I'm designing a capacitive discharge rocket launcher. The reason for the cap discharge is to ignite a series of many devices at once. I did not want to switch the HV with the momentary pushbuttons alone so I decided to drop down to 5v and use transistors. For simplicity I only show 4 buttons. I wanted to add continuity LEDs to show a connection to the firing ignitor and I've placed them in the schematic as shown, but this will not work with the transistor. I'm stumped and need some guidance.
Please note that this is a work in progress and some areas in the circuit still need brainstorming. I'm still relatively new to designing circuits so any help or suggestions with any part of the circuit would be appreciated. If something looks wrong, please tell me.
Best Answer
Eeeek! I'll say that won't work.
1) Your LM311 needs to be powered from 12 volts. So does one set of resistive dividers into the LM311 (in order to set a reference to compare the divided 100 volts to), and so does the "Capacitor Ready" LED. Trust me, an LM311 output stage will not withstand 100 volts. Also, you need a fairly large resistor from the LM311 output to the + input, in order to introduce some hysteresis and get a clean "ready" transition.
2) Your high side switch transistor will not work with a 5 volt converter. The base must be driven higher than the emitter. In this case I suggest you use a 100 volt DC-DC converter with an inhibit input, and drive that.
3) Your low side switch transistor must handle at least 100 volts, and needs a base resistor to the key switch.
4) The 1N4005 will only work if the DC-DC converter has a current limit of less than 1 amp.
5) For the same reason as part 2, your firing transistors will never turn on, although their collector-base voltage rating is high enough that they won't burn out your debouncer. I suggest a topology like
simulate this circuit – Schematic created using CircuitLab
You do not specify either your squib resistance or your peak firing current. Without these, the values of R2 and Q2 cannot be determined. Q2 must have a pulse current capacity equal to the peak firing current, which can probably be approximated as 100 / Rsquib. This will probably be a fairly hefty transistor - for instance, if your original circuit actually worked, the KHS50 only has a pulse current capability of 2 amps. To drive Q2, R2 needs to be about 10 times Rsquib. Not only that, since there is no way to guarantee how long some bozo will hold down the button, it will need to be rated for a lot of power, as it may have to withstand 100 volts indefinitely.
6) Your continuity LEDs need to be tied through high voltage diodes (1N4005s will work) directly to the squib drive lines. This will affect the current limit resistor values, as the extra diode drop needs to be taken into account. When the squib fires, the peak voltage, ~100 volts, will attempt to back-drive the LEDs, and this is A Bad Thing - hence the diode.
7) Finally, all your DC-DC converter - outputs need to be tied to ground.
There may well be other problems, but that will get you started.