This is my first time working with integrated circuits. I have lots of experience with electro-mechanical wiring but such things as 555/556 timers and transistors are new to me. I'm building a paintball gun known in our world as a Spimmy.
My particular gun will operate off a 9V battery and power a 5V 4.7W solenoid/valve that will be activated by the trigger circuit to control the firing of the paintball. The opening and closing of the solenoid must be held to 9ms which will be controlled by a 556 timer. I've worked out the math for the various resistors-potentiometers that are connected to the timer and feel confident about that.
My plan is to buy my parts from All Electronics and, sorting through their stock, I had to use a fairly large capacity transistor to meet the requirements of the 5V 4.7W solenoid. The transistor specs are Pmax 100V, IcMax 6A, VceMax 65W, Hf, 15@ 3V, 4A, VceSat 1.5@600mA, 6A. This is the smallest transistor they carry that seems to work with my solenoid. But I know bigger isn't always better. Is this OK or should I look for something different.
The circuit for the transistor is in the top right corner of the schematic.

 

My hunch is that it'll work fine. Others here will chime in if that's not so. I keep a baggie of transistors I've pulled from old junk, and it rarely matters which one I choose from the baggie, as long as it can handle the current of whatever I'm working on.

Supplying 9v to a 5v relay raises an eyebrow, although once you draw power from the 9v and drop ~0.7 across the transistor, it's probably not so bad. Still, you might want to think about what happens if the trigger fails and holds the solenoid steady "ON".

I think you'd also benefit from switching the solenoid+diode and the transistor positions, so that the solenoid sees positive at all times. Let the transistor open the path to ground. This way, a much smaller voltage at the base of the transistor will do the job, ie. ~0.7v will be full on.

 

I like wayneh's suggestion, but would add 1000 μF in parallel with battery. For smaller transistor , surface mount, I've used FZT849 [ The Electronic Goldmine ], 80V, 7A, β100. Could add a little resistance in series with relay-say 3Ω,1/2 W [ fuse].

 

Are you using a 9 volt rectangular battery, or a set of 6 AA's?

The small 9 volts are very bad about going down in voltage when large currents are drawn. 50mA is a pretty common limit, with large drops in voltage out of the battery when current draw is larger than that. A few minutes of activating that relay and you would have to change batteries. The set of AA's on the other hand could put out more than 1 amp and still stay close to 9 volts.

 

HFE is low which is common for high power transistors. If the HFE is 15 and output current required is 1A then you need at the very minimum 66mA into the base, even then the transistor would get very warm. More current into the base would make it switch on better. You may need another transistor to drive the power transistor, but definately the base resistor is much to high a value.
Unfortunately I don't know much more, but maybe someone can suggest something.

 

All Electronics also shows several darlington power transistors in TO220.

 

Supplying 9v to a 5v relay raises an eyebrow, although once you draw power from the 9v and drop ~0.7 across the transistor, it's probably not so bad. Still, you might want to think about what happens if the trigger fails and holds the solenoid steady "ON".

I think you'd also benefit from switching the solenoid+diode and the transistor positions, so that the solenoid sees positive at all times. Let the transistor open the path to ground. This way, a much smaller voltage at the base of the transistor will do the job, ie. ~0.7v will be full on.

If it did stay steady on I'd be able to turn it off pretty easy. I'll reconfigure the schematic with your suggestion regarding the positions of the components.

I like wayneh's suggestion, but would add 1000 μF in parallel with battery. For smaller transistor , surface mount, I've used FZT849 [ The Electronic Goldmine ], 80V, 7A, β100. Could add a little resistance in series with relay-say 3Ω,1/2 W [ fuse].

I'll rework the schematic to include your suggestion for the capacitor. I'll keep your suggestion for the relay in mind.

Are you using a 9 volt rectangular battery, or a set of 6 AA's?

The small 9 volts are very bad about going down in voltage when large currents are drawn. 50mA is a pretty common limit, with large drops in voltage out of the battery when current draw is larger than that. A few minutes of activating that relay and you would have to change batteries. The set of AA's on the other hand could put out more than 1 amp and still stay close to 9 volts.

