# ATTiny85V + TIP122 transistor to explode giant balloons

#### liberalbid

Joined Apr 27, 2017
24
Hi

I'm working on a device that will detect the vibration of a projectile's impact on a giant latex balloon and then allow sufficient current to pass through a heating element to explode the balloon.

In truth, I have managed to create a working prototype, firstly using an Arduino UNO, and then an Arduino NANO (in an attempt to reduce the physical size). In both cases, the circuit was powered by 9 volts from 6xAA batteries. A piezo disc detected the vibration of impact, the Arduino switched an output pin HIGH connected to the base of the TIP122 which allowed current from the 6xAA to flow through an Estes Solar Rocket Starter as the heating element. Worked well. You can see the result in action here.

In order to shrink/improve the device, even more, I am trying to replace the Arduino NANO with an ATTiny85V and the Rocket Starter with a Remote Control Engine Glow Plug but I am having trouble and hope for some advice.

The Glow Plug can use a 1.5 - 3.0V power supply and draws around 2.5 amps from that supply for the 1.5 seconds that it is switched on.

I have attached a schematic of my test circuit. It doesn't include the piezo sensor because that 'side' of the circuit works as planned. It only includes the 'ignition' side of the circuit which I can't get working.

PB3 on the ATTiny85V switches HIGH when the push button switch is activated. In turn, PB2 is then switched HIGH supplying current to the base of the TIP122 transistor, allowing the flow of current through the Glow Plug Igniter for 1.5 seconds.
The 3V supply on the schematic is provided by a pair of AA batteries.

Problem is, I can't get sufficient current to flow through the Glow Plug and I don't understand why.

I have used various size resistors at R2 without success. I have even removed R2 all together (bad option, I have learned) but still the Glow Plug would not glow.

Am I asking too much of the TIP122? Is there a better option?
The Glow Plug glows immediately if connected directly to even a single 1.5V AA battery, but not if connected via the transistor.

Hopefully

LB

#### Attachments

• 403.4 KB Views: 43

#### DickCappels

Joined Aug 21, 2008
6,941
The TIP122 is not a good candidate for low voltage operation.

For one thing it is a Darlington pair and requires higher base voltage than most transistors. See the the Vbe curve in the chart below.

Another thing is that you loose over 1 volt across the collector and emitter at 2.5 amps -that's the Vce curve.

At 2.5 amps the TIP122 has very high gain but you will need to drive it with several (I suggest 10) millimaps to assure that it saturates. The 10 ma is well within the capabilities of the ATTINY85 but you need some voltage to drop across the base series resistor in order to develop some base current.

At 10 milliamps the ATTINY85 can only be expected to produce about 2.6 volts (above)

That leaves you about 0.8 volts with which to make that 10 milliamps. An 82 ohm resistor might work.

This would not be a safe design for production because these voltages vary among devices and with temperature. With a higher voltage the design would be much more tolerant of variations.

Can you get your power supply up to 4.5 volts or 6V?

#### be80be

Joined Jul 5, 2008
2,051
Better get good AA batteries not all of them could give you 2.5 amps for a second. Duracell Quantum.

#### liberalbid

Joined Apr 27, 2017
24
The TIP122 is not a good candidate for low voltage operation.

...

Can you get your power supply up to 4.5 volts or 6V?
Hi Dick, thanks for such a quick and comprehensive answer.

I chose not to disclose in my OP that two months ago I didn't know anything about electronics. I didn't know Ohm's law let alone what an Arduino was. I've managed to make it this far but your answer shows that I clearly am a long way short of where I need to be to solve this problem!

It would be lovely to keep it down to 3 volts (2AA) in order to keep the overall size down. Do you think this would be possible with something other than the TIP122 transistor?

If not, and I have to increase voltage, then yes, I'll go to whatever value I need, ever mindful that smaller is better.

Perhaps there is a solution other than a transistor, or even a microcontroller? These are just components I've arrived at in quick time.

LB

#### BR-549

Joined Sep 22, 2013
4,938
I hate to be a critic......but there is a long delay between impact and explosion. Have you considered a quicker way of exploding the balloon?

