Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Battery "Bounce"

Status
Not open for further replies.

SirSparks

Member
This is my second (and only other) question for the experts here, my first problem was resolved perfectly thanks to the help I received. First a little background; 6 months ago I decided to leave the rat race and fit out a box truck as a self sufficient home complete with solar electricity and solar hot water.

I have just one remaining annoyance with the PV system; I have a 2000W modified sine wave inverter and a 600 watt true sine wave inverter both 12V/120V and both with over-voltage protection. This means that with an input of above 15.5V they will each trip and need to be manually reset.

The problem with this is that several times during the day due to various loads switching or the PV controller changing mode the voltage will momentarily climb to 16V before being attenuated, this happens in the blink of an eye too fast for a voltmeter to register it BUT the controller records it and the inverters trip. The controller manufacturer refers to this as “battery bounce” and states that is due to the type of battery I am using, however to change battery types is a $1,600 cost I can't afford. When I leave the “home” a 16V surge is likely to shut down the fridge until I return to reset the inverter; Not a good thing so I need a fix.

I estimate that the surge is well under one second duration and only rarely occurs, I need to limit this surge voltage to below 15.5V but without the limiter drawing load below normal charging voltage of up to 14.8V. How many amps this circuit/device should draw as a dummy load will need to be ascertained by trial and error but I expect 300w or 25 amps would do. As this load will only be active for a fraction of a second it will be of no consequence to my overall efficiency.

I was thinking a dummy load resistor in series with a large Zener diode both across the 12V PV output?
If not possible maybe a small zener powering a relay coil which then switches a dummy load?

Any thoughts appreciated.
 
Depends on how much current that over voltage is able to source and it's true duration. A large parallel capacitance may eat the surges if they're of a very short duration and/or low energy. A zener would probably work as well if the duration is short but you have to determine the amount of energy involved or you'll blow up the zener, a load resistance would help take some of the power dissipation off the zener but again you can't really pick one without knowing the amount of power involved.
 
Last edited:
.Unfortunately I can only guess the amperage needed` as the surge only lasts for a fraction of a second it is too fast to measure. What I can say is this; presently I am producing 510 watts of PV power at 13.7V if I now (I just did) switch on a 750 watt toaster the battery voltage immediately drops to 13V so as this 0.7V difference is about what I need to reduce 16V surge to 15.3V then I guess this would be a good dummy load rating. Again I am not sure of the duration but certainly less than one second before it re-settles to normal.
 
Here would be my easiest solution; I have an existing 12V 1800W load which is rarely used (hot water heater instant temperature boost) 1800W is more than enough I am sure. It is operated by a relay the coil pulling a little under one amp 12 volts. So what I need is a zener switching on at about 15.3 volts and back off by 14.8V, with 1 to 5 amps capacity. And such beast
 
A silicon diode has an on voltage of about 0.65 volts ... which would probably be sufficient to drop the peak critical voltage.
Ideally though, you might want to have a method for removing that diode from the circuit when the voltage is not critically high.
... That might require some sort of voltage comparator and a relay.
 
I need an "on" voltage just below 15.5V. 15.3 would be perfect as long as it can shut off again by 14.8V or I'll be taking a lot of hot showers LOL!
 
A relay may be too slow (think several mS) acting to prevent the inverters tripping (assuming they have a rapid response to overvoltage). That leaves two options: 1) somehow slug the inverter response time (but that might result in damage to the inverters), 2) use a rapid-acting shunt regulator consisting of a precision adjustable voltage reference (a zener might do, but they don't all have a very sharp knee voltage) to turn on a power FET to switch a fat load across the supply.
 
I guess what might be useful is some means of determining when the critical voltage ... 15.5 volts, or whatever ... is imminent.
 
The inverters give me a short warning beep, sometimes if the transient is very fast they will just beep but not shut down so I think I could use the relay option, worth a try anyway as it is all existing I'll just need a second control feed to the relay I see your point about the breakdown voltages not being exact though and I need a pretty darn accurate voltage here.
 
SirSparks.. If this pulse is as short as you claim simply putting a lot of capacitance in parallel could reduce the voltage spike. The shorter the pulse is the less capacitance you'll need, and the nice thing about it is it doesn't waste the energy.
 
Adding a small inductance (hash filter) before any added capacitor should help.

You could also consider adding an auto-restart circuit to the inverters. You could wait a few seconds after the output voltage goes to zero and then close a relay contact across the reset switch.
 
I'll certainly give it a try Sceadwian. As I have 585 AH of battery I am guessing it will need to be a large cap. I'll see what I can find, thanks.
 
Not exactly true SirSparks, that's why I was trying to say the exact form of the spike is all important, because if it's a simple high voltage rebound from cell resistance it's a tiny tiny little spike, just enough for the regulator to fault out at by registering, if it's basically a regulator response issue with the change in loads it might be more problematic to fix with a capacitor, again though it depends on exactly what the nature of the spike is.

Crut, interseting term usage there, never heard of a hash filter before. I'd call them chokes.
 
Moved this thread to the new Automotive sub forum.

This is sounding a lot like the sort of problem that exists in automotive electrical systems. There are many automotive grade regulators on the market. Maybe there is some help in that direction.
 
All good food thoughts guys. The PV (Outback Flexmax 80) Controller manufacturer assures me the spike is due to battery type (Automotive instead of solid lead plate RE) so in that case maybe it is a simple fix of a situation caused by internal cell resistance. It is a very long time since I was in electronics ( To give you a clue the BBC transmitter valves/tubes were water cooled), I did an Ebay search looking for a very large cap, thinking very large was maybe 100 mfd. I was amazed to find caps of 4 Farads !!! Something to do with car stereo systems??
 
If it was me I would not hesitate to toss a bank of capacitors on it with several different values from .1 uf all the way up to several tens of thousands of uf.

The small ones will have the least ESR and fastest response times while the bigger ones will progressively trade off response time and internal resistance for shear energy absorbing capacity.

Best part is a progressive capacitor bank like this and of this size for a 12 volt system can be built for only a few tens of dollars at most.
 
There's no guessing there SirSparks... If the couple dollar caps don't work you might want to borrow someone's multi farad caps... Again this is why you need to know what impedance the high voltage event has. The power of the inverter doesn't matter, if this is a mili or micro second switching event...

I'm just guessing and so is everyone else because the 'event' that cause the problem isn't fully described.
 
The PV controller records the event in it's log as a 16V surge but as to time I can only guess although I am sure it is less than one second.

I like the capacitor idea and will purchase some cheap ones on Ebay.
 
If you're sure how have you measured it? Such as the 16V limit? Perhaps it's a 100+volt transient... The limiter TRIPS at 16 volts that doesn't mean it's the limit of the problem... Short duration transients like that don't tend to harm circuits as long as they're non periodic.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top