Hi Guys, I was wondering. If I use a CFL to measure the load power why can't I just rectify
the transformer output to make 240 volt DC to feed the CFL ? Then I can measure that.
Seems like a decent way to go since the CFL's internal circuitry rectifies the AC anyway.
Here is a rough video I made of my existing transformer lighting a 25 watt incandescent globe.
I think it's almost as efficient as it gets before being too close to call.http://www.youtube.com/watch?v=7pzqxQwxVGA
The voltage across the load is a bit higher than 240 and the input power seems very close to
the rating of the bulb minus half the idle power. It appears that a good portion of the idle power
goes to the load when the load is applied.
Please forgive the mumbling and repetition, I'm not trying to bling anybody.
I guess I will need to measure the input/output two ways with the resistor method as well.
So a question is what resistance would be best to use for measuring the input current/output current as direct current.
I think I should be able to use all DC current measurements, if I use a CFL load
The same could be done for LED globes that can use DC. I think simplifying measurements is the way to go.
And especially with CFL's I can't see it affecting the device in any way that is not already done by the CFL.
Also, if I use series capacitors on the transformer AC output I can fairly accurately control the maximum output
at a given frequency, as well as do some interesting things by varying the PW. After some experiment I installed
a small non polarized capacitor between the two primary coil negatives (mosfet drains). Intuition told me to do that
but now I see it done in other schematics so I guess it must be common practice.
No one told me it was a good idea.
I think I might have a fraction too much dead time or something, because of all the spikes. Those spikes can be serious and
I think caused arcing in my 240 volt sockets at times depending on frequency and such.
P.S.Also My secondaries are a bit mismatched so that might be causing some imbalance problems, I need to even them up and
add a couple of layers to increase the unloaded low frequency voltage to about 265 volts or so..