Romanian ZPM (Zero Point Module)

Hi JohnStone,

At switch mode PSUs the storage capacitors need to be specified for severe (1) pulsating currents - else they tend to explode. Standard electrolytic capacitors do not have sophisticated internal connections. So if pulsed with high repetetive currents they experience temperature rise. I experienced it live once when a SMPS exploded 8 capacitors one by one up to burning the machine it was supposed to feed.

Additionally SMPS require capacitors with (2) low inductivity of the leads (low ESR) in order to get the pulsating currents to the full surface of thecapacitor plates. Many electrolytic caps are specified for 50/60Hz only.

I suppose your huge capacitor was not specified for (1) nor (2). So this capacitor might be part of the tank circuit but not with full specified capacity  but some intermediate parasitic R/L/C value.

What kind of capacitors are specified for pulsating currents and also low ESR ? Can you give me a hint about a type or an manufacturer producing them ? How expensive are they comparing to what we have in common stores ?

You might try using a diode in series to your 24V lead (coming from PSU) in order to eliminate backlash to your PSU. 

The fastest diodes I have now in stock are some Schottky diodes, I suppose I actually need some ultra-fast diode to be able to respond to the frequency ranges I'm using (hundreds of KHz).

Your input configuration:

I feel that your setup operates intensively with the power supply as part of the oscillations.

Please be aware that a FET has an intrinsic, reverse and parasitic diode (not very fast) - see schematic symbol .

• In ON state the power will flow from plus to GND - as intended.
• IN OFF state the power might flow reverse through the parasitic diode. 

YoElMiCrO  suggested the insertion of an additional diode in order to overcome this implication.

---------------------

It would be very interesting if you could check if the sweet spot (frequency) of your setup depends on the value or make of this huge capacitor.

-------------------------

I could also try with diode on drain of the MOSFET but this will need more time (modifying the current MOSFET driver) and for this also I need to buy ultra-fast diodes because I don't have any in stock right now. I'll see what I can do depending on the available time. About the capacitor(s) in DC source, I would't make any changes in there just for testing, sorry but I think you understand... 🙂

Fighter, you should not be disappointed. :-)

Unfortunately you tapped into several parasitic properties of electronic components. In fact EVERY component has R/L/C properties or some parasitic diodes .... but usually they do not matter if used in proved standard circuits. High frequency is one ingredient that makes them matter. That's life! Sometimes learning goes the hard way. But you will get support in this forum. Important is - you DO learn.

Actually I'm very dissapointed. That DC source is a lab source, should not care about spikes, pulses or frequencies, its readings should be accurate no matter what. Also that wattmeter being affected in the same way ? I used it to double-check the source's readings. And also those MOSFETS which seem to not be able to handle hundreds of KHz and also sending back to source spikes ? Seems our current electronics available in store are very limited about high-frequencies. What I don't get it is how that current limiter on the DC source is affected also ? I mean that is a physical knob not some software thing. Displaying bad readings I kind of understand. But how is that physical current limited being fooled so the source provide let's ampers while that limiter is set to a few miliampers ? I have some doubts about that current limiter being also affected. I'll figure it out in one way or another.

Thanks for the help and advice ! 🙂