The principle is pretty straightforward: I haven't got the circuit in front of me but here goes...
take one self oscillating power stage that delivers a squarewave to the chopper transformer (we always called them chopper transformers as a generic name in any type of PSU) and regulate its secondary output indirectly by sensing the voltage on a secondary winding. This 'feedback' then controls the main supply voltage applied to said convertor through a series regulator as a means of regulating the PSU. The series regulator feeding the oscillating stage is also a bit unique in that it is a mix between a conventional series pass regulator and a switching stage.
The problem is that it is all far to dependent on the semiconductor device characteristics for reliable operation.
take one self oscillating power stage that delivers a squarewave to the chopper transformer (we always called them chopper transformers as a generic name in any type of PSU) and regulate its secondary output indirectly by sensing the voltage on a secondary winding. This 'feedback' then controls the main supply voltage applied to said convertor through a series regulator as a means of regulating the PSU. The series regulator feeding the oscillating stage is also a bit unique in that it is a mix between a conventional series pass regulator and a switching stage.
The problem is that it is all far to dependent on the semiconductor device characteristics for reliable operation.
There is so much HF energy flying around that it might not be easy for a scope to trigger and latch on to display clearly.
The four power transistors in the output stage get their drive directly from the chopper transformer via small auxiliary windings. That part is all self oscillating... it runs at full tilt all the time. The only way that any control is exercised is by sensing what is going on via another winding and using that to control the series regulator supplying it.
Modern SMPS regulate by varying the duty cycle of the switching transistor. This doesn't, it runs flat out all the time with only the supply voltage as a means of controlling its output.
The four power transistors in the output stage get their drive directly from the chopper transformer via small auxiliary windings. That part is all self oscillating... it runs at full tilt all the time. The only way that any control is exercised is by sensing what is going on via another winding and using that to control the series regulator supplying it.
Modern SMPS regulate by varying the duty cycle of the switching transistor. This doesn't, it runs flat out all the time with only the supply voltage as a means of controlling its output.
Yes, that was a confusion not seeing any pulse width modulation there, having the voltage regulated by sort of classic regulator.
As for the oscillation, my gut feeling is that Q607 shall be an oscillator, discharging to RED 20T winding and driving bases of invertor transistors...
Q607/608 have the look of an old fashioned unijuction type relaxation oscillator but I'm not sure that is how they behave tbh. You can make a UJT from discrete transistors.
Unijunction Transistor and UJT Relaxation Oscillator
Unijunction Transistor and UJT Relaxation Oscillator
Update:
1. OK, it works with MJE15032G (hFe>100 @100mA)
2. Output voltage (no load) is +/-80V
3. Safety bulb still in place (75W) giving 210V at PSU input with no load and just 140V with 50W load bulb at the output.
4. Will try tomorrow getting 100W bulb and try.
5. Possibly have to consider transformer secondary rewinding to get lower output.
1. OK, it works with MJE15032G (hFe>100 @100mA)
2. Output voltage (no load) is +/-80V
3. Safety bulb still in place (75W) giving 210V at PSU input with no load and just 140V with 50W load bulb at the output.
4. Will try tomorrow getting 100W bulb and try.
5. Possibly have to consider transformer secondary rewinding to get lower output.
Yes, on second thought I have denied this idea. Moreover the transformers in 110V and 220V are the same (given the same amount of winding turns - indicated right on the scheme). So basically for today I have two issues:
1. blowing parts if connected without "safety bulb"
2. output voltage is double to the normal.
1. blowing parts if connected without "safety bulb"
2. output voltage is double to the normal.
I would always try and have a small load present on the output, perhaps a 40 watt bulb. Strange things might happen with no load.
The output voltage could also be wrong because it might be running at an incorrect frequency or running at some harmonic of where of should be (fussy on parts again).
The output voltage could also be wrong because it might be running at an incorrect frequency or running at some harmonic of where of should be (fussy on parts again).
If someone still reading...
I did the scoping on secondary winding, got ideal square wave with 52us period (circa 19kHz) and about 170V amplitude. Seems it's pretty close to the numbers on the scheme = 180V amplitude and 42us period. But how they get +/-45 after rectification and caps from square wave 180V? Total misunderstanding... Attaching portion of scheme and the scope dump.
I did the scoping on secondary winding, got ideal square wave with 52us period (circa 19kHz) and about 170V amplitude. Seems it's pretty close to the numbers on the scheme = 180V amplitude and 42us period. But how they get +/-45 after rectification and caps from square wave 180V? Total misunderstanding... Attaching portion of scheme and the scope dump.
Attachments
No, no voltage falling with bulb load. Switching secondary is to feed Class A operation with very high quiescent current.
I have to grab operating 220v unit then, just to check all voltages myself... as well I will revert to other amp unmodified 110v psu just to check what's on secondary winding.
Thank you so much for your help.
I have to grab operating 220v unit then, just to check all voltages myself... as well I will revert to other amp unmodified 110v psu just to check what's on secondary winding.
Thank you so much for your help.
