Hello everyone, quick question. A DCC signal runs in my rail tracks that is close to an AC with 7-10 Hz in average and 20-24V. I want to transform this AC in DC. To reach this goal I put a bridge rectifier with 4 1n4007 and 2 capacitors in parallel of 35V 2200uf each + a resistor of 1kOhm. The calibration of the capacitors is takiibg into account the voltage, the intensity (0.5A - 1A) and the frequency. My calcutions are right? Thanks.
Your DCC signal is 24V (presumably rms) so
the peak dc voltage from the bridge will be (24 * 1.42 ) - 1.4V = 35V
that peak happens every (7Hz = 143msec) 70 msec.
the "formula" for the caps is q = Cv = it
so v (your ripple voltage) will be
v = it/C = 1A * 70 msec / 4.4mF = 16V
does that match your calculations?
Hi John
Hello John more or less! Thanks. I want to operate with a fully secured environment. May I contemplate a 35V 4700 uf?
Can you post a schematic? Whete is the 1kohm resistance in your scheme?
@johnerrington calculated a ripple of 16V. That is a bit more than a ripple in my opinion...
At that freq you will need giant capacitors to smooth out ripple.
At 50 Hz 4000 uF is not uncommon. So you would get to 40000 uF. And that may kill your diodes during startup...
They are rated for 1A...
You need some safety margin here...
35V peak in a 35V capacitor is asking for problems on the longer run...
The schematic is in german : https://digitalbahn.de/shopping/mediafiles/pdf/zweipol_weiche_1v42_schaltplan.pdf
Not a DCC expert, but isn't that a ~7kHz 12-15volt differental digital (square wave) signal.
Then all the answers given are incorrect.
Leo..
GIGO Leo!
I guess the question is incorrect then....
Wawa yes but I run LGB trains 20-24v roughly 7Hz indeed. So what is your view?
Yes my question was incorrect. Apologise. What would be the calibration of the capacitors in that case?
Basically you can divide the numbers by 1000...
...as you have increased the frequency by a factor 1000.
Stuck in a loop?
7kHz might require (1N5819) Schottky diodes instead of the common 1N400x.
And low ESR capacitors instead of common ones.
A 100uF should be enough at that frequency, but small caps can't handle a 1Amp ripple current.
Several 470uF/63volt lowESR caps in parallel would be my choice.
Leo..
Unlike AC power which is a sine wave that periodically reaches zero volts, a DCC signal is roughly a bipolar square wave and only crosses zero very briefly. The frequency could be 7000 Hz, 7Hz, or even 0.007Hz and you would only need a relatively small capacitor to achieve rectification to DC.
The fact that it is G scale has absolutely no bearing on this, nothing is different with that scale except for a higher voltage.
So adding increasingly large filter capacitors is a complete waste.
thanks a lot. When you metion several capacitors what does it mean:2,3 or more and how may I calibrate the right number? I have 5 x 50V 100-470uf electrolytic capacaitors how can I calculate easily ESR ad what is the right level (<1,5 Ohm may be ok?) or do you have specific references? I will defiitely put Schottky diodes forthe bridge. Thanks
Just use the 50volt caps you have. 1000uF total would be ok for that current.
If they get hot, or start to bulge their tops after a few years,
then you know you need to replace them with low ESR caps.
Leo..
Thanks a lot. have a nice day
Are you trying to use the pulse stretching feature of DCC to operate a non-DCC equipped locomotive on a DCC layout? If so you can't use a bridge rectifier. If you do you will get an almost constant voltage. If you are using pulse stretching this way, you have to charge the capacitor with one polarity from the source. So, one diode.
Hello the goal is to power an attiny85 and a track DC turnout motor.