Transistor problem in motion sensor circuit

Thread Starter

LuisGranados

Joined Sep 16, 2019
3
I am a novice, trying to make a simple circuit for a motion sensor closet light. What I did works, but I need to make it brighter to be useful. This ought to be simple, but it's stumping me.

The working circuit that is not bright enough uses a 12V battery, a 7W LED light bulb, and an HC-SR501 motion sensor. The 12V battery, if connected directly to the bulb, is plenty bright. The middle pin of the HC-SR501 sends a signal to the transistor when motion is detected, which then enables the current to light the light. But I had to add a 33R 5W resistor to keep the transistor I first used (a BJT 2N2222) from burning up. That resistor (I think) makes the bulb too dim to be useful.

So I took out the resistor and tried replacing the transistor with a higher-voltage TIP31C, that's supposed to handle 50V. The circuit works, but the light is still too dim. The same result happens when I double the voltage to 24V. The transistor also gets extremely hot -- one began burning the wood on my workbench.

The same result happens with an FJP13009 transistor, which is supposed to handle 700V. Circuit works, but light is still dim, at both 12V and 24V. I tried sticking the 33R resistor between the battery and the transistor, but that did not change the outcome.

I also tried a Vishay IRFZ44 transistor, supposed to handle 60V, and got very strange results. The bulb would glow brightly whether or not the gate was connected. Again, the transistor got very hot. (I thought I was perhaps getting the pin connections wrong, but the only one of the six possible combinations that produced any light at all was the one where the gate pin was irrelevant.)

I'm embarrassed to admit how long I've been fiddling with this. Can someone help me? I'm now trying to learn Spice to simulate the circuit, but that's going slowly.
 

Attachments

ericgibbs

Joined Jan 29, 2010
18,766
hi Luis,
Welcome to AAC.
The problem is that the HCSR can only output 3v via a 1K output resistor, which is only 3V/1k = 3mA.
If you subtract the 0.7Vbe for the transistor, that is only 2.3v/1k = 2.7mA, which is not enough Base current.

The 7W lamp at 12V requires 7W/12V = 0.58Amps!!!
This means the transistor would have a gain of 0.58A/2.7mA = 216 when saturated.!!

You need to modify the circuit.

E
BTW: a COLD 7W lamp will draw approx 5*0.58A while Warming up.
 

Audioguru again

Joined Oct 21, 2019
6,673
Most Mosfets need an input of 10V to fully turn on but a "logic level" Mosfet needs an input of only 5V.
But since you ain't got that much input voltage, you have only 3V then they won't work.
There are some new tiny little Mosfets that might be able to do it but you might not be able to make the tiny connections.
 

TeeKay6

Joined Apr 20, 2019
573
Most Mosfets need an input of 10V to fully turn on but a "logic level" Mosfet needs an input of only 5V.
But since you ain't got that much input voltage, you have only 3V then they won't work.
There are some new tiny little Mosfets that might be able to do it but you might not be able to make the tiny connections.
I did a search at Mouser, finding dozens of suitable devices in SOT-23 packages, many with Vth of 2V or less. As an example, I have attached the datasheet for the first device listed ($0.41 each). A more exhaustive search might turn up other packages (and of course there are other distributors). Unfortunately, we don't know what the forward voltage of the LED bulb is at the rated current. It would be nice to have a mfr/part-number for the "7w LED" bulb to view a datasheet.
 

Attachments

Last edited:

Audioguru again

Joined Oct 21, 2019
6,673
Why are you looking at Vth? That is when it is almost turned off with a current of only 0.25mA. You need to look at the Vgs voltage when it is turned on completely and the spec's show its maximum Rds(on) resistance.
The Mosfet you show has a max Rds(on) resistance of 0.085 ohms when Vgs is 2.5V so with your 3V the max Rds(on) might be 0.08 ohms then with your 0.58A its heating will be 0.08 ohms x 0.58A= 47mW which is fine.
 

MisterBill2

Joined Jan 23, 2018
18,167
The application needs two transistors, the first stage being a fairly sensitive amplifier that boosts the output of the motion sensor up to a much higher voltage, like 10 volts. Then the second stage would handle the power, switching from cut-off to fully saturated while carrying the lamp load. AND, the problem was that the TS kept going for a higher voltage rating instead of a higher current rating.
 

TeeKay6

Joined Apr 20, 2019
573
Why are you looking at Vth? That is when it is almost turned off with a current of only 0.25mA. You need to look at the Vgs voltage when it is turned on completely and the spec's show its maximum Rds(on) resistance.
The Mosfet you show has a max Rds(on) resistance of 0.085 ohms when Vgs is 2.5V so with your 3V the max Rds(on) might be 0.08 ohms then with your 0.58A its heating will be 0.08 ohms x 0.58A= 47mW which is fine.
I am well aware of the meanings of Vth, Ron, etc. In the datasheet both the min/max specs and the typical specs shown in Fig1 & Fig2 show that this device easily meets the requirements. I searched for devices with low Vth because I certainly did know that only 3V drive was available; it would have been stupid to search for a low Ron without also requiring that Vth be low--so my search looked for low values of both. BTW, power=I*I*R, not I*R, yielding 27mW, which is indeed fine.
 

