Another INVERTER we can talk about..

[MP1]

...With Sasi's relay turning the circuit off, it doesn't need complete isolation since the MOSFETs' uA of leakage current is much less than the self-discharge current of the battery.

No need to be sorry, audioguru.
Show the math. Thevinen's theorem would not agree with your statement. This would be like a parasitic draw on the battery. It would also tend to add other problems in the future. All it takes is a simple add on to isolate the battery. Not a big design add-on. Why would anyone not want the added protection? I have to deal with these types of problems in the field all the time. All it takes is a little thinking ahead.

MP

 

[audioguru2]

Hi MP,
Math? But I can't find my old slide-rule!
I'm glad that you agree that Sasi's relay turns-off his circuit, because my new-fangled calculator tells me that a 150A/hr car battery will power the maximum leakage current (25uA each at 55V) of the six MOSFETs for more than 116 years!

Maybe Sasi needs backup only for the short time for his hard-drive to park, and uses just a string of 4A/hr D-cell Ni-Cads. They will power that "parasitic" leakage for more than 3 years!

I think that the little solar-cell (a real one, not just a label) on my calculator can supply the 150uA of leakage current. ;D

Hi Sasi,
Your PCB layouts are very nice. You should post a complete project.

 

[Kasamiko]

Sasi,
How about adding some charging circuit whenever there is a main power? I think replacing the transformer that power the relay with bigger one to supply a charging current to the battery is a nice idea.
What can you say guys? 8)

Rhonn

 

[ante1]

Audioguru,

Sounds good, I am looking forward for a schematic to be posted sometime in the near future.

 

[ante1]

Audioguru,

Oh, I forgot to mention I have a ton of RFG50N05 so it would be nice if they were usable . Its not a must I can get some new if required.

 

[ante1]

I did not analyse the complete circuit but if the battery is 12V then the LM7812 will not work here.

 

[ante1]

sasi,

Well the high point can be 13.75V without losing any water and the low point depends on how long you want your battery to live. Best tip here is the manufacturers recommendations for maximum discharge (minimum voltage). An alternative is to trickle at 13.75V and never shut it down completely.

 

[MP1]

Hi MP,
Math? But I can't find my old slide-rule!
I'm glad that you agree that Sasi's relay turns-off his circuit, because my new-fangled calculator tells me that a 150A/hr car battery will power the maximum leakage current (25uA each at 55V) of the six MOSFETs for more than 116 years!

Maybe Sasi needs backup only for the short time for his hard-drive to park, and uses just a string of 4A/hr D-cell Ni-Cads. They will power that "parasitic" leakage for more than 3 years!

I think that the little solar-cell (a real one, not just a label) on my calculator can supply the 150uA of leakage current. ;D

Hi Sasi,
Your PCB layouts are very nice. You should post a complete project.

Time for you to get a new slide rule. If environmental and meteorological equipment was built upon the same assumptions, the industry would be a mess. I never disagreed that the relay would turn off his power. He still has resistance to ground. If you look at the components as resistors, and note that you have six in parallel with only looking at the transistors, you can use thevinen's theorem and see that you have enough parasitic draw to draw down the battery after sitting for some time. Good design practise always isolates the battery on equipment when it is not in use. This is a standard.
I only metioned it as a courtesy to sasi, since he could use a DPDT relay and take care of this also. Quite frankly, if you guys want to ignore good design practice, it does not bother me a bit. Just don't deny that it is a better design to isolate the battery from the electronics when it is not in use....and at the cost of only one section of the relay!
..and since I am on the subject of design, the circuit could use a method of charging the battery. The availability is there but not used. There should be a step down from the mains and a regulator to keep the battery up when the mains are giving the available power. Even with no draw, the battery will not stay up if not used for long periods. It will need to be charged. This is just a start. This circuit, if used only for emergency back-up, will also need a way to put a load on the battery once in a while and then charge it. An idle battery will not last long. I do not see much long term usefulness for this circuit "as is". Perhaps as a project, it could be considered a building block for something more involved.

