[Amps] Source wanted for HV switch-mode power supply

[David Cutter]

If you haven't done a switcher with feedback before, I would stay with the 
design as it is for the first run.  When you have the bugs out, then think 
about feedback.  You will need more turns on the secondary because 
inevitably you will run at less than 100%.  The problem then becomes one of 
getting the inductor/capacitor combination in the output and the time 
constants in the feedback loop to work in your favour.  The biggest problem 
is not to stabilise at full load with lowest line input (if you are looking 
for stability), but what happens when your load switches from max to min in 
fast sequences, ie keying at various speeds and of course ssb.  What you are 
then trying to avoid is some odd combination which sends the feedback loop 
into instability and the output voltage swinging wildly out of control.

You may not need much in the way of stability anyway, after all, most high 
voltage supplies for linear amplifiers are not stabilised and also have 
considerable ripple.  The advantage of feedback control is that you can 
reduce the large storage caps on the supply input by using feedback control. 
That saves weight, space, cost.

If Manfred reads this, I hope he jumps in and gives you further advice.

73

David
G3UNA


----- Original Message ----- 
From: "Paul Decker" <kg7hf at comcast.net>
To: "KD7QAE" <KD7QAE at ARRL.NET>
Cc: <amps at contesting.com>
Sent: Wednesday, May 20, 2009 9:27 PM
Subject: Re: [Amps] Source wanted for HV switch-mode power supply


>
>
> Hi Tomm ,
>
>
>
> Actually, it is funny you say that. I wanted to redraw the schematic into 
> my schematic program so it could do autoroute and pcb layout. When I did, 
> I found that my pcb layout was limited to 100 pads, which forced me to 
> break the schematic up into individual boards, so that work is already 
> done ;) The original pwm module is about 1"x1", it goes to a signal 
> amplifier board, which goes to the h-bridge. My H-Bridge board is a bit 
> weird on the layout because I used some leftover heat sinks and kept the 
> leftover parts placement. After the HV transformer there is the HV 
> rectifier board. Tim's original design doesn't provide any feedback but 
> that could easily be added and the little pwm module be replaced to allow 
> for a feedback control of some type. For me, the HV transformer was the 
> hardest part and I'm still not convinced I like what I have implemented, 
> it may get some changes.
>
>
>
> As far as a kit, who knows, but I'll try to put design files up on my web 
> page this weekend and if people want them, they can pull them from there.
>
>
>
> 73 for now,
>
> Paul, kg7hf
>