In my continual learning curve mucking around with regulator circuits, I've now turned my attention to a Sziklai or Darlington BJT implementation of a circuit which originally used a MOSFET pass transistor to achieve greater current capability over high performance reg circuits such as the Jung/Didden 'super regulator'. The MOSFET circuit had a goal of being 5A capable. For this circuit, the goal is a more modest 1A thereby allowing implementation with higher gm BJTs.
The thread picks up from the conversation here following the revelation here that the MOSFET's gm differed materially from that implied by its Spice model. Yet again, I am extremely grateful for Mark Johnson's persistent assistance/forbearance.
I've attached my LTspice modelling thus far. Variants modelled:
1. Sziklai configuration
2. An attempt at a Sziklai configuration with push-pull drivers
3. Darlington
4. Darlington with push-pull drive having taken inspiration from the data sheet of Sparkos's discrete voltage regulators (link)
I'm not at all confident I have 2 configured correctly as the 'load off' transient response doesn't seem improved. I've struggled with this.
Modelled line rejection shows the Sziklai outperforming the Darlington (sub 1MHz); interestingly so as the On-Semi data sheet provides the same fT for both the D44VH10 and D45VH10 devices (?). The Darlington configurations seem to have better 'load on' transient response and the Sziklai better 'load off' performance. Perhaps it is poor modelling (likely) but the push-pull drive doesn't seem to do much for the Darlington 'load off' response (versus a simple resistor). The Sziklai also shows considerably lower modelled output impedance.
The thread picks up from the conversation here following the revelation here that the MOSFET's gm differed materially from that implied by its Spice model. Yet again, I am extremely grateful for Mark Johnson's persistent assistance/forbearance.
I've attached my LTspice modelling thus far. Variants modelled:
1. Sziklai configuration
2. An attempt at a Sziklai configuration with push-pull drivers
3. Darlington
4. Darlington with push-pull drive having taken inspiration from the data sheet of Sparkos's discrete voltage regulators (link)
I'm not at all confident I have 2 configured correctly as the 'load off' transient response doesn't seem improved. I've struggled with this.
Modelled line rejection shows the Sziklai outperforming the Darlington (sub 1MHz); interestingly so as the On-Semi data sheet provides the same fT for both the D44VH10 and D45VH10 devices (?). The Darlington configurations seem to have better 'load on' transient response and the Sziklai better 'load off' performance. Perhaps it is poor modelling (likely) but the push-pull drive doesn't seem to do much for the Darlington 'load off' response (versus a simple resistor). The Sziklai also shows considerably lower modelled output impedance.
Attachments
Last edited:
- Status
- Not open for further replies.