is a large number (unlike your power supply due to the current limit). The energy will

E = q * V = ( C*V) * V = C*(V**2)




D-Dog said:
.
.
.
I just want to learn as much as I can from the experience ;)

Due to the current limiting, the charges transfer over to you faster than they can be replenished by the power supply and the result is that the voltage drops during the transfer to much less than 500 V( V=q/C...q on the power supply cap goes to "0" and the voltage drops toward "0"). The voltages stored on you due to running across a carpet with plastic shoes can be very high...perhaps even higher than 500V. Hopefully this clears up your experience.

PeAK
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
What Alan is delicately trying to say is make sure you are also wearing a condom. He doesn't like to talk about his "accident." LOL!

Now that's an accident that will stay with you a while :grin2:

I guess the rule, "always remember your rubbers" work for electricity too although a condom probably isn't rated to too high a voltage... I wonder if a static shock could get through :faint:
 

Alan B

Well-known member
Joined
Nov 19, 2007
Messages
1,963
Location
San Francisco Bay Area
Just paraphrasing some of my training.

The resistance of the body is pretty low, only the skin has high resistivity. People have died from single digit voltages in cases where the skin was penetrated.

Water soaked skin has much lower resistance. 120V and water proves to be very lethal.

If I remember correctly it only takes about 5-7 milliamps to be fatal if the current path disrupts the heart. It just depends where the current goes. If it goes from hand to hand or hand to foot it may traverse the heart. It will always follow the path of lowest resistance.

Burns are caused from heat which is calculated by voltage squared over resistance. 200 watts is clearly possibly with this supply and if that is concentrated in a small volume it will cook quickly.
 

Alan B

Well-known member
Joined
Nov 19, 2007
Messages
1,963
Location
San Francisco Bay Area
I learned today from an electrical safety engineer that the skin resistance breaks down at about 400 volts and drops to a much lower level. This may be a factor here - the resistance measurements used to try to understand the incident were not as low as were experienced in the actual event due to the test voltage.

I do NOT suggest taking new resistance measurements at 500 volts to increase our understanding further. :poke:
 

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
I do NOT suggest taking new resistance measurements at 500 volts to increase our understanding further. :poke:

Perhaps 1000v then? :laughing:

Anyways that makes sense and explains what happened better. I also learned that spilled coffee can cause third degree burns in 2-7 seconds so perhaps I should consider myself lucky :)

The blister is only a callus now and should be gone by next week, however, the memory is for forever. Now I notice myself taking the extra few steps to make sure everything is safe, even if the voltages are harmless. I guess in a way I'm glad it happened :grin2: (only because there was no lasting damage, etc. Not suggesting this is a fun weekend activity... then you are a candidate for blisters/ Darwins)
 
Last edited:

jtr1962

Well-known member
Joined
Nov 22, 2003
Messages
6,206
Location
Flushing, NY
Nobody mentioned it but I should point out that there is a huge difference between AC and DC with regards to lethality. Not to minimize the dangers, but high-voltage AC is much safer to be around than high-voltage DC. Given a choice, I'd rather get a shock from 120 VAC while changing an outlet than from a 30 VDC power supply. First off, AC tends to stay on the surface of your skin rather than penetrating into your body. The higher the frequency the less the penetration. Back when I was in college I accidentally shocked myself on a group of transformers I had wired in series for something like 800 VAC. There was a nasty jolt, but no blisters or other damage.

Second, DC will lock your muscles, possibly preventing you from removing your body from the voltage source. AC won't. You'll generally experience that tingle and reflexively pull away as I have numerous times (due to the old house wiring and lack of knowing which breaker controlled which line, I've gotten shocked with 120 VAC while wiring more times than I care to admit). I now wear gloves when doing any live 120 VAC electrical work, although even this is no 100% guarantee against getting shocked as wires have gone right through the gloves.

The most vivid way to highlight the differences between DC and AC are to examine railway accidents. Overhead catenary is generally thousands of volts AC (in the US 12,000V and 25,000V are common). In fact, the reason it's placed overhead is precisely because the voltage is high enough to jump through a foot or more of air. Now one would think that coming into contact with this much voltage should be uniformly fatal, and yet quite a few have survived, some with remarkably little injury. Why? The current passed mostly on the surface of the body. To be sure, in almost all cases there are horrific burns on the surface. But unless contact is prolonged there will be relatively little internal organ damage. Note the term "relatively" here-don't bother trying this for real as it will still likely get you killed. Besides that, you'll be trespassing on railroad property if you do (RR police might well be worse for your health than overhead wire).

Now let's look at third rail. Generally, third rail in the US in "only" around 600VDC, perhaps as high as 1000VDC on some newer systems like BART. Seems less dangerous than high-voltage catenary, doesn't it? Nope-it's worse despite the fact that the voltage is a fraction of that on the overhead wire. Touch it by mistake and you may never be able to break contact. And while you're in contact you'll be cooked from the inside out by up to a few thousand amps (do the math for how much power that comes to). Unfortunately, the circuit breakers generally won't trip because a body will draw less current than a train normally does. People touching third rail have been found dead with few external burn marks, other than perhaps at the point of entry/exit. Of course, how they look inside is a totally different picture. While the power supply in question was limited to only 200 watts, that's still enough to cause really nasty internal damage with any length of contact, even if it doesn't stop your heart. Think how quickly foods cook on a microwave outputting only a few times that.

In short, high-voltage DC is really nasty stuff. Avoid using it unless you can't do what you're doing any other way.
 
Last edited:

D-Dog

Well-known member
Joined
Nov 10, 2006
Messages
600
Location
NJ
Really, not to discredit what you said because it makes a lot of sense, however, I thought it was the other way around. I'm fairly sure the pulses of AC will lock muscles above around 60ma.

Here is an excerpt from another site which kinda confirms this:


"The first electric chair execution, held in upstate New York, used AC. Exactly why, I am not sure. It is true that AC, especially around 50 to 60 Hz, can paralyse muscles; it causes them to lock up, so that someone who accidentally gets hold of two bare high-voltage wires can find himself unable to let go. DC tends to cause a more `one-way' contraction. Either can kill, but without the `locked muscle' effect, it is easier to free oneself of a DC current. (An old electrician's trick, by the way, is to test a supposedly-dead circuit using the back of the hand, rather than a finger-tip. If the circuit is still energized, the hand will clench shut, because the finger-closing muscles are stronger than the finger-opening ones. The clenched-shut hand will tend to jerk away from the wire, rather than grabbing hold of it. It is an old electrician's trick because the young ones who use it and get shocked live to become old electricians....)"

I don't know why electricians wouldn't use a multimeter instead or test probes with a light... seems safer :)

I think the best advise is to avoid both :) Just stick to not doing dumb stuff (like using a 500v PSU to charge batteries) and you will be fine. That and the 1 hand rule... can't stress that enough especially after what happened.

Edit: and another site I read "Experience tells me that DC grabs, AC throws. A friend of mine suffers from the loss of the use of two fingers and two thumbs from a 12VDC lighting circuit (amperage unknown)"... perhaps DC is the more dangerous one. Either way I don't intend to test it for the forum :p
 
Last edited:
Top