Balloon cutdown device

[mikeselectricstuff]

In terms of battery power to weight. a small lipo, with no protection circuit, can do an amp or two, which should be enough, but bear in mind battery performance can degrade very significantly at low temperatures - different chemistries will doubtless behave differently. Sticking the battery in a lump of polystyrene would probably be a good start.

[nctnico]

The best thing to use is a wirewound resistor. This way you can get the most power from the battery by matching the voltage to the load (getting the best power transfer).

[IanB]

Second you need to look at the datasheet for your nichrome wire and they should have a current vs temperature graph inside, or the numbers that you can use to calculate it. It's very similar to the video Dave put out about resistor self heating. From there you just use a constant current source to regulate the current through the wire to your target temperature and away you go. We used a mosfet to do it, but I don't see why a common 3 terminal linear reg of appropriate current rating wouldn't work equally as well.

I just did an experiment with resistance heating of different thicknesses of steel wire. My results suggest a constant current source is a bad idea. A regulated voltage source pre-set to the calculated voltage would work better.

The reason is the way resistance increases with temperature in a metal wire. Consider any small portion of your wire. The heat dissipated in the wire for any given small section under consideration is given by W = I²R, where R is the resistance of that local section of wire. Since we have a constant current source I² is constant and the power dissipation is therefore proportional to R.

Now suppose that one part of the wire gets hotter than another part (maybe due to a kink or bend, or due to a local narrowing). The resistance of this part will go up, causing the local power dissipation go to up also. This increased power will make that part of the wire get hotter, and so the resistance will go up even more. Very soon (if the constant current source has enough oomph behind it) the local portion of the wire will glow yellow and break.

With a regulated voltage source this won't happen so much, since the increased resistance of the hot wire will reduce the current and counteract the local heating effect.

The effect is worse with thin wires, and much less of a problem with thicker wires. Thicker wires require much more power to melt and they are likely to saturate the constant current source before the voltage can get that high. Also the thicker wires will conduct heat along the length of the wire much better, evening out local hot spots.

Unfortunately thin wires are more suitable for limited power supplies. A thin steel wire (say 32 gauge, I didn't have the exact measurement) glowed red hot and melted at 0.75 A. A thick steel wire (24 gauge) was barely hot to the touch at 5 A.

[jimmc]

Would Nitinol wire (Shape memory wire) be of any use to you?
Lots of hits on Google eg http://www.imagesco.com/articles/nitinol/01.html
It can make an actuator much lighter than a solenoid http://www.imagesco.com/nitinol/files/NM70Super.pdf

Jim

[MikeK]

This 40 AWG nichrome needs at least 8V @1A to get red hot.  My 5V power supply @3A will get it hot, but only just.  My variable 1A supply gets it red hot at 8V.  I need to put this in a fixture with different lengths for testing.

Ian's tests suggests that steel wire gets hotter at lower amps.  So nichrome's main benefit is that is withstands repeated abuse.

[IanB]

This 40 AWG nichrome needs at least 8V @1A to get red hot.  My 5V power supply @3A will get it hot, but only just.  My variable 1A supply gets it red hot at 8V.  I need to put this in a fixture with different lengths for testing.

Ian's tests suggests that steel wire gets hotter at lower amps.  So nichrome's main benefit is that is withstands repeated abuse.

The temperature a given wire reaches depends only on the current, not the voltage.

The situation is best analyzed by considering a unit length of wire.

For our unit length, the power dissipated is given by the formula:

W = I²R

Where I is the current, R is the resistance per unit length (at the operating temperature), and W is the power dissipated per unit length.

We can see that for any unit length of wire, the power is proportional to the square of the current and directly proportional to the resistance.

Now the resistance per unit length is proportional to the cross-sectional area of the wire. (And temperature: resistance goes up with temperature, but if we are designing for a particular temperature we can assume the resistance to be the value that applies at our working design temperature.)

The cross-sectional area of the wire in the AWG system doubles with three steps in wire gauge. So for example, going from 40 AWG to 43 AWG would double the resistance per unit length.

So where does voltage come in? This depends on the length of the wire. If you halve the wire length you halve the total resistance and therefore halve the voltage required to obtain the same current.

Suppose you need 8 V to make 1 ft of wire get red hot. Then you need 4 V to make 6 in of wire get red hot, 2 V for 3 in, and so on.

Combining all these rules together allows you to determine the wire length, the wire diameter, the required current, and the required voltage to make your wire perform as desired. (You can assume that wires of different but similar thickness, within reason, will reach approximately the same temperature when subjected to the same amount of heat dissipation per unit length.)

[IanB]

Just as a quick point of reference, I did an experiment with some 28 gauge steel picture wire, some polypropylene cord, and a "hot wire cuts through cord" arrangement.

I found that a 2-3" length of wire was able to cut through the cord when powered by a 1.5 V AA alkaline battery. It required about 3 A to make the wire hot enough and the alkaline battery could supply this, although an NiMH cell like an eneloop would be more reliable. Note that the wire doesn't need to get red hot, not even smoking hot. Polypropylene has quite a low melting temperature.

