• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Safety Practices, General and Ultra-High Voltage

One thing MORE

A bit of (simplified) Biomedical Engineering.
One of the most important thing mentioned above is the "Use one hand" method - that is clip the return lead on permanently - with the power off, then apply power and probe the various points using the other probe in one hand.

The philosophy of this is to avoid currents across the chest and therefore through the heart. Blood is conductive and major veins and arteries will often be the preferred curent path. It takes just 5mA of current to cause the heart to go into fibrillation. Fibrillation is when the heart is still beating but the sequence of muscle contractions is lost such that the approriate chambers of the heart do not contract in the right sequence to pump blood. The ventricles (lower chamber) must contract first to force blood into the auricles (the upper chamber). The auricles then contracts to force blood out around the body or through the lungs (depending upon which side of the heart you are talking about).

More Massive shocks will cause the heart to lock solid rather than fibrillate and usually (don't rely on it) the heart will then restart in sequence. (This is the principal of operation of a Defibrillator). This is why small shocks can be MUCH more dangerous than more massive shocks.

When a muscle contracts, a muscle cell polarises (Potassium ions move across the muscle cell wall). A continuous contraction is not possible. As the potassium is depleted the muscle cell gradually relaxes and the cell must "repolarise" before it can contract again. Holding a DC voltage stimulus will not allow repolarisation - the electrical stimulus must reverse polarity to allow the cell to repolarise before it can contract again .
A muscle cell takes typically 5 milliseconds to repolarise. If you get connected to AC however the muscle cell will be stimualted and then repolarise in a cycle of 50 or 60 Hz (depending on where you are in the world and the power system used). This make AC VERY MUCH more dangerous as you can actually grab an AC source and find it impossible to let go. In the DC case the muscles will relax after a second or so and you can let go. If less than the 5ms repolarisation time is allowed betwen stimuli then the cell will gradually deplete of potassium and the muscle will gradually relax. This is why aircarft system use 400Hz - As well as it being easier to transform it to other voltages (with a transformer of course) it is actually safer than 50 or 60Hz because a full 5ms repolarisation time does not occur between polarity reversals (ONLY 1.25ms).

The principal of earth leakage breakers/residual current devices is to trip the circuit so that you are exposed to one contraction cycle ONLY - you can get "stuck" on to the power source.

Any shock from DC or AC can upset the hearts timing system such that it goes into fibrillation. The greater majority of so called "Heart Attacks" are not the heart stopping but rather it going into fibrillation such that it cannot pump blood effectively.

If you want to know more about this stuff look up (check my spelling) "Sinoatrial Node" - this is the hearts natural pacemaker which fires contraction cycles AND "Bundle of Hiss" - this is the nerve system which conducts the electrical inpulses from the Sinoatrial Node down to the ventricles and then up to the auricles causing the appropriate heart wall muscles to contract in the correct sequence.

Hope this is of some interest to you. Its a VERY simplified explanation from what I remember from my Biomedical Engineering days BUT (hopefully) helps explain the risks from electric shock, explains why AC is more dangerous than DC and why small shocks MAY be more dangerous than more massive shocks.

Cheers,
Ian
 
I bought a direct coupling 300B amp recently, and modified it somewhat to fit my own taste....

When testing on the table, I touched the cap to feel the temp, then I got shock :eek:

Some electrolytic caps are finished with a plastic top cover & wrapped with heat-shrink on the side; some without the cover so the aluminum top is exposed. The one I touched is the latter. And maybe, the exposed top was connected to the negative leg.

That cap is the 1st C in the power filtering for the output stage. And the output stage in this circuit is stacked on the input stage V+, that is 167VDC.

I got several shocks before, this was the biggest.

Luckily, I survived.
 
Yes.... this was something that went back to my early days of the 1950's..... the electrolytics outer can was (and still is) not polarity defined. The plastic cover sheath is very thin and if can is clip mounted make sure to fit some extra insulating tape 0on exposed parts. PCB mounts make sure cans don't touch.
Modern day electrolytic cans are still isolated and slightly neg pole. Never take it for granted even with DMM checks, the electrolytic can react as a diode, fooling measurements.

richj
 
richwalters said:
Yes.... this was something that went back to my early days of the 1950's..... the electrolytics outer can was (and still is) not polarity defined. The plastic cover sheath is very thin and if can is clip mounted make sure to fit some extra insulating tape 0on exposed parts. PCB mounts make sure cans don't touch.
Modern day electrolytic cans are still isolated and slightly neg pole. Never take it for granted even with DMM checks, the electrolytic can react as a diode, fooling measurements.

richj

some electrolytics were also mounted on a phenolic wafer -- that is, if you didn't have a stamping press to make the mounting holes for the electrolytic you used this socket. the tabs at the bottom ot the cap were twisted and lock it into place and the capacitor tabs "grounded" -- if the ground wasn't perfect there could be a voltage differential between the capacitor and chasis ground.
 
