I need help in designing a capacitor-charging circuit that will accomplish the following: charge a 3,000V/200uF capacitor from a 37V battery. The desired charge time is 5 seconds. The battery is a 329Wh size with >80C discharge capability. The capacitor will not be connected to a load on charging. I would also like to know what the approx. weight of the transformer might be and also, if I wanted to charge multiple capacitors of the same size in parallel, would I need multiple separate charging circuits or could some portion of the first circuit be used for the others (there is only one battery). Thank you for the help!
2
-
\$\begingroup\$ Use some mosfets paralleling ? \$\endgroup\$– Long PhamCommented Apr 9, 2018 at 16:09
-
4\$\begingroup\$ You haven't showed your progress or your idea \$\endgroup\$– Long PhamCommented Apr 9, 2018 at 16:09
Add a comment
|
3 Answers
\$\begingroup\$
\$\endgroup\$
2
Charging a 200 uF capacitor to 3000 volts in 5 seconds allows you to estimate the current needed using this formula: -
$$Q=CV$$ therefore
$$\dfrac{dQ}{dt} = C\dfrac{dV}{dt}$$ And \$\dfrac{dQ}{dt}\$ equals current
So, plugging in values for C, dV and dt we get a charging current of 120 mA.
This is what your secondary voltage has to supply but, given that charging will tend to reduce as voltage gets higher, you should consider increasing the current by maybe 30% and having a circuit that could produce an open circuit voltage of maybe 4000 volts.
This leads me to estimate the output power as something like 500 watts.
So, you need to go to some site like Ferroxcube's and estimate the size of the ferrite core you will need based on power and switching frequency. The site can guide you to a suitable ferrite core material and size and then you'll need to find a particular core made in this material and having the correct overall size for delivering this power.
I would also like to know what the approx. weight of the transformer might be
Investigation of Ferroxcube's producs and materials will give you weight but be under no-illusion - it will take you hours of head-scratching and the learning process can be steep.
if I wanted to charge multiple capacitors of the same size in parallel, would I need multiple separate charging circuits
It depends on step 1 - finding a suitable core.
answered Apr 10, 2018 at 13:47
-
\$\begingroup\$ 500 W is way too much power to handle casually. It exceeds even the power rating of ATX supplies. Better compromise a little and extend the charging time. Maybe the 5 second was a blank guess and there would be no harm even if the capacitor bank takes a whole minute to charge up. \$\endgroup\$– paki engCommented Jun 18, 2020 at 6:46
-
\$\begingroup\$ @AbdullahBaig be careful what you might hint here. To the casual reader it comes over as a criticism of my answer but, on deeper reading it's clearly an observation about the time stated by the OP (5 seconds) being too fast. My answer was to point out to the OP (more subtly) that 5 seconds was onerous. \$\endgroup\$– Andy akaCommented Jun 18, 2020 at 6:55
\$\begingroup\$
\$\endgroup\$
1
This is not a job for an amateur. The voltages and stored energy involved are very hazardous.
The capacitor will hold 1/2 CV^2 = 900 Joules of energy, and thus will absorb an average of 180 W over the 5 second charging time. 900 J is enough energy to lift me well over half my height.
Having said that, capacitor chargers generally use switch-mode fly-back converters. Xenon lamp camera flashes usually use simple blocking oscillators.
-
\$\begingroup\$ It is well understood that this is a very hazardous circuit. This is a circuit for a non-amateur project. \$\endgroup\$– medeangCommented Apr 9, 2018 at 20:09
\$\begingroup\$
\$\endgroup\$
0
The laser jet printers have a very high voltage power supply. See if you can find one that (nearly) fits your requirement.