I'm using a 9V battery. There's no room for AA's in a paintball marker. Pretty much all paintball guns use 9V batteries and solenoids similar in current draw to what I'm using. You have to remember it only has to operate the solenoid for 9ms at a time. Most people get off at least 4000 rounds on one battery which takes at least two to three full game days, or more, but that does add up to 6 total minutes of actually operating the solenoid like you said. After spending $75 to a $120 on 4000 rounds of paintballs, tossing in another battery is peanuts.

but definately the base resistor is much to high a value.

I was going to ask about that but didn't want to make the original post to long. That resistor is one that I haven't run any calculations on and took the 1K from the generic 555 schematics.

All Electronics also shows several darlington power transistors in TO220.

Thanks. I'll look into the Darlington transistors. This one looks promising. http://www.allelectronics.com/make-...PN-DARLINGTON-TRANSISTOR-2A-80V-TO-220/1.html

I appreciate all the replies. I'm leaving tomorrow on a business trip to Russia. I don't know if I'll have time to work on redrawing the schematics before I leave but I'll have plenty of time to work on it on the way. Once I'm settled in to Moscow, I'll post things up.

 

I'd add in a few extra capacitors. The 556 needs 2 decoupling capacitors between it's power supply pins. You should check the datasheet for values, but it might be 100nF and 10uF. Because the 9V battery has limited current capability I'd add a nice big capacitor like 220uF between the power rails as well. These should make things more reliable.

 

Rather than going to Darlingtons, consider using a logic-level power MOSFET. Instead of requiring base current like transistors do, MOSFETs have gates that are sort of like capacitors; you charge the gate to turn the MOSFET on, then discharge the gate to turn them off. You can get rather small MOSFETs that can handle a bunch of current.

Take a look at this MOSFET:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IRLU8743PBF-ND
160A is maximum overkill, but that's just to show you what kind of current these little MOSFETs can sink; the package is much smaller than a TO220 device.

Here's one that would be better for your use:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IRLU7807ZPBF-ND
65A continuous, but the gate charge is very small compared to the previous MOSFET.

To use an n-channel MOSFET like this one, connect the source terminal to ground, the gate to your 555 output pin via a 22 Ohm resistor, and sink current from the relay via the drain terminal. It's a good idea to connect the gate to the source terminal with a 10k resistor to reduce the likelihood of MOSFET damage due to ESD if you remove the timer IC.

 

Hey SGT,

I was just wondering what is the difference between Logic Level MOSFETs and regular MOSFETs?

 

Regular MOSFETs require a voltage difference measured between the source and gate pins to be 10v or more to "turn on".

Logic Level MOSFETs can be "turned on" from standard logic levels (5v).

In a nutshell.

The NTE series Logic Level MOSFETs are compatible with the 5-volt power-supply requirement of logic circuitry. These devices do not require an interface circuit between it and the CMOS logic driver; therefore, the extra cost of the interface circuit power supply is eliminated.

The chief physical structural difference between Logic Level and other MOSFETs, and the electrical reason for its difference in performance, is its gate insulation thickness, which has been reduced from 100nm industry standard to 50nm (500 angstroms), yet which retains the dynamic strength to handle the high voltage applied to power transistors. Since the surface inversion of the MOS channel is determined by the gate-insulator voltage field, the halving of the gate-oxide thickness should be expected to have a major effect on the gate voltage required. In fact, this reduction is the reason for voltage reduction from 10 volts (standard MOSFETs) to 5 volts (Logic Level MOSFETs).

Tight control of the temperature vs. time and oxygen vs. time profiles applied to the silicon substrate during oxide growth assures consistant preformance through the development of good transition regions between the oxide, the silicon below it, and thew polysilicon above it. The reduction in gate insulator thickness makes possible easy ON/OFF control of the Logic Level MOSFETs by CMOS logic alone, and by microprocessors.

Taken from:
http://www.nteinc.com/Web_pgs/LL_MOSFET.html

 

Between work and 9 hours of jet lag it took me a little loner than I expected to rework the drawings. The first one shows the original transistor but with the additional capacitors that have been suggested and with the load repositioned relative to the transistor. The next two show the additional capacitors plus the Darlington transistor and MOSFET in turn. I'd like more input regarding using the Darlington or MOSFET. Also, please make sure that I have the schematic drawn right too.

