I was following the instructions on this page: solarbotics.net/library/circuits/se_t1_mse.html

to try to build a "Miller" solar engine, using a solar panel generating ~4V to drive a small DC motor. This is the motor I used:

dc motor

The schematic of the circuit is as follows, with the 2N3906 and 2.2k resistor removed, and R1 = 0 (as instructed):

the Miller solar engine schematic

I didn't have a 1381 voltage detector, and used a TC54VC3002 instead. Besides that everything was exactly the same. But after connecting the solar panel to the circuit it just won't work, no matter how I tried.

For debugging, I tried replacing the solar panel with a 9V battery and the circuit works well, the motor sucking ~150mA of current, but the 2N3904 was getting extremely hot. So I switched in a TIP120 for the 2N3904 instead. What happened was that when I switched the solar panel back for the battery, the motor actually turned!

Here's the working circuit with the TIP120 in place of 2N3904:

I am very befuddled by this and hope that someone can explain to me, what is the difference between using a 2N3904 and a TIP120 in this circumstance, such that one won't work and the other one does?

  • \$\begingroup\$ What value R1 did you use? \$\endgroup\$ – WhatRoughBeast Mar 7 '16 at 3:18
  • \$\begingroup\$ @WhatRoughBeast I didn't use R1 actually, coz the author of the instructions said it's not necessary when driving a small dc motor. \$\endgroup\$ – Gao Yuan Mar 7 '16 at 3:43

Right. Your problem is that you're using a motor which needs more current than the one in the article. In order to turn on the NPN hard (which will provide minimum voltage across the transistor and maximum voltage across the motor, you need to supply more current from the TC54 than it could provide. With inadequate base drive there was a fair amount of current through the motor, but not enough to get it turning. At the same time, the voltage across the transistor was high enough that the power dissipated in the transistor (voltage times current) became objectionable.

When you switched to a TIP120, you moved to a Darlington transistor, which has much greater gain. The greater gain allowed the detector output current to turn the transistor on much harder than the 2N3906. This allowed more voltage across the motor, and more current through the motor.

I suggest that you measure the motor current when using the TIP120, and I expect you'll find significantly more than you got with the 2N 3906.

There is one drawback to using the TIP120. Since it's a Darlington, its saturation voltage will be about 0.8 volts more than a 3906. In other words, if you could drive the 2N3906 harder, you'd find the collector-emitter voltage to be about 0.2 volts, but the TIP120 won't go lower than about 1 volt. This means that, if you could drive the 2N3906 harder, you'd find it stays cooler and provides more current in the motor than the TIP120. Since you couldn't drive the 2N3906 hard enough, though, this potential superiority of the 3906 never became an issue.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.