[time-nuts] Using a common power supply among few time standards

Taka Kamiya tkamiya9 at yahoo.com
Mon Feb 10 05:11:23 UTC 2020


I spent last two days boxing  up LPRO-101 into a 2U rack case.  It has Rb, dist amp, interface for LED, and EFC knob on front.  While doing so, I gave power supply concern I have been discussing careful thought.  I believe I came to something that is workable and fits my needs.  

I started with 24V linear power supply (open frame type).  This is the only power supply in the box.  From there, it branches to Rb, a small dist amp, and a simple interface circuit.  Dist amp is 12V rated and interface is 5V.  I modified the dist amp by including a 3 terminal regulator.  Now, it's a 24V device.  I crated a 5V source right at Rb's interface board to be used as EFC power.  This is the only purpose of this branch.  I also created a separate 5V for the simple interface circuit. The reason for two 5V supply is that I didn't want to mix the source for EFC and digital circuit together.

Pretty quickly, I found out going from 24V to 5V incurs huge loss resulting in heat and death of 3 terminal regulator for that branch.  EFC circuit was fine but digital interface board killed the regulator. Put a heat sink on it and issue is resolved.  A bit surprising but it shouldn't be....  The circuit only uses 25mA or so but voltage drop is huge.
Basically, I'm doing the same thing S100 bus computers did.  I will have to consider use of small smp board for large voltage drops.  I didn't this time, as its noise profile is not known to me.  Can't really go wrong with linear regulators.  My only remaining concern is if regulator fails short, then what happens??  It will kill the particular device, which I'm fine with, but what else?  Perhaps simple fuse is in order for every branch.  

This method is a lot simpler than hap-haphazardly creating multiple voltage at power supply and routing all over the place.  It became impossible to trace ground loops.  Above method makes it really easy as, as far as power supply is concerned, everything is a 24V device.
Thanks everyone for discussions.  I learned a lot in this process.
--------------------------------------- 
(Mr.) Taka Kamiya
KB4EMF / ex JF2DKG
 

    On Saturday, February 8, 2020, 9:52:02 PM EST, Charles Steinmetz <csteinmetz at yandex.com> wrote:  
 
 Chris wrote:

> How about using the open frame linear supplies like the ones from Power One or Lambda?

I have had very good results with Deltron open-frame linear supplies. 
They have selectable mains voltage and low output noise (1mVrms/5mVp-p). 
  I have used many (i.e., hundreds) of the Model W303E triple-output 
supply.  These have one, 5v/8A section and two sections, each selectable 
between 5v/12v/15v at 0.7A/1.7A/1.5A.  [*NOTE* there is an overall 55W 
output limit, so they will not do all of that at once.]  Each section 
can be adjusted +/- 5%.  These show up regularly on ebay.

As for switchers, I have been satisfied with the performance of 
Integrated Power Designs (IPD) products.  [IPD is a different company 
than Power Designs, who made some of the best lab and bench power 
supplies of the 1960s through 1990s.]  One in particular may be of 
interest to time nuts: the Model SRW-65-4006.  This is a four-output 
supply (5v/5A, 24v/1A, and +/-15v/2A) [*NOTE* again, there is a 65W 
overall limit, so it won't do all of that at once].  It is available 
open-frame or with a chassis and cover.  For installation close to other 
circuitry, and particularly inside a piece of equipment, I strongly 
recommend the model with both chassis and cover (options CH and CO). 
Check ebay and look at distributors.

Assume you will need to post-filter the switcher output, either with L/C 
filters or low noise linear regulators.  Note that I did not say "LDOs". 
  I don't know why, but everyone seems to be stuck on LDOs these days. 
Active regulators (including LDOs) *always* perform better with more 
headroom (up to 4v or so).  The only excuse for designing LDO is if you 
have a gun to your head from a power budget you can't otherwise control. 
  [Here, "LDO" is mostly a matter of how you use the part, not of its 
claimed capabilities -- "LDO" regulators work much better with at least 
4v of headroom.]

Best regards,

Charles



_______________________________________________
time-nuts mailing list -- time-nuts at lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.
  


More information about the Time-nuts_lists.febo.com mailing list