 

Several problems: 556-pin 7 should go to gnd. remove cap. Pin 14 should go to + 9V, remove cap.& add .1mF pin 14 to gnd. Add pull-up resistors,10k, from +9V to pin 6 & 8, A & B triggers. On FET ckt. change 22k gate, term.1, resistor to1k;move 10k from terminal 2 to terminal 3, source. On darlington , gain 1000 to 2000 +-. ' would go with 3.3k base R. FET maybe NTE 2985. May possibly have trouble with switch bounce[ 0-20 ms] with a 8 ms pulse.

 

Bernard, I reworked the MOSFET schematic. I believe I was confused by some earlier instructions. And I may still have some confusion so please look and see what you think?

 

The 1k resistor on the MOSFET gate should be changed to <100 Ohms, preferably between 22 and 39.

The diode across the solenoid needs to be rated for the solenoid's current; a 1N5817 1A 20V Schottky diode would be a good candidate, or any fast diode rated for >= 1A. A 1N4148/1N914 would likely get zapped pretty quickly. You could also use a couple of 12v Zener diodes back-to-back in series.

Tip: Since you're using both sides for multivibrators, you can connect CTRL-A and CTRL-B together, and just use one 10nF cap for the both of them.

The delay between the first round and the second round will be longer than subsequent rounds in one auto burst. That's a hazard due to the way the 555 timer works, and in this implementation will be tough to get around.

The A-side trigger (pin 6) is not wired properly. The B-side trigger isn't wired properly either.

 

I don't want to get in "trouble" for suggesting this (as sort of happened on another post)... Is this is a one-off for you (for fun) or are you planning on making a bunch of them? The reason I ask is if you have a few dollars to spend a microcontroller (like the "Basic Stamp") would allow you to forget ALL of the capacitors, timers and such and leave you with the "Stamp" (if you want to go that way, there are others) and a couple of switches and your output driver (transistor, FET, whatever). You would have the freedom to program the "timers" for anything you want without changing components, drive leds directly, etc. It would also open the "world of programming" to you... I'd be happy to help (as would a bunch of others here I'm sure).

A BS-1 is $29.00 and a carrier board (with 9V battery clip) is $4.99 and the Serial Adapter is $4.99 (no, I have nothing to do with them).

Mike T.

 

As The Sgt. sais, the triggers, each need to connect to +9V thru the 10k's, the PB SW connects to pin6, other SW to pin8. Welcome back.

 

Boy. You Guys have my head spinning every time I post. Not that I'm complaining. Just trying to get my brain wrapped around this.

SgtWookie I made the changes you suggested to the schematic.

I don't want to get in "trouble" for suggesting this (as sort of happened on another post)... Is this is a one-off for you (for fun) or are you planning on making a bunch of them? The reason I ask is if you have a few dollars to spend a microcontroller (like the "Basic Stamp") would allow you to forget ALL of the capacitors, timers and such and leave you with the "Stamp" (if you want to go that way, there are others) and a couple of switches and your output driver (transistor, FET, whatever). You would have the freedom to program the "timers" for anything you want without changing components, drive leds directly, etc. It would also open the "world of programming" to you... I'd be happy to help (as would a bunch of others here I'm sure).

A BS-1 is $29.00 and a carrier board (with 9V battery clip) is $4.99 and the Serial Adapter is $4.99 (no, I have nothing to do with them).

Mike T.

That might end up being the sane thing to do. The fact is, there is a paintball supplier that sells a programed board that does everything I want it to do and more for $50. I got the idea for using a 555 timer off a paintball forum and it got my creative juices flowing since it's cheap and I'm doing this for fun. Obviously, this is turning out to be more complicated than I originally thought but that's the way all projects go, isn't it?

As The Sgt. sais, the triggers, each need to connect to +9V thru the 10k's, the PB SW connects to pin6, other SW to pin8. Welcome back.

Thanks for the welcome back, although I've left again. I don't normally do this much international traveling at once but I'm in Kuala Lumpur, Malaysia this week.

I'm confused because I don't understand where I went wrong with the trigger connections. I've been using the schematics found at places like Wikipedia and kpsec. Did I make a mistake in my connections according to their schematics or do I just need to do it different?

 

A "logic level" MOSFET only needs 2.5V - 4V (varies by device) to switch fully on whereas most regular ones can require up to 10V on the gate to turn fully on.

 

The trigger inputs are still messed up.