Keantoken's CFP cap multiplier

Hey guys. There is a problem with the DirtyPCB boards. Due to problematic layout, they have a problem with oscillation.

attachment.php


I have wanted to make my own boards for the KM, but when these PCBs came out I figured that ship had sailed. But the layout is mostly done. I figure now would be a good time to print if you guys are interested.

attachment.php


The KM design for this board will be different. It has better filtering and doesn't have all those protection diodes - they never worked anyway. The PCB can be mounted flat to a heatsink with a standoff or if you want to use a floating heatsink, mount it under the board so it's away from the capacitors.

To fill the board you simply look at the circled pins. They must be the same polarity as your output voltage. So if your output voltage is positive, all the circled pins are positive. It fits terminal blocks and works up to 50V. It could also be used with D4xH11G for 2A output or so.
 

Attachments

  • DirtyKM-oscillation.png
    DirtyKM-oscillation.png
    290.6 KB · Views: 1,959
  • PKM-compact-layout.png
    PKM-compact-layout.png
    22.7 KB · Views: 1,937
Last edited:
Sorry I missed this. Can you tell me why you made those changes? If anything the 10nF cap should go on the B-E of the BD140. And I think I have seen more than one KM using the BD140 that oscillated.

The KM should have capacitors close to both the input and output, but 100nF film caps are unnecessary and sometimes will even cause oscillation.


100nF for decreasing the Esr of the Elkos at higher frequencies.
10nF for the BC337 is for prevent oscillation that maybe occours when the transistor gets older.

Maybe BD140 needs 10nF too, but my K-multi just went wrong today morning.
The output voltage is 6,5V with 14,2V input.
I'll change the BD140 to a TIP42, and later to the D45H11.
 
100nF for decreasing the Esr of the Elkos at higher frequencies.
10nF for the BC337 is for prevent oscillation that maybe occours when the transistor gets older.

Maybe BD140 needs 10nF too, but my K-multi just went wrong today morning.
The output voltage is 6,5V with 14,2V input.
I'll change the BD140 to a TIP42, and later to the D45H11.

Remove the 10nF caps, in some circuits they are a last resort but are rarely optimal for a fast or low noise circuit. The BC337 should not age, if so then they have been abused.

The 100nF can actually cause oscillating by making a peak in the source impedance. I suggest instead you use Wima 10uF or Vishay MKP416 1uF. This will go much farther towards the goal of low ESR film bypass.

wima 10uf WIMA 10 uF 5 mm Film Capacitors | Mouser
 
Remove the 10nF caps, in some circuits they are a last resort but are rarely optimal for a fast or low noise circuit. The BC337 should not age, if so then they have been abused.

The 100nF can actually cause oscillating by making a peak in the source impedance. I suggest instead you use Wima 10uF or Vishay MKP416 1uF. This will go much farther towards the goal of low ESR film bypass.

wima 10uf WIMA 10 uF 5 mm Film Capacitors | Mouser


10uF Wima parallel to each elko (100uF & 820uF)?
 
OK.

And what could cause that the drop of the K-multi increased.
With 14,2V with BD140 the output was 6,5V.
I've changed the BD to TIP42C, and the drop decreased, the output goes up to 10,7 V.
I've measured the BD140 after solder out, and it was good.

Now I've increased the input voltage, but don't knwo if it cause any problem.
 
OK.

And what could cause that the drop of the K-multi increased.
With 14,2V with BD140 the output was 6,5V.
I've changed the BD to TIP42C, and the drop decreased, the output goes up to 10,7 V.
I've measured the BD140 after solder out, and it was good.

Now I've increased the input voltage, but don't knwo if it cause any problem.

Sounds like your circuit is overloaded, or maybe Q1 has C and E swapped. Also check the 47R resistor with a DMM to make sure it's the right value.
 
I would be interested on several boards too, for 45V and 100V versions.
Any ETA on these?

I sent some of the 50V PCBs to print and they are at the fab right now. If they turn out well and test fine I can sell some as soon as I get them (could be 2 week or longer).

I will have to make a different PCB for the 100V version, but you could make one by soldering on a bunch of diodes on to the 50V PCB. How much input ripple would you expect?