RESONANCE EFFECTS FOR EVERYONE TO SHARE

[gotoluc]

OK, good, I'm glad you aren't going quiet.

First, some more results.
First, my unit seems to oscillate at 1/4 the clock frequency, not 1/2. That is, if I give it a 1 kHz square wave input (say) the output bridge gives a 250 Hz square wave output. I didn't notice this last night as I was using both scope channels to look at the outputs. Are you sure yours does 1/2? Could this difference be because I am using both flipflops in the 4013 and you are only using one?
Second, I looked at a larger freq range. The no-load leakage current peaks at a drive freq of about 1.8 MHz or so, then goes down again, but the whole unit shuts down at a little over 2 MHz. I'm sort of bummed at this, I was hoping it would get to 3 MHz.
Third, I am using the IR2104 which has the shutdown or inhibit pin. When the shutdown is selected, the leakage current goes away and I just see the 26 mA for the chips. So the problem, if that's what it is, is in the output of the driver chips or the mosfets themselves. The first thing I'd try is bigger gate resistors. But that's just a guess.
Fourth, I forget, maybe it will come back to me in a minute.

Second, your circuit and the cap charging. I'd like to see an exact diagram of the set-up before I stick my foot too far in my mouth, but a couple things strike me as possibilities.
First, the meter may not be fast enough to respond to the current surge. It's all over pretty quick, I think.
Second, you are only charging the output cap stack to a relatively low voltage. As you know the energy on a cap goes as the square of the voltage. I'm not sure what to use as the input energy, though, which is why I'd like to see a whole diagram of the circuit, including meters, variacs, load resistors, etc. In other words, the whole setup.

Third, sorry about all the firsts, seconds, and so forth. It's been a long day, and it will be a long night.

Yes, that is what I had at first until I made the mods to the 4013. Do the mods and it should be 1/2 and not 1/4

There is a switching speed limit to the IR2103 and it seems to be at 1Mhz and if your 4013 is wired right your signal generator should be at 2Mhz to get 1Mhz at the IR2103.

I think Groundloop should answer your comments on the MOSFET's as I have minimal electronic knowledge.

So you think the current maybe too fast for the amp meter ...  do you also think the current is too fast for the bulb I put in series

I will do a hand drawn schematic of my test setup and maybe someone that has some time can clean it up.

Thanks for sharing

Luc

[Groundloop]

@TinselKoala,

The output frequency will be 1/2 the input frequency if you wire the 4013 IC as shown in the new drawing.
The maximum switching frequency is set by the IR2103 (or IR2104 in you case). It is also set by the
maximum speed the HEXFETs can do. But the main limiting factor is the drivers. When Luc asked me
to design the circuit he stated that up to 1 -2 MHz was enough. If you want a circuit that can switch much
faster, then this will be another ball game. This switch is designed as simple as possible so that people with
little experience with electronic can build this circuit. The gate resistors must be small enough to ensure
a fast turn on and turn off time for the HEXFETs, but also big enough (in value) so that the current to and
from the driver is limited to what the drivers can do. So around 50 ohm will keep the current low enough so that
the IR2103 will not heat up too much. Hope this answers some of your questions.

@Gotoluc,

Mail me the hand drawing and I will clean up the drawing for you.

Groundloop.

[gotoluc]

Hi everyone,

here is the complete circuit under test at this time. If anyone would care to help by cleaning and adding what you believe is needed for a clear understanding of the circuit please post that you will take on this task so not to have more than one working on this.

Thanks

Luc

Added: I forgot to add the voltage meters at the input cap and output caps.

[gotoluc]

Okay, I see Groundloop has offered to do the cleanup of the circuit.

Thanks for all the help Groundloop

Luc

[TinselKoala]

OK, thanks Groundloop, that's what I figured.

Gotoluc, OK, I understand your diagram.

So you are looking at the ammeter on the supply side of the bridge rectifier which is keeping the 6000 mFd cap charged to 50 volts. There's no ammeter on the output side of this cap, before it goes into your h-bridge?
Because it seems to me that this location is where you should be monitoring input current, not before the rectifier. It's not surprising to me that the surge current wouldn't show on the meter where you have it. That's what those big caps are for, anyway!! Like in car audio systems--they provide a big current surge for that bass note, then charge up again for the next one, without letting the input power sag too much.

And let me confirm: you are charging, until the voltmeter on the secondary caps reads a certain value in the 3.5 volt range? Then you disconnect? Where in the circuit is the "switch" or interrupter wire that you close in the video?