The neatest way would be H2 and a spark. But that wouldn't be PC.

Perhaps a straight pin solenoid.

#### liberalbid

Joined Apr 27, 2017
24
Better get good AA batteries not all of them could give you 2.5 amps for a second. Duracell Quantum.
I appreciate the advice, thanks, be80be.

I failed to mention that in five hours of use (typical event length), the glow plug will only be activated on five occasions at most (10.5 seconds total duration). This infrequent activation might be why we've enjoyed success with many different AA batteries in testing. Our very agricultural testing method of connecting a glow plug to a single AA battery every 15 seconds has yielded many dozens of 'glows' from that single battery.

LB

#### liberalbid

Joined Apr 27, 2017
24
I hate to be a critic......but there is a long delay between impact and explosion. Have you considered a quicker way of exploding the balloon?

The neatest way would be H2 and a spark. But that wouldn't be PC.

Perhaps a straight pin solenoid.
Not at all, BE-549, I appreciate (and need!) the criticism.

My very earliest tests were with solenoids. Unfortunately, the travel distance of the pin was insufficient to always puncture the balloon. The latex (unless very highly stretched which required greater inflation pressure than we want) had too much give. I needed reliability and certainty that a hit would result in explosion. The move to a heating element has brought that about, but as you quite rightly observe, at a cost of a delay in explosion.

All that said, I'm open to any idea at all to bring this about.

Thanks again.

LB

#### BR-549

Joined Sep 22, 2013
4,938
OK then.......maybe a semi transparent balloon.......for a couple of LEDs to give it internal glow waiting for explosion.....that's even more dramatic. A flash of light right before the bang.

Maybe even try a flash after explosion.

Edit: you could add speakers......and broadcast a transverse thunder effect waiting on explosion too. A thunder clap going towards the balloon.

#### DickCappels

Joined Aug 21, 2008
6,941
I think that part of what is behind b380be's intention in post #3 is that when you put a load on a battery the voltage at the terminals goes down because there is resistance in the battery itself. When you start talking 2.5 amps some of the flashlight cells fall to their knees. In one project I was trying to pull about 2 amps out of a carbon-zinc D cell. I finally got close to that goal but the voltage had dropped to about half. b380be made a specific recommendation, probably representative of the type of cell you should be looking at.

Either of these two On Semiconductor P-channel MOSFETs might be able to do the job if you can get good batteries that can keep the voltage up when the current is drawn. The transistors are in stock Newark/Element 14/ Farnell -guess the name depends on which geographical region you are in.

CPH3350-TL-W
http://www.farnell.com/datasheets/2...5.1566298488.1512026256-1500129595.1509899757

MCH3377-TL-W
http://www.farnell.com/datasheets/2...4.1566298488.1512026256-1500129595.1509899757

Being P-Channel MOSFETs, they are "upside-down" from your NPN bipolar transistor. To use these you would connect the source to battery +, connect the load to the drain, and the other end of the load to ground. The gate would normally be kept at battery + and connected to ground to turn it on.

The weak point in this is still the power source.

If all the ATTINY85 is doing is acting as a two second timer, you can do this with a much less expensive one-shot such as the LMC555C or the NE555. I imagine that with 8kk or flash you can do a lot more than that.

#### liberalbid

Joined Apr 27, 2017
24
OK then.......maybe a semi transparent balloon.......for a couple of LEDs to give it internal glow waiting for explosion.....that's even more dramatic. A flash of light right before the bang.

Maybe even try a flash after explosion.

Edit: you could add speakers......and broadcast a transverse thunder effect waiting on explosion too. A thunder clap going towards the balloon.
BR-549, don't underestimate how seriously I'll consider your ideas!!

LB

#### liberalbid

Joined Apr 27, 2017
24
I think that part of what is behind b380be's intention in post #3 is that when you put a load on a battery the voltage at the terminals goes down because there is resistance in the battery itself.
...
The weak point in this is still the power source.
It is all coming back to the power, isn't it. Thank you for adding to be80be's advice. I'm fast realising that 2xAA might just be fanciful thinking.

