To deliver 5.5V at 200mA from 0.7V, the battery would have to put out over 1.6A. More likely 3A at the efficiency you are likely to get. Can your dead battery do that? You are dreaming if you think you can extract that kind of power from a 9V battery that has run down below about 7V.

I didnt state exactly 0.7v i amended that and said at least some point below 5.5V of the battery it stops runing not sure what voltage battery fail from original circuit but it would be lower than circuit posted by sghioto

 

Well OK, good luck with that.

Thanks for showing me your circuit I might use it after all its really useful not lot of components and does the job

 

It was just for illustration so you see how you can vary the duty.

For automatic feedback control you change this trimpot with transistor amplifier. As the Vout rises too much this transistor pull the top of R2 lower and increases the Toff (lower the duty) so the Vout falls back to normal.View attachment 354330

I am also looking into your circuit. However, I have the above attached image with questions in red. Can you please help me resolve them?

 

Regarding diodes, normaly the EB junction (not a BE) of bjt has very high impedance (>10Mohm), but all changes once this EB junction is stressed with voltage above 7V. In that case the EB junction is almost shorted.
These diodes removes this stress because they will block this high voltage. I must add it’s a bit dirty solution to add this diodes but it works.
Without diodes the timing of multivibrator is corrupted because EB junction will quickly discharge the caps in every cycle to about 7V immediately.
When I use multivibrator (i really like it) I supply it in most cases with low voltage rather (like 6V), so I don’t have to use this diodes. In that case the EB breakdown cannot occure.

 

Regarding diodes, normaly the EB junction (not a BE) of bjt has very high impedance (>10Mohm), but all changes once this EB junction is stressed with voltage above 7V. In that case the EB junction is almost shorted.
These diodes removes this stress because they will block this high voltage. I must add it’s a bit dirty solution to add this diodes but it works.
Without diodes the timing of multivibrator is corrupted because EB junction will quickly discharge the caps in every cycle to about 7V immediately.
When I use multivibrator (i really like it) I supply it in most cases with low voltage rather (like 6V), so I don’t have to use this diodes. In that case the EB breakdown cannot occure.

What about rest of the red questions? Aside from diodes

 

https://forum.allaboutcircuits.com/threads/2-transistor-buck-circuit.180784/
https://www.romanblack.com/smps/smps.htm

In those links is their any square wave generator?

 

What about rest of the red questions? Aside from diodes

Regarding diodes, normaly the EB junction (not a BE) of bjt has very high impedance (>10Mohm), but all changes once this EB junction is stressed with voltage above 7V. In that case the EB junction is almost shorted.
These diodes removes this stress because they will block this high voltage. I must add it’s a bit dirty solution to add this diodes but it works.
Without diodes the timing of multivibrator is corrupted because EB junction will quickly discharge the caps in every cycle to about 7V immediately.
When I use multivibrator (i really like it) I supply it in most cases with low voltage rather (like 6V), so I don’t have to use this diodes. In that case the EB breakdown cannot occure.

I didn't get an response from you to post #65?