MP

 

[ante1]

The most common UPS:s where I work just have trickle charge at 27.6Volts and these batterys (Dual Panasonic 12V 14Ah) is changed every five years. When measuring voltage and voltage under load (40 A for 15s) on these recycles only 5 out of 50 show any signs of degradation. However there is a charging method called reverse pulse charge that can be used if one considers idling a threat to the battery. Sometimes a high current pulse is used, once every hour (or at any suitable interval) a high current pulse is pushed through the battery causing the voltage to raise and stir the electrolyte. The length of this pulse can vary as well as the strength. The number one reason for batterys to lose there capacity is sulphating and one cause for this is the separation of the acid in the electrolyte. This means that on the bottom of the cells the gravity of the electrolyte can be 1.28 and on the surface 1.22. The sulphating starts at the bottom of the cells and eventually causes shorting of the plates. This will not happen if you can stir the electrolyte from time to time. On some bigger batterys they are even pumping air through plastic pipes that discharges at the bottom of the cells during charging. This is like the pumping of air in a fishtank only it

 

[audioguru2]

Hi Sasi,
On your latest circuit, pin 7 of the 324 appears to always be high, and therefore doesn't do anything, just like in the original circuit. I think that pin 6 is missing a voltage reference which would allow pin 7 to go high only when the voltage on pin 5 is higher than the reference. Then the relay would activate and the LED would turn on when the battery voltage is too high.

 

[ante1]

Since this is not a serious thread anyway I think it

View attachment 35740

 

[audioguru2]

Another inverter?
Wait 'til you see the "Class-D" one that I'm working on. Sine-wave output with very high efficiency (runs cool).
When it's shut-down, it will be protected from the environment and BS by keeping it indoors, and covering its "sensitive" bits in case you spill your drink in it or other disasters.
I was going to use the National Semi. one, but it's being discontinued next year, and many people on another forum had theirs blow-up! They were even worried about dropping the IC (that's "OOps" dropping, not voltage dropping). ;D

 

[MP1]

ante, do you have enough for everyone... ;D ;D

MP

 

[audioguru2]

Who is counting?

 

[ante1]

MP,

Yea, sure it

 

[nvd]

Can anyone tell me about the speed of the inverter with which it energises? In the case of power failure, is it fast enough in providing power? Will my computer restart if it trips ON to provide power in the case of emergency?

 

[audioguru2]

Hi NVD,
Welcome to our forum.
The cheap and simple square-wave output inverters use a heavy relay to change-over the power source when the mains power fails. They must detect the failure and move those heavy contacts. Of course the power will be disconnected for some time and your computer might reset.

Your computer and some other electronic devices might not operate with a square-wave power source anyway since their power supply might rely on the higher peak voltage of a sine-wave power source.
A square-wave inverter is fine for lights and some power tools. Some people reported that they power international TVs fine because those TVs run on nearly any voltage.

Computers usually use an Uninterruptable Power Source (UPS) for backup power. They usually contain a sine-wave output inverter that is running and powering the computer all the time. The only thing they use mains for is to charge their battery.

 

[Alun]

Just and idea, but how about connecting capacitors in paralell with the transformer to form tuned circuits to make sine waves?

 

[audioguru2]

Hi Alun,
Your idea might work if the transformer doesn't have a load and you operate the Mosfets in class-C so they don't load-down the tuned circuit much. If you time the class-C pulses correctly at zero crossings, you might even get mega-volts out of the thingy!
You would probably need enormous caps to tune a big transformer to the mains frequency.

How do those very expensive tuned transformers in UPS's work?
They even regulate the output voltage.

 

[Alun]

Hi Alun,
Your idea might work if the transformer doesn't have a load and you operate the Mosfets in class-C so they don't load-down the tuned circuit much.

Yes, the duty cycle needs to be a bit lower than 50% so both transistors aren't on at the same time at any given point.

If you time the class-C pulses correctly at zero crossings, you might even get mega-volts out of the thingy!

I doubt it because this is parallel resonance and you get current magnification not voltage.

You would probably need enormous caps to tune a big transformer to the mains frequency.

Well they would only need to be big enough to carry the off load current. You can also buy pretty big capacitors for power factor correction and audio crosover networks.

An over-sized primary coil and core may be of some benefit too because loading of the secondary will reduce the Q and thus the quality of the sinewave. This doesn't really matter too much as bigger appliances are far less likely to effected by the unusual waveform anyway.

The capacitors will also reduce the off load current and oncrease the efficiency significantly because the power factor will be a lot higher and it will be easier on the transistors too because the capacitors will absorb the back EMF and send it back to the coil.

How do those very expensive tuned transformers in UPS's work?
They even regulate the output voltage.

I don't know, they probably use capacitors in a similar way to my circuit and may be some kind of active feedback device to regulate the output.

I'm planning to do some experiments, I'll start of with a single transistor rather than jumping in at the deep end with the push-pull driver.