RC Multi Rocket Launch Ckt (Questions)

SgtWookie

Joined Jul 17, 2007
22,230
My objection to MOSFETs is their sensitivity to static electricity.

However, if put into the circuit with Zeners used as clamps from source to gate, and source and drain, they should be reasonably bullet-proof.

I've been eyeing these IRLU8721's for awhile:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IRLU8721PBF-ND
Very low Rds(on) @ 8.4m Ohms, rated for 65A, 30v, logic level gate, in a compact TO251-3 package. Gate charge is only 13nC's. Even a 4017 could snap them on pretty quickly.
 

Thread Starter

Rent2Own

Joined Sep 2, 2011
12
Because I cannot erase lines I've made with my crappy, free circuit drawing program (for mac) I could not go back and add safety switches and LEDs to indicate continuity and "armed". I will redraw the whole thing later to add them. (there are notes on the circuit)

The coil and switch symbols are all I could find to represent a relay. I realize it's kind of crowded.

The diodes, I read, are needed to protect the IC from induced current as the relay coil discharges. (Right?)
---------------
*Do the arrows in the debouncing circuit represent two inputs (+) from the same voltage supply? (Can it be 6V feeding the IC?)
*Is the debouncing latch/switch able to be wired several yards away as my fire switch?
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Do you have Wine installed on your Mac? If so, you may be able to run LTSpice under Wine. Some of our members are running LTSpice under Wine on their Unix/Linux platforms; I don't know if it would work under Wine running under OS X, but I don't know if it won't either.

If you want to try it, here's a tutorial on Wine:
http://davidbaumgold.com/tutorials/wine-mac/

After you get Wine working, you can try downloading and installing LTSpice. The download section on Linear Technology's website is here:
http://www.linear.com/designtools/software/#LTspice

You'll also need some extra libraries and symbols, but take things one step at a time.
Get Wine installed and working first.
Then get LTSpice installed, and try some of the educational simulations in the examples\educational folder to see how they work.

It will make things a lot easier if you have software on your computer that can read and simulate schematics that other people created.
 

Wendy

Joined Mar 24, 2008
23,415
That debounce will not work as is. It should look like this...



With an inverter (which could be another NAND gate) you could use a momentary contact switch. You would still need the RC circuit on the momentary contact however. A flip flop style debounce needs the dual input.

.
 

Attachments

Thread Starter

Rent2Own

Joined Sep 2, 2011
12
I'm trying to see the difference in our debounce circuits.

Is it that my switch/latch is different than yours? My NAND gates, resistors, & and wiring seems to be the same.

Also, you're saying I need a second switch for that circuit to work? Where, in your circuit, is the RC circuit addition you mentioned? (or is it implied that it's just needed?)

Sorry. So many questions..
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
R2O,
The switch in your circuit is drawn in a non-standard manner. It looks like it is enclosed in a box that is grounded, and that the switch can either be open, or connecting the upper and lower wires together. I'm not sure if you meant for the resistors to be connected to GND, or the European +V and -V supply symbols, or exactly what.

If you meant for the the resistors to be connected to GND, then the NAND outputs will start off high, and stay that way. If you meant for the resistors to be connected to +V, then the circuit will oscillate at high frequency.

Bill's circuit is not without problems either. The resistors need to be connected to GND, and the switch common to +V. Otherwise, when the switch is "in flight", the circuit will oscillate at high frequency.

Have a look at the attached circuits and simulations. The first is a "bounce-free switch"; as long as the switch is in one state or the other when the circuit is powered up, Q will be opposite of Q\, and it won't oscillate when the switch is changing states.

The 2nd schematic is practically identical with a few subtle changes; the resistors are now connected to 5v instead of GND, and the switch controls which NAND input is grounded instead of being connected to 5v. I've set the "sel" of the switch to .5, which means the switch is in-flight (the common terminal not connected to either NC or NO). The bottom plot is solid red, which means that the Q output is oscillating at some frequency higher than can be discerned from the selected time scale. (Q\ is also oscillating, but the Q output plot is covering it up)
 

Attachments

Last edited:

Thread Starter

Rent2Own

Joined Sep 2, 2011
12
Well some changes definitely need to be made to the debouncing circuit then.

Sgt,
Were you saying I need two zener diodes per MOSFET if I were to use them? (the diodes to protect from ESD)
 

SgtWookie

Joined Jul 17, 2007
22,230
I edited my prior post to add a couple of schematics and simulations.

And yes, I mentioned adding a couple of Zener diodes. If Vgs exceeds ±20v, the MOSFET will be destroyed. You can use 15v Zener diodes from the gate to the source terminals to keep that from happening.
 