Given that the Glow Plugs only want 1.5 volts (the manufacturers don't warranty them beyond 1.5v), you can imagine how my electronically-uneducated brain thinks 3.0 volts should be more than sufficient. But it is because of the switching role of the transistor...

I will try both your other PNP suggestions, thanks for taking the time to link, and explain.

If all the ATTINY85 is doing is acting as a two second timer, you can do this with a much less expensive one-shot such as the LMC555C or the NE555. I imagine that with 8kk or flash you can do a lot more than that.
In the complete circuit, there is no switch, instead the ATTiny85 listens for an analog input from a piezo sensor, compares it to a threshold value, and if that value is exceeded then switches on the output to the base of the transistor.

#### be80be

Joined Jul 5, 2008
2,051
I'd break the glass off one of these

It would get hot way faster then a glow plug the down side is its a one shot deal it will burn out but any.
2n3904 should start a fire. And way less the 2.5 amps.

#### liberalbid

Joined Apr 27, 2017
24
I'd break the glass off one of these

It would get hot way faster then a glow plug the down side is its a one shot deal it will burn out but any.
2n3904 should start a fire. And way less the 2.5 amps.
Indeed, be80be, all earlier versions used one of many types/versions of rocket igniters. One was very effective at only 0.4amps. Of course they were one-shot deals as well and it that expense (and mild fire risk) that led me to consider the fully contained glow plug. I might have to rethink. ..

#### liberalbid

Joined Apr 27, 2017
24
If I have to stay with as lower voltage as possible because of size contraints (say 4.5v, 3xAA), is there any advantage in using a 3v solid state relay rather than a transistor?

LB

#### DickCappels

Joined Aug 21, 2008
6,941
There is an advantage to using MOSFETs in place of the bipolar NPN in the case of 3 volts. That is because MOSFETs switch as a function of gate voltage as opposed to bipolar transistors switching as a function of base current.

If you investigate solidstate relays I think you will be lucky to find one that fits your application as well as one of the MOSFETs mentioned in post #9.

#### philba

Joined Aug 17, 2017
960
What ever transistor you use (I agree on the mosfet approach), try putting a relatively large capacitor across + and - to boost the current and speed up the element heating.

#### Bernard

Joined Aug 7, 2008
5,740
A miniature bulb would work; in 1950s I used # 49's as substitute for blasting caps when testing blasters.
Other choices might be: 48, 49, 2V @ 60 mA, 112, 1.2V, @ 220 mA, 123, 1.2V, @300 mA, 129, 1.3 V @ 150 ma , 131, 1.3V @ 130 mA. # 123 might survive an explosion @ 2 to 2.5 V , 250 ms. ?
What is balloon atmosphere?
In surface mount I use FDD6530A, N ch., 36 mohm, 20 V, 21 A, & for P ch.,
FDS9435A, 30 V, 5 A.

Last edited:

#### Bernard

Joined Aug 7, 2008
5,740
Found # 131 @ Granger for US $7.40 / 10. Better yet, # 123 @ Genesis Lamp Corp.,$ 5.20 / 10.
I wonder how heating bulb and dunking it in water would work ??

Last edited:

#### liberalbid

Joined Apr 27, 2017
24
A miniature bulb would work; in 1950s I used # 49's as substitute for blasting caps when testing blasters.
Other choices might be: 48, 49, 2V @ 60 mA, 112, 1.2V, @ 220 mA, 123, 1.2V, @300 mA, 129, 1.3 V @ 150 ma , 131, 1.3V @ 130 mA. # 123 might survive an explosion @ 2 to 2.5 V , 250 ms. ?
What is balloon atmosphere?
In surface mount I use FDD6530A, N ch., 36 mohm, 20 V, 21 A, & for P ch.,
FDS9435A, 30 V, 5 A.
Bernard, I never even considered bulbs. Are you thinking that they would develop enough heat to pop the balloon with the bulb intact and pressed against the balloon? Or would I have to remove the glass in order to expose the filament?

Exposing the filament might not be ideal. Glass still in place would be better. Unless that glass was to shatter upon popping the balloon ... not a great result on a school oval.

Thank you.

LB