Homebrew PSU question - UK Vintage Radio Repair and Restoration Discussion Forum

julie_m · 28th Jul 2009, 9:49 am

Am I right in thinking that, if I was building an inverter with a centre-tapped primary driven by open-collector transistors (or open-drain MOSFETs), I shouldn't fit flyback diodes from each collector/drain to CT?

jimmc101 · 28th Jul 2009, 10:25 am

Quote:

Originally Posted by ajs_derby

Am I right in thinking that, if I was building an inverter with a centre-tapped primary driven by open-collector transistors (or open-drain MOSFETs), I shouldn't fit flyback diodes from each collector/drain to CT?

Absolutely correct.
Consider what voltages are on the transformer, the end connected to the conducting device will be at approx. 0v, the centre tap is connected to the supply so the other end of the winding must be at twice the supply.
Not a good idea to try and clamp this to the supply.

Jim

julie_m · 28th Jul 2009, 10:41 am

So, as long as I choose transistors that will withstand at least twice the supply voltage across C-E (or D-S, but I'm thinking bi-polar for the moment), then the autotransformer action of the primary will limit the voltage rise on disconnection?

kalee20 · 28th Jul 2009, 10:46 am

Definitely not across the transformer.

However, it is good practice to connect a (fast recovery) diode across the switching transistors (assuming NPN, connect anode to collector, cathode to emitter). These will stop the collector trying to swing negative, if there is any resonance effects with stray capacitance and transformer leakage inductance.

MOSFETs effectively have an diode structure which clamps this anyway; bipolars (NPN's and PNP's) don't. And after the collector has gone negative, the base can be filled with charge carriers which hold the transistor 'on' for a few microseconds when the collector tries to swing positive again.

Herald1360 · 30th Jul 2009, 9:40 pm

Quote:

Originally Posted by kalee20

However, it is good practice to connect a (fast recovery) diode across the switching transistors (assuming NPN, connect anode to collector, cathode to emitter). These will stop the collector trying to swing negative, if there is any resonance effects with stray capacitance and transformer leakage inductance.

Shouldn't that be the other way round- I think what you said = dead short across the transistor.....

jimmc101 · 30th Jul 2009, 10:43 pm

Well spotted!

Jim

kalee20 · 31st Jul 2009, 2:14 pm

Oops!

28th Jul 2009, 9:49 am	#1
julie_m Dekatron Join Date: May 2008 Location: Derby, UK. Posts: 7,735	Homebrew PSU question Am I right in thinking that, if I was building an inverter with a centre-tapped primary driven by open-collector transistors (or open-drain MOSFETs), I shouldn't fit flyback diodes from each collector/drain to CT? __________________ If I have seen further than others, it is because I was standing on a pile of failed experiments.

28th Jul 2009, 10:41 am	#3
julie_m Dekatron Join Date: May 2008 Location: Derby, UK. Posts: 7,735	Re: Homebrew PSU question So, as long as I choose transistors that will withstand at least twice the supply voltage across C-E (or D-S, but I'm thinking bi-polar for the moment), then the autotransformer action of the primary will limit the voltage rise on disconnection? __________________ If I have seen further than others, it is because I was standing on a pile of failed experiments.

28th Jul 2009, 10:46 am	#4
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,087	Re: Homebrew PSU question Definitely not across the transformer. However, it is good practice to connect a (fast recovery) diode across the switching transistors (assuming NPN, connect anode to collector, cathode to emitter). These will stop the collector trying to swing negative, if there is any resonance effects with stray capacitance and transformer leakage inductance. MOSFETs effectively have an diode structure which clamps this anyway; bipolars (NPN's and PNP's) don't. And after the collector has gone negative, the base can be filled with charge carriers which hold the transistor 'on' for a few microseconds when the collector tries to swing positive again.

30th Jul 2009, 10:43 pm	#6
jimmc101 Heptode Join Date: Mar 2009 Location: Portsmouth, Hampshire, UK. Posts: 674	Re: Homebrew PSU question Well spotted! Jim

31st Jul 2009, 2:14 pm	#7
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,087	Re: Homebrew PSU question Oops!