# Calculating Size of Resistor needed?

The good news is that you do not have to worry about how high the coil spikes would have been, as they will be suppressed. The diode voltage rating is for normal operation, allowing for the fact that vehicle supplies tend to be full of nasty transients. Yes, higher rating is better.

The bad news is that the base drive currents appear to have been set equal to the collector currents. This seems excessive, and would result in large dissipations in R1 and R3. R1 would run about 9Watts! You would therefore need to use physically big resistors.

Can't you assume a more reasonable current gain of say 10? I note the transistor you have chosen is an RF power type which may not work well as a saturated switch. Don't panic though, it is in the popular T0-220 case with the usual pin-out, so could easily be changed.

Last edited:
I'm actually going to use TIP41C transistors, but they didn't have a vertical version of it in Eagle, so I just used a TO-220 case NPN transistor for kicks.

P.s. You're talking way over my head. :) I thought we did assume 10* current gain? I don't know anything about calculating this stuff though. How did you get 9W, and what's this about base current = collector current and how is that bad?

You can suppress a coil transient with a resistor, believe it or not.

For a 12v coil the suppressor diode PIV needs to be 12v or slightly more.

Whome9
Yes, thanks for pointing that out. A suitably chosen resistor overcomes the slow turn-off, and is probably cheaper than a Zener. Probably it's still better to put the resistor in series with a diode to avoid wasting current while the coil is on.

That said, I do not agree that it is safe to use a diode with barely sufficient PIV to sustain the normal supply voltage (up to roughly 14.5V for a "12V" lead-acid battery on charge). This is especially true in an automotive environment where there can be large transient voltages on the supply. You can get 100V devices cheaply enough, so why risk it?

Wetmelon
Good, TIP41 would be a more typical choice. To get sensible resistor values though, you must use a reasonable current gain value. At present you have R3=16 ohms, so it passes roughly 700mA, most of which gets into the base. I believe someone said that the collector current Ic is 500mA for the bigger solenoid. Where is the factor of 10?

This really isn't rocket science, but I appreciate that it may be confusing if you have not studied these things before. Possibly you might do better to ask for help from someone closer to hand. Could you ask a lecturer, or perhaps another student who is studying Electronics as a major subject?

DOH! how did I manage to bung up such a simple calculation?

R1 = Supply Voltage / ( Maximum Current Required / Minimum HFE * 1.3 )
R1 = 12V / (.3/10?*1.3) = ~310ohm
R2 = 310*10 = ~3100ohm

R3 = Supply Voltage / ( Maximum Current Required / Minimum HFE * 1.3 )
R3 = 12V / (.5/10?*1.3) = ~185ohm
R4 = 185*10 = ~1850 ohm

SOOO using 185ohm, V = IR = 12V/185 = 65mA. * 10 / 1.3 = 500mA on the head.

Fail. If you check the calculations that I did earlier in the thread, I completely missed the Minimum hFE. Don't know where it went, but it went! (And it seems that hFE is the forward current gain in a transistor?)

As for people near me, I'm currently at home 800 miles away from people in my program... My dad has an electronics degree from Algonquin from 1972... and hasn't used it since, so he's not much help. Though he did teach me how to solder.

Edit: I checked on the standard resistor values. Looks like there's decade multiples of 18 and 20. What would be more appropriate? Going up to 200 (a 7.5% deviation) or down to 180 (a -2.7% deviation)?

Last edited:
So far you're on track. I would insert three R1's and three R2's, one for each transistor, and run two of the R1 inputs to a single push button switch input. This might even make the board layout simpler. Your design might work as is, but I haven't thought about it.

Direct current power dissipation is P = V*I. Each resistor must be rated to dissipate enough power in the design, and the transistor must be rated to dissipate power when it is turned on. For example, the spec sheet says saturation voltage V = Vce(SAT) = 1.5V and you plan to pass collector current of iC = 0.5 amps. So the transitor without a heat sink must safely dissipate 0.75 watts. Check the spec sheet for the power dissipation rating at various temperatures. Then run a similar calculation to verify power ratings for each resistor.

The best practice is to test the circuit performance before soldering it up to a board. In this case some meter measurements of voltage and current, and verify the transistor saturates with the actual loads. You can defer the board design until after the circuit is tested, unless you wish to keep revising the board layout as you go along.

300mA Solenoid:
P = IV = .3 * 1.5 = .45W

From the TIP41C Datasheet:
Collector Dissipation (TC=25°C) 65W
Collector Dissipation (Ta=25°C) 2W

So it can handle it bare with no problem.

Resistors:

For R1 (300mA solenoid)
VBE(sat) = 2.0V
P = 2.0V * .065A = .13W

For R2
10V = I * 1850ohm
P = 10V * .0055A = .055W

For R3 (500mA solenoid)
VBE(sat) = 2.0V
P = 2.0V * .039A = .078W

For R4
10V = I * 3100ohm
P = 10V * .00325A = .0325W

But I need someone to double check these to make sure I have any clue what I'm doing. I'm pretty sure I have R1 and R3 right, but I am really unsure of R2 and R4.

Well done, you're getting the hang of it now... Except for some resistor ratings!

Noting that a bit less than the full supply voltage is across R1 and R3, you might like to round them down to the next preferred value. Use the preferred values when making the resistor wattage checks.

