[SOLVED] Pic 16f876a Pure Sinewave inverter

[doc007]

Hello Taner,
My output signal is a square wave 50 Hz and not a sine wave !!!!!!
Have you one explication .
Do you think that my problem is hard or soft ?
Thank you for the answer
Pierre

 

[@t]

Hello Taner,
My output signal is a square wave 50 Hz and not a sine wave !!!!!!
Have you one explication .
Do you think that my problem is hard or soft ?
Thank you for the answer
Pierre

Hello Pierre, do you use your ASM code or the original C code?
And hardwarewise, did you try a higher impedance trafo, (to unload the driverstage) and see what the waveform is then?
Is the driverstage ok? Can you control the A and B output levels from the Fet's manually, by controlling the Vo lines from TP250? (after removing TP250 or the HC257).
Greetings, At.

 

[doc007]

Hello @t
Thank you for your intervention.
I use the original C code from Taner without a and b voltage check .( with the check from a and b the inverter doesn't work.)
With a higher impedance trafo, the wave form is the same.
How can I do the control of A and B output levels manually ?
Pierre

 

[doc007]

Hello@t
I have removed the HC257.
Here are the tensions mesured :
VCC U3 , U4 : 9.84 V Is it enough for boostrap ?
VCC U5 , U6 : 12.70 V
Vo U3 and U4 :8.95 V
Vo U5, U6 :some mV
A :8.7V
B: 8.8V
Thanks for your answer.
Pierre

 

[Taner]

Hello doc007 and all my friends,
sorry for my mistake again, but U5 and U6 VCC pins should be connected to +10V, not to +U.
Please correct this pin connections in your schematics.

Best regards,
Taner

 

[doc007]

Hello Taner
Taner wrote

Hello doc007 and all my friends,
sorry for my mistake again, but U5 and U6 VCC pins should be connected to +10V, not to +U.
Please correct this pin connections in your schematics.

Must cross also by diodes D6 and D7 for VCC U5 et U6
Pierre

 

[Taner]

Hello Pierre

Must cross also by diodes D6 and D7 for VCC U5 et U6

Not, U5 and U6 VCC pins should be directly connected to +10V.

Taner

 

[doc007]

Hello Friends,
After modification of the schematic There are the tensions: HC 257 removed.
Vo U6 : 360 mV
Vo U4 : 8.87 V
Vo U3 : 3.8V
Vo U5 : some mV
can the problem be on the driverstage commanded by U3 , U5 ?
Pierre

 

[@t]

Hello @t
With a higher impedance trafo, the wave form is the same.
How can I do the control of A and B output levels manually ?
Pierre

Hello Pierre, to do a static test of the drivers: (HC257 removed)
Instead of the trafo, use two 12v (ca 1A or so) lamps, connect them from A to GND and from A to +U, they will light at half intensity (ca. 6v. each).
Now check if the Fet drivers are capable of shorting one out, so that the other will light at full brightness.
Pick 9 and 12 (pins 2 and 3 from U4 and U6) and connect them (in series with 470 ohm) to gnd and +5, and then the other way, +5 and gnd. The output lamps should follow.
Do the same for the other stage, B. Then check if you can increase the output load (with other lamps) to 10 or maybe 50Amps. (600W.)

If this does not switch as expected, we should go into more detail.
Good luck, At.

If you can connect a regulated voltage (or current) to pins 2 and 3, you should be able to control the lamps in various 'dim' stages, to simulate the analog levels from a sinewave.

 

[doc007]

Hello At,
I have made the static test of the drivers.
the output stage does not switch as expected

For stage B
+5 on pin7 and gnd on pin4 no switch
+5 on pin4 and gnd on pin 7 normal switch

For stage A
+5 on pin12 and gnd on pin9 no switch
+5 on pin9 and gnd on pin 12 normal switch

It seems to be that opto U3 and O4 doen't work correctly or can it be the mosfet Q1 or Q2 or Q3 or Q4 ?
What do you think ?
Pierre

 

[@t]

Hello At,
I have made the static test of the drivers.
the output stage does not switch as expected

For stage B: +5 on pin7 and gnd on pin4 no switch
For stage A: +5 on pin12 and gnd on pin9 no switch

It seems to be that opto U3 and O4 doen't work correctly or can it be the mosfet Q1 or Q2 or Q3 or Q4 ?
What do you think ? Pierre

Hello Pierre,
You're right, the two upper (high) sides don't react, it can ofcourse be either the opto coupler or the Fet's. I assume the high side is 'open' (non conducting)?
Otherwise you could blowup the lowside too, if that is switched on.
So the question is: If no input signal is applied, are the two lamps lit equally then?
If so, then the Fet's are 'open'. Problem is, you can't turn them on. We can do this by applying a positive voltage to their gate's.
If you can take out U3 and U4 (or free their pins 6 and 7), you could connect +12 or +10 via a protection resistor (100 ohm) to the gate of each Fet (one at the time) to see if A or B goes up (to +U) or the upper lamp goes off.
If a Fet does not follow your action, it's blown-up.

