| Author |
Message
|
| (In)Sanity |
 Posted: Sun Feb 17, 2008 9:27 pm Post subject: My little solar experiment |
 |
|
Forum Cosmic Wizard
Joined: 19 Oct 2004
Posts: 2145
Location: Phoenix AZ
|
So I have myself 40 watts of Monocrystalline solar panels (2x20) and have been experimenting a bit with running one of my servers off that power during a part of each day. Now the first thing someone will say is, it's only 40 watts..what good is 40 watts. Well not a whole lot, but still useful.
The setup I have right now does the following.
1. Solar panels charge a bank of sealed lead acid batteries. (5ah x 10) or 50ah (more like 30 as they never get what is rated).
2. The batteries are connected to an APC 750 VA Battery backup with a heavy duty diode (32 amps) acting as a barrier.
3. A circuit I built monitors battery voltage and when it reaches a set limit (14.2) it activates a relay that in turn kills AC power to the APC UPS. It then continues to monitor until the voltage reaches 12.2 - 12.3 (adjustable) and then disengages the relay.
The process repeats each time the batteries are charged.
The problem I'm having is that the monitor circuit is being triggered by random events. An example would be, turning on/off the overhead light in the room. Touching the battery pack with a meter lead to see what the voltage is at.
It appears my circuit has some major noise issues. Static and EMI are setting it off.
So, I would be most appreciative of any comments as to correct this problem. Everything works perfect when connected to a variable supply, it turns on and off exactly as it should and doesn't false trigger. I left it sit for a day just to check that. Once connected to my real environment it's not doing what is expected.
I've tried adding an LC filter to the input to help reduce noise.
Here is the circuit, it's best viewed outside the browser as the lines tend to vanish.
Click on the image to view it at its original size
This is kind of a Radioshack project, in other words it's not at all optimal choice of parts.
_________________
"I don't know, I was really drunk at the time" |
|
| Back to top |
|
 |
| Harold14370 |
| Posted: Sun Feb 17, 2008 10:29 pm Post subject: |
|
|
Forum Professor
Joined: 13 Apr 2007
Posts: 1579
Location: Pennsylvania
|
Oh, sure. You have time to play with this but we still have no latex. 
What kind of wiring did you use to make the connections to your monitor circuit. Is it twisted pair or at least do you have the leads twisted together? |
|
| Back to top |
|
|
| (In)Sanity |
| Posted: Mon Feb 18, 2008 7:05 am Post subject: |
|
|
Forum Cosmic Wizard
Joined: 19 Oct 2004
Posts: 2145
Location: Phoenix AZ
|
| Harold14370 wrote: |
Oh, sure. You have time to play with this but we still have no latex.
What kind of wiring did you use to make the connections to your monitor circuit. Is it twisted pair or at least do you have the leads twisted together? |
Guess I'm busted on that one 
Because the "monitor" portion is also the power supply to the circuit, I used standard single pair (red/black) parallel run automotive style wire in the 24 awg range with alligator clips on the ends for fast connect.
This circuit only connects to the batteries, however the UPS also connects to the batteries. I have to wonder if I don't have some kind of ground fault loop. The circuit appears to be much more stable since I've added in the LC filter. It still will freak out if you touch the battery pack once in a while. I assume that is static.
_________________
"I don't know, I was really drunk at the time" |
|
| Back to top |
|
|
| Harold14370 |
| Posted: Mon Feb 18, 2008 4:52 pm Post subject: |
|
|
Forum Professor
Joined: 13 Apr 2007
Posts: 1579
Location: Pennsylvania
|
| I'm thinking the alligator clips could be the source of your problem if that is what you are using to make the connection to the battery. If there is a high resistance connection, the leads could pick up some noise that they would not pick up if you had a good, solid voltage there. |
|
| Back to top |
|
|
| (In)Sanity |
| Posted: Mon Feb 18, 2008 9:21 pm Post subject: |
|
|
Forum Cosmic Wizard
Joined: 19 Oct 2004
Posts: 2145
Location: Phoenix AZ
|
Oddly enough it appears to be working well today. The added LC filter may have taken care of the problem. The alligator clips are pretty well connected, unlikely to be of an issue. The circuit only draws 6 ma, so unless the connection was downright loose it would not cause a problem. The voltage also has to rise to trigger the circuit, one would think a bad connection would cause more drops then spikes. It's non inductive.
The next problem will be to see if the batteries I'm using are going to take the abuse of constant charge/discharge cycles. The more solar watts I add the more the batteries will cycle.
_________________
"I don't know, I was really drunk at the time" |
|
| Back to top |
|
|
|
|