KJ6EAD

Joined Apr 30, 2011
1,581
Can you feed the counter outputs through an inverter to a monostable for a fixed time firing pulse?

http://www.falstad.com/circuit/#%24+3+0.0020+1.7725424121461644+50+12.0+50%0A165+312+264+328+264+3+0.0%0Ag+600+368+600+384+0%0Ar+312+280+264+280+0+220000.0%0Aw+312+280+312+328+0%0Ac+344+344+344+368+0+1.0E-8+8.0%0Ac+312+328+312+368+0+1.0E-6+5.454297531930366E-4%0Aw+344+224+264+224+0%0As+32+368+32+312+0+1+true%0Aw+312+368+344+368+0%0Aw+376+248+376+280+0%0A163+80+296+96+296+1+10+0.0+0.0+0.0+0.0+0.0+0.0+5.0+0.0+0.0+0.0%0AR+240+328+240+352+0+0+40.0+12.0+0.0+0.0+0.5%0Ac+80+312+32+312+0+1.0E-6+9.99999999995893E-4%0Aw+256+312+256+280+0%0Aw+240+280+256+280+0%0Aw+312+368+32+368+0%0Ap+568+312+552+328+0%0Aw+632+248+632+280+0%0Aw+600+224+520+224+0%0Ac+568+328+568+368+0+1.0E-6+5.454297531930366E-4%0Ac+600+344+600+368+0+1.0E-8+8.000000000000002%0Aw+568+280+568+328+0%0Ar+568+280+520+280+0+220000.0%0A165+568+264+584+264+3+0.0%0Aw+32+224+264+224+0%0AI+256+312+312+312+0+0.5%0Aw+568+368+600+368+0%0Ap+80+312+80+328+0%0Ap+376+296+376+312+0%0Aw+472+368+568+368+0%0Aw+384+312+384+240+0%0AR+600+224+600+200+0+0+40.0+12.0+0.0+0.0+0.5%0AI+512+312+568+312+0+0.5%0Aw+384+240+224+240+0%0Aw+224+240+224+280+0%0AI+384+312+440+312+0+0.5%0A165+440+264+456+264+3+0.0%0Ar+440+280+392+280+0+220000.0%0Aw+440+280+440+328+0%0Ac+472+344+472+368+0+1.0E-8+8.0%0Ac+440+328+440+368+0+1.0E-6+5.454297531930366E-4%0Aw+472+224+392+224+0%0Aw+504+248+504+280+0%0Ap+440+312+424+328+0%0Ap+504+296+504+312+0%0Ap+632+296+632+312+0%0Aw+512+312+512+232+0%0Aw+512+232+208+232+0%0Aw+208+232+208+280+0%0Aw+440+368+472+368+0%0Aw+440+368+344+368+0%0Aw+344+248+376+248+0%0Aw+472+248+504+248+0%0Aw+600+248+632+248+0%0Aw+472+224+520+224+0%0Aw+344+224+392+224+0%0Ar+32+312+32+224+0+10000.0%0Aw+264+280+264+224+0%0Aw+344+248+344+224+0%0Aw+392+280+392+224+0%0Aw+472+248+472+224+0%0Aw+520+280+520+224+0%0Aw+600+248+600+224+0%0Ao+27+64+0+34+20.0+9.765625E-5+0+-1%0Ao+28+64+0+34+20.0+9.765625E-5+0+-1%0Ao+44+64+0+34+20.0+9.765625E-5+0+-1%0Ao+45+64+0+34+20.0+9.765625E-5+0+-1%0A
 
Last edited:

Wendy

Joined Mar 24, 2008
23,415
R2O,
The switch in your circuit is drawn in a non-standard manner. It looks like it is enclosed in a box that is grounded, and that the switch can either be open, or connecting the upper and lower wires together. I'm not sure if you meant for the resistors to be connected to GND, or the European +V and -V supply symbols, or exactly what.

If you meant for the the resistors to be connected to GND, then the NAND outputs will start off high, and stay that way. If you meant for the resistors to be connected to +V, then the circuit will oscillate at high frequency.

Bill's circuit is not without problems either. The resistors need to be connected to GND, and the switch common to +V. Otherwise, when the switch is "in flight", the circuit will oscillate at high frequency.

Have a look at the attached circuits and simulations. The first is a "bounce-free switch"; as long as the switch is in one state or the other when the circuit is powered up, Q will be opposite of Q\, and it won't oscillate when the switch is changing states.

The 2nd schematic is practically identical with a few subtle changes; the resistors are now connected to 5v instead of GND, and the switch controls which NAND input is grounded instead of being connected to 5v. I've set the "sel" of the switch to .5, which means the switch is in-flight (the common terminal not connected to either NC or NO). The bottom plot is solid red, which means that the Q output is oscillating at some frequency higher than can be discerned from the selected time scale. (Q\ is also oscillating, but the Q output plot is covering it up)
You got this one wrong Wookie, at least as concerns the debounce circuit. If both of the inputs to the NAND flip flop go low (which will happen while the switch is in transition) it is no longer a flip flop, and both gates go high. The circuit triggers on a low, and forces the flip flop to change states.

Generally I use dual NOR gates to avoid this (and use lows on the resistors), but I was going with the OP circuit, which could work but needed a SPDT switch.

This applies to the circuit you drew. Imagine what happens if the switch was in the center position, if both resistors are ground you have lost your logic state. If the switch bounces, which is to say goes between the center position and the positive contact the outputs will follow the bounce. In other words, it won't work

Only when the Set Reset latch has memory does the debounce work.

 

Wendy

Joined Mar 24, 2008
23,415
Can you feed the counter outputs through an inverter to a monostable for a fixed time firing pulse?
I helped another OP with a very similar circuit. It does work in theory, but I wound up building it for him on a protoboard. It is surprisingly vulnerable to power supply cross coupling, which is only solved with decoupling caps. They did fix the problem nicely.

Parts count counts, always, as well as reliability.
 
Top