If it all works out ok, then we have to take a look at the TLP250.
I hope you succeed, good luck, At.

note: Because Q1 and Q2 and Q3/Q4 have their gates connected to each other, the test is somewhat rough, you can't distinguish between the two without separating the resistors (R41/R43 and 42/44), but that's no problem for this moment.

 

[doc007]

Hello At
Yes the Fet's are open.
In effect if no signal applied, the two lamps lit equally

Added after 20 minutes:

Here are the tension of alim TPL250
VCC U3 , U4 : 9.84 V Is it enough for boostrap ?
VCC U5 , U6 : 9.95

The normal supply of TPL250 is min. 10 V. Perhaps is 9.84 V to low to activate U3 and U4 ?

Added after 2 hours 53 minutes:

Hello friends
Isn't it a problem with the alim of U3 and U4 ??
The GND od fhese two opto is respectively on A and B.
Therefore the alim tension of U3 and U4 is only 5,11V.

Must i not pick the GND of U3 and U4 on general ground ??

 

[@t]

Hello At
Yes the Fet's are open. In effect if no signal applied, the two lamps lit equally

Added after 20 minutes:

Here are the tension of alim TPL250
VCC U3 , U4 : 9.84 V Is it enough for boostrap ?
VCC U5 , U6 : 9.95

The normal supply of TPL250 is min. 10 V. Perhaps is 9.84 V to low to activate U3 and U4 ?

Added after 2 hours 53 minutes:

Hello friends
Isn't it a problem with the alim of U3 and U4 ??
The GND od fhese two opto is respectively on A and B.
Therefore the alim tension of U3 and U4 is only 5,11V.

Must i not pick the GND of U3 and U4 on general ground ??

No not GND, control of the high drivers is with respect to point A or B.
Dont worry about 0,1 volt with resp. to 10 or 5,11v. that's not the problem. First check if your driver Fet's are ok. They might be blown-up.
If that is the fact, you should find out why that happened.
Only if the Fet's are ok, its time to fix the opto circuit, to drive them.

Resumé: Pos on TLP250 pin2 (and zero on pin 3) should control the Fet.
edit: In dynamic mode the supply is pumped-up to about +U (-0.7v). Now staticly you should try it out, or temporarily connect pin5 to GND. (Use the 1A lamps).

Greetings, At.

 

[doc007]

Hello At
I have taked out U3 and U4 , and connected +10 via 100 ohm) to the gate of Q1 (lamp connected between B and +U : the upper lamp goes not off.same result with Q2
Lamp connected between A and +U same result when activation of gate Q3 and Q4.
Therefore if I have well understood the Fet Q1 Q2 Q3 and Q4 seems to be blown-up. They must be changed.
Pierre

 

[@t]

Hello At
I have taked out U3 and U4 , and connected +10 via 100 ohm) to the gate of Q1 (lamp connected between B and +U : the upper lamp goes not off.same result with Q2
Lamp connected between A and +U same result when activation of gate Q3 and Q4.
Therefore if I have well understood the Fet Q1 Q2 Q3 and Q4 seems to be blown-up. They must be changed.
Pierre

Yes, unfortunately they are gone, for some reason, probably while your first transformer was connected (very low impedance, high current).
I presume you have two lamps connected in this test (also the one to GND).
I suggest you first go on testing with the two lamps only. Let's hear how it goes with new Fet's, and then via the TLP250 (connected again).
Greetings, At.

 

[doc007]

Hello At
I am going to command new fets and test the fets with two lamps without TLP 250 ( taking out U1, U2, U3 and U4 )
Thanks for your good advice .
I contact you after the tests
Pierre

 

[@t]

Hello At
I am going to command new fets // I contact you after the tests
Pierre

Ok Pierre, but because you only draw 1A now, with the lamps, you can replace one of the lower Fet's (Q5 or 6) to the upper position (Q1 or 2), to see what happens.
At.
see 'edit' in post 17 april about supply voltage of TLP250 (10v.or more).

 

[farshadtitanic]

hi to all
is the final inv working fine?

 

[farshadtitanic]

mr taner plz helppppppp
i nedd detail for your inverter.i decide to make it but i cant undersatnd this.trans ,....
if its possible to send details to may mail:

titanicfarshad2@yahoo.com
tanx a lot

 

[doc007]

Hello At
I have received my new fets.
in order to limite wrong maneuvers here is how I will proceed.
I have taked out U1 U2 U3 and U4 ( R41 to R48 unsolderd on the side of TLP)

- Q1 and Q5 lamps between (B and +U ) and ( B and gnd )
Can I test with +10V on R41 and than on R45 to see they switch ?
If it is OK I will make the test with Q3 and Q7 lamps between (A and +U ) and ( A and gnd ) ...
It seems to you correct ?
Pierre