| Created by |
|
| Created | July 03, 2014 |
| Last modified | April 30, 2015 |
| Tags | No tags. |
Not provided.
This circuit provides a constant current load through Q1 and Q2. A source to be tested should be put in place of VTEST.
Starting from the left, we have a PWM signal that will come from an Arduino IO pin. This PWM signal gets smoothed out by an RC filter to produce a smooth analog voltage. This voltage is fed into a unity gain buffer. This buffer is not required but I plan to use an LM358 which is a dual op-amp, so I might as well make use of it. OA2 receives this analog voltage signal and tries to maintain the same voltage across R2 by biasing Q1/Q2 appropriately.
Thus, by changing the PWM duty cycle at the input, you can change how much current this load is drawing through Q1/Q2.
|
I like this circuit and need something similar to test solar panels current at different loads. Newark.com dosn't have any 'D45VH7' so I think I'm going to get this transistor if you think that will work. I like how you have it connected to a microcontroller to control the effective load and measure the voltage and current. |
May 10, 2015 |
|
Hi Garrett, glad you like the circuit. The transistor you linked would work but is kinda overkill for this. Those TO-3 packages aren't any better at getting rid of heat than T-220s like the one I use. You could use a cheaper transistor and it would work just as well. I can help find a good transistor if you like. |
May 10, 2015 |
|
So I just made up the load part of the circuit and it works well. I used a TO-3 transistor just because I never have used that package style before, though I would give it a try. A few things I noticed: 1) The op amp only responds to input voltage 0-2.5 (powered by 5) 2) The effective load can be changed with differed load resistor (obvious but its nice to be able to test different ranges by just switching out resistors 3) The first transistor BC338 is by far the hottest component to the point where I think I'm at risk of damaging it. Should I use another transistor? I was under the impression like that was going to be a low power transistor. |
May 25, 2015 |
|
Scratch that last comment. I had a little mix up switching out resistors and antecedently unplugged the big transistor, the little guy didn't like that. Anyways thanks so much for this circuit I have some solar panel testing to do and this will be a big help. |
May 25, 2015 |
|
Thanks for the feedback. 1) This is probably because your load transistors can't pull enough current to make the output go higher than 2.5V. Try using a higher value for R2 and see if this changes. 2) That's a good observation. You could even swap out different load resistors electrically using some transistors as switches, although this would make it harder to get an accurate resistance. 3) I see you figured out why your BC338 was overheating. This could happen in normal operation too if the transistor you use for Q2 is not very good. Basically if Q2 has low gain, Q1 has to do more work and it could end up getting pretty hot this way. Let me know how your solar panel testing goes! I'm planning to add more test modes to the software for things like solar panels and power supplies, like a ramp mode where it starts at 0 and increases the current gradually so you can plot the characteristics of your test device. |
May 25, 2015 |
|
I like the idea of a ramp mode, when I built the circuit on perfboard I kept that mod in mind so I could add it in the future if needed. This weekend I tested my 1.5W 12V solar panel. Here is the power vs voltage plot there was a lot of variation probably due to sun intensity and temperature, but the idea is there. Here is just constant current draw vs. time to show the veration over ~5 minutes. Thanks again for the circuit |
June 08, 2015 |
Please sign in or create an account to comment.
Only the circuit's creator can access stored revision history.
