But it does (just in a different way), and I suspect this is part of the reason. The resistance of each wire increases with increasing temperatures. I do now suspect that is the reason for the popping. But it is the power supply that provided an impedance (resistance to current change) that was the reason.
I am coming to the same conclusion that it is the power source and the varying resistance due to temperature of the NiCh wire. Either through induction, or as you also indicated that it was not really a constant voltage device but constant voltage at a given load. As the load decreased due to higher resistance, voltage increased which then caused the other wires to also heat to that high resistance point, causing the voltage to spike, overloading the array it total.
Though just to state this is pretty thick, thicker then I used, which may have been as thin as 36.
36 gauge nichrome wire is 26 ohms per foot. Max operating temperature of about 1000 C, melting at about 1400C. About a foot is 0.15 grams. (Sorry about the mixed units. )
Temperature coefficient is 450 J/kgC So you need of the order of about 100 Joules to melt a foot of wire. Ignore heat losses if it happens really fast. That is still a reasonable bit of energy. Say it takes one second. So 25 ohms, two amps. That requires 50 volts across the wire. 100 volts and it dies in about a quarter of a second.
This puts some bounds on what the power supply would need to deliver. With multiple lengths of wire we are into significant current demands as well as no small voltage. So much so that it starts to smell a lot like a direct connection to the mains.
Which gets us back to the power source. Whatever it was, it was not behaving well, and lethal dangers were present.
Wait, was the multiple rung setup using a single nichrome wire? If so, I’m thinking the temp diff between the rod-wound sections and the rung sections may have contributed to failure.
36 gauge is awful thin. I won’t have anything that narrow but I can assemble this anyway to see what happens. Do you have any idea of the size steel rods you used? Not sure it matters much but I have 1/4" and 5/16" available which would seem appropriate and much heavier stuff not so apt. Also what are the lengths of the wires, the shortest one and the next one to establish the angle of the rods?
I think the power supply will be significant. One thing that occurs to me if a Variac was used, if that shortest wire burns up there will be a voltage surge from the Variac. More so than at low power than simpler transformers because the Variac maintains high magnetic flux in the entire core even when only a small portion of the windings are active at low voltage. This allows variacs to output higher voltage than the input. That surge could be the start of a cascading effect as each successive surge blows out the next larger wire. I have no idea if there’s enough voltage in such surges to matter even with thinner wire. It’s one of many possibilities for a power supply issue, although I think a lack of proper fusing is the most likely reason. I am thinking the triac based variable AC supply I use most often is very stable and I’ll see wires burn out one after another as voltage increases.
These seem about right. Though like you I don’t suspect it makes much difference.
Longest NiCh span (top) about 16 in, shortest about 13 in (bottom).
I also suspect the power supply.
Everything was using a single NiCh wire, that single wire would be wrapped around the rods and go back and forth between them. With the single rung setup which worked the NiCh wire could glow as desired by adjusting the power to the point of popping it if I wanted to. When it glowed I noted that the wire section near the rod was cooler by less/no glowing which seemed to be due to heat transfer to the steel rods. I don’t suspect that was the issue as even NiCh space heaters demonstrate this ‘cool edge’ when in use. Even when a mid span or turn section of a space heater is supported by a spring to preserve tension it does not glow like the rest.
I am wondering if the power supply became unstable when the load was first applied. The initial load impedance is very low, resulting in a high inrush current. So am wondering if there was some oscillation going on with the voltage (overshoot, undershoot, overshoot, etc.) during the first second or so.
Well yes but not because of the paralleling of wires, and the change in resistance is up not down. So the opposite change to what @MrDibble was implying.
Nichrome has a quite low temperature coefficient of resistivity, of the order of 150ppm/C. That is about 50 times less than copper. The caveat is that this is at low temperatures, how it behaves near its melting point may be another matter. Overall however, its change in resistance as it heats may only account for say a 20% increase in resistance. I doubt that this has much to do with the OP’s experience.
Not that it matters for the “why” discussion, but what was the gradient across the air stream accomplishing?
It was to simulate different atmospheric conditions to see how it interacted with a rising column of smoke in a stack. The V shape was for normal atmospheric conditions as air cools as one ascends.
I don’t have much 20 gauge or higher wire. I’ll see what I can do. With 18 and 19 gauge. You think it was much thinner wire but I’m not sure it’s practical to use a much higher gauge. I know it could be a problem under tension but tension is not necessary here.
I was planning to start with just three wires and use a pot on the power supply that allows fine voltage control on the triac supply. If wire starts burning out I’m not going to waste it. If not, depending on what happens I have a Variac transformer to try out also.
I have a bunch of 36 AWG nichrome. And a variac. What was the desired experiment again? Get a 13" and 16" wire, run in parallel, and crank up the voltage until one goes poof?
He had more wires within that range. I think 2 or 3 would show any problems. There may not be any problems with a stable property fused power supply, and in that case I’m slightly interested in his original idea to create a rising column of air
Absolutely no excitement of any kind:
Wires were 13" and 16", 36 AWG. The short wire breaking had no effect on the long one. Voltage control via fairly beefy variac.
Not exactly a surprise.
Did you measure much? Would be interesting to know roughly what the temperature coefficient of resistivity is close to melting. Or even just resistance just on the point of failure.
Hooked up an oscope and measured voltage, current, and V/I (i.e., R). Still uploading, but it will appear here shortly:
The resistance definitely goes up as it heats, but not dramatically so.
Wow, that was great to see it finally working, and it worked just as I have hoped. To me this confirms it was the power supply and the concept was sound to produce the desire outcome. Thanks.
Did it seem like when the first wire popped the second one moved just a bit?