[ARC5] comments on a DC - DC B+ converter -- capacitor ripple currents

[J Mcvey]

 Les,

Not surprised at the input current measurements. 12v at 40 watts implies 3.33... amps and accounting for 90% efficiency of the supply, brings the draw to 3.7 amps.
The good news is that the radios don't draw a steady 85 ma, in fact maybe half of that or less . Only high volume spikes would kick it up a bit.

I ran a radio with one of those supplies @24V input for about an hour and didn't notice any untoward heating.



    On Thursday, April 19, 2018, 9:33:45 PM EDT, Leslie Smith <vk2bcu at operamail.com> wrote:  
 
 
Hello David & others,

I read your article describing how a DC-DC converter might be used as a small B+ supply.  Noise from the switching circuit was a problem, but you solved that problem.  It seems to me many readers of this list share common goals (putting a "command" unit on the air) and at the same time share common problems.  Getting a power supply  (for either a rx or tx) seems of some interest.  and so I posted a brief comment on that topic.

I got some very useful comments in response and that surprised me.  I expected the general reaction would be "Nothing new here" because that was my view about the meagre info I posted.  Beyond mere V(in) vs V(out) and power capacity it is clear that some measure of reliability would be useful.  A card has little value if the rated power (40W according to eBay) lasts only for one year.  So some readers made a comment about capacitor rating and ripple current.  Another (not mentioned) is the rating of the transformer (or choke) that "does the trick".  In my initial test I found the inductor (or transformer) appeared to be hotter than the heat-sink on the power transistor.  More than that, I wonder about the wire gauge used in the inductor.  40W at 120V (for example) suggests gauge #26 or #27 should be used.  

On the point of "something new" I have more info now - I measured the static performance when delivering power into various loads at voltages between 45 and 250V.  I connected the converter to a 2A supply and found it still tripped the over-current circuit under some conditions. 

I would be most interested to hear comment about capacitor rating and ripple.    I thank Bob & Scott for their observations.    Scott gave typical values (22uF/260mA) and I note that the output cap on this board is 10uF/400V.  0.5x capacitance suggests 130mA ripple rating.  I tested the current capacity at 250V/4k load or (about) 60mA.  Seemed OK for 1/2 hour or so.  Taking the load up to 85mA (about 20W) was OK, but the next step above that (30W) tripped the over-current protection circuit in my Marconi Instruments supply.  At any rate, observations of this type should be presented in graphical form; I haven't done that work yet.  I might say that the efficiency drops noticeably under load.

I noted the posting where ripple current was judged to be up to 3x the load current.  That surprised me.  I made some calculations and would have put that figure at 5x or 8x the load current - but this figure was for lower voltage higher current supplies with (say) capacitance values of 5 to 10 thousand "muffs".  Also my figure ignored the resistance and inductance of the transformer winding.  

I would be interested to read more comment on ripple current in the filter caps and how to test the power supply more thoroughly.  A simple static test might show the card working well for a month or a year (or longer).  Estimating the service life of the card is another matter, and testing the same at full load (or discovering the real capacity of the card) may take months, even if tested at +65C (or some elevated temperature).

I'm interested in any comment about how the life may be estimated and measured.

Les Smith.
VK2BCU "at" OperaMail "dot" com.
______________________________________________________________
ARC5 mailing list
Home: http://mailman.qth.net/mailman/listinfo/arc5
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:ARC5 at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20180420/916694de/attachment-0001.html>