diyAudio (
-   Group Buys (
-   -   The SLB (Smooth Like Butter) Active Rect/CRC/Cap Mx Class A Power Supply GB (

jameshillj 14th April 2019 04:06 AM

... and donuts/space for Caddocks, etc?

xrk971 14th April 2019 11:05 AM

Sorry, don’t understand what you are asking about donuts and caddocks?

jameshillj 14th April 2019 11:33 AM

Not to worry, just idle speculation - please ignore.

Mark Johnson 14th April 2019 12:09 PM

1 Attachment(s)

Originally Posted by jhofland (
... [in a power supply CRC filter] the resistors themselves dissipate enough power to get pretty toasty.

Although resistor manufacturers publish datasheets, few people read them. Which is too bad, because then there would be less angst and surprise when dissipating one watt in a 3 watt rated resistor produces a very hot booboo device. If the datasheet curves are accurate, thermal resistance theta (degrees C per watt) is constant; the curves are straight lines of constant slope, passing through the origin.

Oh and if you dare to dissipate 3 watts in a "3W" resistor, the resistor body will be 175 degrees C above ambient according to those curves. Hot enough to make the PCB turn brown. Hot enough to melt 63/37 solder!


xrk971 14th April 2019 12:24 PM

Which “people” are you referring to? We are using KOA Speer non inductive cement filled rectangular metal plate 5W resistors. A pair of 0.1ohm in series passing 5A is dissipating 2.5W ea - well under rating. This has all been sorted out with the verification build on the single rail.

Data sheet is page 211:

Mark Johnson 14th April 2019 01:04 PM

1 Attachment(s)
Just referring to resistor buyers / resistor users who don't bother to read resistor datasheets.

I couldn't find a temperature-vs-dissipated_power graph, or a theta spec, for the BPR resistors on page 211 of that KOA Speer catalog. But the previous resistor in the catalog, type "BWR" on page 205, looks physically very similar and I was able to find its graph, copied below.

If using a BWR resistor rated 5W, dissipating 2.5W in it will result in a temperature rise of 65 to 75 degrees C above ambient, depending on whether you measure at the bottom {which cooks the PCB} or the top {which burns people's fingertips.} Add the room ambient air temperature (~ 24C) and you're close to the boiling point of water.

It sounds like you have already investigated this, using either fingertips or thermometers, and deemed the resistor temperature acceptable. If so, congratulations. Too many people fail to do this, and receive a high temperature surprise early in the production of assembled units within chassis. The unhappiest surprise is when these folks operate a resistor "within rating" and yet it melts its own solder.


xrk971 14th April 2019 02:35 PM

It's definitely not something you want to grab with your fingers when it is running, but doesn't melt its own solder at least and we are giving it room for air flow. If folks are concerned about keeping temps low, use four 0.22R in series parallel and dissipation will be about 1.25W each for maybe 65C temps.

Mark Johnson 14th April 2019 03:38 PM

1 Attachment(s)
Here is a screen grab made by diyAudio member "ZenMod" sixteen years ago (!)

Somebody asked Nelson Pass, "When you touch various places on a Class A amplifier (heatsink etc), what physical sensations correspond to different temperatures?"

or some question like that.

NP replied, "Count seconds and see how long you can maintain contact before the pain becomes unbearable", and then he gave this handy table, attached below.


jameshillj 18th April 2019 06:14 AM

A suggestion, as per the layout on post #12,

If you moved the LT4320 between the 2nd and 3rd fet (ie. move the first fet into old position of the LT 'around the corner' and so on...) the control tracks would be shorter and more uniform - same with the -ve rail - not sure where you'd fit the 1uf C5 & C6.

I see the board layout is 142 x 100mm, yes? I struggle a bit with the black background (not as young as I used to be!)

xrk971 18th April 2019 07:12 AM

I think in the current layout, Jhofland is trying to keep the loop area for the high current flow as small as possible to reduce emitted EMI. Putting the LT4320 in bewteen the FETs would improve symmetry of the gate control traces, but that is secondary to minimizing generation of noise from large ripple currents around the MOSFETs.

For the same reason, the SMT implementation also has the LT4320 mounted off to the side:

The SMT version is so much more compact.

All times are GMT. The time now is 04:36 PM.

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