diyAudio

diyAudio (https://www.diyaudio.com/forums/index.php)
-   Power Supplies (https://www.diyaudio.com/forums/power-supplies/)
-   -   Juma's Easy-Peasy Capacitance Multiplier (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier.html)

prasi 16th October 2017 03:53 PM

1 Attachment(s)
Quote:

Originally Posted by jameshillj (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5214091.html#post5214091)
Very neatly done, Prasi - and all inside 16 sq inches!

Maybe next time, move the central ground point to the output away from the charging caps and their 'noise', particularly for the filter caps - this then becomes the new 'central ground point', not near to the diode/ripple cap point - a bit of a contentious point in some circles, I think

Maybe also, could shift a couple components to mount underneath (or 'piggyback' a couple [ie, R1 and C1], or SMD some larger size resistors) to free up enough room for the return of the AC snubbers, and maybe leds for 'working' indicators ...

Just saying, no criticism

HI James,
Thank you for your suggestions. Reworked the layout, see attached. Is this what you are suggesting when you say 'move central gnd to output'. By moving to o/p side, It now almost looks like a proper layout for LM317/337 type ckt. Should cap multipliers too should get their ground reference from o/p?
Added 3mm LED indicators, with smd resistor on bottom. I am trying to avoid SMD especially high power components (R1/R8) as they are not easily available locally.

Quote:

Originally Posted by AKSA (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5214115.html#post5214115)
Lovely work, Prasi, you are getting better and better.......
Congratulations!!

HD

Thank you Hugh, Cheers! I am hoping its useful to many people.

regards
Prasi

SemperFi 16th October 2017 06:37 PM

1 Attachment(s)
Cool circuit. I've always used a very similar ckt the only difference being a resistor to GND from the 10kohm to the gate since I always thought I needed to drop the gate slightly from Vrail to ensure compliance. Didn't know this is what is called a cap-multiplier.

This site is simply awesome. Always something to learn.

MOSFETs with low RdsON always have higher parasitic capacitances than FETs with higher RdsON. Since you are not operating with the MOSFET in the 'ON' state, but rather linearly, the RdsON is pointless to use when selecting FETs.
Tho this is a follower, the high capacitance of a powerful FET such as IXTK200N10P will give poorer attenuation with frequency than a light-weight FET such as IRF510.
Todays FETs are unfortunately made to switch fast in SMPS service and are not as good in linear circuits (such as for pass duty in regulators). Even todays IRF610s or IRF240s are probably not as robust as same parts 10years ago. But they switch better.

I'd look for low capacitances and large package when selecting FETs for this duty.

xrk971 16th October 2017 06:48 PM

The important figure is how much ripple attenuation at line AC frequency of 50Hz/60Hz. I seem to recall about -50dB when I ran the LTSpice sim for this with IRFP240 and also added an LC filter at the output. -50dB makes a huge difference with hum suppression though.

SemperFi 16th October 2017 08:15 PM

Quote:

Originally Posted by xrk971 (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5214943.html#post5214943)
The important figure is how much ripple attenuation at line AC frequency of 50Hz/60Hz. I seem to recall about -50dB when I ran the LTSpice sim for this with IRFP240 and also added an LC filter at the output. -50dB makes a huge difference with hum suppression though.

True that. But isn't it 100/120Hz? or are we using half wave rectification?

Mark Johnson 16th October 2017 08:33 PM

When you analyze the circuit to discover the reason why some of the 60Hz / 120Hz input appears on the output,

you may agree with me that the easiest way to improve this is to cascode the output stage.

And if you are determined to use MOSFETs, you may agree with me that the headroom considerations disappear when you give yourself permission to build a dickson charge pump.

SemperFi 16th October 2017 09:03 PM

Quote:

Originally Posted by Mark Johnson (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5215028.html#post5215028)
When you analyze the circuit to discover the reason why some of the 60Hz / 120Hz input appears on the output,

you may agree with me that the easiest way to improve this is to cascode the output stage.

And if you are determined to use MOSFETs, you may agree with me that the headroom considerations disappear when you give yourself permission to build a dickson charge pump.

Sorry man, I havent read the whole thread. The interesting part is the simple cap-multiplier. The more intricate ckts look more like voltage regulators to me.

My intention was to give my dime (inflation so no longer 2cents) ragarding MOSFET selection.

Sorry if I jumped in and interupted, the thread has probably evolved way past the simplest C-x.

xrk971 16th October 2017 09:45 PM

Quote:

Originally Posted by Mark Johnson (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5215028.html#post5215028)
When you analyze the circuit to discover the reason why some of the 60Hz / 120Hz input appears on the output,

you may agree with me that the easiest way to improve this is to cascode the output stage.

And if you are determined to use MOSFETs, you may agree with me that the headroom considerations disappear when you give yourself permission to build a dickson charge pump.

Please show us how to make such a device with charge pump, cascode, and low dropout. I'm all ears. I hope it works easily though as the Juma circuit is one of the few that never fails to work, albeit, with large dropout voltage. I have tried building more complex 2 BJT based cap Mx before. They just failed to even work. Maybe just me not properly making circuit?

Mark Johnson 16th October 2017 10:18 PM

1 Attachment(s)
First do an analysis and discover why some of the sinusoidal input, finds its way onto the output; why isn't there infinite attenuation?

You might determine that infinite amounts of bypass capacitance on the MOSFET gate, does not translate into infinite amounts of attenuation (figure 1). Or, you might discover that I am completely misguided and totally wrong. If so then my claim "cascodes make it better" becomes highly dubious.

_

xrk971 17th October 2017 01:37 AM

That's a useful plot because it says if we just want to attenuate 60Hz hum, anything larger than 30uF doesn't but you anything.

Very few things are infinite. Might even be something like a limit of an infintite wall to quantum tunneling. In this case the intrinsic blocking of a hexFET mosfet is -45dB.

prasi 17th October 2017 06:39 AM

3 Attachment(s)
Quote:

Originally Posted by xrk971 (https://www.diyaudio.com/forums/power-supplies/297921-jumas-easy-peasy-capacitance-multiplier-post5214663.html#post5214663)
+1, Prasi is a master of layout, and he gives us all of this for FREE! Bravo.

Thank you, Prasi!

https://www.diyaudio.com/forums/atta...ode-bridge-png



It should be pointed out that this slick 100mm square circuit, connected to a trafo with two secondary windings (a standard config of trafos), now gives essentially qnty 4x isolated power supplies because the power fluctuations on one is independent of another now, so the two amp boards will see essentially a dual mono-block PSU for super channel separation. All this without expense of second trafo, second CRC PSU, etc, and better overall ripple suppression as if the caps were huge 200mF deals.


Thank you x,
here are the schematic, stuffing guide and gerbers. LED resistors mount from bottom.

regards
Prasi


All times are GMT. The time now is 09:46 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio

Wiki