Keantoken's CFP cap multiplier

[keantoken]

D11 on my webpage for the HV Kmultiplier.

 

[RCruz]

I am using the high-power amps with 100V rails version.

The corresponding diodes might be D5 and D7..... I did not use those in my breadboard build

 

[kinku]

what you all think about this design?

 

[maouna]

what you all think about this design?

Conserning cap multiplier section,use 3 x 1N4148 in series for your D8 as shown on web page The K Multiplier.

 

[kinku]

Thanks Mauona.
What will be a good replacement 2SA1903? Can I use FZT751?

 

[RCruz]

Hi Kean

Reworked the layout now using your simplified schemetic. Trying to follow your board layout I included a copper pour connected to GND.... your comments are most welcome here as I do not have enough experience on how to use gnd planes.

Best Ricardo

 

[kinku]

LT3042 output cap as KeantokenCFP

hi guys.Check this out
I might be doing something wrong but the results are very good.
Can someone please verify this.

 

[keantoken]

Looks reasonable to me. PSRRs add when you put regulators/KMs in series. Of course at some point EMI will cause more noise than is getting through your regulators.

 

[keantoken]

Reworked the layout now using your simplified schemetic. Trying to follow your board layout I included a copper pour connected to GND.... your comments are most welcome here as I do not have enough experience on how to use gnd planes.

Few here use ground planes, the important thing is to just not create large current loops when small current loops will do. If you can do a good one-sided layout, then using ground planes starts being a good idea. I don't think they are much of an improvement the way you are using them.

It's hard to figure out your layout without seeing the whole board. And I'm not really in a good position to teach this stuff. If others here can't find flaws with the board, then you will probably be satisfied with it. There is no guarantee that my advice won't just confuse you and cause you to make a worse layout.

 

[av-trouvaille]

Earlier you mentioned for the standard KM the safety diodes can be omitted. Would this also apply for the HV-version?

 

[keantoken]

On the HV version those aren't safety diodes, they are startup diodes.

 

[keantoken]

For the next week I will be gone to a funeral gathering, so I won't be able to respond.

 

[kinku]

Looks reasonable to me. PSRRs add when you put regulators/KMs in series. Of course at some point EMI will cause more noise than is getting through your regulators.

Thanks Kean. I know it is still simulation but if I don't add a Cap after the CFP the output impedence is significantly high. Why is that?

 

[keantoken]

I put 820u for C1 in the original schematic, that would help. Like any C multiplier, there is some rise in output impedance at LF. It sounds fine though apparently.

Using BC550C/560C with higher Hfe and loading with 100mA or so would reduce it further.

 

[kinku]

Kean, Thank you! What are the qualities of Transistor that works best in your CFP? Other than HFE?And which Transistor Q3 or Q4 in my schematic?
I may need about 110 mA for my project.

 

[kinku]

Also is it possible to use Keantoken CFP to make a virtual dual supply,using positive regulators?

 

[kinku]

zener follower with CFP

while waiting for a reply I experimented with another design and got this. But for some reason I have to use 10x capacitance to get the same PSRR.
Please get the models from here
http://www.cordellaudio.com/book/Cordell-Models.txt

 

[keantoken]

C7 has 820F, not 820uF.

 

[kinku]

oops.That fixed it. Kean what do you think about making virtual ground with your CFP as in post #1056

 

[AndrewT]

I am using separated paths for the zobel gnd and other grounds... they will all join star ground on the psu board. Somehow I believe zobel return should not be connected directly to other gnd. Am I wrong ?

In fact I see the zobel as another load to the power amp and as we normally route speaker return to the psu star, I decided to do the same for the zobel.

The amplifier's output Zobel is predominantly an HF load.
10ohms+100nF has an F-3dB frequency of 159kHz.
The load presented to the amplifier at this frequency is ~14ohms.
Significant VHF current can only flow through the output Zobel if the impedance at these high frequencies is maintained to a very low level.
This cannot happen if you take the output Zobel route on a tour around the amplifier to tie into some remote star.

The impedance can only be kept low by placing the output Zobel in a short compact LOW LOOP AREA route around the output devices and HF decoupling.