[IanB]

Based on previous discussions in the thread, here is a picture of a "T" shaped cutter concept. It seems logical to me and it comes rapidly to mind if I think about how to engineer such a system.

The T shaped rigid tubing provides stability and protection, and prevents the cord from pulling on the wire and snapping it.

The spring loaded cutter wire ensures the hot wire will pull through the cord and separate it when activated.

The hot wire can just be cheap ordinary steel wire of an appropriate thinness. It is fixed to the electrical terminals using small bolts and compression washers. To install a new wire you just hook the cord out, feed the wire through the loop in the cord, and reassemble.

[MikeK]

Excellent!  I had trouble visualizing the arrangement, but now I understand it.

[T4P]

You can use a very thin nichrome wire or a steel wire ( of course, 32awg or similar, wirewrap wire or the sort ) with a big arse D cell.
Those $3 wire foam cutters you buy off the shelf usually use this arrangement but always remember to throw it in a block of polyprop

[mikeselectricstuff]

Now the resistance per unit length is proportional to the cross-sectional area of the wire.

And temperature. This can be a significant issue in heating applications.
For a true constant current drive, power will increase with increasing temperature, and chances are a hotspot will form somewhere.
For constant voltage, power decreases with temperature, so it is self-regulating.

[IanB]

Now the resistance per unit length is proportional to the cross-sectional area of the wire.

And temperature. This can be a significant issue in heating applications.

Indeed, it is surprising how much the resistance goes up even with seemingly small temperature increases. When I turned up the voltage the temperature increase dropped the current almost back to the initial value. Hot metal wires (or bulb filaments) make somewhat effective current regulators.

For a true constant current drive, power will increase with increasing temperature, and chances are a hotspot will form somewhere.
For constant voltage, power decreases with temperature, so it is self-regulating.

Very true. In my experiments constant current drive was a disaster. As soon as a hot spot formed the wire melted and broke at that point. With constant voltage I could get the whole wire to glow evenly red hot.

[MikeK]

You can use a very thin nichrome wire or a steel wire ( of course, 32awg or similar, wirewrap wire or the sort ) with a big arse D cell.

Well, a D cell is bigger and heavier than I'd like.  I might wind up with the 3V photo battery (1/2 AA size?).  It's small and light, but I need to try a smaller piece of nichrome as I'm not able to get a 4 inch section to heat up.

[SeanB]

You must have a dead toaster around, an ideal source of nichrome wire. That or an old kettle element ( a little messy but a nice finely coiled wire inside, even has an insulating powder).

[jerry507]

You're absolutely correct Ian about the local hotspot issue. The reason it's not a problem in the application is that it's completely UNNECESSARY to reach the red hot temperatures. In fact this is quite bad. With nichrome wire, as opposed to steel wire, you can get the wire quite hot in absolute temperature and the metal won't glow. Nichrome withstands higher temperatures without corrosion or fatigue hence a longer lifespan and why it's commonly called resistance wire. If you've heated the wire up to red hot then it will become more brittle and it will fail faster.

As I mentioned, we use a CC source to control the temperature more easily because at normal operating temperatures the wire won't be red and won't form excessive hotspots. Since you're regulating the largest term of the power dissipation, you're more readily controlling the power dissipated and therefore the temperature rise in the wire.

If you're only planning on using the wire once, it's still not quite as reliable, but not terrible. We use the same device many times because we found that it it was more reliable to rig it up and do a really good job inspecting it, and then not mess with it much.

[MikeK]

Wikipedia is wonderful!

The attached image is from http://en.wikipedia.org/wiki/Nichrome

[ejeffrey]

With nichrome wire, as opposed to steel wire, you can get the wire quite hot in absolute temperature and the metal won't glow.

Red hot is essentially the same temperature for any material. It is quite acceptable (but unnecessary in this application) to heat nichrome to red hot, as you can see from any toaster.

[kaptain_zero]

Here is a typical small Lipo pack, 3 cell so 11.1V nominal and rated for 4.5A draw, holds 180mAH and weighs 19g which could be reduced a little bit by removing the connectors if making it single use.

http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=23339

Regards

Christian

[jerry507]

With nichrome wire, as opposed to steel wire, you can get the wire quite hot in absolute temperature and the metal won't glow.

Red hot is essentially the same temperature for any material. It is quite acceptable (but unnecessary in this application) to heat nichrome to red hot, as you can see from any toaster.

Yes, you're correct. What I was trying to say is if you get nichrome red hot the chromium in the wire prevents the wire from oxidizing. In addition, the melting point is much higher which isn't exactly important in this application. If you're looping the wire though, you're going to create significantly more hot spots as you're working the metal to that shape. In any case, you don't need to be anywhere near red hot to melt the materials we're typically looking at. Dacron will melt at between 200 and 250C IIRC, and red hot is usually around 700C IIRC. Obviously there is a huge gap there. The lower temperatures cause significantly less fatigue to the wire and it will last longer as well as be more reliable. So from that table from Wikipedia you're looking at around 1A for a typical size wire to get to the necessary temperatures. It's very reasonable.

[chrome]

Why not use something like this:
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=13365

It's made for that and can be reused super easy, just add a small servo and done, you don't need all this temperature stuff just give it a position and bam done.