Thread Starter

LuisGranados

Joined Sep 16, 2019
3
Thanks so much for your prompt and helpful responses. I've redone the circuit as attached, and am testing it with the various transistors I have on hand for the new "middle transistor" designed to amplify the current from the HRSC before it gets to the transistor that controls the 12V or 24V for the bulb. In case none of them work, which is likely, I've ordered the DMG3402LQ transistor that was suggested. and will try that as both a replacement for the transistor I had and as the new amplifying "middle" transistor. I will post something when I've completed this.

I do not understand datasheets very well. For the DMG3402LQ datasheet, at Gate Threshold Voltage, its says Min=0.6V and Max=1.4V. Does this mean there is a problem if the gate receives more than 1.4V? It also says "Source Drain Diode Forward Voltage" is 1.16V. Does this mean that the current when it is triggered on is only 1.16V? I guess I'll find out when I try it.

The LED bulb, for someone who asked, is InLED Part No. 105, according to Amazon. The best information I could get said "12V 7W, AC/DC, Low Voltage, Standard Base, 60W Equivalent."
 

Attachments

TeeKay6

Joined Apr 20, 2019
573
Thanks so much for your prompt and helpful responses. I've redone the circuit as attached, and am testing it with the various transistors I have on hand for the new "middle transistor" designed to amplify the current from the HRSC before it gets to the transistor that controls the 12V or 24V for the bulb. In case none of them work, which is likely, I've ordered the DMG3402LQ transistor that was suggested. and will try that as both a replacement for the transistor I had and as the new amplifying "middle" transistor. I will post something when I've completed this.

I do not understand datasheets very well. For the DMG3402LQ datasheet, at Gate Threshold Voltage, its says Min=0.6V and Max=1.4V. Does this mean there is a problem if the gate receives more than 1.4V? It also says "Source Drain Diode Forward Voltage" is 1.16V. Does this mean that the current when it is triggered on is only 1.16V? I guess I'll find out when I try it.

The LED bulb, for someone who asked, is InLED Part No. 105, according to Amazon. The best information I could get said "12V 7W, AC/DC, Low Voltage, Standard Base, 60W Equivalent."
If you intend to continue with electronic projects, you really need to learn about the devices you will be using. There is plenty of info--at all levels of detail--online. This site, AAC, offers a very good, free, electronic "textbook" for beginners (use the AAC search function to find "textbook"). That said, here is some specific info in response to your comments.

The gate threshold voltage of a MOSFET is the voltage (measured gate-to-source) at which the MOSFET just begins to conduct current and is usually specified for a current of less than 1mA. Because all devices vary somewhat, Vth is given as a range. Depending on the specific application, either the min or max Vth may be useful to a designer. To see how the device performs at higher gate-to-source voltages, most datasheets offer graphs showing typical drain current vs gate-to-source voltages over a range from Vth to the max allowed for the device. For the DMG3402LQ datasheet, see Figures 1 & 2. At the top of page 2 note that the max allowed gate-to-source voltage (Vgs) is +/-12V. Figures 1 & 2 show that much less than 12V is sufficient for your application.

MOSFETs have an "intrinsic" (i.e. inherent in the design; you cannot remove it) diode between the drain and the source. This diode behaves like an ordinary discrete rectifier in parallel with the MOSFET. When current flows in any diode/rectifier in the forward (the "on" direction, in the direction of the arrow symbol) direction, a voltage drop is produced across the diode terminals (in this case the drain and the source). For the DMG3402LQ, the datasheet shows on page 2 that when a forward current of 2A flows through the intrinsic diode, a voltage drop of 1.16V max will be produced (at Ta=25C). In your circuit no forward current will ever flow through the diode (it will always be "reverse biased") so this spec is not relevant.

As clarification, I did not "recommend" this device. As I stated, it was the first in a long list of possibly suitable devices. There could be many other suitable choices at even lower cost. I hope you looked closely enough to realize that the SOT-23 is a very small device and some skill is needed for soldering. A more extensive search might find a device in an easier to use package (probably at slightly higher cost).
 
Last edited:

Thread Starter

LuisGranados

Joined Sep 16, 2019
3
I spent an afternoon trying to solder a tiny surface-mount Mosfet. Long story short: I'm not going to try that again.

Then I wound up getting a Darlington transistor, 2N5306. This promised a very high gain (7000) from a low base signal. It worked insofar as getting me a bright light, but it got extremely hot, and after a brief time the light began to flicker (perhaps because the transistor was melting the plastic on the connectors). So I tried sticking a 1-ohm high-wattage resistor in front of it to protect it from current coming straight from the battery, and voila -- everything works, the light is plenty bright, and nothing gets very hot. I will now be able to tell the green shirts from the blue shirts in my closet.

Thanks again to everyone who helped.
 

Audioguru again

Joined Oct 21, 2019
6,673
The light bulb has a 12V current of 0.58A but the little Darlington transistor has a maximum allowed current of only 0.5A.
It got too hot because it was overloaded.
 

MisterBill2

Joined Jan 23, 2018
18,167
Put two of those darlington transistors in parallel, put a 1 ohm resistor in series with each emitter, and the system should work without overheating.
 

Hemi

Joined Mar 17, 2012
30
Personally, I would have used an easily solderable FQP30N06L as a low side switch.

Just remember that the motion sensor draws something like 20-50mA so your 12V battery will have a constant drain on it while it's looking for motion.
 

Audioguru again

Joined Oct 21, 2019
6,673
The IRLZ44 Mosfet is spec'd to turn on well when its Vgs is 4V and some of them are almost turned off (250uA threshold current) when its Vgs is 2V.
You are simply lucky that your Mosfet does not have minimum spec's.
 
Top