Losing power through solar controlled transister.

Thread Starter

Grashwi

Joined Nov 25, 2015
23
First, the values. I've added a 47ohm resister in series with a red LED as the load. So now, it looks almost identical to the schematic above. My voltages are now, E/Gnd=2.64 B/Gnd=1.93 C/Gnd=2.27

Alec, what you are suggesting is intriguing if for no other reason as it isn't something I would have had a clue about and what happening has me clueless. Is there a way to test this? Is there a way to solve this or is it just potentially a fact that I have to live with? If you are suggesting is that this may be a result of the nature of the panels I'm using there may not be a reasonable solution at these power levels.

Grant
 

GopherT

Joined Nov 23, 2012
8,012
This circuit is pretty crappy. The base must be 0.6 v below the emitter voltage for current to start flowing. Above that, the base needs to let current flow FROM the transistor to GROUND. That cannot happen because the photocell is really a diode PN junction that is reverse biased. So, to get 20 mA to flow to the diode, you should have 2 mA from the base to ground. That is not happening (even if the photo cell was connecting the base to ground directly) because the 5k resistor is limiting current flow to sub-milliamp levels.
 

Thread Starter

Grashwi

Joined Nov 25, 2015
23
Do you have a suggestion as to how to make this work or are you saying that the concept is fundamentally flawed? I have enough battery and solar charging power to just let this run 24/7 if this solar panel approach cannot work, but this circuit seemed like an elegant solution for solar applications. You are saying this isn't at all great. If this is on the 'ain't happening' list do you know of an efficient alternative?
 

GopherT

Joined Nov 23, 2012
8,012
Do you have a suggestion as to how to make this work or are you saying that the concept is fundamentally flawed? I have enough battery and solar charging power to just let this run 24/7 if this solar panel approach cannot work, but this circuit seemed like an elegant solution for solar applications. You are saying this isn't at all great. If this is on the 'ain't happening' list do you know of an efficient alternative?
I do not have an immediate suggestion but the lack of real current flow from base to ground is killing this idea for now. I think it will be easier with an NPN. I will think about it. I am sure others have a suggestion. For now, just get rid of the transistor and let the diode determine the power source so it can run 24/7 if you need more current.
 

Thread Starter

Grashwi

Joined Nov 25, 2015
23
Actually, your comment about possibly using an NPN is interesting. I had toyed with using the above circuit to actually hold open a 2v relay during the day and then having the relay released when sunlight fails. But, for the reason Gopher has proposed, that didn't work either because I couldn't get enough power to the relay. I wonder is going the NPN approach would allow me to hold that relay open during daylight and then release it during night. I think that means that I'll be consuming power to hold the relay open when I'd rather be pushing it all into the batteries but I'll have to do some math to find out if I come out ahead or not.
 

dl324

Joined Mar 30, 2015
12,447
The circuit posted by @ronv should work better. It avoids the base drive issue by grounding the base resistor and using solar cell charging voltage to turn the transistor off.

I was trying to confirm the lack of base drive, but you didn't provide measurements.
 

Thread Starter

Grashwi

Joined Nov 25, 2015
23
Thank you all. This has been a great help. A last couple of questions and I'll go down the road Ronv suggested. In his post 14 I'm not familiar with the symbol in the lower left corner. The circle with the +- symbols in it. Also, what is the PULSE label under that. And, while I'm at it, what does the Rser=2 by the battery mean.

Yeah, that's how much a novice I am....
 

ronv

Joined Nov 12, 2008
3,770
ok, you're talking to a real amateur here. This is a single pulse value? I'm looking for the transistor to act as a switch that would simply funnel full amperage through at saturation for the duration of the base being no longer under bias, when the solar panels are not charging. Idealy to applies bias to the basis of the transistor during daylight, blocking any flow, and then allows uninterrupted current flow when no bias is applied to the transistor. My measureable problem is that I cannot get full amperage through the transistor upon saturation even though I get voltage. I appreciate your input but can't understand how it applies to my issue. I only want the transistor to act as a switch, acting a couple of times a day. Not in some Mhz frequency. Why does the transistor cost me such a huge cost in amperage?
Make believe it was a very short day. :D
For the transistor to turn on the voltage on the base must be lower than the voltage on the emitter.
As Alec has pointed out the solar panel doesn't "pull" it down in the dark.
 

Thread Starter

Grashwi

Joined Nov 25, 2015
23
I've recrafted the circuit using Ronv's version and it's a great improvement. Thanks to all of you for your help. As to effectively using this there is still some fundamental flaw that I don't understand. If I put all of this together including my capacitor charging circuit and joule thief with two UV leds I compared results. Putting two 1.2vdc batteries in series and running through this circuit I can charge my capacitor to 200volts. That's actually enough to knock off mosquitos but it's marginal. If however I use a single 1.2vdc battery and drive the charger/joule thief directly I drive the capacitor to nearly 400 volts which is better. So, half the voltage in gets me twice the capacitor charge out. The transistor is this circuit is robbing me of 3/4 of my input power and I have no idea why. This approach does however reopen the option of using an intermediate relay to hold the circuit off during daylight but then release the battery current at night. It's almost as much current as the zapper though so I have to do the math to see if it's worth it at all or if running this 24/7 isn't the easy button.
Thanks to all for your help. You've put up with novice questions and provided great guidance. If I run into issues in the future I wouldn't hesitate to post them here.
Grant
 

GopherT

Joined Nov 23, 2012
8,012
I've recrafted the circuit using Ronv's version and it's a great improvement. Thanks to all of you for your help. As to effectively using this there is still some fundamental flaw that I don't understand. If I put all of this together including my capacitor charging circuit and joule thief with two UV leds I compared results. Putting two 1.2vdc batteries in series and running through this circuit I can charge my capacitor to 200volts. That's actually enough to knock off mosquitos but it's marginal. If however I use a single 1.2vdc battery and drive the charger/joule thief directly I drive the capacitor to nearly 400 volts which is better. So, half the voltage in gets me twice the capacitor charge out. The transistor is this circuit is robbing me of 3/4 of my input power and I have no idea why. This approach does however reopen the option of using an intermediate relay to hold the circuit off during daylight but then release the battery current at night. It's almost as much current as the zapper though so I have to do the math to see if it's worth it at all or if running this 24/7 isn't the easy button.
Thanks to all for your help. You've put up with novice questions and provided great guidance. If I run into issues in the future I wouldn't hesitate to post them here.
Grant
You can also use a comparitor or op amp like this. Some low quiescent current op amps use almost no power. Here is an option..

I use 200 mV for dark and 2 V for sunlight.

LM393, for example can run off of 2 v. Power it from the NiMH batteries.

image.jpg image.jpg
 

Thread Starter

Grashwi

Joined Nov 25, 2015
23
Wow. You folks rock. This is going to be a long term hobby for me so at some point I will know what a comparator or op amp is. Then, when I understand them, I'll consider deploying them. Right now though I've got the 'muscle' I need to deliver the mosquito control I want with the capacitor charging/delivery and joule thief driven UV leds. I have enough battery and charging ability to put up with 24/7 operation so inelegant or not, I'm going to go with that. I've printed everything and when my comfort zone includes suggestions such as yours I may circle back around and redesign this.
Thanks for all the help I've gotten from you folks.
Grant
 
Top