[Top] [All Lists]

[AMPS] Additional Info Voltage Doubler/Filter Capacitance

To: <>
Subject: [AMPS] Additional Info Voltage Doubler/Filter Capacitance
From: Ian White, G3SEK" < (Ian White, G3SEK)
Date: Fri, 21 Sep 2001 17:38:48 +0100
Tom Rauch wrote:
>> Kurt,
>> There is full info on modelling voltage doublers on the Duncan amps =
>> site. You do not have to download and store, you can simply open the
>> site and = do the exercise
>I didn't seem to be able to do that Barry. I had to download the file.
>I could only run the bridge rectifier program

Double-click at the *top left* or *bottom left* corner of the schematic
to change the rectifier configuration. Double-click *on* the rectifier
to select the diodes from the drop-down list.

Not exactly obvious, I agree...

Same applies to the filter: double-clicking *above* it lets you select
capacitor input, choke input etc. Double-click *on* the individual
components to change their values.

Double-click above the load to see the very useful option of defining a
constant load current. It can also be stepped up or down after a
selected time, so you can simulation at two current levels on the same

>, and it very clearly gave 
>me incorrect answers.

>In one example I modelled a full wave rectifier with 20 ohms ESR 
>(the maximum it allows) 

For HV power supplies, it's useful to define a new "diode" called
"1N5408*6" by adding the following line to RECTIFIERS.TXT:

1N5408*6, SS,  0.0438,  20, 9.353, 6000, 200,  99

You can then select this by double-clicking on the rectifier symbol
(save the new .TXT file first, and restart the program to load it).

It seems that solid-state rectifiers are best modeled by assuming a low
value for Rs and letting the program compute the voltage drop from the
non-linear V/I law. Try modeling a low-voltage PSU where these drops
really matter, and you'll see what I mean.

>with 2600 volts at the transformer, and it 
>said 3600 volts dc output. That is clearly incorrect.  

At low current, what's wrong with that? It's very close to 1.41*2600. 
At higher currents the predicted voltage falls and shows more ripple -
by how much will depend mainly on the series resistance of the
transformer (secondary + primary*n^2) and the capacitor value.

>Anyone else have bad results?

Quite the contrary. It seems right on the money, within the limitations
of the transformer model (no leakage reactance) and the simple  diode
model (which doesn't matter for HV supplies).

If you have a file that's causing trouble, Tom, please can you save it
(when you quite the program), put it on your website and post a URL

73 from Ian G3SEK          Editor, 'The VHF/UHF DX Book'
                          'In Practice' columnist for RadCom (RSGB)

FAQ on WWW:     
Administrative requests:

<Prev in Thread] Current Thread [Next in Thread>