I need help about DC-AC Inverter

ronv

Joined Nov 12, 2008
3,770
Maybe what we can do is replace the 12 volt regulator with a 5 or 6 volt one (no diode), but move the positive supply for the op amp (comparator) straight to the +12 form the battery.
Which FET are you using the IRFP250 or the 540?
 

Alec_t

Joined Sep 17, 2013
14,280
With 12k, the ref voltage at the inverting inputs of the comparators is ~7.8V.
If the multivibrator supply is now only 9V that puts the comparison point on a rather curved rising part of the waveform. Perhaps better to increase the 12k to 22k to move the ref down to 6V where there is a sharper rising edge; or even swap the 12k with the 22k (R6) to move the ref down to 4.6V?
 

ronv

Joined Nov 12, 2008
3,770
If you leave the reference voltage on the regulator it would be stable and 50%. Only move the power for the 358 to +12.
With a 9 volt regulator the base emitter voltage is still out of spec. - but better. ;)
 

Attachments

Thread Starter

waldopulanco

Joined Nov 14, 2013
26
I have change some value and voltage regulator to 7808!! and the + supply for LM358 is direct to +12v and the voltage devider or reference voltage is from +8v regulator!

but what is better for R5 2.2k, 10k, 12k or 22k?

here is Screenshot:
 

Attachments

Last edited:

ronv

Joined Nov 12, 2008
3,770
22k will center the threshold voltage so the square waves will be symmetrical. You seem to have lots of regulators, why don't you use a 7805 so the reverse bias on the transistors won't be out of spec???? The base voltage on a BC547 should not be more than 6 volts below the emitter. In this circuit it goes negative almost to the supply voltage. See the simulation in post 26.
 

Thread Starter

waldopulanco

Joined Nov 14, 2013
26
22k will center the threshold voltage so the square waves will be symmetrical. You seem to have lots of regulators, why don't you use a 7805 so the reverse bias on the transistors won't be out of spec???? The base voltage on a BC547 should not be more than 6 volts below the emitter. In this circuit it goes negative almost to the supply voltage. See the simulation in post 26.
ok i will try!! how about R5 and R6 is the same value 22k?
 

bertus

Joined Apr 5, 2008
22,270
Hello,

With a smaller value of R5 the pulses will be separated, due to the rounded top of the pulses on the multivibrator..
Take a look at the simulation for the signals on out 3 and out 4.
This "dead" time will assure that not both fets are conducting at the same time.

Bertus
 

Thread Starter

waldopulanco

Joined Nov 14, 2013
26
So what is the best value that i can place on R5 and R6?
i did not test it on 78L05! my 78L05 regulator was gone!! i have only 78L08 and 78L12!! i test it on 78L08 and it is much better than diode or 78L12!!
 

bertus

Joined Apr 5, 2008
22,270
Hello,

When R5 is 2k2 the dead time will be larger as when R5 is 10k.
You could try both values and see if the transformer is getting warm at higher load.
(when you actualy are building the circuit).

Bertus
 
Top