Bleeder

A 100k/2W resistor across (each) cap in the power supply.

Look at the attached schematic (coming from Frank de Grove). The bleeder is the 100k resistor in the upper left corner...

Erik
 

Attachments

  • 12bh7a_srpp_wcf.gif
    12bh7a_srpp_wcf.gif
    8.2 KB · Views: 1,710
Hi Amirmk

bleeding resistor

Go to post 74 and 75

Those are the shematics of a power supply and an active linestage. The first schematic (post 74) shows the power supply. Note that after the diode bridge there is a 330uF cap and parallel to this cap there is a 1M resistor. This is bleeder one! the second schematic (post 75) shows the active stage, but look at the upper right corner. There is a diode coming from the power supply and connected to two 330uF caps. Parallel to those is a 100k resistor, which is bleeding resistor nummer 2! This is, I think, is the answer for your last question...if you have two capacitors parallel, one bleeding resistor is enough.

And why are there two bleeding resistors in this circuit? Because the capacitor after the diode bridge is isolated from the other two capacitors, after the regulator. When the cap is charged, but the cathode of the EL86 is cold (happens when you turn the amp off) this tube won't be conducing anymore...so power keeps storaged in this first cap (which is bad). The diode between power supply and active stage is there because you don't want current storaged in the 2 330uF capacitors flowing back to the power supply (could also happen when you turn off the amp). And when it can't flow back, it needs another path to ground...given through the 100k resistor. I hope I haven't said anu nonsense here...

Hope this helps

Erik
 
I've got a confession to make....with 34 yrs tube amp experience I actually got stung by a fixed bias supply (neg volts -80).....carelessness ? No.....Due to using smd components for a bias watchdog circuit I was forced to use reading glasses to see what goes where.......and forgot to turn the bias juice off.

Wearing glasses slows one down.....one sees infront and forgets the sides.

richj
 
Look at what you're wearing, what you're sitting on and what you're working on top of.

Lots of chairs and desks have metal components that will provide a pathway to ground, as do some items of clothing; like the metal buttons /rivets in jeans (Small contact area = very painful shock). Wearing shorts will increase the chance of such a pathway forming but no normal clothing fabric is good enough when you're working with 1000V+. The voltage may be able to cross the fabric alone. When you sit down you compress the fabric and make it easier. Any kind of sweat or moisture in the fabric (Socks, back of chair, seat) and it's easier again.

Such a pathway would negate the benefits of any one handed rule.

I only started thinking about this when I noticed the metal edges on a desk, on top of which sat a piece of equipment that regularly gave me shocks. Almost no desks are anywhere near strong enough to stand a good knock and I realised that if I had my legs under the desk I'd either;

a.) Not be able to get away from the shock, perhaps making it worse in the attempt by making better contact with the desk as I tried to stand up. Or...
b.) Be able to stand up etc but pull everything off the desk onto myself, perhaps with the desk following, again producing the risk of only making things worse.

Way to fix it? Put some carpet around your desk and chair. Choose a desk and chair that don't have any exposed metal components. If there are exposed areas, a piece of metal framework over your thighs for example, cover them over with foam lagging or something like that. Check the feet / casters of the desk and chair to see if they connect the frame to the ground, of if they could should you drop or leave something on them.

A lot of major accidents happen with a smaller accident that primes the second, e.g. you may get a small shock through you clothes that makes you jump, accidentally putting your hands directly on the circuit in an attempt to get away.

I remember reading a perfect example set by a guy who did something similar with a lathe. If you don't have high pressure coolant on a lathe, the swarf can spool up around the chuck. Temptingly, it tends to catch and sit still. One instantly finds oneself wishing to reach in and pull it away, despite the fact the stuff probably has an edge on it like a razor blade. Said guy did attempt to remove the swarf. Usually, as soon as you touch it, the whole lot catches on the chuck and starts spinning again. In the shock, not wanting to get his hand caught in the spinning swarf, he whips his hand away, which then gets impaled on the drill bit in the tailstock.
 
d:-....I see so many old fashioned good wooden chairs end up in the skip......they may not look trendy but they make excellent sitting insulators for general lab work. The worst are swivel office chairs full of metal, even a metal coat hanger fitted behind. I did a resistance check of the fabric of these posh chairs and they conduct. Avoid them.