As for the resistor dissipations, please note that for the circuit in your diagram, most of the supply falls across R1 and R3. It's the other resistors that get VBE! Lets do a sanity check: The 500mA IC transistor has about 65mA base current, so if we have about 10V across R3 it would dissipate roughly 650mW.

Actually, you should assume the maximum possible battery voltage. If you are using a lead-acid battery, this could reach about 14.5V on charge. For R1 and R3 dissipation, don't assume the absolute maximum VBE. Something more like 1V VBE may be more appropriate. Thus you could have about 13.5V on a 180 ohm R3, giving a shade over 1W. Remember P=V2/R

Don't forget the anti back-emf diodes, and you might think to add series resistors if space permits.

Now I have to go and try to fix the lights on my Christmas tree, so that's all for now.

You need to take into account the Safe Operating Area of the transistor. National Semiconductor has application notes on this subject.

Sorry I forgot you're not EE so my instructions did not explain how to find the voltage across each component. But you have the right idea.

Instead of building and testing the circuit (or before doing so) you could build it in a free student version of SPICE. Depends on whether you want to handle the learning curve.

I've never used 5spice, but it appears to be free and easy to use:

The TIP41 is a special part so you'd need to import a device model.

Result: http://www.onsemi.com/PowerSolutions/supportDoc.do?type=models&part=TIP41

I'm not sure how to model the solenoid coils however, you may need an inductor value and a resistor value which are unknown ...

The point of simulation would be to study the circuit performance in terms of voltage, current, power ratings, and transistor saturation before building anything in hardware.

Attachment shows the rough equivalent circuit in the base-emitter loop when the button is pushed. The base-emitter (internal) diode turns on and has a small voltage of 0.7 volts in small transistors, it shouldn't be much more in the TIP41, but I don't know how much. Anyway each R1 may see significantly more voltage than in your calculation.

#### Attachments

• npn_on.gif
1.4 KB · Views: 440
Last edited by a moderator:
Someone just mentioned the dreaded Safe Operating Area. It's a valid enough point, but I think a TIP41 is man enough to handle a 500mA relay on a 12V supply, provided you don't forget the coil transient protection.

TIP41 looks good for up to 40V continuously for currents up to 1A. See Fig.3 on Page2 of the attached data-sheet.

#### Attachments

• TIP41_data.pdf
36.6 KB · Views: 112
I decided to try 5Spice and ran this saturation model of the TIP41C with success. A load resistor RLoad = 24 ohms will draw 0.500 amps at 12V, so I put this in as the coil resistance.

R1 = 330 ohm 0.5 watt rating 5% tolerance.
R2 = 3.3k ohm 0.25 watt rating 5% tolerance.

Base current is just 0.034 amps with the switch closed (shown as 12V battery). Power dissipation is about 0.38 watts in R1, so use 0.5 watt rated component. Power dissipation in the saturated transistor is a few tenths of a watt. When I set the input battery to zero volts the transistor turns off and a nano-amp current flows in the load resistor.

You can use the same values of R1, R2, and D1 (diode) for both the 0.300 ampere (40 ohm load) solenoid and the 0.500 ampere (24 ohm load) solenoid and the transistor should saturate properly in both cases.

The only cautionary issue is that simulated saturation voltage was quite low, less than a tenth of a volt, and I am not sure why the datasheet has such a high saturation voltage listed. It is probably because the collector current is 6A under the spec sheet saturation test. Yes, I think that's why, and with collector current at less than 1/10 that value in the solenoid design the simulator is probably fairly accurate.

#### Attachments

• Tiptest.png
2.8 KB · Views: 407
SystemTheory:
Hurray! My calculations were rather close, then! I suppose this clears me for breadboard stage testing :P

From what I know of the car's electrical system, we get (surprisingly) little transient voltage. Our main problem is electrical/magnetic fields created by the alternator...

OTOH, if you would like to explain to me how to include protection in the circuit without throwing off the calculations too much, feel free :)

Surge Protection.

Unless the solenoids all have with internal anti-back emf measures (check, don't guess), you need to include a flywheel path for each coil. As a minimum, connect a diode like 1N4002 etc. in parallel with each coil with the cathode (bar end of symbol) to positive supply and the anode (triangle end) to the collector.
The coils will turn off more briskly if you can add a resistor in series with each diode, of value roughly equal to the coil's resistances (say, 24 and 39 ohms) and rated to carry the coil current briefly.

For protection against external transients, consider adding a Transient Voltage Suppressor between Collector and Emitter of each transistor, something like 1.5KE33A (aka 1N6283A), might fit the bill - see attached datasheet.

The above measures will have no effect on your calculations so far.

Finally, there is something that has to be said. FOR YOUR OWN SAFETY AND THAT OF OTHERS, PLEASE ENSURE THAT YOUR DESIGN IS CHECKED BY A COMPETENT PERSON BEFORE IT IS TESTED IN PRACTICE. However sensible any advice you obtain from this forum may appear to be, you don't really know where it comes from. Neither do we know exactly your situation. So do yourself a favour, get it checked by someone you can be sure to be competent.

#### Attachments

• 1N6267A-D.pdf
77.6 KB · Views: 142